Delineating the Cardio-Myogenic Hierarchy During Mouse Embryonic Development

Delineating the Cardio-Myogenic Hierarchy during Mouse Embryonic Development

Charles Yoon

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Institute of Biomaterials and Biomedical Engineering University of Toronto

Delineating the Cardio-Myogenic Hierarchy during Mouse Embryonic Development

Charles Yoon

Doctor of Philosophy

Institute of Biomaterials and Biomedical Engineering University of Toronto

2018

Abstract

The identification of cell surface proteins on stem cells or stem cell derivatives is a key strategy for the functional characterization, isolation, and understanding of stem cell population dynamics.

We have turned to cell surface mass spectrometry to increase the candidate pool of membrane proteins on cardiac progenitor cells (CPCs). Specifically, we examined the expression of surface markers on CPCs using an integrated mass spectrometry and microarray-based approach. We analyzed the genome and surface proteome of cardiac progenitors generated from the stage- specific differentiation of mouse (m) and human (h) pluripotent stem cells (PSCs). We have identified and characterized Frizzled 4 (FZD4) as a new marker for lateral plate mesoderm (LPM).

We also utilized FZD4 as a marker, in conjunction with fetal liver kinase 1 (FLK1) and platelet derived growth factor alpha (PDGFRA) and demonstrated an increase in CPC purity and a subsequent increase in cardiomyocyte (CM) enrichment. Additionally, we have found FZD4 is also expressed in the hPSC system and results in a similar enrichment in CM. Furthermore, we showed that NORRIN can be presented to the FZD4 receptor to induce Wingless-related integration site (WNT) signaling-mediated proliferation, resulting in an increase in CM output ii

from CPCs. This demonstrates the value in knowing the set of surface markers present on a cell at a specific stage of development and the potential to leverage that knowledge into more efficient cell differentiation protocols. Further validation of the function of FZD4 is also being established in a preliminary in vivo study. The identified surface markers also have the potential to isolate cell types not only within the CPC stage, but within the PSC, epiblast, and primitive streak stages as well. These markers have been compiled into a preliminary cell-cell communication network model overlaid with an initial alternative slicing analysis to determine potential mechanisms that impact cell signaling during early CPC development. In summary, the application of a systems biology approach as demonstrated in this thesis, greatly expanded the number of surface markers available and upon further characterization and validation, can improve our understanding of cardiac biology.

iii

Acknowledgments

I would like to thank all those who have made this long journey a memorable one. This dissertation would not have been possible without all your support, kindness, and goodwill.

First, I would to thank my supervisor, Peter Zandstra, for providing me with the opportunity to pursue a prestigious degree in a leading world-class academic environment. I also would like to thank him for his support and mentorship, and especially his infinite patience, for he has given me countless chances to learn from my mistakes and to improve upon myself. Being in this supportive environment gave me room to grow, not just academically, but personally as well, and I will keep these lessons with me always.

Second, to my committee members, Gordon Keller, Andrew Emili, and Anthony Gramolini, as well as my collaborator, Bernd Wollscheid, for giving me your time and valuable feedback in guiding me through my degree as you helped me develop my critical thinking skills. Also, the support from your respective labs were instrumental in the development of my project and I would also like to thank your lab members, especially Damaris Bausch-Fluck, Andreas Frei, Steve Kattman, Nicole Dubois, Alec Witty, Johannes Hewel, and Hongbo Guo, for their time and assistance.

Third, to my friends and lab mates, there isn’t much to say other than it has been an absolute blast. Special thanks to Emanuel Nazareth, Joel Ostblom, Manu Tewary, Shreya Shukla, Nimalan Thavandiran, Nika Shakiba, Dave Fluri, Liz Csaszar, Jen Ma, and everyone else for making the lab environment a fun and happy place to spend countless hours fussing over experiments and poring over data. A special thank you to Ting Yin for being our lab mom and making sure we were all well fed and clean up after ourselves. The lab wouldn’t have been able to run without you.

Finally, to my parents John and Agnes Yoon, thank you for always being there through thick and thin throughout my personal journey and always giving me support when I need. Your tutelage has made me to be what I am today and your insistence on higher levels of education have provided me the opportunity to pursue my passions.

Table of Contents

Acknowledgments...... iv

Table of Contents ...... v

List of Tables ...... viii

List of Figures ...... ix

List of Appendices ...... xi

List of Abbreviations ...... xii

INTRODUCTION ...... 1

1.1 Motivation ...... 2

1.2 Mouse Cardiac Development ...... 3

1.2.1 Cardiac Morphogenesis in vivo ...... 3

1.2.2 Signaling during Heart Development ...... 5

1.2.3 Transcriptional Regulation of Cardiac Cells...... 11

1.2.4 Identification of Cardiac Progenitors and CPC Markers in vitro ...... 12

1.3 Bioreactor Technology and Stem Cell Production ...... 14

1.4 Mass Spectrometry-Based Surface Profiling of Pluripotent Stem Cells and Derivatives ...... 17

1.4.1 Liquid Chromatography - Tandem Mass Spectrometry ...... 17

1.4.2 Methods to Isolate Cell Surface Proteins ...... 19

1.4.3 Cell Surface Proteomics in Stem Cells ...... 21

1.5 Transcriptomic Analysis of Pluripotent Stem Cells and their Derivatives ...... 24

1.5.1 Microarray Analysis and the Advent of Deep Sequencing ...... 24

1.5.2 Expression Profiling in Cardiac Development ...... 26

1.6 Thesis Goals and Approach ...... 27

FZD4 marks lateral plate mesoderm and signals with NORRIN to increase cardiomyocyte induction from pluripotent stem cell-derived cardiac progenitors ...... 28

2.1 Abstract ...... 29

2.2 Introduction ...... 29

2.3 Results ...... 31

2.3.1 Integrated Mass Spectrometry and Microarray Analysis Identifies Early Mesoderm Surface Markers ...... 31

2.3.2 FZD4 Identified as a Potential Marker of a Sub-Population Enriched for Cardiac Progenitors ...... 35

2.3.3 Gene Expression Analysis on Sorted CPC Sub-Populations Indicates that FZD4 Marks Pre-Cardiac Mesoderm in CPCs ...... 38

2.3.4 FZD4-Expressing CPCs Yield Higher CM Outputs ...... 40

2.3.5 Greater CM Output from FZD4+ CPCs is Also Observed during hPSC Differentiation ...... 43

2.3.6 Canonical FZD4-NORRIN Signaling Enhances CM Output ...... 45

2.4 Discussion ...... 47

2.5 Conclusion ...... 50

2.6 Experimental Procedures ...... 50

2.6.1 PSC Culture and Bioreactor Differentiation ...... 50

2.6.2 Flow Cytometry, Cell Sorting, and Immunocytochemistry ...... 51

2.6.3 Microscopy and Image Analysis ...... 52

2.6.4 Quantitative PCR Analysis ...... 52

2.6.5 Microarray Analysis...... 52

2.6.6 Cell Surface Capture (CSC) ...... 53

2.6.7 Mass Spectrometry (MS) Analysis ...... 53

2.6.8 Label-Free Quantification ...... 54

2.6.9 Technical Validation of Candidate Proteins ...... 54

2.6.10 Statistical Analysis ...... 55

2.6.11 Accession Numbers ...... 56

2.7 Supplemental Materials ...... 57

2.7.1 Supplemental Tables ...... 57

2.8 Acknowledgments...... 60

Discussion ...... 61

3.1 Evaluation of Output Lineages from FZD4+ and FZD4- Cells ...... 63

3.2 In vivo Validation of FZD4 ...... 65

3.3 FZD4-mediated WNT Signaling Dynamics during CPC Development ...... 70

3.4 Generation of Cell-Cell Communication Network to Interrogate CPC Population Dynamics ...... 73

3.5 Alternative Splicing is a Key Feature in Cardiac Progenitor Cell Specification ...... 75

3.6 Systems Biology Approach Towards Understanding CPC Biology...... 78

3.7 Identification of Surface Markers Improves Isolation of Pure CPCs and Improves Cell Therapies in Regenerative Medicine ...... 80

Conclusions and Future Work ...... 84

4.1 Thesis Summary...... 85

4.2 Impact and Future Studies ...... 86

References ...... 89

Appendices ...... 118

vii

List of Tables

Table 2-1: Antibodies used in this study...... 57

Table 2-2: qPCR primers used in this study...... 57

Table 2-3: Candidate proteins identified by mass spectrometry and microarray, related to Figure 2-2...... 59

Table 2-4: Protein Quantification and Clustering, related to Figure 2-1...... 60

Table 3-1: Differently expressed genes and their breakdown of predicted alternative exons...... 76

viii

List of Figures

Figure 1-1: Schematic describing the morphological features during development...... 4

Figure 1-2: Schematic diagram of the WNT signaling pathway ...... 8

Figure 2-1: Cell surface mass spectrometry yields protein abundance profiles...... 32

Figure 2-2: Microarray analysis yields sub-population specific surface markers...... 34

Figure 2-3: GO analysis of CPC sub-population relative to CM, related to Figure 2-2...... 35

Figure 2-4: Validation of candidate surface markers, related to Error! Reference source not found...... 36

Figure 2-5: Flow cytometry and qPCR validation of proteins...... 37

Figure 2-6: Flow cytometry and qPCR validation of purity of sorted populations, related to Error! Reference source not found...... 39

Figure 2-7: Differentiation of cardiac progenitors into cardiomyocytes, related to Figure 2-8. .. 41

Figure 2-8: FZD4 activation further enriches CPC in the FLK1+PDGFRA+ population increasing subsequent CM yield...... 42

Figure 2-9: FZD4 enriches CPC in hPSC-derived cardiomyocytes...... 44

Figure 2-10: Exogenous addition of WNT ligands, related to Figure 2-11...... 46

Figure 2-11: Model of FZD4 abundance in the context of early cardiac differentiation...... 47

Figure 2-12: FZD4 antibody titration and negative isotype control, related to Figure 3...... 55

Figure 3-1: Preliminary attempts at whole mount embryo staining...... 67

Figure 3-2: High-throughput imaging of embryo sections and controls...... 68

Figure 3-3: Updated WNT signaling stages based on FZD4 marker...... 72

Figure 3-4: Cell-cell Interaction Network generated from microarray and proteomic data...... 74

Figure 3-5: Identification of predicted alternative splicing events...... 77

Figure 3-6: Systems level approach to understand biology...... 80

List of Appendices

Appendix I: Protein Quantification and Clustering...... 118

List of Abbreviations

7-AAD 7-Aminoactinomycin D ALK activin-like kinase AMHC atrial-myosin heavy chain ANOVA analysis of variance AP alternative promoter APC adenomatous polyposis coli AQUA absolute quantification of proteins AS alternative splicing bHLH basic helix-loop-helix BMP bone morphogenic protein BRY brachyury CaCN calcineurin CamKII calcium/calmodulin-dependent kinase II CCC cell-cell communication CD cluster of differentiation CHF congestive heart failure CKIT kit oncogene CM cardiomyocyte CPC cardiac progenitor cell CSC cell surface capture CTNT cardiac troponin T CX connexin CXCR C-X-C chemokine receptor DAPI 4',6-diamidino-2-phenylindole DDA data dependent acquisition DIA data independent acquisition DKK Dickkoph DNA deoxyribonucleic acid DVL dishevelled EB embryoid body EC endothelial cell ECM extracellular matrix EDTA ethylenediaminetetraacetic acid EdU 5-ethynyl-2´-deoxyuridine EGF epidermal growth factor EMT epithelial-to-mesemchymal transition EOMES eomesodermin EPHB2 ephrin-b2 ERK extracellular signal-regulated kinase ESI electrospray ionization FBS fetal bovine serum FEVR familial exudative vitreoretinopathy FGF fibroblast growth factor xii

FHF first heart field FLK1 fetal liver kinase 1 FZD frizzled GATA GATA-binding protein GEO genome expression omnibus GFRA glial cell line-derived neurotrophic factor receptor alpha GO gene ontology GSK3B glycogen synthase kinase 3b HAND heart and neural crest derivatives expressed HCN hyperpolarization-activated cyclic nucleotide-gated channel HER human epidermal growth factor receptor HRT hairy-related transcription factor ICAT isotope-coded affinity tag iPSC induced pluripotent cell IRX Iroquois homeobox gene ISL islet ITG integrin iTRAQ isobaric tags for relative and absolute quantitation IWP inhibitors of WNT production JNK jun N-terminal kinase KDR kinase insert domain receptor LC liquid chromatography LEF lymphoid enhancing binding factor LGR4 leucine rich repeat containing G protein-coupled receptor 4) LMO LIM domain only LPAR lysophosphatidic acid receptor 4 LPM lateral plate mesoderm LRP low-density-lipoprotein-related protein MALDI matrix-assisted laser desorption/ionization MAPK mitogen-activated protein kinase MAQC microarray quality control MEOX mesenchyme homeobox MESP1 mesoderm posterior basic helix-loop-helix transcription factor 1 MHC myosin heavy chain MI myocardial infarction MLC myosin light chain MRI magnetic resonance imaging mRNA messenger ribonucleic acid MS mass spectrometry MudPIT multidimensional protein identification technology NKX NK2 transcription factor-related NORRIN norrie protein NSC neural stem cell OFT outflow tract PAX paired box PCP planar cell polarity xiii

PDGFRA platelet derived growth factor alpha PECAM platelet endothelial cell adhesion molecule PKC protein kinase C PRIDE proteomics identifications database PSC pluripotent stem cell qPCR quantitative polymerase chain reaction RALDH2 retinaldehyde dehydrogenase 2 RGD arginine-glycine-asparagine RNA ribonucleic acid ROCK rho kinase SCA stem cell antigen SCoPE-MS single cell proteomics by mass spectrometry SDF stromal cell-derived factor SHF second heart field SHF second heart field SI splicing index SILAC stable isotope labeling by amino acids in cell culture SIRPA signal regulatory protein alpha SMAD mothers against decapentaplegic SMC smooth muscle cell SRM selected reaction monitoring SRp serine/arginine protein TBX T-box transcription factor TCF/LEF transcription factor TGFB transforming growth factor β TMEM transmembrane protein TNFA tumour necrosis factor alpha TPP Trans-Proteomic Pipeline VCAM vascular cell adhesion molecule VMHC ventricle-myosin heavy chain WNT wingless-related integration site

xiv

INTRODUCTION

1.1 Motivation

Comprehensive knowledge of the cell surface proteins on stem cells or stem cell derivatives is critical in order to determine the population dynamics of stem cells (Nunomura, 2005b). Extensive analysis of the identified stem cell types can enhance our understanding of the signaling events that stimulate fate decisions during branching points in the differentiation process. In mouse cardiac development, the heart is derived from cardiac progenitor cells (CPCs) which emerge from the lateral plate mesoderm (LPM) of the embryo (Kinder et al., 1999; Rana et al., 2013). Although there have been many studies detailing numerous marker combinations to specifically label CPCs (Kattman et al., 2006; Moretti et al., 2006; Wu et al., 2006a), many of these markers are not surface markers, which increases the difficulty of performing live cell-sorting studies for a comprehensive characterization of cell types.

In addition to the potential applications in biology, improved surface maker panels to isolate viable CPCs would be beneficial to the field of regenerative medicine. Myocardial infarction (MI) is now the leading cause of congestive heart failure (CHF) and death in the world (Mackay, J., Mensah, G. A., Mendis, S., & Greenlund, 2004). Coronary occlusion and the resultant myocardial ischemia rapidly result in myocardial necrosis followed by scar formation. As a result, loss of cardiomyocytes (CMs) in the adult heart is irreversible and leads to reduced cardiac function. While several studies have now shown that cell transplantation results in small improvements in the infarct area (reviewed in (Grigoropoulos and Mathur, 2006; Melo et al., 2004)), major challenges such as increasing cell survival, engraftment and functional integration with the host tissue remain.

Some have challenged this idea of an ideal cell type. Ye et al. have compared injection of multiple cell types and have showed no significant difference in outcome with different sources and claim the beneficial effects appear to be via paracrine influences (Ye and Yeghiazarians, 2015). However, these results are confounded by the more mature cell types used and not having a standardized method of generating cells at the same time. Additionally, neither study has adequately demonstrated the generation of new myocytes at a high enough number or in the appropriate animal models to significantly demonstrate the benefits due to the increased contractile force.

Mouse embryonic stem cells (mPSCs) are a promising source of cells as they can differentiate into CM (reviewed in (Amit et al., 2000; Cameron et al., 2006a; Gerecht-Nir et al., 2004a; Sachinidis et al., 2003)) as well as into the recently identified CPCs (Cai et al., 2003; Passier et al., 2005a). Although multiple CPC stages can be generated, which one or which combination that will result in effective treatment of cardiac disease is uncertain. Additionally, the ability to isolate and enrich for specific cell types is limited, especially at the progenitor cell stage. Taken together, there is a great need for an improvement in the number of surface markers available in order to specifically isolate and enrich for cell types for use in both biological and clinical applications. In order to address this problem, new markers that allow the isolation of CPCs have to be identified and new assays to screen the potential of these cells need to be developed. With this information, newly identified progenitor cells can be integrated into a cellular interaction network, which can then further our knowledge of cardiac development through perturbation experiments and model simulations.

1.2 Mouse Cardiac Development

1.2.1 Cardiac Morphogenesis in vivo

The heart is the first organ to functionally develop in a vertebrate embryo. In the mouse embryo, cells destined to form the heart emerge from the mesoderm, one of the three primary germ layers (ectoderm, endoderm, and mesoderm) that are patterned during gastrulation at embryonic day (E) 6.0 (Tam and Loebel, 2007), (Figure 1-1). The initiation of gastrulation is marked by the primitive streak, which is patterned along the anterior-posterior axis, and lineage tracing studies have shown it can be divided roughly into posterior, mid, and anterior regions. Cells destined to become cardiac tissue first emerge from the mid-anterior streak and form the splanchnic mesodermal layer (also known as LPM) in one of the earliest fate decisions in the gastrulating embryo (Harvey, 2002; Srivastava and Olson, 2000). As the LPM is specified, a subsequent development of the somatic (paraxial) mesoderm occurs (Kinder et al., 1999). At ~E6.5 the cardiogenic mesodermal cells in the LPM migrate laterally towards the anterior-proximal side of the epiblast to form an epithelial layer of differentiating CMs. These cells form a crescent shaped structure termed the cardiac crescent at E7.5 (Schoenwolf and Garcia-Martinez, 1995; Tam et al., 1997). At around E8.0, two waves of cell migration bring cardiac precursors to the midline where the heart tube is formed. The first wave, brings established CPCs ventrally to form the heart tube proper (Sizarov et al., 2011). As the cells that contribute initially are the first precursors to differentiate, they are known as the

Figure 1-1: Schematic describing the morphological features during development.

Lateral plate mesoderm migrates (E6.5) from the primitive streak laterally and develop into the cardiac crescent on the anterior side of the epiblast. The cardiac crescent develops into the cardiac tube (E8.5) and after further specification, generates the four-chambered heart (E10.5). The contributions of the primary heart field (blue) and the secondary heart field (red) are represented throughout developmental stages of the heart. Both heart fields contribute cells to the atria (yellow). first heart field (FHF) and give rise to the left ventricle and atria. The second wave of precursors remain largely undifferentiated and maintain high levels of proliferation and gradually add to the heart, contributing to the elongation of the heart tube. Due to the delayed nature of the differentiation of these cells, the progenitors are referred to as the second heart field (SHF) and give rise to the right ventricle and outflow tract. The SHF is also distinguished molecularly from the FHF by the expression by Islet1 (ISL1) and fibroblast growth factor (FGH) 10, which have

5 been determined through lineage tracing studies (Abu-issa et al., 2004; Buckingham et al., 2005; Cai et al., 2003; Kelly and Buckingham, 2002). With the continuation of this process, by E8.5 the closed heart tube is formed, and the tube undergoes looping and the chambers are well defined by E10.5, until finally the fully formed fetal heart is generated by E14.5.

1.2.2 Signaling during Heart Development

The patterning of early heart tissue is tightly regulated process that is controlled through a complex interaction of major signaling pathways. Studies with different model organisms have provided supporting evidence that the Activin/Nodal, bone morphogenic protein (BMP), and Wingless/Int- 1 (WNT) signaling pathways all play key roles in the generation of the cardiovascular system.

1.2.2.1 Activin/Nodal and BMP Signaling

The Activin/Nodal and BMP signaling pathways are both members of the transforming growth factor β (TGFβ) superfamily. This family, in general, signals through the biding of heteromeric complexes of serine/threonine kinase receptors. The receptors are divided into two families: type I (also commonly referred to as Activin-like kinase (ALK)) and type II receptors (Goumans and Mummery, 2000). Once an active ligand/receptor complex is formed, the downstream mothers against decapentaplegic (SMAD) proteins are activated through phosphorylation and propagate the signal from cytoplasm to the cell nucleus. In the nucleus, the phosphorylated SMADs form a complex with SMAD4 and interact with target transcription factors (Chang et al., 2002). For Activin/Nodal signaling, the ligands first bind to type II receptors which will then phosphorylate and activate the type I receptor in the complex. The type I receptor will then phosphorylate SMAD2/3/6 and translocate into the nucleus to activate Activin/Nodal-specific target genes (Attisano et al., 1993, 1996; ten Dijke et al., 1994). BMP ligands follow a similar pathway, except they can bind to both type I or type II receptors to initiate receptor complex formation, and their type I receptors phosphorylate SMAD1/5/8 before translocating into the nucleus to activate BMP- specific target genes (Liu et al., 1995; Nishitoh et al., 1996). While these pathways share similar components of the signaling cascade, the combination of type I and type II receptors along with the specific ligand yield vastly different biological responses.

Nodal signaling regulates mesoderm specification and the development of the anterior posterior axis during the formation of the primitive streak during gastrulation in the mouse (Conlon et al., 1994). This effect is further evidenced by the lack of proper axial mesendoderm specification upon

6 deletion of SMAD2 and/or SMAD3 (Dunn et al., 2004; Vincent et al., 2003). Nodal signaling is also required for proper left-right asymmetric development of the heart as evidenced by loss of function studies in mouse embryos (Brem et al., 2002). A more direct contribution of Nodal signaling to the development of cardiac mesoderm is demonstrated by the induction of multiple mesoderm regions in the Xenopus embryo upon addiction of Nodal/Activin at various thresholds (Green et al., 1992). While these data suggest a direct role of Nodal in cardiac tissue formation, recent studies have demonstrated that high levels of Nodal activate endoderm formation, and suggest that Nodal signals activate cardiac mesoderm indirectly through the generation of endoderm (D’Amour et al., 2005; Tada et al., 2005; Yasunaga et al., 2005). Nodal is not only required for visceral endoderm formation (Mesnard et al., 2006), but signals from the anterior visceral endoderm have also been shown to initiate and support heart formation in mouse embryos (Arai et al., 1997; Nijmeijer et al., 2009). An additional key step in the formation of CMs is the inhibition of Nodal signaling once CPC commitment has taken place (Cai et al., 2013; Shiratori et al., 2001). The inhibition is mediated by soluble compounds, Cerberus and Lefty, that bind to Nodal and block downstream signaling (Perea-Gomez et al., 2002). Interestingly, activation of heart formation from Nodal also initiates the expression of Cerberus (Foley et al., 2007) and taken together, suggests Nodal signaling is required in general for mesoderm and endoderm formation, but local inhibition of Nodal is required to specify cardiac tissue.

BMP ligands are expressed throughout both the early and late stages of cardiac development. During the gastrulation stage, BMP2 and BMP4 is secreted from the endodermal and mesodermal layer respectively in the mouse and are required for gastrulation and formation of the primitive streak (Winnier et al., 1995). Additionally, cardiac induction potential has been demonstrated using both ligands on non-cardiac explant tissue (Lough et al., 1996). The type I BMP receptor (BMPRIA, also known as ALK3), binds BMP2/4 and is also required for gastrulation and subsequent mesoderm generation (Mishina et al., 1995). BMP4 also plays a role in lateral plate mesoderm (LPM) specification from the primitive streak as BMP4 knock out mice die without any mesoderm formation between E6.0 and E9.0 (Winnier et al., 1995). In chicken embryos, BMP signaling is involved in cardiogenesis and promotes FHF and SHF generation (Schlange et al., 2000; Schultheiss et al., 1997; Tirosh-Finkel et al., 2006), and Klaus et al. demonstrated BMPRIA signaling is required for the generation of cardiac progenitors in the FHF (Klaus et al., 2007). While BMP5/6/7 individually have no significant effect on heart development, collectively there

7 may be overlapping functions during late-stage cardiac development of the outflow tract (Liu et al., 2004b). BMP10, in contrast, is considered an important regulator of cardiac function due to its requirement for the development of ventricular trabeculae (Chen et al., 2004). Similar to Nodal signaling, BMP inhibition through soluble factors such as Noggin and Chordin are required in later stages of cardiac differentiation (Re’em-Kalma et al., 1995; Zimmerman et al., 1996). These factors bind to BMP ligands and attenuate the downstream signaling cascade. While BMP2/4 signaling is required to promote cardiac development, there is a simultaneous suppression of cardiac differentiation that must be inhibited. These studies also indicate that the timing and context of BMP signaling is critical for the efficient development of cardiac mesoderm and subsequent differentiation into cardiomyocytes.

1.2.2.2 WNT Signaling in Cardiac Development

The WNT signaling pathway is widely involved in various processes throughout the developing mouse embryo including anterior-posterior axis formation (Huelsken et al., 2000), gastrulation, epithelial-to-mesenchymal transition (EMT), mesoderm patterning (Lee and Frasch, 2000; Lekven et al., 2001), and lineage specification. WNT ligands are secreted glycoproteins that interact with Frizzled (FZD) receptors, which have seven transmembrane-spanning domains and have similar homology to G-protein coupled receptors (Yang-Snyder et al., 1996). The WNT signaling pathway can traditionally be divided into canonical (or b-catenin dependent) and non-canonical (or b- catenin independent) signaling pathways, (Figure 1-2).

The classic canonical WNT signaling pathway is initiated by WNT ligands binding to FZD receptors. When unbound, b-catenin is phosphorylated in a destruction complex consisting of the scaffolding proteins Axin, adenomatous polyposis coli (APC), and serine/threonine kinase glycogen synthase kinase 3b (GSK3B) and then depredated through an ubiquitin-mediated proteasome pathway (Gordon and Nusse, 2006). However, upon binding of WNT to the FZD receptors and the low-density-lipoprotein-related protein 5/6 (LRP5/6) co-receptor (He et al., 2004), phosphoprotein Dishevelled (DVL) is activated and inhibits GSK3B within the destruction complex (Wallingford and Habas, 2005), leading to stabilized levels of cytoplasmic b-catenin.

The b-catenin then translocates into the nucleus where it binds co-transcription factors transcription factor/lymphoid enhancing binding factor (TCF/LEF) to activate downstream target

Figure 1-2: Schematic diagram of the WNT signaling pathway

The canonical and non-canonical WNT signaling pathway is schematically represented indicating the major components to each pathway.

9 genes linked to cell proliferation control (Gordon and Nusse, 2006; Moon et al., 2004). The non- canonical WNT pathway acts independent of b-catenin and can be further divided into two distinct branches: the planar cell polarity (PCP) pathway or the Ca2+ pathway. In the PCP pathway, signal is transduced via DVL to small GTPases, Rho and Rac, and initiate mainly the Rho kinase (ROCK) and jun N-terminal kinase (JNK) pathways (Seifert and Mlodzik, 2007). In the Ca2+ pathway, G- protein-mediated increase in intracellular calcium levels activates calcium sensitive enzymes calcium/calmodulin-dependent kinase II (CamKII), calcineurin (CaCN), and protein kinase C (PKC) (Kohn and Moon, 2005). While these pathways have been described as distinct pathways, many WNT signaling components overlap and interact intracellularly, blurring the lines between each of the defined pathways and the downstream response phenotypes.

WNT ligands and FZD receptors are expressed throughout the embryo during cardiac development. During early stage development, canonical WNT signaling is essential to anterior- posterior axis, primitive streak, and mesoderm formation (Huelsken et al., 2000; Lindsley et al., 2006). Liu et al. have shown that WNT3 knock out mice do not form the primitive streak or mesoderm tissue (Liu et al., 1999). The migration and segregation of the lateral plate and paraxial mesoderm as they emerge from the primitive streak is controlled by WNT3a and WNT5a signaling (Sweetman et al., 2008a). Sweetman et al. showed that WNT5a signals through the non-canonical WNT pathway to cause migration of cells from the posterior streak, while WNT3a, in addition to providing canonical WNT signals for mesoderm production (Dunty et al., 2008; Yamaguchi, 2001), mediates the production of paraxial mesoderm by inhibiting WNT5a signaling. This data is further supported by Tan et al. who showed inhibition of GSK3B, which initiates the canonical WNT signaling pathway, generates the primitive streak and subsequently lateral plate mesoderm, while extended inhibition generated endodermal tissue (Tan et al., 2013a). Interestingly, further extension of GSK3B inhibition generated endodermal cells, and combined with the cell fate change from endoderm to mesoderm upon ablation of b-catenin (Lickert et al., 2002), these data indicate that WNT signaling is involved in both direct and indirect roles in the generation of cardiac mesoderm.

In order for cardiac specification of the progenitor cells to occur in the cardiac crescent, canonical signaling via WNT3a and WNT8c in the anterior LPM must subsequently be inhibited (Marvin et al., 2001; Schneider and Mercola, 2001). Natural inhibitors to the WNT pathway include Dickkoph (DKK) proteins (Glinka et al., 1998), and soluble FZD-related proteins (Hoang et al., 1998). DKK

10 proteins inhibit WNT signaling through blocking WNT/LRP interactions, while soluble FZD receptors reduce effective WNT ligand availability via competitive binding. Synthetic inhibitors such as inhibitors of WNT production 2 (IWP2), which block Porcupine and interfere with the release of WNT ligands (Chen et al., 2009), and XAV939, an inhibitor of an Axin regulator (Huang et al., 2009), are also commonly used to inhibit the WNT signaling pathway. Constitutively active b-catenin during the anterior migration of LPM, suppressed cardiac tube formation, while inhibition of canonical WNT through the addition of DKK1 initiated cardiogenesis, specifically that of the FHF (Bondue et al., 2008; David et al., 2008). In the chick embryo, WNT3a and WNT1 are secreted from the neural tube and abrogate cardiac formation, further highlighting the requirement of canonical WNT inhibition at this stage of cardiac development (Tzahor and Lassar, 2001). Non-canonical WNT signaling also plays a role in cardiac specification from mesoderm tissue. WNT11 is essential for cardiac formation and signals through PKC and JNK pathways, independent of b-catenin (Pandur et al., 2002). Specifically, WNT11 and WNT5a are both required to initiate non-canonical WNT signaling and are directly involved in the generation of the SHF and outflow tract (OFT) (Cohen et al., 2012; Schleiffarth et al., 2007).

Once specification has occurred, canonical WNT activation is again required in the later stages of cardiac development in a proliferative capacity to expand the cardiogenic population. Ai et al. have shown that canonical WNT signaling functions in a positive fashion to promote right ventricular cell expansion (Ai et al., 2007). While the precise WNT ligand responsible for SHF expansion remains unclear, WNT3a, WNT2a, WNT2b, and WNT8 are likely candidates based on expression profiles in the embryo (Cohen et al., 2008a; Eisenberg and Eisenberg, 2006). Finally, during terminal differentiation of CMs, canonical WNT signaling needs to be inhibited once again, demonstrated by the lowered frequency of beating cells upon the late addition of WNT3a (Naito et al., 2006). In the Xenopus embryo, depletion of canonical WNT signaling via WNT6 also increases the frequency of cardiac structures (Lavery et al., 2008). WNT11 expression is also shown to increase during terminal differentiation of CMs and increases the number of myocytes generated (Ueno et al., 2007a), indicating that the presence of non-canonical WNT signaling is required for late-stage CM development.

The kinetics and type (i.e. canonical versus non-canonical) of WNT activation is critical in guiding cardiac induction and differentiation (Sumi et al., 2008). Additionally, recent evidence suggests that WNT signaling should be considered as one large complex signaling network. As mentioned

11 above, WNT11 and WNT5a contribute to SHF formation through activation of the non-canonical pathway and simultaneous suppression of the canonical signaling pathway (Cohen et al., 2012; Schleiffarth et al., 2007). These studies indicate that the WNT signaling pathways are interconnected and there exists extensive cross-talk resulting in cell responses from ligand-receptor pairs that are highly context dependent.

1.2.3 Transcriptional Regulation of Cardiac Cells

Precise regulation and control of transcription factors is required for the correct development of the cardiac lineage. Additionally, not only do transcription factors regulate migration, patterning, and specification, they serve as a form of cell-type identification and can be used to distinguish between different cell types.

The generation of the primitive streak is the earliest sign of gastrulation and is marked by brachyury (BRY) and contains the precursors to the mesoderm and endoderm lineages (Rivera- Pérez and Magnuson, 2005; Wilkinson et al., 1990). A T-box transcription factor eomesodermin (EOMES) is expressed in the primitive streak and is responsible for the process of EMT and mesoderm migration (Arnold et al., 2008). EOMES marks the earliest cardiac progenitor population and promotes cardiac differentiation through activation of mesoderm posterior Basic helix-loop-helix transcription factor 1 (MESP1) (Costello et al., 2011). MESP1 is also considered a marker of the earliest cardiac mesoderm and lineage tracing studies have shown that MESP1+ cells contribute extensively to the adult heart (Saga et al., 1999). MESP1 expression has been found to initiate features of EMT and promote development of cardiac precursors (Lindsley et al., 2008). MESP1 is expressed during the primitive streak stage and is quickly downregulated as the cells migrate out of the streak in the early gastrulating embryo (Bondue et al., 2008; Liu et al., 2007; Ueno et al., 2007a). Additionally, overexpression of MESP1 greatly increases the generation of cardiac progenitors including derivatives of the FHF and SHF (David et al., 2008). Functional studies have shown that MESP1+ cells can differentiate into all types of CM present in the heart (atrial, ventricular, and pace maker) (Bondue et al., 2008; David et al., 2008; Lindsley et al., 2008). MESP1 also marks additional lineages such as trunk mesenchyme and vasculature, indicating that MESP1 characterizes mesoderm more broadly and not specific to cardiac mesoderm (Yoshida et al., 2008).

The cells within the posterior cardiac crescent, the FHF, is characterized by the T-box transcription factor 5 (TBX5) and hyperpolarization-activated cyclic nucleotide-gated channel 4 (HCN4), and fate mapping experiments have demonstrated that TBX5+HCN4+ cells contribute to the left ventricle and atrial precursors (Bruneau et al., 2001; Später et al., 2013; Takeuchi et al., 2003). Originating from the SHF in the anterior crescent, lineage tracing revealed Islet1+ (ISL1+) cells that contributed to the right ventricle and OFT (Cai et al., 2003; Zaffran et al., 2004). ISL1 is also a broad marker as it labels cardiac neural crest (Engleka et al., 2012) and its been demonstrated that ISL1 expressing cells have the ability to differentiate into smooth muscle, and endothelial cells in addition to cardiac cells (Moretti et al., 2006). The homeodomain transcription factor NK2 transcription factor-related, locus 5 (NKX2.5) is critical for the terminal differentiation of CM (Lyons et al., 1995). Mice that have interrupted expression of NKX2.5 demonstrate abnormal heart morphogenesis and function. Interestingly, the function of NKX2.5 is to critically regulate SHF proliferation and OFT morphology (Prall et al., 2007). In the absence of NKX2.5, over proliferation of progenitor cells led to over specification and subsequent failure of SHF development, indicating that NKX2.5 controls the transition between induction and proliferation. GATA-binding protein 4 (GATA4) also controls cardiac differentiation and GATA4+ cells are present in pre-cardiac LPM. GATA4 null mice lack a primitive heart tube, indicating that it is required for the midline fusion stage of development (Kuo et al., 1997; Molkentin et al., 1997).

Many of these transcription factors interact in order to regulate downstream expression and orchestrate cardiac development. Chromatin localization studies have shown GATA4, NKX2.5, TBX5, and others, collaborate in order to direct cardiac differentiation (He et al., 2011). This suggests that the complex regulation and control of cardiac differentiation is the result of a coordinated effort from a core set of transcription factors and different sets of core transcription factors may be present during the various stages of heart development.

1.2.4 Identification of Cardiac Progenitors and CPC Markers in vitro

In vitro differentiation of pluripotent stem cells (PSCs) to CMs occurs in a pattern paralleling in vivo embryonic cardiogenesis (Boheler et al., 2002; Kinder et al., 1999) where it has been shown that there are cardiac precursors fated to become CMs (Brand, 2003). Due to these similarities, the markers used to identify cell lineages in vivo can largely be applied in vitro to PSC differentiation. Recent studies have revealed the presence of a close family of multipotent progenitor cells that

13 give rise to the CM, endothelial cell (EC), and vascular smooth muscle cell (SMC) lineages of the heart. Kattman et al. have shown that a BRY+FLK1+ population can give rise to two separate progenitor populations that arise sequentially (Kattman et al., 2006). The first population gives rise to the hemangioblast and subsequent endothelial and hematopoietic lineages. The second is enriched for early cardiovascular progenitors that display CM, SMC, and EC lineage potential. Wu et al. (Wu et al., 2006b) have shown that NKX2.5+ and C-KIT+ cells are capable of clonal expansion and differentiation into cardiomyocytes, conduction system cells and smooth muscle cells in vitro. However, these cells did not give rise to cells of the endothelial lineage, indicating a more differentiated progenitor population that had already segregated from the endothelial fate. Finally, Moretti et al. (Moretti et al., 2006) have isolated and shown that ISL1+ cells are enriched for cardiovascular progenitors. More specifically, ISL1+NKX2.5+FLK1+ cells give rise to all three (CM, SMC, EC) lineages, while ISL1+FLK1+ cells produce SMC and EC, and ISL1+NKX2.5+ cells produce CM and SMC lineages.

However, despite the above studies identifying CPC markers, our ability to map out these sub- populations and their spatiotemporal relationship within the developmental hierarchy of cardiac cells remains limited. Surface markers that can specifically isolate progenitor cell types are required in order to enable studies that can investigate these relationships. FLK1, mentioned in the above studies, is expressed on the cell surface as they exit the primitive streak and migrate laterally to form the cardiac crescent (Ema et al., 2006). Lineage tracing studies have demonstrated a direct contribution of FLK1+ cells to the endocardium and a portion of the myocardium (Ema et al., 2006; Motoike et al., 2003) Another marker, PDGFRA, has been shown to be co-expressed with FLK1 and can be used label CPCs (Kataoka et al., 1997). PDGFRA has also been found on CPCs in the early cardiac crescent and labels differing regions within the FHF and SHF (Prall et al., 2007). Nelson et al. discovered a C-X-C chemokine receptor type 4 (CXCR4)+ /FLK1+ population as cardiopoietic lineage markers (Nelson et al., 2008). The authors also reported the CXCR4+FLK1+ population expresses FHF genes such as MESP1, GATA4, and TBX5. Further supporting the role of CXCR4 in cardiac differentiation in mPSCs, the addition of the associated ligand, stromal cell-derived factor 1 (SDF1), enhanced and accelerated cardiogenesis (Chiriac et al., 2010). More recently, Ishida et al. have discovered that glial cell line-derived neurotrophic factor receptor alpha 2 (GFRA2) identifies cardiac progenitors in mPSCs and also validated GFRA2 expression in both the FHF and SHF (Ishida et al., 2016). Interestingly, Lee et al. have

14 identified atrial and ventricular progenitors using retinaldehyde dehydrogenase 2 (RALDH2) and Glycophorin A (CD235a) respectively (Lee et al., 2017). While it is difficult to incorporate these findings into the FHF and SHF model, the authors suggest that these progenitors are present within the early migrating cardiac mesoderm. Finally, in terms of later stage cardiac development, signal regulatory protein alpha (SIRPA) has been discovered in the human system to be expressed on cardiac cells from the earliest cardiac pre-cursor stage of contraction to more mature CMs (Dubois et al., 2011). While SIRPA is considered a pan-cardiac marker, vascular cell adhesion molecule-1 (VCAM1) has also been reported to specifically mark CMs and not CPCs (Elliott et al., 2011; Pontén et al., 2013).

The proteins mentioned are a short list of cell surface markers available to enrich for CPCs (albeit in a non-specific manner) relative to the high level of complexity and cell types involved during cardiac development. While the current set of surface markers yield information regarding pre- cardiac sub-populations and downstream cardiac cell fate, the sub-populations remain largely heterogenous and the spatial and temporal relationships between the identified populations remain unclear. Therefore, the identification and validation of new cell surface antigens to enrich for CPCs and more mature CMs would not only increase the granularity of known cell types, but significantly accelerate our understanding of the hierarchy of cell types during cardiac development.

1.3 Bioreactor Technology and Stem Cell Production

The use of PSCs and their derivatives in high-throughput discovery assays and regenerative medicine requires the scalable production of well characterized cells at a high level of purity. A conservative estimate of the number of undifferentiated hPSCs required to produce sufficient CMs for heart repair is 109 cells, and this figure assumes a 1:1 PSC to CM conversion efficiency (Mummery, 2005). Surface mass spectrometry (MS) assays also require large numbers of cells (at least 108 cells for some applications) in order to achieve high enough sensitivity to detect low abundant targets (Wollscheid et al., 2009a). For both cases, cell production needs to be able to achieve the required cell number while maintaining phenotypic purity and general cellular health.

2D culture systems are traditionally used to maintain undifferentiated PSCs and can be scaled “out” using multiple culture dishes as the culture unit size remains constant while the number of parallel unit operations is multiplied (Eaker et al., 2013). However, despite some progress in this

15 area with the incorporation of sensors for oxygen and pH, real-time monitoring, and feedback- based control, 2D expansion remains a costly space and labour intensive procedure (Rodrigues et al., 2011). As an alternative, 3D culture or suspension culture has quickly become the preferred approach to large-scale stem cell production (Wang et al., 2014). The overall microenvironment and environmental cues in spherical, multicellular aggregates seem to support the maintenance of PSCs and also provide a seamless transfer towards differentiation into multiple lineages through the replacement of maintenance media with specialized differentiation media (Chen et al., 2015; Pampaloni et al., 2007).

Bioreactors are the vessels commonly used for dynamic suspension cultures to generate large numbers of cells while maintaining a controlled cell environment and to address the technical challenges due to complex growth kinetics and environmental heterogeneity typical of static conditions (Amit et al., 2010; Singh et al., 2010). Due to the improved mixing and maintenance of homogeneous culture conditions throughout the vessel, bioreactors allow for the production of stem cells at a higher density, which improves efficiency and enables high target cell numbers (Zweigerdt, 2009). Furthermore, bioreactor systems are well equipped with sensors and feedback control systems in order to continuously monitor environmental parameters (oxygen, pH, temperature), nutrients (glucose), and metabolites (lactate), which facilitate scale up in a controlled manner. In terms of stem cell maintenance, Kehoe et al. were able to culture mPSCs in serum-free suspension culture and scale up without losing pluripotency genes (Kehoe et al., 2008). Bauwens et al. found that during differentiation into cardiomyocytes, input mPSCs were 4% more efficient when oxygen tension was lowered, and now have included oxygen tension as a key factor to monitor during bioreactor scale-up (Bauwens et al., 2005). In hPSCs, culture parameters were optimized to both expand undifferentiated stem cells and differentiate into CMs in a one-step suspension culture system (Chen et al., 2015). These studies demonstrate the capability of suspension bioreactors to provide controlled culture conditions in order to generate large numbers of target cell types of desired purity.

The adoption of automated bioprocessing systems is becoming a necessity as they can significantly reduce outcome variability and high failure rates as seen in current manual protocols. Additionally, such systems can easily be applied to suspension bioreactor systems mentioned previously to process cells at scale for research and therapeutic uses. Starting with 2D systems, automatic PSC maintenance systems have been developed to expand hPSCs in traditional T-flasks and have

16 demonstrated capability of scaling cell numbers to clinically relevant levels (Thomas et al., 2009). More recently, mPSCs have been expanded in microplates and further differentiated into CMs using embryoid body (EB) differentiation protocols, albeit requiring a manual transfer step between the maintenance and differentiation systems (Kowalski et al., 2012). Improving upon these systems, a fully integrated automation platform for PSC maintenance and subsequent differentiation into the neural lineage was developed and demonstrated a significant increase in the reproducibility of cell outcome (Hussain et al., 2013). This system not only automated the entire culture process, but the authors included a centrifugation capability in order to utilize enzymatic dissociation protocols.

As automated systems become more sophisticated, they can not only be coupled with real-time monitoring of culture/environmental conditions and feedback controls but can also incorporate medium exchange strategies to further optimize conditions for cellular growth and overall health. To achieve higher yields, repeated batch medium exchanges have successfully resulted in increase in cell production in many suspension bioreactor systems (Amit et al., 2010; Chen et al., 2012; Krawetz et al., 2010; Singh et al., 2010). However, this method of medium of exchange is not easily coupled with automated systems and results in sudden changes in culture conditions upon media exchange. Profusion systems, on the other hand, avoid these issues by providing a continuous supply of fresh medium while maintaining cell retention and are easily compatible with automation and improved feedback control. High cell densities were reported in such profusion systems for both mPSC (Baptista et al., 2013; Niebruegge et al., 2008) and hPSC (Kropp et al., 2016; Serra et al., 2010) expansion. The authors demonstrated that profusion resulted in more homogenous cultures and even higher cell densities of viable cells compared to repeated batch systems.

The final component to a fully automated cell production system would be to incorporate non- invasive online measurement technology for real-time monitoring and quality control (dos Santos et al., 2013). Technology such as 3D microscopy of aggregates (Lorbeer et al., 2011) and Raman spectroscopy (Brauchle et al., 2016) to monitor culture parameters and can be used to substantially expand the options available for cell monitoring during expansion. In summary, the production of target cell types in high cell numbers is essential for the future development of regenerative medicine protocols and automated bioreactor systems equipped with real-time monitoring and feedback controls can be used achieve these goals.

1.4 Mass Spectrometry-Based Surface Profiling of Pluripotent Stem Cells and Derivatives

1.4.1 Liquid Chromatography - Tandem Mass Spectrometry

Mass spectrometry (MS) was largely confined to limited role in molecular biology in the 20th century due to its inability to ionize and vaporize labile proteins. However, with the discovery of electrospray ionization (ESI) (Fenn et al., 1989) and matrix-assisted laser desorption/ionization (MALDI) (Hillenkamp et al., 1991), mass spectrometers were now able to analyze intact proteins and unshackled the potential of MS analysis in biology. Hunt et al. developed an online liquid chromatography (LC)-MS system using ESI and was able to identify femtomole concentrations of protein in complex mixtures (Hunt et al., 1992). He used this setup to analyze peptides bound to MHC class-I proteins and is one of the early demonstrations of the utility of MS in a biologically relevant context. The incorporation of protein identification using peptide mass maps through database searching (Henzel et al., 1993; James et al., 1993; Mann et al., 1993; Yates et al., 1993) to the MS analysis pipeline further increased the speed of protein identification and ultimately the analytical power of MS. Most notable was the algorithm proposed by Yates et al. which identified the correct protein based on the greatest number of peptide matches, and was developed into the SEQUEST (Eng et al., 1994a) search engine currently used in many MS analysis pipelines. Taken together, the LC-MS hardware and database search engine software form the cornerstone for all MS-based biological analysis today.

To collect proteomic data, MS analysis modes can be divided into three major categories: data- dependent acquisition (DDA), targeted analysis, and data-independent acquisition (DIA). DDA analysis, also commonly termed discovery (or shotgun) proteomics, is mainly used to achieve complete and unbiased coverage of the sample proteome (Henzel et al., 1993). Targeted analysis, or selected reaction monitoring (SRM), is used to generate highly sensitive and reproducible analysis of a subset of known panel of proteins (Lange et al., 2008). Finally, data independent analysis (DIA) is applied to obtain reproducible and greater protein coverage than DDA and SRM methods (Chapman et al., 2014). While each method has advantages and limitations, the generally preferred method for proteome discovery is DDA.

The main benefit of DDA-based MS is the unbiased and data-driven nature of the studies. In other words, a priori knowledge of the protein composition of a given sample is not required and in

18 general, enables protein identification without the need for additional protein-based assays. Furthermore, the entire proteome of the complex sample can be interrogated at once. Less than a decade ago, a comprehensive quantification of the full yeast proteome (de Godoy et al., 2008) or the identification of 10,000 proteins in human cell lines (Beck et al., 2011; Nagaraj et al., 2011) would have taken days or even weeks to collect. Current DDA protocols have been improved and optimized in order to achieve a similar depth of coverage within a fraction of the time (Hebert et al., 2014; Kulak et al., 2014). The continued development of innovative and robust proteomic workflows will only improve the ease and efficiency by which entire proteomes are interrogated.

In addition to the identification of proteins within a given proteome, MS-based proteomics can simultaneously yield quantitative information regarding the levels of protein abundance within a sample (Bantscheff et al., 2007; Ong and Mann, 2005). The main methods of quantification currently in use are metabolic labeling, chemical labeling, spiked peptides, and label-free quantification. Metabolic labeling is the process by which a stable isotope signature is integrated within proteins during cell growth and division. Popularized by Mann et al. (Ong et al., 2002), stable isotope labeling by amino acids in cell culture (SILAC) is considered to be the most accurate of all quantification methods due to its early introduction and integration into live cells. Chemical labelling also consists of integrating stable isotopes in proteins, however, it is applied directly to protein or peptide solutions and is either enzymatically integrated or targeted based on specific chemical moieties. Two examples of chemical labeling methods are isotope-coded affinity tag (ICAT) (Gygi et al., 1999) and isobaric tags for relative and absolute quantitation (iTRAQ) (Ross et al., 2004). Both labeling methods quantify proteins based on the abundance difference between the molecular weights of the isotopic labels. In terms of absolute quantification of proteins, spiking in isotopically labeled peptides has been historically used in MS (Desiderio and Kai, 1983). More broadly referred to as absolute quantification of proteins (AQUA) (Gerber et al., 2003), protein samples are compared to known quantities of added peptides. Finally, label-free quantification consists of two commonly used strategies: quantification based on the signal intensity of a given peptide (Ono et al., 2006), or quantification based on the number of spectra (or spectral counts) generated by a given peptide (Liu et al., 2004a).

While label-free quantification is the least accurate of the previously listed quantification methods, due to the large number of systematic and non-systematic errors that can be introduced and are reflected in the final data, there are many benefits as well. Label-free methods can be applied to a

19 limitless number of samples, which is an advantage over the isotope labeling methods which are limited to 2-8 experiments for comparison. Additionally, unlike isotope-based quantification, label-free quantification does not introduce additional complexity into the sample, which may negatively influence analysis depth. And finally, label-free methods provide a higher dynamic range, which is desirable when quantifying low-abundance proteins (Wolf-Yadlin et al., 2007). Interestingly, it is reported that performances between ion intensity-based quantification versus spectral counts are comparable (Bubis et al., 2017; Fabre et al., 2014; Griffin et al., 2010), however, numerous studies have validated the advantages of label-free quantification in general, mainly due to simple nature of the sample preparation (Patel et al., 2009). Others have successfully applied label-free quantification MS in a wide variety of applications from quantifying a complete proteome of up to 5000 proteins (Distler et al., 2016; Huang et al., 2015) to revealing novel proteins in human sweat (Csősz et al., 2015) to even single-cell MS in Xenopus embryos (Lombard-Banek et al., 2016).

1.4.2 Methods to Isolate Cell Surface Proteins

Despite the increasing sophistication of MS methods, achieving complete proteome coverage while maintaining a high dynamic range of detection remains a challenge. As a strategy to increase proteome coverage, extensive sample fractionation methods such as Multidimensional Protein Identification Technology (MudPIT) (Washburn et al., 2001) can be incorporated to reduce the input sample complexity and thus, improve protein coverage of the sample. Kislinger et al. utilized MudPIT to great effect to analyze the difference between normal and diseased heart tissue and identified > 1200 high confidence proteins (Kislinger et al., 2005). The authors, however, also highlighted potential issues and optimizations required for successful isolation of protein compartments to minimize cross-contamination. In terms of cell surface proteins, the difficulty of isolation is compounded by the fact that cell surface proteins are insoluble and integral to the cell membrane. Traditional methods include density-gradient centrifugation (Castle, 2003; Huber et al., 2003), lectin-based methods (Kaji et al., 2006; Lewandrowski, 2005), cell surface shaving (Wu et al., 2003), antibody-based enrichment (Watarai et al., 2005), and two-phase separation (Elortza et al., 2003). A major disadvantage that is common to these techniques is the lack of direct labeling of the surface protein. Such a targeted approach would enable an increased signal-to-noise ratio by utilizing a unique feature of surface proteins to separate them from the background cell lysate. As a result, several techniques have been developed to specifically target cell surface proteins

20 using chemical modifications: silica bead coating, cell surface biotinylation, and chemical capture of glycosylated cell surface proteins.

Pioneered by Chaney et al., silica particles were used to isolate surface membrane proteins and achieved a reported 10-17 fold purification with low levels of cytoplasmic contamination (Chaney and Jacobson, 1983). The method was predicated on the positively charged silica particles binding to the negatively charged cell surface, which was then subsequently cross-linked to form silica sheets. These sheets were then able to be separated from the cell lysate through centrifugation, after which the isolated surface proteins were eluted from the beads for processing and MS analysis. This method can also be applied in vivo by perfusing the silica beads through the vasculature. Durr et al. employed this technique on rats to assess the membrane proteins of endothelial cells in the rat lung (Durr et al., 2004). The authors identified 450 proteins on the luminal endothelial cells and reported 41% of proteins identified are not detected in vitro. Arjunan et al. performed a similar study on mouse heart endothelial cells, but with minor success (Arjunan et al., 2009). The authors reported increased contamination from cytoplasmic proteins due to the harsh methods employed to isolate the endothelial cells from fibrous heart tissue. While these studies demonstrated the utility of identifying surface proteins in their native environment in vivo, the effectiveness of silica bead coating is limited exclusively to luminal endothelial cells and is tissue-specific in terms of maintaining a high enrichment of cell surface proteins.

In contrast to silica bead coating, chemical modification of surface proteins is fast becoming an attractive method of labeling target proteins. The chemical modification provides a tag that can be used to resolve surface proteins from untagged cytoplasmic and nuclear proteins using affinity purification. A popular surface labeling method is through the biotinylation of lysine residues of surface proteins. The biotin conjugated proteins/peptides are then captured using avidin/streptavidin-coated solid supports while the untagged proteins/peptides are washed away. The bound proteins/peptides are then eluted from the support material, processed, and analyzed with MS. Nunomura et al. applied biotinylation enrichment to mouse PSCs and identified 235 surface proteins out of 324 captured proteins (Nunomura, 2005b). Similar to silica bead coating, biotinylation of surface proteins can also be applied in vivo to capture cell surface proteins in the vasculature. Rybak et al. perfused mice with a biotin tag and were able to identify organ-specific protein signatures as well as detect qualitative and quantitative differences in surface protein expression between normal and tumorigenic organs (Rybak et al., 2005). The diversity of biotin

21 chemistry enables multiple configurations of the biotinylation reagent providing control of physical properties such as solubility, membrane impermeability, cleavable linkers, and target functional group (Elia, 2008; Gauthier et al., 2004). However, potential issues for this method also include cross-contamination of proteins in other cellular compartment due to the common reaction moieties on all proteins and a partial penetration of the labeling reagent into the cytoplasm due to apoptotic or necrotic cells.

To address this, a novel cell surface capture (CSC) protocol (Wollscheid et al., 2009a) employs a three-step tandem affinity labelling strategy to highly enrich for plasma membrane proteins. First, the glycan structures on surface glycoproteins are oxidized with sodium meta-periodate and covalently bound to the bi-functional linker molecule biocytin hydrazide. Second, cells are lysed, and the membrane patches are collected and subsequently digested with trypsin. Third, the peptides containing labeled glycans are immobilized on streptavidin beads. The peptides are then released from the beads using PNGaseF, resulting in a peptide mixture enriched for cell surface proteins. Due to the PNGaseF cleavage enzyme, the asparagine residue, on which the glycans on N-linked glycoproteins are attached, is deamidated resulting in a conversion to aspartic acid, and a corresponding mass shift, which is detectable within the mass spectrometer. This mass shift in conjunction with the predictive glycosylation motif results in a high confidence identification of a bona fide surface glycopeptides.

1.4.3 Cell Surface Proteomics in Stem Cells

Global cell surface profiling of stem cells in an important step in the capturing the breadth of surface markers available on a cell (Yang et al., 2008a). Membrane proteins on the surface of the cell are involved in many critical functions involving cell–cell interactions and cell signaling via surface receptors (Wu and Yates, 2003). One method of identifying cell surface proteins is using antibody-based methods to screen for positive hits. Dubois et al. successfully applied this strategy to identify signal regulatory protein α (SIRPA) on hPSCs from a panel of 370 known cluster of differentiation (CD) antibodies in order to identify markers to isolate CMs from hPSC differentiation cultures (Dubois et al., 2011). Maesner et al. also designed custom antibody panels using prior knowledge of markers expressed on satellite cells in skeletal muscle to assess distinct or overlapping phenotype subsets of the heterogeneous satellite cell cultures (Maesner et al., 2016). With a target library of 377 cell surface proteins, Collier et al. profiled human naïve and primed

22 pluripotent cells and were able to compare both profiles to human blastocysts (Collier et al., 2017). As demonstrated, antibody-based methods can be effective and yield positive results, however, these studies are ultimately restricted to the known subset of surface markers. In contrast, MS- based proteomic studies are not limited by the current knowledge base and can perform an unbiased interrogation of the target proteome.

MS-based proteomic studies can comprehensively interrogate and identify proteins samples without a priori knowledge of the number and types of proteins expected in the sample. Additionally, the ability to accurately identify and quantify surface protein abundance can be used to understand and highlight important differences between diseased and normal cells/tissues. Shin et al. applied surface protein enrichment methods on multiple cancer cell lines, including neuroblastoma, adenocarcinoma, leukemia, and ovarian tumour cells (Shin et al., 2003). The authors identified surface profiles for each cancer line and found line-specific proteins that aligned with the individual characteristics of each cell line. The authors also reported a common set of chaperone proteins highly expressed across all cell lines. Cell surface labeling can also be used to dynamically track the abundance level of surface proteins in response to stimuli. Schiess et al. identified 202 glycoproteins on Drosophila cells and demonstrated the ability to detect cell surface protein dynamics in response to insulin stimulation, including receptor internalization and downstream intracellular signaling networks (Schiess et al., 2008).

Surface protein signatures can also be used as markers to distinguish between different cell types. Cell surface analysis of mPSCs and hPSCs have previously revealed a large set of diverse markers and signaling molecules associated with mPSC maintenance and development. As mentioned previously, Nunomura et al. applied surface MS analysis to the D3 mPSC line in order to investigate the interactions between surface proteins and extracellular ligands (Nunomura, 2005b). The authors reported that mPSCs express receptors from the major signaling pathways that control cell maintenance of undifferentiated cells as well as those that initiate differentiation. This result was corroborated by Nagano et al., who performed a similar study on E14-1 mPSCs and reported the identification of 260 membrane proteins (Nagano et al., 2005). The authors also reported a similar variety of receptors and surface proteins expressed. Comparisons made by Munoz et al. have identified extensive overlap between the surface proteome of hPSCs and induced PSCs

(iPSCs) (Munoz et al., 2011). In a similar study, Boheler et al. also confirms a high degree of overlap between the two surface proteomes and validate proteins that are unique to hPSCs (Boheler

23 et al., 2014a). The authors even identify and validate a negative selection marker in order to enable selective removal of undifferentiated hPSCs from a differentiation culture. Rugg-Gunn et al. identified surface markers that are able to specifically identify and separate mPSCs, epiblast, trophoblast, and extraembryonic endoderm stem cell linages. These results allow for the selective isolation of cells in the early embryo and enable new approaches to study cell fate specification (Rugg-Gunn et al., 2012). Using the identified surface markers, the authors were able to sort out each respective cell type from blastocysts in vivo to further investigate mechanisms that regulate cell fate decisions in the early embryo.

The dynamic nature of surface protein profiles during differentiation can also be interrogated using cell surface proteomics. To investigate the mechanisms that characterize the emergence and maturation of early neural stem cells (NSCs), DeVeale et al. profiled the cell surface throughout the transition from mPSCs to primitive NSCs to finally definitive NSCs (DeVeale et al., 2014). Surface profiling revealed key signaling and adhesion factors that played significant roles during the differentiation into definitive NSCs. Applied to the immune system, Kalxdorf et al. investigated the surface profile of monocytes as they differentiated into macrophage-like cells

(Kalxdorf et al., 2017). The authors were able to identify time-dependent changes in membrane protein abundance and demonstrated key functionality by inhibiting a specific receptor family resulting in severely compromised macrophage differentiation. In the skeletal muscle system, Gundry et al. report that surface protein profiling revealed key markers that scaled with myotube differentiation (Gundry et al., 2009). The markers were validated and revealed the presence of cell intermediates for further investigation. Surface proteomics can also be applied to terminally differentiated cell types to further characterize cell function. Sharma et al. identified transmembrane protein 65 (TMEM65) as being critical for the function of Connexin 43 (CX43) in CMs (Sharma et al., 2015). The authors functionally validated this marker through knockdown studies in both in vitro and in vivo mouse systems.

The results from these studies are collected into MS data repositories such as the Cell Surface Protein Atlas (Bausch-Fluck et al., 2015a) and the proteomics identifications database (PRIDE) (Vizcaíno et al., 2014), which provide a global space to submit and disseminate MS-based data. With the continued advance of MS technology and membrane enrichment techniques, high throughput interrogation of the complete surface proteome will soon be possible. MS technology can then enable increasingly sophisticated studies that address more complex questions dealing

24 with the dynamic nature of protein composition and quantities during both development and disease progression.

1.5 Transcriptomic Analysis of Pluripotent Stem Cells and their Derivatives

1.5.1 Microarray Analysis and the Advent of Deep Sequencing

The ubiquitous spread of microarray technology has significantly transformed molecular biology. From its origins as a tool for transcript-level analysis (Schena et al., 1995), microarrays have become common place in many studies. Applications of microarrays have yielded a range of critical information including, transcriptomic differences between different cell types and tissues (Chan et al., 2009; Kai et al., 2005), gene expression changes during differentiation (Arbeitman et al., 2002; Spellman et al., 1998), disease phenotypes (Alizadeh et al., 2000; Golub et al., 1999), and evolutionary changes between different species (Brem et al., 2002; Zhang et al., 2007). While there are numerous applications for microarray technology such as determining genotype, micro ribonucleic acid (RNA) expression, and deoxyribonucleic acid (DNA) copy number to name a few, gene expression profiling remains the most common application by far.

Modern microarrays consist of an array of oligonucleotide probes representing transcripts that are to be investigated (Pease et al., 1994). In most commercially available arrays, the probes are designed based on known genome sequence and known or predicted open reading frames, and usually have multiple probes assigned to one target gene. Transcripts extracted from cell samples are then labeled with a fluorescent dye and hybridized to the probes on the array. The transcripts that correspond with the complementary probe bind to their target. While the fluorescent intensity is not proportional to the amount of bound genetic material (Held et al., 2006), the intensity per probe spot is analyzed and measured as gene expression.

The majority of variability in microarray studies have come initially come from probe design and equipment differences. Production of microarrays from different manufacturers and laboratories appeared to produce differing results using the same samples. Tan et al. reported considerable variance across three platforms using a single RNA sample (Tan et al., 2003). The equipment being used to analyze the microarrays were also a major factor in the quantification of gene expression data. Irizarry et al. extended the previous study to span multiple laboratories and compared data performance across different platforms as well as different labs (Irizarry et al., 2005). The authors

25 found that there was a large difference in data collected between different labs, however, amongst the top performing labs, the data was consistent. Even environmental factors impacted the variance in data collected from microarrays. Fare et al. reported differential ozone susceptibility of fluorophores used to quantify microarray data and suggested maintaining ozone levels below 2ppb to ensure no degradation of fluorophores in microarray analysis facilities (Fare et al., 2003).

To address the issues of biases and inter-lab and inter-manufacturer variability, organizations such as the MicroArray Quality Control (MAQC) consortium has led multiple studies to evaluate and standardize microarray-based experiments (MAQC Consortium et al., 2006). In addition to the hardware variability, the MAQC and others have developed standardized computational methods to address the systematic variation between laboratories. The MAQC looked specifically at computational methods used to quantify microarray data and recommend a set of metrics that do not show batch variability in upwards of 75% of the cases evaluated (Luo et al., 2010). These findings are also summarized by Reimers and Tarca et al., who highlight significant challenges and solutions in the data-processing aspect of microarray analysis (Reimers, 2010; Tarca et al., 2006). With these major efforts, the underlying variance and biases of microarray experiments have largely been studied and understood resulting in a standardized method to employ stable analytical solutions to microarray data.

While microarray technology has been undergoing revisions and overhauls to standardize and minimize variance between studies, DNA sequencing technology has been quietly progressing. Now, the superiority of microarrays for transcriptome profiling is being challenged by deep sequencing. The main barrier for the wide usage of sequencing technology was due to the immense time investment, and the associated costs, required per base sequenced. Early efforts required years of work across multiple laboratories and institutions to finally yield the Drosophila genome (Adams et al., 2000). With the recent developments in sequencing technology, the same fly genome can be sequenced within several weeks (Mardis, 2008). These techniques have since then been adapted and extended to sequencing the transcriptome, which is referred to as RNA-Seq.

RNA-Seq directly sequences the transcripts present in a sample instead of using hybridization to capture transcripts with target probes. The sequenced transcripts are then mapped back to a reference genome and based on the number of sequences produced, a gene expression value is inferred. The fundamental difference in approaches provide RNA-Seq with a number of

26 advantages over microarrays. RNA-Seq does not require a priori knowledge of the genome, resulting in the ability to sequence junctions between exons for each gene and detect RNA editing events (Mortazavi et al., 2008; Nagalakshmi et al., 2008). Additionally, probe design for microarrays are heavily dependent on annotated genes and as a result, microarrays will miss potential genes whereas RNA-Seq only requires the sequenced strand to be present in the genome (Telonis-Scott et al., 2009). Furthermore, RNA-Seq can be used to sequence samples that do not have full genome sequences available (Grabherr et al., 2011). These advantages are summarized in a study by Zhao et al., who found that RNA-Seq was far superior to microarrays in detecting low abundant transcripts and differentiating critical isoforms (Zhao et al., 2014). However, despite these advantages, microarrays are still a common choice amongst researchers. RNA-Seq methods are much more expensive than microarrays, and due to the challenges in data analysis and storage, microarrays still add value to global transcriptomic profiling. However, much like the microarray, RNA-Seq methods and analyses will likely be standardized and once these barriers are overcome, will become the predominant platform for transcriptomic analysis.

1.5.2 Expression Profiling in Cardiac Development

Several groups have performed microarray and RNA-Seq studies on PSC-derived CMs. Nelson et al. profiled mPSCs that were differentiated into CMs using a serum-based protocol. At specific stages, the authors used microarrays to assess the transcriptome and identified 306 genes specifically upregulated with cardiogenic differentiation and discovered a CXCR4+FLK1+ population enriched for the cardiopoietic lineage. Using a similar strategy but with an alternative method of differentiation, Faustino et al. also profiled mPSCs as they differentiated into CMs (Faustino et al., 2008). Using microarrays, the authors identified 65 upregulated genes associated with the cardiac phenotype. In both studies, the authors generated a cardiogenic network that modeled cardiac differentiation and identified critical nodes and manipulation points that controlled cardiac output. In parallel, in the human system, Xu et al. profiled key stages during cardiac differentiation where the CMs were generated through antibody selection (Xu et al., 2009). The authors identified four genes as potential markers of various stages of cardiac development and validated the genes in vitro and gene orthologs in vivo and found predominant localization within the mouse heart. A similar analysis was done by Hartogh et al., where CMs were generated in a directed manner using cytokines to initiate cardiac differentiation (Hartogh et al., 2016). Using network analysis, the authors identified leucine rich repeat containing G protein-coupled receptor

4 (LGR4) as a surface marker for cardiac progenitors. Taken together, these studies indicate potential for microarray and subsequent RNA-Seq analysis to identify key genes regulating cardiac development. Additionally, much like the data generated through MS, genomics data are stored in online repositories like the Genome Expression Omnibus (GEO) (Edgar et al., 2002). With the continued improvement of microarray and RNA-Seq technology, genome and proteome data can ultimately be combined to further produce insights into cardiac development.

1.6 Thesis Goals and Approach

The goal of this thesis is to achieve a greater understanding of the hierarchy of cardiac progenitor cell types and interrogate the intercellular communication dynamics between these cells. To achieve these goals, this work will also address the lack of surface markers that can be used to specifically identify sub-populations of cardiac progenitors. Bioreactor technologies capable of producing large numbers of PSC-derived CPCS and mature CMs will enable novel proteomic strategies to identify surface proteins/markers at distinct stages of cardiac development. These markers, through isolation and functional validation of resultant cell populations, will characterize sub-populations with distinct cardiomyogenic potentials. This cell differentiation hierarchy can provide insight as to how cell-cell interactions and non-autonomous signaling impact differentiation.

Specifically, I posit that cell surface proteomics will yield novel cell surface markers during early cardiac differentiation. I will test this hypothesis by performing cell-surface antigen capture and proteomics to identify known and novel cell surface markers on mesoderm/cardiac cells that were produced in serum-free bioreactors. The generation of CPCs will be defined throughout this thesis as FLK1+PDGFRA+ expressing cells. Bioinformatic analysis of mass spectrometry and additional microarray expression profiles will be used to identify lists of interesting candidate markers and validated using flow cytometry and quantitative polymerase chain reaction (qPCR). Promising markers will then be selected for further validation through single-cell sorting and characterized for cardiac potential.

FZD4 MARKS LATERAL PLATE MESODERM AND SIGNALS WITH NORRIN TO INCREASE CARDIOMYOCYTE INDUCTION FROM PLURIPOTENT STEM CELL-DERIVED CARDIAC PROGENITORS

This chapter has been published online in Stem Cell Reports (Yoon et al., 2018). Co-authors include Hannah Song, Ting Yin, Damaris Bausch-Fluck, Andreas P Frei, Steven Kattman, Nicole Dubois, Alec D. Witty, Johannes A. Hewel, Hongbo Guo, Andrew Emili, Bernd Wollscheid, Gordon Keller, and Peter W. Zandstra

Author Contributions C.Y. initiated the project with guidance from P.W.Z. All analyses were done by C.Y. Mouse embryonic stem cell experiments were designed and performed by C.Y. Sorting studies were aided by T.Y. Microarray data preparation was performed by C.Y., H.S., S.K., N.D., and A.D.W. Cell Surface Capture was performed by C.Y. with help from D.B. Mass spectrometry was performed by C.Y. with help from J.A.H., H.G., and A.P.F. Human embryonic stem cell differentiation was performed by H.S. C.Y. prepared the figures and wrote the manuscript with input from all authors.

2.1 Abstract

In order to understand the spatiotemporal dynamics of stem cell differentiation, identification of cell surface markers is a key factor in determining the functional characteristics and dynamics of stem cells and their derivatives. Here, using an integrated mass spectrometry and microarray-based approach, we analyzed the surface proteome and transcriptome of CPCs generated from the stage- specific differentiation of mouse and human pluripotent stem cells. Through bioinformatic analysis, we have identified and characterized FZD4 as a marker for lateral plate mesoderm. Additionally, we utilized FZD4, in conjunction with FLK1 and PDGFRA, to further purify CPCs and increase CM enrichment in both mouse and human systems. Moreover, we have shown that NORRIN presented to FZD4 further increases CM output via proliferation through the canonical WNT pathway. Taken together, these findings demonstrate an important role for FZD4 in mammalian cardiac development.

2.2 Introduction

Understanding the composition of cell surface proteins on stem cells or their derivatives is critical in the development of strategies in order to elucidate functional characteristics and cell interaction dynamics (Nunomura, 2005a). Cardiac progenitor cells are derived from the lateral plate mesoderm and ultimately form the adult heart (Kinder et al., 1999; Rana et al., 2013). Reports on the enrichment of cardiac progenitor cells (CPCs) (Cai et al., 2003; Passier et al., 2005b) show that it is possible to use in vitro cardiomyocyte (CM) differentiation of mouse embryonic stem cells (mPSCs) (Amit et al., 2000; Cameron et al., 2006b; Gerecht-Nir et al., 2004b; Sachinidis et al., 2003) as a model system to study cardiac development. A brachury (BRY)+, fetal liver kinase 1 (FLK1)+ population has been shown to mark two separate mesodermal progenitor populations that arise sequentially; the first gives rise to the hemangioblast and subsequent endothelial and hematopoietic lineages; and the second is enriched for early cardiovascular progenitors that display cardiomyocyte (CM), smooth muscle cell (SMC), and endothelial cell (EC) lineage potential (Kattman et al., 2006). Another type of progenitor cell expressing NK2 transcription factor related, locus 5 (NKX2.5) and kit oncogene (C-KIT) is capable of clonal expansion and differentiation to CMs, conduction system cells and SMCs in vitro (Wu et al., 2006a). However, these cells do not give rise to cells of the endothelial lineage, suggesting that they represent a more differentiated progenitor population that has already segregated from the endothelial fate. Finally, isolated islet

1 (ISL1)+ cells have been shown to be enriched for CPCs (Moretti et al., 2006). More specifically, ISL1+NKX2.5+FLK1+ cells give rise to CM, SMC, and EC lineages, while ISL1+FLK1+NKX2.5- cells produce SMCs and ECs, and ISL1+FLK1-NKX2.5+ cells produce CM and SMC lineages. It is clear that a number of genes have been implicated in early cardiogenesis, however only a small fraction encodes surface markers. These include FLK1, platelet derived growth factor alpha (PDGFRA) (Bondue et al., 2011), and more recently, C-X-C chemokine receptor type 4 (CXCR4) (Nelson et al., 2008). Accordingly, there is a need for additional surface markers that can identify additional cardiogenic sub-populations, which would allow for the enrichment of pure of CPCs and thus, enable a better understanding of cell dynamics during heart development.

Cell surface analysis of mPSCs and hPSCs have previously revealed a large set of diverse markers and signaling molecules associated with mPSC maintenance and development (Bausch-Fluck et al., 2015b; Boheler et al., 2014b; Zhang et al., 2009). We have extended this analysis to encompass mPSC differentiation to cardiac mesoderm and ultimately to CMs. Using mass spectrometry, we identified 246 surface markers during key stages of mesoderm specification and early cardiac development in vitro. We also performed microarray analysis on the CPC sub-populations isolated by surface markers and cross-referenced the proteomic data to identify candidate proteins specific to CPCs. These proteins were further validated using qRT-PCR and flow cytometry leading to the selection of five promising CPC marker candidates. In this context, we focused on the surface receptor frizzled 4 (FZD4) for further analysis. Supporting our technical validation, FZD4 is known to be involved in the Wingless (WNT) signaling pathway, which is known to play an essential role in cardiac development (Cadigan and Nusse, 1997; Cohen et al., 2008b; Gessert and Kühl, 2010a).

Here, we demonstrated that day (d) 3.75 mPSC derived FZD4+ cells are enriched for LPM, which gives rise to cardiomyocytes, while the FZD4- population is enriched for paraxial mesoderm, which gives rise to somites. We also have shown in both mPSC and hPSC systems, the CPC population can be further segregated into FZD4+ and FZD4- sub-populations. Furthermore, we have provided evidence that activation of the WNT signaling pathway with the addition of the ligand NORRIN further increases the cardiogenic output of the FZD4+ cell population. In conclusion, FZD4 is a marker that can be used to segregate pre-cardiac mesoderm and FZD4-

NORRIN signaling increases CM output through canonical WNT activation, which further validates the role of FZD4 in cardiac induction.

2.3 Results

2.3.1 Integrated Mass Spectrometry and Microarray Analysis Identifies Early Mesoderm Surface Markers

A staged protocol was applied in bioreactors to differentiate mPSCs to cardiomyocytes (Figure 2-1A). The stages were tracked by flow cytometry to characterize and assess purity starting with mPSCs (OCT4+SOX2+) followed by epiblast-like cells (CD24+CD40+), then primitive streak cells (BRY+), then mesoderm containing CPCs (FLK1+PDGFRA+), and finally CMs (cardiac troponin T; CTNT+) (Figure 2-1A I-V). We then used mass spectrometry (MS) in order to identify and characterize surface protein abundance profiles corresponding to the CPC stage of development (Figure 2-1A IV). MS analysis of surface proteins has been generally challenging due to the limited abundance of surface proteins relative to intracellular proteins, and difficulty in isolating hydrophobic membrane-bound proteins (Josic and Clifton, 2007; Macher and Yen, 2007). To address these concerns, we utilized a tag-based method (Wollscheid et al., 2009b) to specifically label and exclusively isolate surface proteins for analysis (Figure 2-1B). Protein abundance at the cell surface was compared between time points using label-free quantification. Unsupervised k-means clustering was performed revealing five distinct temporal patterns (Figure 2-1C). Cluster 1 contained 246 proteins whose abundance increased at the CPC stage and included known CPC markers, such as FLK1 and PDGFRA. Cluster 1 was also enriched for receptors relevant to mesoderm differentiation based on gene ontology (GO) analysis. However, the d3.75 time point, associated with the emergence of CPCs, is a mixed population and thus, it was difficult to determine which of the highly expressed surface proteins are associated with the FLK1+PDGFRA+ sub-population. In order to better understand the underlying heterogeneity, we next performed a microarray time course focusing on the CPC and CM stages.

The CPC stage cell population consists mostly of the cardiogenic FLK1+PDGFRA+ and hemogenic FLK1+PDGFRA- sub-population (Kattman et al., 2006). We sorted the two sub- populations from the CPC stage and compared their transcriptome to that of the CTNT+ CMs (Figure 2-2A). Hierarchal clustering of the gene expression data showed distinct expression patterns unique to each of the sub-populations. When both of these sub-populations were compared

Figure 2-1: Cell surface mass spectrometry yields protein abundance profiles. A) Cell production strategy involving embryoid body differentiation with cytokine addition at d2 and d3 of differentiation. Cell types were assessed using flow cytometry with stage-specific markers (I-V). The number in the box represents the percent positive value. B) Surface proteins were labeled and processed into peptides using Cell Surface Capture procedure before MS analysis. C) Unsupervised k-means clustering of cell surface MS data was separated into 5 main clusters (Red = upregulated, White = neutral, Blue = downregulated; n = 4-5 in each group, pooled from 6 independent experiments). Line graph representations of each cluster show distinct profiles of up and down regulation.

33 to CMs, both showed similar pathway enrichment, however, the FLK1+PDGFRA+ sub-population showed 10-fold greater significance in cardiovascular pathway enrichment compared to the FLK1- PDGFRA+ population (Figure 2-3). Overall, 871 of 16,755 genes were differentially expressed (either up- or down-regulated) when the FLK1+PDGFRA+ sub-population was compared to the FLK1+PDGFRA- sub-population. While these two sub-populations were relatively similar in gene expression and were enriched for similar cardiac pathways, the up-regulated pathways specific to cardiogenesis showed a higher enrichment, indicated by a larger -log(P value), (Figure 2-2B) suggesting that the CPCs reside in the FLK1+PDGFRA+ sub-population. As stated previously, MS analysis at the CPC stage was performed on a heterogeneous population. In order to determine which sub-population is associated with the identified surface markers, we applied gene-level segregation between hemogenic and cardiogenic sub-populations with respect to surface protein abundance by cross-referencing the genes with the identified proteins based on common patterns. As a result, the 246 surface proteins identified in cluster 1, were divided into three categories: 47 proteins uniquely expressed in the FLK1+PDGFRA+ sub-population, 34 proteins in the FLK1+PDGFRA- sub-population, and 165 proteins common to both, (Figure 2-2C). Thus, by integrating the specificity of the MS-based protein identification with the purity of microarray analysis of sorted sub-populations, we have identified a list of 47 surface markers that are uniquely enriched in the FLK1+PDGFRA+ CPC sub-population. We next set out to independently analyze and validate this data set.

Figure 2-2: Microarray analysis yields sub-population specific surface markers. A) Gene analysis was performed on the sorted FLK1+PDGFRA-, FLK1+PDGFRA+, and CTNT+ populations in triplicate. Unsupervised clustering yielded distinct expression patterns unique to each cell type (Red = upregulated, White = neutral, Blue = downregulated; n = 3 in each group, pooled from 4 independent experiments). B) In a comparison between the FLK1+PDGFRA+ and FLK1+PDGFRA- populations, pathway enrichment analysis of the significantly upregulated and downregulated genes revealed pathways relevant to CM development. C) Proteins identified by MS from cluster 1 were further segregated as either uniquely or commonly expressed on FLK1+PDGFRA+ and FLK1+PDGFRA- cells using microarray data.

Figure 2-3: GO analysis of CPC sub-population relative to CM, related to Figure 2-2.

A) CM compared to FLK1+PDGFRA+ sub-population. B) CM compared to FLK1+PDGFRA- sub-population.

2.3.2 FZD4 Identified as a Potential Marker of a Sub-Population Enriched for Cardiac Progenitors

The 47 proteins that were enriched in the cardiogenic sub-population were assessed and filtered based on literature search, messenger RNA (mRNA) expression and protein abundance correlation (Figure 2-4A), and antibody availability and efficacy (Figure 2-4B). These analyses identified FZD4, integrin b5 (ITGB5), lysophosphatidic acid receptor 4 (LPAR4), and PLEXINB1 as top candidates for further validation. Co-staining with these four candidate markers resulted in further segregation of cardiac progenitors within the heterogeneous FLK1+PDGFRA+ population (Figure 2-4C). Since FZD4 is involved in the WNT signaling pathway, which is developmentally

Figure 2-4: Validation of candidate surface markers, related to Figure 2-5.

A) The 47 surface proteins unique to the cardiac progenitor population were validated with qPCR and 32 proteins with significant correlation between transcriptomic and proteomic data were selected (n = 3 for each group, mean ± SEM). B) Proteins were validated using a flow cytometry time course, and antibodies that showed non-specific binding were discarded, resulting in 10 proteins C) The resultant proteins with good quality antibodies were then assessed for their ability to resolve the cardiac progenitor population yielding 4 candidate markers. The number in the box represents the percent positive value.

37 relevant during early cardiac differentiation (Cohen et al., 2008b; Gessert and Kühl, 2010a; Ueno et al., 2007b), its abundance was further evaluated by flow cytometry (Figure 2-5Error! Reference source not found.A). FZD4 abundance frequency decreased during primitive streak formation, increased at the cardiac progenitor stage, and decreased again when the CM population began to emerge. These observations suggested a potential role for FZD4 during CPC development and we chose to further investigate it as a potential marker of a sub-population with cardiac potential.

Figure 2-5: Flow cytometry and qPCR validation of proteins.

A) FZD4 protein abundance time course was assessed using flow cytometry (n = 4 for each group, mean ± SEM). B) FLK1+PDGFRA+ sub-population concomitantly stained with FZD4 show FZD4+ and FZD4- sub-populations. The number in the box represents the percent positive value. C) qPCR interrogation of lateral plate and paraxial mesoderm genes on sorted FLK1+PDGFRA+FZD4+ and FLK1+PDGFRA+FZD4- populations. FLK1+PDGFRA+FZD4+ cells showed an enrichment of lateral plate mesoderm genes while FLK1+PDGFRA+FZD4- cells showed an enrichment of paraxial mesoderm genes (n = 4 for each group, mean ± SEM). Statistical analysis was performed using a two-sided Mann- Whitney U test, * p<0.05.

2.3.3 Gene Expression Analysis on Sorted CPC Sub-Populations Indicates that FZD4 Marks Pre-Cardiac Mesoderm in CPCs

We next set out to examine the role of FZD4 in CPC development. We first used a sorting strategy to determine if the FZD4+ and FZD4- populations captured a bifurcation point in early cardiac differentiation (Figure 2-5Error! Reference source not found.B). Cardiac progenitors were generated (as described in Figure 2-1A) and the gated FLK1+PDGFRA+ population was sorted into FZD4+ and FZD4- sub-populations (Figure 2-6A). Enrichment of FLK1, PDGFRA, and FZD4 abundance relative to the mock sorted sample within each compartment was confirmed using both flow cytometry (Figure 2-6B) and qPCR analysis (Figure 2-6C) which demonstrated high purities and minimal contamination from adjacent sub-populations.

We analyzed the FZD4+ and FZD4- populations directly after being sorted for genes that are expressed as the cells migrate from the primitive streak towards the anterior lateral region of the embryo, specifically looking at the bifurcation point segregating lateral plate and paraxial mesoderm. Cardiomyocytes are derived from the lateral plate mesoderm and are marked by LIM domain only 2 (Lmo2), Platelet endothelial cell adhesion molecule 1 (Pecam1), NK2 homeobox 5 (Nkx2.5), and Isl1, while paraxial mesoderm is marked by mesenchyme homeobox 1 (Meox1), T- box 6 (Tbx6), transcription factor 15 (Tcf15), and paired box 1 (Pax1) (Cheung et al., 2012). We observed that the FZD4+ population expressed higher levels of lateral plate mesoderm markers, while the FZD4- population expressed higher levels of paraxial mesoderm markers (Figure 2-5C). These results indicated FZD4 may distinguish between lateral plate and paraxial mesoderm and be used to further purify the early mesoderm population for CPCs.

Figure 2-6: Flow cytometry and qPCR validation of purity of sorted populations, related to Error! Reference source not found..

A) Gating strategy to sort cardiac progenitor cells. Five sub-populations were sorted from the cardiac progenitor stage (FLK1-PDGFRA-, FLK1-PDGFRA+, FLK1+PDGFRA+, FLK1+PDGFRA+FZD4+, and FLK1+PDGFRA+FZD4-) and differentiated into cardiomyocytes. Purity of sorted samples were assessed using B) flow cytometry and C) qPCR (n = 4 for each group, mean ± SEM).

2.3.4 FZD4-Expressing CPCs Yield Higher CM Outputs

Given that our gene expression analysis showed that the FZD4+ CPC sub-population was enriched for LPM markers, we decided to examine the cardiogenic potential of the FZD4+ expressing cells in vitro. Sorted FZD4+ and FZD4- populations were seeded onto a 384 well plate and cultured for 3 days until beating CMs were observed (Figure 2-7A). Bright-field imaging revealed that the FLK1+PDGFRA+FZD4+ derived fraction maintained robust beating and web-like networks, while beating was generally not observed in the FLK1+PDGFRA+FZD4- derived fraction, which displayed static cell monolayers (Figure 2-7B). This observation was consistent with gene expression (Ctnt, alpha myosin heavy chain (α-mhc), Isl1, and Pecam1) (Figure 2-7C), flow cytometry (CTNT) (Figure 2-8A), and immunofluorescence (CTNT) (Figure 2-8B) analyses of the sorted populations.

To compare the level of cardiac induction between the FZD4+ and FZD4- sub-populations, we quantified the frequency of CTNT abundance and used 5-ethynyl-2´-deoxyuridine (EdU) to estimate cell proliferation (Figure 2-8C). Compared to the mock sorted controls that showed a CTNT+ frequency of 31.1 ± 8.6%, the CPC (FLK1+PDGFRA+) population showed a significantly enriched CTNT+ output of 53.9 ± 4.3%, and the non-CPC (FLK1-PDGFRA-) showed a lower output of 5.0 ± 2.0%. Additionally, the FLK1+PDGFRA+FZD4+ (62.1 ± 6.0%) and FLK1+PDGFRA+FZD4- (38.7 ± 6.4%) condition showed a significant increase and decrease of CTNT+ abundance respectively, relative to the FLK1+PDGFRA+ population. We also observed this trend in the CTNT+EdU+ sub-population indicating that there is an increased baseline proliferation of CTNT+ cells in the FLK1+PDGFRA+FZD4+ sub-population. This suggests that the increased CTNT+ expression in FZD4+ sorted populations may be due to a proliferation-based mechanism.

Figure 2-7: Differentiation of cardiac progenitors into cardiomyocytes, related to Figure 2-8.

A) d3.75 CPCs are sorted into 5 sub-populations and seeded into individual wells and cultured for 3 days and assessed for CTNT. B) Bright-field images show beating CM in the FLK1+PDGFRA+FZD4+ condition. C) qPCR measurements of cardiac markers (Ctnt, α-mhc, Isl1) and endothelial cell marker (Pecam1). Expression is normalized relative to the unsort condition. The FZD4+ compartment expressed high amounts of Ctnt and α-mhc relative to the other compartments (n = 4 for each group, mean ± SEM). D) Gating strategy to evaluate cardiomyocytes using flow cytometry.

Figure 2-8: FZD4 activation further enriches CPC in the FLK1+PDGFRA+ population increasing subsequent CM yield. A) Cardiac output of sorted progenitor sub-populations based on CTNT abundance measured by flow cytometry (The number in the box represents the percent positive value.) and B) immunofluorescence staining CTNT-expressing cells (green) and DAPI nuclear stain (blue), scale bar represents 500 µm. C) Quantified from flow cytometry, d7 CTNT+ percentage from

43 d3.75 input CPC populations sorted using CPC-associated markers FLK1, PDGFRA, and FZD4. CTNT abundance is significantly higher in the FLK1+PDGFRA+FZD4+ population relative to both the FLK1+PDGFRA+ and FLK1+PDGFRA+FZD4- population (n = 4 for each group, mean ± SEM). Statistical analysis was performed using a two-sided Mann-Whitney U test, * p<0.05. D) I) d3.75 (input) CPC marker abundance was correlated to the cardiomyocyte percentage (CTNT+) of the d7 (output) populations. II) In the mock sorted d7 output, the CTNT abundance has a higher correlation with the FLK1+PDGFRA+FZD4+ input marker than FLK1+PDGFRA+. III) FZD4 alone also exhibited a higher correlation than FLK1+PDGFRA+ with the output CTNT abundance in the mock sorted sample (n = 4 for each group). Statistical analysis was performed using a two-sided Mann-Whitney U test, * p<0.05.

To explore how FZD4 abundance correlates with output CTNT response, we compared the different d3.75 (input) CPC markers to cardiomyocyte percentage (CTNT+) of the d7 (output) populations (Figure 2-8D I). A high correlation between an input CPC marker and output CTNT abundance can be indicative of the predictive power of the CPC marker. In the mock sorted d7 output, CTNT abundance correlated more highly with the FLK1+PDGFRA+FZD4+ input marker than FLK1+PDGFRA+ alone (Figure 2-8D II). Moreover, FZD4 alone also exhibited a higher correlation than FLK1+PDGFRA+ with the output CTNT abundance in the mock sorted sample (Figure 2-8D III). The other output populations showed relatively no correlation with input markers, which can be expected since these populations are sorted, essentially negating any positive or negative biases the input markers may have had. Taken together, these results indicate that at the CPC stage, FZD4 can be used either alone or in combination with FLK1 and PDGFRA to predict the number of CMs after 3 days of differentiation, and that higher FZD4 abundance at the CPC stage will lead to greater CM output.

2.3.5 Greater CM Output from FZD4+ CPCs is Also Observed during hPSC Differentiation

To determine whether FZD4 abundance patterns and signaling effects were conserved in the human system, hPSC-derived CPCs were generated in a similar BMP/ACTIVIN induced system (Yang et al., 2008b). In the human system, CPCs are also identified using the surface markers kinase insert domain receptor (KDR; analog to FLK1 in mouse) and PDGFRA. However, in contrast to mouse CPCs, the KDR-PDGFRA+ sub-population contains the CPCs. Co-staining with FZD4 reveals a similar segregation of the cardiogenic KDR-PDGFRA+ population into a FZD4+ and FZD4- sub-populations, consistent with what we have observed in the mouse system (Figure 2-9A). The FZD4+ and FZD4- fractions were sorted and re-seeded, and after 7 days, abundance of CTNT and the cell proliferation marker, Ki67, was analyzed using immunofluorescence

Figure 2-9: FZD4 enriches CPC in hPSC-derived cardiomyocytes. A) FZD4+ and FZD4- sub-populations were sorted from PDGFRA+ cardiac progenitor cells and differentiated into cardiomyocytes. The number in the box represents the percent positive value. B) Intracellular staining for CTNT (green), DAPI nuclear stain (blue), and Ki67 (purple) showed the FZD4+ compartment exhibited an increased number of cardiomyocytes as compared to the FZD4- population, scale bar represents 500 µm, and C) quantified using CellProfiler (n = 3 for each group, mean ± SEM). Statistical analysis was performed using a two-sided Mann-Whitney U test, * p<0.05.

(Figure 2-9B) and quantified using image analysis (Figure 2-9C). We observed a similar increase in the CTNT+ percentage in the PDGFRA+FZD4+ sub-population relative to the FZD4- sub- population. This trend is also seen in the Ki67+ percentage within the CMs, suggesting that the increase in CM frequency is due to increased proliferation. These results indicate that FZD4 enriches for CMs and suggests the increase in CMs is due to increased proliferation, which is consistent with our observations in the mouse system.

2.3.6 Canonical FZD4-NORRIN Signaling Enhances CM Output

We next examined whether the addition of ligands that bind FZD4 would have biological consequences on CM differentiation. We added ligands that bind with FZD4 (WNT3A, WNT5A, WNT7A, and NORRIN) to both mock sorted and sorted sub-populations at the CPC stage (d3.75) and measured the subsequent CTNT abundance after 3 days (Figure 2-10A). To determine optimal conditions for CM induction, individual dose-response curves were generated for each ligand (Figure 2-10B). The CM output was reported as a ratio relative to the base condition where no ligand was applied (Figure 2-11A). The ligands WNT3A, WNT5A, WNT7A showed no significant difference with the ligand-free base condition. The addition of NORRIN, on the other hand, led to an increase in CTNT abundance in both mock sorted and FLK1+PDGFRA+FZD4+ sub-populations. This data shows that the addition of the exogenous WNT ligand, NORRIN, further increases CTNT output in the FZD4+ cells relative to the absence of WNT stimulation, demonstrating a functional consequence for the abundance of FZD4 on CPCs. This provides evidence to suggest that NORRIN signals through the canonical WNT pathway to enhance cardiogenesis (Figure 2-11B). Given that our flow cytometry and qPCR results revealed a lack of Fzd4 expression at the CM stage compared to the CPC stage (Figure 2-5A), it is likely that FZD4- NORRIN signaling occurs during or immediately after the specification of FLK1+PDGFRA+ CPCs and acts to increase proliferation.

Figure 2-10: Exogenous addition of WNT ligands, related to Figure 2-11.

A) FZD4 ligands WNT3A, WNT5A, WNT7A, and NORRIN were added to activate the WNT signaling pathway. B) Dose response curves of each ligand with and without IWP2. Optimal dose was determined to be: NORRIN [100 ng/mL], WNT3A [10 ng/mL], WNT5A [10 ng/mL], and WNT7A [100 ng/mL].

Figure 2-11: Model of FZD4 abundance in the context of early cardiac differentiation. A) The addition of NORRIN further enhances CTNT response in both mock sorted and sorted FLK1+PDGFRA+FZD4+ populations (n = 3 for each group, mean ± SEM). Statistical analysis was performed using a two-sided Mann-Whitney U test, * p<0.05. B) Proposed model indicating the hierarchy of cells during differentiation and the relevant cell populations with their markers. 2.4 Discussion

In vitro analysis of LPM during differentiation has typically required the use of intracellular markers (such as Lmo2, Nkx2.5, and Isl1), which are not compatible for live cell sorting for further interrogation of enriched LPM sub-populations. Identification of membrane-expressed proteins that mark LPM would not only contribute to the functional analyses of in vitro mesoderm development, but also enable live cell sorting of cardiac progenitors which could then be enriched for further cell fate studies and the development of efficient cell manufacturing protocols. During embryogenesis, the surface protein FZD4 is expressed in LPM, from which cardiac mesoderm

48 develops, defining one of the earlier specification points in cardiogenesis (Tan et al., 2013a). We have shown evidence that receptor FZD4 can initiate the WNT signaling pathway, which has been shown in chick embryos to be required for LPM development during gastrulation (Sweetman et al., 2008b). FZD4 has also been shown to map to a chromosomal region important for cardiac development (DeRossi et al., 2000). Additionally, Abdul-Ghani et al. has demonstrated that blocking FZD4 results in reduced cardiac induction (Abdul-Ghani et al., 2011), further implicating the importance of FZD4 in cardiac development. Here, we report that abundance of the surface marker FZD4 distinguishes lateral plate from paraxial mesoderm and marks a pre-cardiac mesoderm population which can be used to more specifically select the CPC population during in vitro mPSC and hPSC differentiation.

While our results show that FZD4 abundance can be used to purify the CPC population and obtain subsequent CM enrichment, we also observed the presence of CM in the FZD4- sub-population (38.2 ± 8.3% CTNT+ percentage). One possible cause may be the regeneration of the FZD4- population during culture post-sort. While the purity of the sorted FZD4+/- populations is > 95% as measured by both flow cytometry (Figure 2-6B) and qPCR (Figure 2-6C), the FZD4- population may generate a secondary FZD4+ sub-population, albeit at a lower frequency, resulting in a lower CM population. In addition, it is also possible that the FZD4+ and FZD4- sub- populations are enriched for different types of progenitor cells which could result in different types of CMs. Finally, this may indicate that another marker in addition to FZD4 may be required to fully purify the CPC sub-population.

The FZD4 receptor is involved in WNT signaling which is a key pathway during cardiogenesis (Cohen et al., 2008b), consistent with our demonstration that NORRIN enhanced the CM output from our CPC sub-population by increasing CM proliferation. It is known that the kinetics and type (i.e. canonical versus non-canonical) of WNT activation is critical in guiding cardiac induction and differentiation (Sumi et al., 2008). Early cardiac induction relies on canonical signaling to induce primitive streak formation and gastrulation. Subsequent canonical signaling must be inhibited in order for cardiac specification to occur as the lateral plate mesoderm moves laterally across the embryo. Canonical WNT activation is again required in later stages of cardiac development in a proliferative capacity to expand the cardiogenic population (Gessert and Kühl, 2010a). In this study, we have shown that FZD4 is not only expressed on LPM, but also involved in increasing CM induction through increased proliferation. Unexpectedly, when our

49 differentiation cultures were presented with WNT ligands at d3 (the equivalent of gastrulation in the embryo), there were no differences observed in CM output. However, upon addition of NORRIN at the CPC stage, there was an increase in CM enrichment due to an increase in proliferation. In order to clearly distinguish between proliferation and selection mechanisms in CM development, further in-depth survival and proliferation studies are required. Additionally, a move towards network-based WNT activation studies (Moon and Gough, 2016) would provide further evidence of the intricate role WNT signaling plays during this stage of differentiation.

Cardiac development is fairly conserved across multiple organisms, especially between the mouse and human systems (Brade et al., 2006). We have demonstrated that FZD4-signaling promotes increased CM differentiation efficiency in the human system. During the CPC stage of hPSC differentiation, FZD4-NORRIN signaling increased CM output through increased proliferation. A previous study has also confirmed the increase in FZD4 abundance during the LPM stage in hPSCs and also implicated the role of FZD4 through the non-canonical WNT signaling pathway (Mazzotta et al., 2016). While our work indicates a canonical WNT signaling mechanism, both pathways may play a sequential role where immediately after progenitor specification canonical WNT signaling is required, then subsequent non-canonical signaling dominates to further promote CM specification. Further studies are required to tease out the intricate timing required for these signaling pathways to occur. In both cases, not only can FZD4 be used as a marker for enrichment analysis, but the FZD4 pathway can also be exploited to increase CM yield from hPSCs. Furthermore, several studies to develop clinical protocols for hPSC-derived CPC and CM therapies (Fernandes et al., 2015; Trounson and DeWitt, 2016) have highlighted the need for good quality markers to identify and isolate the cell types required at high purity. To that end, our findings suggest that FZD4 is a highly promising candidate marker that can provide increased CPC purity and be incorporated into the development of cardiac cell-based therapeutic applications.

Our sorting studies confirmed FZD4 can be used in conjunction with previously known CPC markers, FLK1 and PDGFRA, to further resolve the CPC population and increase CM output. In addition to FZD4, our MS analysis identified other markers based on similar abundance patterns which may also be of importance in the CPC stage of heart development. We already have initial validation of the abundance of three markers (LPAR4, ITGB5, and PLEXINB1) in terms of alignment with the proteomic data, and while further validation in terms of antibody specificity is required, these markers, including FZD4, can be used in combination to further resolve the

50 different sub-populations present in the CPC population. For example, LPAR4 has been shown to be present in all four chambers of the developing rat heart (Wang et al., 2012), which may indicate a similar function in mouse cardiac development. In addition, ITGB5 contributes to transforming growth factor β-mediated cell adhesion to extracellular matrix and cell movement in mouse and human epithelial cells (Bianchi-Smiraglia et al., 2013), which may be involved during ingression of the LPM. PLEXINB1 is largely implicated in mouse neuronal development (Fazzari et al., 2007), however it can still play a role as a negative marker in improving the purification of the CPC sub-population from the general heterogeneous population. Applying this process broadly, our analysis generated potential surface markers for additional stages of development (stem cell, epiblast, and primitive streak) and may provide further insights into the multiple cell types involved during the entire process of stem cell maintenance, CPC differentiation, and CM specification.

2.5 Conclusion

We have identified and characterized FZD4 as a marker for LPM and were able to utilize it as a marker, in conjunction with FLK1 and PDGFRA, to further purify CPCs and increase CM enrichment. Additionally, we have shown FZD4 is also expressed in the hPSC system and allows for a similar enrichment in CM. Finally, NORRIN can be presented to the FZD4 receptor to induce WNT signaling-mediated proliferation which results in an increase in CM output from CPCs. Taken together, these findings demonstrate a role for FZD4 in contributing to our understanding of the biology of mouse and human cardiac development.

2.6 Experimental Procedures

2.6.1 PSC Culture and Bioreactor Differentiation

The mouse PSC line (E14.1, 129/Ola) that expresses green fluorescent protein driven by Brachyury expression (Fehling, 2003) was generously provided by Dr. Gordon Keller. PSCs were maintained on 0.2% gelatin (Sigma) coated tissue culture polystyrene (Fischer) in Dulbecco's modified eagle medium/nutrient mixture F-12 (Thermo Fisher Scientific) and Neurobasal medium (Thermo Fisher Scientific) supplemented with 1X B-27 supplement (Thermo Fisher Scientific), 1X N-2 supplement (Thermo Fisher Scientific), 2 mM Glutamax (Thermo Fisher Scientific), 100 U/mL penicillin-streptomycin (Thermo Fisher Scientific), 0.05% bovine serum albumin (BSA;

Wisent), 1.5×10-4 M monothioglycerol (MTG; Sigma), 107 units/mL leukemia inhibitory factor (Millipore), and 10 ng/mL bone morphogenic protein 4 (BMP4; R&D).

For the generation of embryoid bodies (EBs), PSCs were dissociated into single cells with TrypLE Express Enzyme (1X) (Thermo Fisher Scientific) and plated at 750,000 cells per 10 mL in a 100 mm Petri dish (BD Biosciences) and rotated at 60 RPM on a shaker plate. The differentiation medium used consisted of Iscove's modified Dulbecco's medium (Thermo Fisher Scientific) and Ham's F-12 nutrient mix (Thermo Fisher Scientific) supplemented with 1X B-27 supplement minus ascorbic acid (Thermo Fisher Scientific), 1X N-2 supplement, 2 mM Glutamax, 100 U/mL penicillin-streptomycin, 0.05% BSA, 1.5×10-4 M MTG, and 0.5 mM ascorbic acid (Sigma).

At day (d) 2, EBs are harvested and dissociated into single cells using TrypLE (Thermo Fisher Scientific) and re-seeded into 100 mm Petri dishes (BD Biosciences) with differentiation medium further supplemented with 1 ng/mL BMP4, 2 ng/mL ACTIVIN A (R&D), and 3 ng/mL WNT3A (R&D). At d3, 2µM inhibitor of WNT production (IWP2; Reagents Direct) is added to each Petri dish. At d3.75, cells are dissociated into single cells using TrypLE or 1mM Ethylenediaminetetraacetic acid (EDTA; Sigma) and seeded onto 0.2% gelatin coated 96 well (Corning) or 384 well (Greiner) tissue culture plates at 200,000 cells and 50,000 cells respectively in StemPro-34 (Thermo Fisher Scientific), 2 mM Glutamax, 0.5 mM ascorbic acid, 150 µg/mL transferrin (Roche), 100 U/mL penicillin-streptomycin, and 2µM IWP2. At d7, cells are harvested and analyzed.

Cells were incubated in a humidified 5% (v/v) CO2 air environment at 37°C.

2.6.2 Flow Cytometry, Cell Sorting, and Immunocytochemistry

EBs and cardiac cultures were harvested and dissociated to single cells in 1mM EDTA or TrypLE and stained with the appropriate markers listed in Table 2-1. For cell surface markers, cells were stained in 0.5% fetal bovine serum (FBS; Gibco) in Hank’s balanced salt solution (Thermo Fisher Scientific) and 7-Aminoactinomycin D (7-AAD; Thermo Fisher Scientific) was used to denote live/dead cells. For intracellular proteins, cells were stained with LIVE/DEAD fixable near-IR dead cell stain kit for 633 or 635 nm excitation (Thermo Fisher Scientific) to denote live/dead cells, then fixed with 0.37% formaldehyde (Sigma) and permeabilized with 0.5% Saponin (Sigma). Cell proliferation assays were performed according to the manufacturer's protocol using the Click-

52 iT Plus EdU Alexa Fluor 647 Flow Cytometry Assay Kit (Thermo Fisher Scientific). Stained cells were analyzed using an LSRFortessa (BD Biosciences). A gating strategy to mark live/dead cells along with singlet cells was used to minimize noise in the output signal (Figure 2-7D).

Cells were sorted using the FACSAria II (BD Biosciences) or FACSAria III (BD Biosciences), and analyzed using LSRFortessa (BD Biosciences).

All flow cytometry data was analyzed using FlowJo software (vX.0.7; Treestar).

For immunocytochemistry, cells were fixed in 0.37% formaldehyde, and permeabilized using 0.1% Triton X-100 (Sigma) in phosphate buffer saline (PBS; Thermo Fisher Scientific).

2.6.3 Microscopy and Image Analysis

Brightfield images were captured using Olympus inverted microscope (Olympus) using cellSens software (Olympus). High throughput image capture of fluorescent images were acquired at 10X and stitched together using Cellomics Arrayscan (Thermo Fisher Scientific). Image analysis was performed using Cell Profiler (Carpenter et al., 2006).

2.6.4 Quantitative PCR Analysis

Total RNA was extracted from samples using RNeasy mini or micro kits (Qiagen) and cDNA was generated using Superscript III reverse transcriptase (Thermo Fisher Scientific) according to the manufacturer’s protocols. cDNA was mixed with primers and SYBR Green Master Mix (Roche) and run on QuantStudio real-time PCR software (Applied Biosystems). Relative quantitative expression of genes was determined by the delta–delta cycle threshold method with GAPDH as an internal reference control. Primer sequences are listed in Table 2-2.

2.6.5 Microarray Analysis

Samples of mRNA were submitted to The Centre for Applied Genomics, and Affymetrix Mouse Exon 1.0 ST arrays were used for the hybridization. Arrays were scanned and the expression data was obtained in the form of .CEL files. Data was analyzed using R (v3.3.1) Bioconductor suite to extract, normalize, and summarize gene expression data.

2.6.6 Cell Surface Capture (CSC)

CSC was performed as described previously (Wollscheid et al., 2009b). In brief, cells or EBs were dissociated using 1 mM EDTA (Thermo Fisher Scientific) and 10% FBS (Hyclone). Exposed extracellular aldehydes were then oxidized with 1.6 mM sodium meta-periodate (Thermo Fisher Scientific) and reacted for 1 hour with 5 mM biocytin hydrazide (Biotium). The cells were then washed and lysed by sonication in a hypotonic lysis buffer, and the nuclei were pelleted by centrifugation. The supernatant containing the membranes and the cytoplasm were subjected to ultra-centrifugation. The microsomal pellet was collected and solubilized by addition of 0.1% RapiGest (Waters) and sonication. After overnight trypsin (Thermo Fisher Scientific) digestion, the biontinylated peptides were coupled to streptavidin beads (Thermo Fisher Scientific), thoroughly washed, and enzymatically released by PNGaseF (NEB). Peptides were then cleaned- up over C18-tips (The NestGroup) and subjected to liquid chromatography tandem mass spectrometry (LC-MS/MS).

2.6.7 Mass Spectrometry (MS) Analysis

Samples were analyzed on an EASY-nLC 1000 nano liquid chromatography system (Thermo Fisher Scientific) connected to a Orbitrap Velos (Thermo Fisher Scientific) mass spectrometer, which was equipped with a nanoelectrospray ion source (Thermo Fisher Scientific). Peptide separation was carried out on an RP-HPLC column (75 µm × 10 cm) packed in-house with C18 resin (Magic C18 AQ 3 µm, Michrom BioResources) using a linear gradient from 90% solvent A (water, 0.2% formic acid, 1% acetonitrile) and 10% solvent B (water, 0.2% formic acid, 80% acetonitrile) to 65% solvent A and 35% solvent B for 60 minutes at a flow rate of 0.2 µL/minute. The data acquisition mode was set to acquire one high-resolution first stage scan (ms1) in the ion cyclotron resonance cell followed by 10 collision induced dissociation mass scans in the linear ion trap. For a high-resolution mass scan, 106 ions were accumulated over a maximum time of 500 ms, and the full width at half maximum resolution was set to 60,000 (at m/z 300). Only mass signals exceeding 500 ion counts triggered a second stage (ms2) attempt, and 104 ions were acquired for an ms2 scan over a maximum time of 250 ms. The normalized collision energy was set to 35% and one microscan was acquired for each spectrum. Singly charged ions were excluded from triggering ms2 scans.

All acquired mass spectra were searched against the International Protein Index database (Version 3.26) using the SEQUEST algorithm (Eng et al., 1994b). The SEQUEST database search criteria included: 0.2 Da mass tolerance for the precursor ion, 0.5 Da mass tolerance for the fragment ions, variable modifications of 0.984016 Da for asparagines (representing formerly N-glycosylated asparagines after deamidation through the PNGaseF treatment) and 15.994915 Da for methionines (covering rapidly oxidizing methionines), carbamidomethylation as static modification for cysteines, at least one tryptic terminus, and two missed cleavage sites. Statistical analysis of the data, including peptide and protein identification was performed using the Trans-Proteomic Pipeline v4.3 (TPP; Seattle Proteome Center) including PeptideProphet and ProteinProphet (Keller et al., 2005). Peptides and proteins were detected and quantified at a confidence score of >1.3 using the TPP in combination with Progenesis software. The ProteinProphet probability score was set such that the false discovery rate (FDR) was less than 1% as determined by ProteinProphet.

2.6.8 Label-Free Quantification

Protein quantification was performed using Progenesis (Nonlinear Dynamics). After manually improving the alignment, peptides were quantified based on ms1 intensity, and filtered for sequences that have a UniProt accession number. Overall protein abundance was estimated using a 10% trimmed mean of all identified peptide abundances. Significance of differential expression in protein abundance was assessed by ANOVA.

2.6.9 Technical Validation of Candidate Proteins

From literature, identified proteins consisted of a mixture of mesoderm, endoderm, and neural related markers in addition to previously unannotated markers relevant to the CPC cell system (Table 2-3). Next, temporal mRNA expression profiles were used to validate profiles generated using mass spectrometry. Samples of mRNA were collected at each stage of cardiac development (outlined in Figure 2-1A) for qRT-PCR analysis of gene expression for the 47 proteins associated with CPC. The Pearson correlation coefficient was calculated from the normalized mRNA expression and protein expression values (Figure 2-4A). 32 proteins with a significant Pearson correlation coefficient (p<0.05) were selected for antibody validation using flow cytometry. 14 antibodies that were available and rated for flow cytometry were first titrated against their reported positive control cell type. Those that displayed a positive signal were then used in a differentiation time course (d0 – d3.75). The antibodies that displayed 0% or 100% positive staining across all

55 time points (indicative of non-specific binding) were discarded. The final panel of 10 candidate proteins (ANTXR1, CXCR4, EPHB2, FZD4, ITGA4, ITGB5, LPAR4, NCAM1, NT5E, AND PLEXINB1) were selected based on antibody availability and differential expression (Figure 2-4B). These markers were then co-stained with FLK1 and PDGFRA to determine co-expression and their capacity to further resolve the FLK1+PDGFRA+ population. Using supervised gating, we found that the FLK1+PDGFRA+ population could be separated based on four markers (ITGB5, LPAR4, FZD4, and PLEXINB1) (Figure 2-4C). After the selection of FZD4 as a candidate marker, further validation was performed to ensure optimal antibody binding (Figure 2-12A) using a staining index metric (Telford et al., 2009) and measured non-specific background abundance using an isotype control (Figure 2-12B).

2.6.10 Statistical Analysis

All statistical tests were performed in R (v3.3.1) using a two-sided Mann-Whitney U test or Kruskal-Wallis one-way analysis of variance with significance level α = 0.05.

Figure 2-12: FZD4 antibody titration and negative isotype control, related to Figure 3.

A) FZD4 antibody titration using staining index to determine optimal antibody dilution. B) Staining controls using negative isotype to determine FZD4 positive gating strategy. The number in the box represents the percent positive value.

2.6.11 Accession Numbers

The microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus (Edgar et al., 2002) (GSE103560). The proteomic data have been deposited in ProteomeCentral (PXD007684).

2.7 Supplemental Materials

2.7.1 Supplemental Tables

Table 2-1: Antibodies used in this study. Antibody Name Vendor Alexa Fluor® 488 goat anti-Rat IgG (H+L) ThermoFischer Scientific Alexa Fluor® 647 goat anti-Rat IgG (H+L) ThermoFischer Scientific Anti-Integrin beta 5 antibody Abcam Anti-Mouse CD140a (PDGF Receptor a) Biotin eBioscience Anti-Mouse CD326 (EpCAM) APC eBioscience Anti-P2Y9 (LPAR4) antibody Abcam Anti-Plexin B1 antibody Abcam APC Anti-Mouse CD24 BD Biosciences APC Rat anti-Mouse CXCR4 (CD184) BD Biosciences Brilliant Violet 421 Anti-Mouse CD40 BD Biosciences DyLight™ 405 Goat anti-rat IgG Biolegend FZD4-Fab TRAC Human ACTIVIN RIA Affinity Purified Polyclonal Ab, Goat IgG R&D Systems Human TEM8/ANTXR1 Affinity Purified Polyclonal Ab, Goat IgG R&D Systems Human/Mouse DCBLD2/ESDN Affinity Purified Polyclonal Ab, Sheep IgG R&D Systems Human/Mouse EphB2 Phycoerythrin MAb (Clone 512012), Rat IgG2A R&D Systems Human/Mouse Frizzled-4 Antibody R&D Systems Human/Mouse SOX2 Antibody R&D Systems PE Rat Anti-Mouse Flk-1 BD Biosciences PE Rat Anti-Mouse NT5E (CD73) BD Biosciences Purified Mouse Anti-BMPR-II BD Biosciences Purified Mouse Anti-Oct-3/4 BD Biosciences Purified Mouse Anti-R-Cadherin BD Biosciences Purified Rat Anti-Mouse ITGA4 (CD49d) BD Biosciences Purified Rat Anti-Mouse NCAM-1 (CD56) BD Biosciences Purified Rat Anti-Mouse SIRPA (CD172a) BD Biosciences Troponin T, Cardiac Isoform Ab-1, Mouse Monoclonal Antibody ThermoFischer Scientific

Table 2-2: qPCR primers used in this study. Gene Forward Reverse α-mhc GCCCAGTACCTCCGAAAGTC GCCTTAACATACTCCTCCTTGTC Acvr1 GTGGAAGATTACAAGCCACCA GGGTCTGAGAACCATCTGTTAGG Acvr2b ACCCCCAGGTGTACTTCTG CATGGCCGTAGGGAGGTTTC Alpl CCAACTCTTTTGTGCCAGAGA GGCTACATTGGTGTTGAGCTTTT Anpep ATGGAAGGAGGCGTCAAGAAA CGGATAGGGCTTGGACTCTTT Antxr1 TGGACAAGTCAGGAAGTGTGC TGATGAATCTATGAGCCAACTGC Apc CTTGTGGCCCAGTTAAAATCTGA CGCTTTTGAGGGTTGATTCCT Asah1 CGTGGACAGAAGATTGCAGAA TGGTGCCTTTTGAGCCAATAAT Atp1b2 GGCAGGTGGTTGAGGAGTG GGGGTATGGTCAGAGACGGT Axin1 CTCCAAGCAGAGGACAAAATCA GGATGGGTTCCCCACAGAAATA Axin2 TGACTCTCCTTCCAGATCCCA TGCCCACACTAGGCTGACA B-Catenin ATGGAGCCGGACAGAAAAGC CTTGCCACTCAGGGAAGGA Bmpr2 TTGGGATAGGTGAGAGTCGAAT TGTTTCACAAGATTGATGTCCCC B-Tubb CACCTGCAAGCCGGTCAAT TCCCCATGATAGGTCCCAGTG Cacna2d1 GTCACACTGGATTTTCTCGATGC GGGTTTCTGAATATCTGGCCTGA Cadm1 CAGCCTGTGATGGTAACTTGG AGGAGGGATAGTTGTGGGGG Cd97 CTCCCCGAGCAGACAACTAC CAATGGTTTTGCCCGGAGAT Cdh4 CAGGCCACTGACATGGAAGG ATGATTCGGTAGACGGCGTTC

Cdh6 CAGCCCTACCCAACTTTCTCA GAACGGCTCAGCTCATTCC Celsr1 TCGCTGACTTCGGTGCTTG TTACCAGCTCTACCCAAACGG C-myc ATGCCCCTCAACGTGAACTTC CGCAACATAGGATGGAGAGCA Cnnm2 AAGTGGCCCACCGTGAAAG CGCTTCTACTTCTGTTGCTAGG Cnnm4 CTGCACATCCTTCTCGTTATGG TGCGAGCATACTTTCTCTCCTT Ctnt CAGAGGAGGCCAACGTAGAAG CTCCATCGGGGATCTTGGGT Cyclind1 GCGTACCCTGACACCAATCTC CTCCTCTTCGCACTTCTGCTC Dcbld2 ACACACTGTACTAGGCCCTGA CGTCCTGACTCGAATCTCCCA Dvl1 ATGGCGGAGACCAAAATCATC AACTTGGCATTGTCATCGAAGA Dvl2 GGTGTAGGCGAGACGAAGG GCTGCAAAACGCTCTTGAAATC Efna5 ACACGTCCAAAGGGTTCAAGA GTACGGTGTCATTTGTTGGTCT Enpp3 CAGAGGAGCCCATTAAGAAAGAC GTGCGATGAGTCAAAGCATTTT Epcam GCGGCTCAGAGAGACTGTG CCAAGCATTTAGACGCCAGTTT Ephb2 GCGGCTACGACGAGAACAT GGCTAAGTCAAAATCAGCCTCA Erbb2 GAGACAGAGCTAAGGAAGCTGA ACGGGGATTTTCACGTTCTCC Fam38b AATCAAACCAACATTCCCCTTCA CAGGTAGACGAGCAAAGGAGA Flk1 TTTGGCAAATACAACCCTTCAGA GCAGAAGATACTGTCACCACC Fzd10 CATGCCCAACCTGATGGGTC GCCACCTGAATTTGAACTGCTC Fzd4 TGCCAGAACCTCGGCTACA ATGAGCGGCGTGAAAGTTGT Gapdh AGGTCGGTGTGAACGGATTTG TGTAGACCATGTAGTTGAGGTCA Gsk3b TGGCAGCAAGGTAACCACAG CGGTTCTTAAATCGCTTGTCCTG Il17ra AGTGTTTCCTCTACCCAGCAC GAAAACCGCCACCGCTTAC Il17rd AACAGCGGACTGCACAACAT GCAAGCGTACTGGCTGATG Isl1 ATGATGGTGGTTTACAGGCTAAC TCGATGCTACTTCACTGCCAG Itga4 GATGCTGTTGTTGTACTTCGGG ACCACTGAGGCATTAGAGAGC Itgb5 GCTGCTGTCTGCAAGGAGAA AAGCAAGGCAAGCGATGGA Lef1 TGTTTATCCCATCACGGGTGG CATGGAAGTGTCGCCTGACAG Lifr AGCTCTGACCCTCCTGCAT TGGGTGACAAGAATGGAACCT Lmo2 ATGTCCTCGGCCATCGAAAG CGGTCCCCTATGTTCTGCTG Lpar4 AGTGCCTCCCTGTTTGTCTTC GCCAGTGGCGATTAAAGTTGTAA Lrp5 AAGGGTGCTGTGTACTGGAC AGAAGAGAACCTTACGGGACG Lrp6 TTGTTGCTTTATGCAAACAGACG GTTCGTTTAATGGCTTCTTCGC Mapk8 AGCAGAAGCAAACGTGACAAC GCTGCACACACTATTCCTTGAG Meox1 GAAACCCCCACTCAGAAGATAGC TCGTTGAAGATTCGCTCAGTC Ncam1 AGCGCAGGTGCAGTTTGAT ACAAAGAGCTTTTACGGACTGG Ndp GCATCCATTTCTATGCTCTCCC GGTGTCTCATGCAGCGTTG Nfatc1 GACCCGGAGTTCGACTTCG TGACACTAGGGGACACATAACTG Nkx2.5 GACAAAGCCGAGACGGATGG CTGTCGCTTGCACTTGTAGC Nptn CGCTGCTCAGAACGAACCAA GCTGGAAGTGAGGTTACACTG Nt5e GGACATTTGACCTCGTCCAAT GGGCACTCGACACTTGGTG Odz3 CGGGAAAAGGAAAGGCGCTAT CTTCGAGTTGCGGATTCACAC Pax1 CCGCCTACGAATCGTGGAG CCCGCAGTTGCCTACTGATG Pdgfra ACACGTTTGAGCTGTCAACC CCCGACCACACAAGAACAGG Pdgfrb TTCCAGGAGTGATACCAGCTT AGGGGGCGTGATGACTAGG Pecam1 CTGCCAGTCCGAAAATGGAAC CTTCATCCACCGGGGCTATC Plxnb1 CACACATCTACTACACTTGGCAA CAATCCCGGCTGTCATTCAC Prickle1 ACCTGGAGTATGCTGGCAC CACAGTGGATTTTTCCATCCTGA Ptgfrn CCCTGCAATGTCAGCGACTAT CGTTGGCAGTTCTCCTCAACA Robo1 GAGCCTGCTCACTTTTACCTC GGTCTGAAGGGTGTTCAACAAT Sdk2 GTGACCAAGTGGCAGTCTCC GTTGCTCAGGATGGGCTAAGG Sirpa CACGGGGACAGAAGTGAAGG TGCAGTTGAGAATGGTCGAATC Slc29a1 CAGCCTCAGGACAGGTATAAGG GTTTGTGAAATACTTGGTTGCGG Slc39a6 GTCACACGGTTGCTGGTAAAA GGGCGAGATCCTTTCCCTAGA Slco3a1 AGGTGTCCTGCTTCTCCAAC GTCAACACGCTCACCAGGTAG Stim1 GGCGTGGAAATCATCAGAAGT TCAGTACAGTCCCTGTCATGG

Tbx6 ATGTACCATCCACGAGAGTTGT GGTAGCGGTAACCCTCTGTC Tcf15 GGGCAGCTGCTTGAAAGTGA CTCCGGTCCTTACACAACGC Tcf7 AGCTTTCTCCACTCTACGAACA AATCCAGAGAGATCGGGGGTC Tgfbr3 GGTGTGAACTGTCACCGATCA GTTTAGGATGTGAACCTCCCTTG Thsd7a AGGTGCCCACCCTCTATCTG TGTATGTAACGTAGTCCAGCCT Tmem132c TCAGAGCCGAGACTGCATTCT GCCCATAGCTGACGTTTAATACC Tmem87a TGGCATGGAAGGAGTCCTCA GAGAGGGCCAGGCTTACTATC Tmtc4 TCCCAAGTACGTTCATGCCAT GTTTTAGGTGACGGGAAACTGG Vangl2 ACTCGGGCTATTCCTACAAGT TGATTTATCTCCACGACTCCCAT Vcam1 AGTTGGGGATTCGGTTGTTCT CCCCTCATTCCTTACCACCC Wnt11 GCTGGCACTGTCCAAGACTC CTCCCGTGTACCTCTCTCCA Wnt5a CAACTGGCAGGACTTTCTCAA CATCTCCGATGCCGGAACT Wnt5b CTGCTGACTGACGCCAACT CCTGATACAACTGACACAGCTTT Wnt7a CCTTGTTGCGCTTGTTCTCC GGCGGGGCAATCCACATAG

Table 2-3: Candidate proteins identified by mass spectrometry and microarray, related to Figure 2-2.

List of 47 proteins identified to be uniquely expressed in the FLK1+PDGFRA+ sub-population of cardiac progenitors. Classification was done based on literature search. FLK1+PDGFRA+ Description Classification ACVR1 Activin a receptor, type 1 Unannotated ACVR2B Activin a receptor, type 2b Muscle ALPL Alkaline phosphatase, liver/bone/kidney Non-specific ANPEP Alanyl (membrane) aminopeptidase Cardiac/Blood ALPL Anthrax toxin receptor 1 Endothelial ANPEP Atpase, na+/k+ transporting, beta 2 polypeptide Unannotated ANTXR1 Anthrax toxin receptor 1 Endothelial ASAH1 Calcium channel, voltage-dependent, alpha2/delta subunit 1 Unannotated ATP1B2 Cell adhesion molecule 1 Unannotated CACNA2D1 Cd97 antigen Unannotated CADM1 Cadherin 4 Unannotated CD97 Cadherin 6 Neural CDH4 Cadherin, egf lag seven-pass g-type receptor 1 Neural CDH6 Cyclin m2 Unannotated CELSR1 Cyclin m4 Unannotated DCBLD2 Discoidin, cub and lccl domain containing 2 Neural EFNA5 Ephrin a5 Neural ENPP3 Ectonucleotide pyrophosphatase/phosphodiesterase 3 Unannotated EPCAM Epithelial cell adhesion molecule Unannotated EPHB2 Eph receptor b2 Unannotated FZD10 Frizzled 10 Neural FZD4 Frizzled 4 Cardiac/Blood IL17RA Interleukin 17 receptor a Unannotated IL17RD Interleukin 17 receptor d Non-specific ITGB5 Integrin beta 5 Non-specific KDR Kinase insert domain Cardiac/Blood LIFR Leukemia inhibitory factor receptor Unannotated LPAR4 Lysophosphatidic acid receptor 4 Cardiac/Blood NCAM1 Neural cell adhesion molecule 1 Cardiac/Blood NPTN Neuroplastin Unannotated NT5E 5' nucleotidase, ecto Cardiac/Blood PDGFRA Platelet derived growth factor alpha Cardiac/Blood PDGFRB Platelet derived growth factor receptor, beta polypeptide Cardiac/Blood

PECAM1 Platelet/endothelial cell adhesion molecule 1 Endothelial PLXNB1 Plexin b1 Unannotated PTGFRN Prostaglandin f2 receptor negative regulator Unannotated SIRPA Signal-regulatory protein alpha Cardiac/Blood SLC29A1 Solute carrier family 29 (nucleoside transporters), member 1 Unannotated SLC39A6 Solute carrier family 39 (metal ion transporter), member 6 Unannotated SLCO3A1 Solute carrier organic anion transporter family, member 3a1 Unannotated STIM1 Similar to stromal interaction molecule 1; stromal interaction molecule 1 Non-specific TGFBR3 Transforming growth factor beta receptor 3 Muscle THSD7A Thrombospondin, type I, domain containing 7a Cardiac/Blood TMEM87A Transmembrane protein 87a Unannotated TMTC4 Transmembrane and tetratricopeptide repeat containing 4 Unannotated VCAM1 Vascular cell adhesion molecule 1 Unannotated WNT5B Wingless-related mmtv integration site 5b Cardiac/Blood

Table 2-4: Protein Quantification and Clustering, related to Figure 2-1.

Table of protein intensities determined by mass spectrometry and clustered using k-means clustering. Please see Appendices

Appendix I for complete table. 2.8 Acknowledgments

The author is grateful to all co-authors and members of the P.W.Z lab for helpful feedback on the manuscript. He thanks the following labs for technical support: A.E. (mass spectrometry), B.W. (cell surface capture), and G.K. (mouse pluripotent stem cell differentiation). This research was funded by the Heart & Stroke Foundation, CIHR, and Medicine by Design. The authors declare there is no conflict of interest.

DISCUSSION

In this thesis I present findings describing the identification of surface markers on mPSC-derived cardiac progenitors and the characterization of this progenitor sub-population in the context of cardiac differentiation. Specifically, I examined the expression of surface markers on CPCs using an integrated mass spectrometry and microarray-based approach. Subsequent bioinformatics analysis of the genome and surface proteome of cardiac progenitors generated from the stage- specific differentiation of mPSCs, yielded a list of candidate surface markers, which were phenotypically validated using qPCR and flow cytometry, and through differentiation assays with sorted cell populations. After filtration based on protein-gene correlation, antibody availability, and literature review, I focused on the FZD4 receptor, a member of the WNT signaling pathway, as a new surface marker for LPM. I then utilized FZD4 as a marker, in conjunction with FLK1 and PDGFRA, to demonstrate an increase in CPC purity and a subsequent increase in CM enrichment. In terms of mechanistic studies, I showed that NORRIN can be presented to the FZD4 receptor to induce WNT signaling-mediated proliferation which resulted in an increase in CM output from CPCs. This demonstrates the value in knowing the set of surface markers present on a cell at a specific stage of development and the potential to leverage that knowledge into more efficient differentiation protocols. Additionally, I have shown that FZD4 is expressed in the hPSC system and allows for a similar enrichment in CM. This finding highlights the utility of FZD4 in identifying and sorting for CPCs that can potentially contribute to human cell transplantation studies. Finally, this study has greatly increased the potential pool of surface markers present not only on CPCs, but also on mPCSs and each stage of development in between. Moreover, the transcriptomic and proteomic data gathered can be further mined for additional markers and insights. Once validated and characterized, these markers can also contribute to our understanding of the biology of mouse and human development.

The results presented in this thesis also raises a number of questions, specifically what is the composition of the output CMs in both FZD4+/- sub-populations, and what is the extent of involvement of FZD4-NORRIN signaling during early LPM generation and CPC specification. As an initial approach, the composition of the CM generated can be further interrogated through the application of a more detailed panel of makers which can identify specific cardiac phenotypes. Additional validation through in vivo staining would further bolster evidence of the role FZD4 plays during early cardiac differentiation. In vivo data would provide spatial localization information of FZD4 and NORRIN during development and not only validate marker expression

63 but lend insight into the function of FZD4 at specific stages. Moreover, studies designed to increase the resolution of the effects of FZD4-NORRIN signaling during this dynamic transition stage would help address how FZD4 signaling adds to the current understanding of WNT signaling in the field. Furthermore, the characterization of CPCs brings up the question of applications towards cell therapeutics. Identification of FZD4 and other markers may contribute to the current discussion regarding appropriate cell types for cell transplantation in heart disease. Finally, the datasets provided in this thesis can be analyzed for additional markers, alternate splice variants, and signaling analysis at the network level. These questions are discussed in more depth within their relevant sections below.

3.1 Evaluation of Output Lineages from FZD4+ and FZD4- Cells

The data generated in this thesis, briefly discussed in Section 2.4, showed CM differentiation in both FZD4+ and (at a lower frequency) FZD4- sub-populations for both mouse and human systems (Figure 2-8C, Figure 2-9C). Initial discussion centered around a potential temporal effect on the kinetics of FZD4 which would then result in a delayed expression of FZD4 in the FZD4- culture, subsequently generating a secondary FZD4+ sub-population post-sort. An alternate interpretation would be that FZD4 is also delineating between different cardiac precursors, resulting in different CM sub-types being generated. To further investigate this phenomenon a broader panel of markers to characterize CMs on a number of axes, such as atrial/ventricular, left/right, FHF/SHF, and even on a functional level using electrophysiology, would need to be applied.

In terms of some of the markers that are differentially expressed between atrial and ventricular CMs, hairy-related transcription factors (HRTs) are transcriptional repressors and are involved in the formation of the atrioventricular boundary of the heart (Kokubo et al., 2005, 2007), where HRT1 expression is specific to the atria, while HRT2 is localized in the ventricles (Fischer and Gessler, 2003; Nakagawa et al., 1999). Iroquois homeobox gene (IRX) 4 plays an important role in establishing chamber-specific expression in the developing heart by upregulating ventricle- myosin heavy chain (VMHC) 1 while downregulating atrial myosin heavy chain (AMHC) 1 (Bao et al., 1999). Structural proteins are directly associated with contractile properties and are also divided between atrial and ventricular CMs. Myosin light chain 2 (MLC-2) is a structural protein involved in sarcomere formation and is important for the contractility of CMs, where MLC-2a is restricted to the atrium (Kubalak et al., 1994) while MLC-2v expression is present only in the

64 ventricles (O’Brien et al., 1993). Functionally, atrial and ventricular CMs, along with nodal-like cells, can also be differentiated using electrophysiological measurements which characterize the contractions and action potentials of beating CMs (He et al., 2003; Maltsev et al., 1993). To name a few, atrial CMs have lower and shorter action potentials (Giles and Imaizumi, 1988) and have a slightly less negative resting membrane potential than ventricular CMs (Schram et al., 2002). Left/right symmetry can also be distinguished using heart and neural crest derivatives expressed (HAND) 1 and HAND2, which are basic helix-loop-helix (bHLH) transcription factors that are expressed during heart development (Srivastava et al., 1995). While both HAND1 and HAND2 are expressed in the linear heart tube, HAND1 is expressed in cells destined for the left ventricle (Biben and Harvey, 1997) while HAND2 is high expressed in the developing right ventricle (Thomas et al., 1998). A marker for the FHF is TBX5, which appears early during development (Bruneau et al., 1999), and HCN4 (Später et al., 2013). In contrast, ISL1 is a LIM homeodomain transcription factor, and is primarily expressed in the SHF and gives rise to the majority of the cells in the heart (Cai et al., 2003). The SHF can also be marked by TBX1 which is required for OFT development (Jerome and Papaioannou, 2001) and its two downstream activated ligand FGF8 and FGF10 (Hu et al., 2004; Vitelli et al., 2002).

While the markers listed above are not an exhaustive list of identifying factors differentiating the CM sub-types, these marker panels can identify the overall proportion of each sub-type for each population and distinctions can be made in terms of which type of CM is being produced by each sub-population. Additionally, these panels can be standardized and used to quickly screen and assess lineage restrictions for the other candidate markers identified in this thesis. It is interesting to note that the CMs generated by both FZD4+/- sub-populations are likely ventricular as the induction of atrial CMs requires retinoic acid signaling, which was not present within the differentiating culture conditions (Lee et al., 2017). Furthermore, based on the evidence of proliferation linked with b-catenin mediated WNT signaling, it is likely that the CMs derived from the FZD4+ sub-population are likely part of the SHF as the overexpression of b-catenin results in an expansion of SHF-derived tissues due to proliferative effects (Ai et al., 2007; Kwon et al., 2007). Further interrogation of the resulting CM population would yield insights into the stage of development these cells represent in addition to the purity and ultimate utility of the generated CMs.

An additional consideration would be the relatively low dynamic range of CM induction presented in both the mouse and human system. Updates and improvements to CM differentiation protocols have raised the CM induction percentage to 60-90% using directed differentiation from biomolecules gleaned from in vivo studies or small molecule equivalents (Burridge et al., 2011; Kattman et al., 2011; Lian et al., 2013; Ren et al., 2011; Yuasa et al., 2005). Optimization of the culture conditions to improve the baseline CM induction may amplify the differences between the FZD4+ and FZD4- sub-populations and result in a more conclusive interpretation. Moreover, expansion of the above panel of CM markers to include non-CM cell types, such as endothelial and fibroblast lineages, would develop a more comprehensive analysis of the output cell composition for each sub-population. Having a broad range of identification markers spanning multiple linages enables further studies regarding the clonality of the isolated FZD4+/- sub- populations and can determine the single-cell modality of a FZD4+/- cell as have been done with other cardiac markers such as FLK1 and ISL1 (Kattman et al., 2006; Moretti et al., 2006). Clonal analysis can identify true progenitors at the single-cell level and contribute to our understanding of the role and function of the progenitor cell in a hierarchal context during development.

3.2 In vivo Validation of FZD4

In addition to the functional outputs of FZD4 expressing cells, identification of spatial expression patterns of FZD4 in the mouse embryo in vivo is important for further validation of the marker and potential insight into the molecular mechanisms during early cardiac development. Confirmation of the spatial and temporal expression patterns of FZD4 and its associated ligand NORRIN, will not only offer further support for the marker identification methods used in this thesis, but will also provide insights into the function of FZD4-NORRIN binding during early CPC development. These results would also provide strong evidence for the credibility of the other candidate markers identified in this study and provide support for their individual validation as well.

Previous studies have looked at FZD4 expression patterns in the chick embryo. Paxton et al. have investigated differential gene expression using microarrays and further validated potential candidates with in situ hybridization (Paxton et al., 2010). The authors have found that FZD4 was expressed in the bilateral mesodermal patches, and interestingly have also observed NORRIN expression in the primitive streak, which may suggest a potential role for NORRIN signaling in early cardiac mesoderm development. Unfortunately, in the mouse, similar studies are unavailable

66 for the time frame and regions of interest relevant to early cardiac development. Databases that are repositories for in situ spatial gene expression data such as eMouseAtlas (Richardson et al., 2014) and Mouse Genome Informatics (Ringwald et al., 2001), only show results for Fzd4 staining either early stage zygotes, or after late stage E9.0 embryos when the heart is predominantly fully formed. However, recent efforts have been made by Peng et al. to create a digital reconstruction of the mouse embryo in mid-gastrulation (E7.0) (Peng et al., 2016). The authors sectioned the embryo into slices, then using laser capture microdissection, groups of cells in specific spatial regions were collected for RNA-seq. The data from each cluster of cells were then be grouped and rendered into a 3D reconstruction of the gastrulating embryo to generate an in silico spatial transcriptomic model (iTranscriptome). Referencing the online database, Fzd4 gene expression was determined to be present in the lateral plate region of the embryo, providing further support for the utility and function of FZD4 during LPM specification.

To provide supporting evidence at the protein level, our strategy involved obtaining mouse embryos at E6.5, E7.0, and E7.5 to capture gastrulation at the mid-late primitive streak, emerging LPM, and subsequent cardiac crescent formation. We would use BRY as a spatial reference point and co-stain with FZD4 to investigate the spatial and temporal patterning of FZD4 in vivo. We initially attempted to stain wholemount embryos in collaboration with Dr. Sevan Hopyan’s lab, however, we had trouble obtaining clear images due to permeabilization of the embryo and antibody specificity. E6.5 and E7.5 mouse embryos were obtained and stained for FZD4, FLK1, and DAPI (Figure 3-1). While DAPI clearly labeled all the nuclei in both E6.5 and E7.5 sections, FZD4 and FLK1 labeled cells non-specifically in E6.5 as both marker expression was not expected in the headfold region of the embryo. In E7.5, both antibodies did not seem to penetrate the dorsal region as evidenced by the thick band of staining along the outer rim of the embryo indicating poor antibody penetration into the center of the embryo.

In order to address the potential permeabilization issue, we looked towards generating embryo sections via cry section and re-create the embryo through digital reconstruction of the resultant images. To facilitate high-throughput imaging of embryo sections, we combined standard cyrosection protocols using glass slides, and 96 well plate form factors to create a custom 96 well plate, (Figure 3-2A-B). Embryos were obtained and flash frozen in mounting medium at -80°C. Frozen embryos were then sectioned transversally along the proximal-dorsal axis. Each section was 10um in thickness and was deposited on a glass slide with gridlines marked with spacings

67 appropriate for a standard 96-well plate. Glass slides were allowed to thoroughly dry overnight at room temperature. Epoxy resin was carefully coated on the bottom of a bottomless 96-well plate and glass slides were inverted and adhered to the plate and cured overnight at room temperature. Each well containing a section was fixed with formaldehyde before continuing on to a standard immunocytochemistry protocol. High throughput fluorescent images were acquired using the Cellomics Arrayscan platform (Thermo Fisher Scientific).

The custom plate improved the efficiency of staining as well as enabling the usage of high- throughput imaging instruments. As a control to measure both efficacy of permeabilization and feasibility of the custom 96 well plate, we tested the protocol on in-house cardiac microtissues as representative of embryos, (Figure 3-2C). Both representative images taken at 10X and 20X respectively showed clear DAPI and CTNT stain which not only indicates the successful

Figure 3-1: Preliminary attempts at whole mount embryo staining.

E6.5 and E7.5 embryos were stained with FZD4 (green), FLK1 (red), and DAPI (cyan) and a representative sample of each optical section was taken. A: anterior, P: posterior, D: dorsal, Pr: proximal, scale bars represent 1mm.

Figure 3-2: High-throughput imaging of embryo sections and controls.

A) Schematic showing glass slides with embryo sections being glued underneath a bottomless 96-well plate. B) Representative images of the glass slides and fully constructed plate. C) Cardiac microtissues were obtained and sectioned, staining for DAPI (blue) and cardiac troponin T (CTNT). D) E7.0 embryos were sectioned and stained with FZD4 (grey), BRY (green), and DAPI (blue) and a representative sample of each optical section was taken. A: anterior, P: posterior, scale bars represent 1mm.

69 permeabilization of each section, but also demonstrates the feasibility of the custom 96-well plate platform to image embryo sections in a high-throughput manner.

Based on the success of the control samples, we proceeded to measure FZD4 expression in vivo. To aid in positioning and to clearly demarcate spatial orientation and boundaries, we used BRY expression as a clear reference point. Unfortunately, while the control samples were imaged successfully, the associated staining quality of the in vivo samples was poor, (Figure 3-2D). While the DAPI stain was again quite successful, both BRY and FZD4 exhibited non-specific binding characteristics. These sections were both from E7.0 embryos and the area sectioned was from the middle of the embryo. We would expect BRY to only be present on the posterior side of the embryo, and yet here, and also on the other sections, BRY stain was visible. Additionally, FZD4 showed a similar pattern where all sections displayed similar staining patterns. In combination with the control conditions, we speculate that this may be due to incompatibilities with our staining protocol and inappropriate antibody specificity. Moving forward, we are investigating alternative antibodies with possible compatibilities with in vivo samples and have begun collaborations with Dr. Nicole Dubois at the Icahn School of Medicine at Mount Sinai. We anticipate their expertise in whole mount and cry section staining of the early stage mouse embryo will provide the required insight and guidance to achieve in vivo validation of FZD4 in the gastrulating embryo.

An especially important advantage of in vivo studies is the ability to trace cells during differentiation to determine the contribution of their progeny to downstream lineages. This principle depends fundamentally on the concept that a developmental fate can be assigned to each cell at the early stages of an embryo, based on the seminal work of Whitman and Wilson on leech and ring worm embryos (Whitman, 1887; Wilson, 1892). Restated, lineage tracing aims to determine which cells in the early embryo will give rise to which parts of a given structure during development. These interrogations are also extended to determining the origins of a stem cell population through clonal analyses which investigate the derivatives of single cells and can provide insight into the segregation of distinct cell lineages. The most widely used methodology to label specific cells is site-specific genetic recombination, and mainly through a Cre-Lox-based strategy (Barker et al., 2007; Ito et al., 2005; Tata et al., 2013). Briefly, the system consists of an enzyme, Cre recombinase, which recombines a pair of target sequences called Lox. The Cre recombinase is placed under the control of a cell type or tissue-specific promoter and crossed with a ubiquitously expressed reporter line which is prevented by a stop cassette flanked by two Lox sites. The

70 transcription of the cell type or tissue-specific promoter in a cell will activate the Cre recombinase, which will then excise the stop cassette enabling the ubiquitous expression of the reporter. As a result, all progeny of a cell expressing the promoter will express the reporter and enable spatial and temporal localization throughout the embryo. Some notable examples of key lineage tracing experiments are for MESP1, which is an important transcription factor expressed in cardiac mesoderm, and was one of the earliest targets to determine cardiac lineage specification phenotypes (Saga et al., 1999). Early findings attributed FLK1 to the early stages of the vascular and hematopoietic lineages, however, lineage tracing studies have provided evidence that the FLK1 mesodermal progenitors also contribute to the muscle lineage including cardiomyocytes (Ema et al., 2006; Motoike et al., 2003). Reconstructing the lineage of cells during development is a key factor to understanding how cell linages are developed in both in vitro stem cell culture and in the embryo. Similar studies done with FZD4 may yield additional information through clonal analysis and determining the extent of the contribution to the cardiac linage.

3.3 FZD4-mediated WNT Signaling Dynamics during CPC Development

WNT signaling during cardiac progenitor development is a complex and dynamic process whereby both canonical and non-canonical WNT signaling play diverse and often opposing roles. The data presented in this thesis have identified and characterized FZD4 as a new marker for LPM. Previous studies have largely relied on intracellular markers, LMO2, PECAM1, NKX2.5, and ISL1 (Iyer et al., 2015; Zhou et al., 2008) to mark and differentiate the LPM from paraxial mesoderm. Additionally, surface markers such as FLK1, PDGFRA, and CD34 have also been used to denote LPM phenotypes (Tan et al., 2013b). However, these markers are generally non-specific and show a high degree of overlap between different cell types. We have provided evidence that FZD4 can be used to further segregate a FLK1+PDGFRA+ population and thus increase CPC purity and the specificity of cell type identification. Careful investigation of the timing of expression and spatial localization of each of these markers both individually and in combination are needed to better determine the specific boundaries and capabilities of each marker and/or marker sets to uniquely identify a given cell type.

In addition to the identification and characterization of FZD4 as a marker, we also reported a proliferative role of canonical WNT signaling through FZD4 during early cardiac progenitor

71 differentiation. We showed that NORRIN can be presented to the FZD4 receptor to induce WNT signaling-mediated proliferation which resulted in an increase in CM output from CPCs. This demonstrates the value in knowing the set of surface markers present on a cell at a specific stage of development and the potential to leverage that knowledge into more efficient differentiation protocols. While a previous study has also confirmed the increase in FZD4 abundance during the lateral plate mesoderm stage in hPSCs, the authors have suggested a non-canonical WNT signaling role of FZD4 (Mazzotta et al., 2016). However, our data suggests an alternative explanation based on evidenced provided by Ai et al. and Kwon et al., which show that the overexpression of b- catenin resulted in an expansion of the SHF. The authors further linked this phenotype with an increase in cell proliferation (Ai et al., 2007; Kwon et al., 2007). The data presented in this thesis also supports a proliferative role of FZD4 signaling, in addition to identifying FZD4 as the receptor associated with the canonical WNT signaling pathway.

The apparent contradiction of FZD4 signaling during early mesoderm specification can be attributed to the context-dependent responses of the FZD4 receptor to various WNT ligands. WNT5a, WNT7a, and WNT11 have been shown to interact with FZD4 to initiate non-canonical WNT signaling via the JNK-mediated pathway (Gessert and Kühl, 2010b; Yang, 2012), while NORRIN and WNT3a activate the canonical WNT signaling pathway (Hendrickx and Leyns, 2008). NORRIN, while being a non-WNT ligand, has been previously described to bind to FZD4 in the presence of the co-receptor LRP5 with nanomolar affinity, strongly activating the canonical WNT pathway (Clevers, 2004; Xu et al., 2004). NORRIN signaling primarily is associated with the promotion of angiogenesis as overexpression induces the overgrowth of capillaries in the eye (Ohlmann et al., 2005). Abnormalities in NORRIN and FZD4 are commonly associated with familial exudative vitreoretinopathy (FEVR), which results in retinal hypovascularization and include complications such as retinal detachment and irreversible scarring of the retina (Robitaille et al., 2002). In chick embryos, both FZD4 and NORRIN have been shown to be present in overlapping spatial areas from late-stage gastrulation to early cardiac specification, suggesting a functional relevance of NORRIN-FZD4 signaling during cardiac differentiation (Paxton et al., 2010). Taken together, these data may indicate both a canonical and non-canonical role of FZD4 signaling. The induction of the non-canonical signaling pathway may be, as previous studies suggest, to promote specification of CPCs into CM, while the canonical signaling pathway induces proliferation of CPCs resulting in a downstream increase of overall CM, (Figure 3-3). FZD4

72 signaling, and WNT signaling overall, is an extremely context-dependent pathway and a careful look into the timing of the canonical and non-canonical signaling aspects of the WNT pathway is warranted. To further enhance our understanding of the role FZD4 plays in this transient population, generation of a response surface model of WNT ligands on the FZD4 receptor at various stages of development may provide important insights as to the impact of FZD4 signaling in the context of other progenitor cell types.

Another observation that was quite striking was the difference in morphology exhibited between the CMs derived from the FZD4+ versus FZD4- sub-population (Figure 2-8B). The difference in morphology also may reflect a difference in migratory potential between the two sub-populations. Extracellular matrix (ECM) proteins are secreted beginning at the earliest stages of embryonic development, involved in multiple biochemical contexts, and are modified and remodeled throughout development (Pulina et al., 2011). Fibronectin is a key component during the migration of cardiac mesoderm from the primitive streak to the midline to form the cardiac crescent and is

Figure 3-3: Updated WNT signaling stages based on FZD4 marker.

The main phases of cardiac development where alternating roles of canonical and non-canonical WNT signaling are depicted. The role of FZD4-NORRIN signaling is associated with CPC proliferation and downstream CM proliferation.

73 expressed on the basal surface of the LPM and at the midline (Trinh and Stainier, 2004). ITGB5, which has been previously identified in this thesis, may be an interesting candidate due to its function as a binding protein and upon activation, interaction with the WNT signaling pathway. ITGB5 binds to the arginine-glycine-asparagine (RGD) sequence on fibronectin and subsequently activates the mitogen-activated protein kinase - extracellular signal-regulated kinase (MAPK- ERK) pathway. The integrin-mediated crosstalk with cadherin-dependent b-catenin signaling servers to increase canonical WNT pathway activation to increase cardiac differentiation (Sa et al., 2014). These connections further highlight the value of the other candidate proteins identified in this study and demonstrates the need for further investigation into the association or interaction with FZD4.

3.4 Generation of Cell-Cell Communication Network to Interrogate CPC Population Dynamics

Cell-cell communication is generally important for an organism as it strictly governs the behaviour and resultant phenotype of each cell. During development, such communication is critical as the development and function is ultimately controlled by neighboring signals (Eichmann et al., 1997; Gale et al., 1996). This dynamic cell-cell interaction drives the development of the embryo throughout all stages of development. The onset of gastrulation is determined by the cell-cell interactions between the epiblast and visceral endoderm region (Rivera-Pérez and Magnuson, 2005). This results in the symmetry-breaking event that leads to the establishment of the primitive streak on one side of the epiblast and the subsequent formation of the anterior-posterior axis. Early cardiac progenitor cells also require interaction with the neighboring visceral endoderm to subsequently differentiate into cardiomyocytes (Arai et al., 1997). Additionally, cell-cell interactions via direct contact through Notch signaling also regulate cell motility and differentiation (Timmerman et al., 2004). In the SHF and FHF stages of development, Notch receptor and ligands are present suggesting active signaling during early cardiac specification. Not only are the interactions between cardiac sub-lineages, such as epicardial and endocardial, important, but also between non-cardiac cells, including fibroblasts and endothelial cells. Neuregulin 1, expressed on endocardium is released as a paracrine signal that activates human epidermal growth factor receptor (HER) 2/4 receptor complex on CMs, which is required for trabeculation of the primitive heart (Gassmann et al., 1995). Communication between embryonic fibroblasts, fibronectin, collagen, and heparin-binding epidermal growth factor (EGF)-like growth

74 factor collaboratively promote CM proliferation (Ieda et al., 2009). Interacting cell populations can form a cellular interactome and can create similar networks as with genes or proteins that interact with one another. Interactions formed during cardiac development will occur in a spatiotemporal manner and require different cell populations recruited at different times. As a preliminary step towards investigating the cellular interactome during cardiac development, I focused specifically on cell-cell communication during early CPC development.

In order to further understand the signaling both within the CPC stage and during the transition to CM, we generated a cell-cell communication (CCC) network using an algorithm developed in- house (Qiao et al., 2014). Using MS data previously generated, we compiled a list of surface receptors and their complementary ligands as input to generate a CCC network between two sub- populations at the CPC stage and a differentiated CM stage (Figure 3-4). Preliminary analysis indicates a greater number of ligands considered to be autocrine in nature versus ligand signaling between cell types. This may indicate a high degree of self-regulation from each of these cell types, however, further network topology analysis is necessary to accurately assess this hypothesis.

After the CCC network is constructed, certain properties of the network can be determined. One, is to test whether each node (or cell type) is modular, which would indicate the interactions

Figure 3-4: Cell-cell Interaction Network generated from microarray and proteomic data.

75 between the cell types to be cell type specific. Second, we can classify the interactions between each node depending on the number of ligands presented versus the number of receptors present. This analysis would yield control nodes or groups of cells that direct or initiate the feedback loops, and, to further validate these interactions, high-content in vitro validation of the receptors and ligands would be required. Once validated, the construction of a cellular interactome may provide insight into how different cell sub-populations cross talk with one another in order to promote differentiation, proliferation, or cell motility. Understanding these interactions can contribute towards general understanding of cardiac developmental biology and is critical towards the design of efficient differentiation protocols or effective therapeutic solutions.

3.5 Alternative Splicing is a Key Feature in Cardiac Progenitor Cell Specification

Precise spatial and temporal regulation of gene expression is required during development and the focus has traditionally been on the genome and transcriptome. However, new regulatory domains have begun to emerge as being critical in cell differentiation. Namely, alternative splicing (AS), a post-transcriptional regulatory mechanism of gene expression that allows generation of more than one unique mRNA strand from a single gene (Black, 2003; Chen and Manley, 2009). Recent reports indicate that upwards of 80% of pre-mRNAs are alternatively spliced (Pan et al., 2008). These AS events have been reported to impact both mouse and human heart development in early cardiac progenitors (Cooper, 2005; Emig et al., 2010), early fetal heart (Revil et al., 2010), and well into the adult heart (Sheng and Jin, 2014). FGF signaling occurs during gastrulation in the mouse, specifically FGF8 which is expressed in the epiblast and subsequently in the emerging primitive streak (Crossley and Martin, 1995). FGF8 can produce many different isoforms (Gemel et al., 1996), namely FGF8a and FGF8b which exhibit distinct bioactivities. FGF8b has a more potent signaling activity due to its higher affinity for FGF receptors, and result in different phenotypes upon expression in the gastrulating embryo and in the mid-hindbrain during development (Fletcher et al., 2006; Guo and Li, 2007; Olsen et al., 2006). Splicing factors also contribute to the regulation of AS and knockout of serine/arginine protein 38 (SRp38), a general splicing repressor (Shin and Manley, 2002), results in embryos that die with multiple cardiac defects (Feng et al., 2009). Disruption of SRp38 resulted in a reduction of the appropriate isoform of triadin and calsequestrin 2, which play key roles in excitation-contraction coupling, indicating that SRp38-regulated AS plays a critical role in embryonic heart development. ECM proteins such

as fibronectin can also be regulated through AS to generate multiple protein sequences with functional variants which are stage- and tissue-specific. The absence of such AS segments of fibronectin can result in multiple cardiovascular defects (Astrof et al., 2007). Further investigation of AS dynamics during cardiac differentiation would thus, enable a better understanding of cell dynamics during heart development.

To identify AS events during CPC specification to CM, we used an open-source application called AltAnalyze (Emig et al., 2010), which uses the splicing index (SI) approach to calculate differential exon expression that is corrected for overall gene expression (Gardina et al., 2006; Srinivasan et al., 2005). Comparing the FLK1+PDGFRA+ to FLK1+PDGFRA- sample, of the 1128 differentially expressed genes, 74% (832) were predicted to have at least one alternative exon

(Table 3-1). There were similar percentage There were similar percentage of genes predicted to have at least one alternate exon in both the CM compared with either FLK1+PDGFRA+ or FLK1+PDGFRA- compartments (69% and 71% respectively). However, due to the significantly larger number of differentially expressed genes, the number of AS genes is 2-3x greater. Once again, comparing the FLK1+PDGFRA+ vs. FLK1+PDGFRA- samples, of the genes that are predicted to have at least one alternative exon, there were 28% (231) genes with AS, 5% (45) AP, 5% (38) both AS and AP, and the rest with no evidence of alternative regulation. Similar percentages for the other two pair-wise comparisons. Thus, our analysis predicts ~1-6% of genes exampled are alternatively spliced relative to the ~3-10% differentially expressed genes.

Numerous splicing events were identified within the CPC population (between FLK1+PDGFRA+ and FLK1-PDGFRA+ compartments) and even more in the CPC specification stage to CM. From

Table 3-1: Differently expressed genes and their breakdown of predicted alternative exons.

FLK1+PDGFRA+ vs. CTNT+ vs. CTNT+ vs.

FLK1+PDGFRA- FLK1+PDGFRA+ FLK1+PDGFRA- Differentially expressed genes 1128 3140 2776 Genes with alternative exons 832 2164 1984 Alternative splicing (AS) 231 865 765 Alternative promoter (AP) 45 184 137 AS and AP 38 162 116 No AS nor AP 518 953 966

77 the 2164 genes with predicted alternative exons in the CM vs FLK1+PDGFRA+ comparison, there were 230 identified surface proteins based on GO cellular compartment annotations. Comparing these genes with a list of proteins upregulated during the CPC stage (previously identified), yielded 21 surface proteins that were alternatively spliced indicated by p<0.05 for the SI metric (Figure 3-5A). We selected ephrin-B2 (EPHB2) for deeper analysis. EPHB2 is required for angiogenesis and has a role in cardiovascular development, specifically the demarcation of arterial/venous domains (Diehl et al., 2005). Expression profiles (Figure 3-5B) and exon profiles (Figure 3-5C) are shown. The maps indicated that EPHB2 is alternatively spliced and the isoform present in CM has significantly decreased expression in 2 central exons.

Analysis of previously identified motif changes corresponding to specific functionality were also mapped by AltAnalyze (Figure 3-5D). The green diamonds indicate an up-regulation of

Figure 3-5: Identification of predicted alternative splicing events.

A) AltAnalyze predictions of alternatively spliced genes based on significant SI score mapped onto the surface proteome. B) Expression profile and C) exon profile of EPHB2 protein. D) Signaling network of EPHB2 protein.

78 interacting factors upon specification into CM which may indicate the initiation of downstream target genes. Our study has reported 21 surface proteins with predicted splice variants. Future studies will involve validation of these predictions at the transcript level for accurate exon expression values. Optimal flanking, isoform-specific, and/or constitutive primers need to be designed and each sample amplified using PCR and visualized on DNA-agarose gels.

3.6 Systems Biology Approach Towards Understanding CPC Biology

We demonstrated a systems biology approach to investigating the cell type hierarchy of cardiac progenitor cells during development. Starting with omics data on both surface proteomic and transcriptomic levels, we have drilled down and selected a key surface protein, FZD4, relevant to cardiac development. This surface molecule was then technically validated and characterized. The final step would then be to cycle back to the omics level and repeat the analysis but with FZD4 included. As demonstrated earlier with ITGB5, this thesis has generated a number of candidate markers in addition to FZD4 that warrant validation and have potential to have a significant biological impact, not only related to overall cardiac development, but to the earliest stages of development (from PSC to epiblast to gastrulation) as well.

Previous studies characterizing, and elucidating mechanisms have been able to converge on many fundamental signaling pathways and core transcriptional networks driving cardiac differentiation. However, these studies have largely been isolated and partially defined and there remain very few studies approaching development in a more holistic, systems biology view. Similar to work provided in this thesis, Faustino et al. analyzed gene expression in differentiating mouse stem cells using microarrays and isolated distinct patterns of up and downregulation of genes at specifics stages of cardiac differentiation (Faustino and Terzic, 2008). More recently, Scialdone et al. applying single cell RNA-Seq to early mesoderm differentiation in mouse embryos during gastrulation, also analyzed expression clusters related to transcriptional programs in cardiac differentiation (Scialdone et al., 2016). Such high throughput screening methods provide non- hypothesis driven research leading to increased identification of novel factors associated with cardiac development. The study presented in this thesis has greatly increased the potential pool of surface markers present CPCs as well those present on the PSC, epiblast, and primitive streak stages as well. Another potential application of the datasets presented herein can be towards

79 investigating differences between identified surface proteins relative to the differential expression of transcripts identified by both microarray and RNA-Seq datasets. The increasing volume of high throughput datasets must then be combined and incorporated into a coherent process whereby in silico models can be generated to monitor all the relevant parameters.

With the continued improvement of MS technology and optimization of methods for surface membrane protein enrichment, a true systems biology approach to analysing cardiac development comprehensively can be possible. In terms of equipment, mass spectrometers are routinely sequencing upwards of 5000 proteins (Distler et al., 2016; Huang et al., 2015) and revealing novel proteins in both human (Csősz et al., 2015) and Xenopus (Lombard-Banek et al., 2016) systems. Recently, there has been a major effort towards miniaturizing the sample requirement for MS down to single cell levels. Budnik et al. have developed Single Cell ProtEomics by Mass Spectrometry (SCoPE-MS), and validated its ability to identify distinct human cancer cell types (Budnik et al., 2017). The authors have also used SCoPE-MS to quantify over 1000 proteins in differentiating mPSCs. These new techniques can also be applied to interrogate the secretome during cardiac differentiation from the media the cells are incubated in. A de novo analysis of secreted factors present during differentiation can bolster the ligand side of the cell-cell communications network. Finally, single cell RNA-Seq (instead of microarrays) can be utilized to offer a complementary transcriptomic analysis in concert with the proteomic analysis. Scialdone et al. applied single cell RNA-Seq to early mesoderm differentiation in mouse embryos during gastrulation (Scialdone et al., 2016).

The highly integrated approaches herein have contributed to providing fundamentally important biological insight for cardiac development in both mouse and human. Detailed knowledge of the cardiac differentiation hierarchy will accelerate the development of protocols that efficiently generate the desired cell populations that then can be classified with a larger set of markers to ensure specificity. These sub-populations can then be systematically assayed to determine their in vivo regenerative potential and ultimately applications towards regenerative medicine. Future studies can validate and characterize these markers, and ultimately using a systems biology approach, (Figure 3-6), further our understanding of the biology of mouse and human development.

Figure 3-6: Systems level approach to understand biology.

Systems level biology study where multi-level high throughput information can be combined to create a model which is then validated and tested to feedback into the original model. 3.7 Identification of Surface Markers Improves Isolation of Pure CPCs and Improves Cell Therapies in Regenerative Medicine

The aims of this thesis and the data presented herein focus on the development of cardiac progenitors of embryonic origins and the subsequent derivation of CMs. We have shown FZD4 is expressed in the hPSC system and allows for a similar enrichment in CMs. This finding highlights the utility of FZD4 in identifying and sorting for CPCs that can potentially contribute to human cell transplantation studies.

Stem cell therapy for heart repair and regeneration overall has been motivated as a new treatment of myocardial infarction and congestive heart failure for the past 10-20 years. The major hypotheses regarding the mechanism of action are direct cardiomyogenic differentiation and indirect stimulation of endogenous repair (Behfar et al., 2014). The first generation of

81 transplantation studies used skeletal myoblasts, endothelial progenitor cells, and mesenchymal stem cells, among others, and obtained widely varying results despite promising pre-clinical studies (Fisher et al., 2015; Gyöngyösi et al., 2015, 2016). Some of these concerns are beginning to be addressed by the use of endogenous CPCs or PSC-derived CPCs in the second generation approach (Li et al., 2012; Oskouei et al., 2012; Rossini et al., 2011). In particular, CPCs isolated from biopsies of the adult heart, herein termed adult CPCs, express C-KIT, ISL1, and stem cell antigen (SCA) 1 and have been shown to be multipotent and clonogenic (Messina et al., 2004). Initial clinical results from adult CPCs are promising and suggest treatment potential, however larger cohort studies must be done and the mechanism of action must be determined in order to prove effectiveness (Bolli et al., 2011; Malliaras et al., 2014). PSC-derived CPCs, such as the ones described in this thesis, were also capable of differentiating and contributing CMs, endothelial cells, and smooth muscle cells upon transplantation and demonstrated engraftment with native mature CMs in the mouse (Mauritz et al., 2011; Yang et al., 2008c). While PSC-derived CMs can be another source of cells to transplant and have demonstrated some potential in coupling with the host tissue and improving contractility in mouse models (van Laake et al., 2007; Laflamme et al., 2007), the transplanted CMs are lineage committed and are unable to differentiate into the multiple cell types present in the heart. Moreover, the use of PSC-derived CPCs have also progressed to clinical trials (Menasché et al., 2015), and while the preliminary results seem promising, additional results are required in order to make an appropriate conclusion. While CPC transplantations may show the potential for remuscularization, these effects can largely be explained by paracrine effects and are not truly regenerative in nature as the engraftment is not accompanied by scar resorption and regeneration (Ong et al., 2015; Shiba et al., 2016). To address this barrier, a third-generation strategy is required and would tackle these limitations by engineering combinatorial cell transplantations which would address immune modulation, scar resorption, remuscularization, re- vascularization, and reinnervation (Hatzistergos and Vedenko, 2017). Appropriate matching of combinations of cell types with complementary roles may more efficiently regulate regeneration pathways in comparison to previous transplantation methods (Golpanian et al., 2016). This hypothesis has produced encouraging results and is currently in phase II clinical trials where a combination of bone marrow-derived mesenchymal stem cells and adult CPCs are combined and transplanted into ischemic cardiomyopathy patients.

Regardless of the selection of the appropriate cell type(s), cell delivery remains a major challenge. Most stem cell studies attempt to deliver stem cells directly into the area of infarction through injection or infusion. However, such designs may not be suitable for a complex organ especially in the context of ischemic cardiomyopathy with increased oxidative stress, inflammation, and decreased perfusion. As a result engraftment after transplantation into damaged tissue is inadequate and transplanted cells are negatively affected by the harsh environments of the infarct (Cho et al., 2012). To address this challenge, many studies are employing tissue-engineering techniques to support cell transplantation with biomaterials in scaffolds and regulatory factors. A notable example is the construction of engineered cell sheets of CMs that are positioned over the infarct area, which demonstrated beneficial effects stemming from the structural support of the graft and cytokine production (Xiong et al., 2013). Upon transplantation of the appropriate cell type or graft, imaging the transplanted cells to track engraftment and position over time becomes increasingly important. The analysis of transplanted cells allows for detailed interrogation of cell composition and characterization of genes and surface markers. Unfortunately, during in vivo transplantations and during pre-clinical trials, sample collection requires euthanizing the animal at fixed time intervals, which increase the complexity of the study and introduces a need for a non- invasive method to track cell position, growth, and overall function (Orlic et al., 2002). Magnetic resonance imaging (MRI) is often used in most studies to track stem cells in vivo based on its safety profile and high 3D imaging capabilities. Depending on the contrast agents used, MRI can permit tracking of stem cells for up to 6 weeks and can detect numbers low as thousands of cells (van den Bos et al., 2003; Frank et al., 2003; Garot et al., 2003), enabling MRI to be used to successfully track stem cell transplants over a long period of time.

Despite the challenges posed by cell survival, scar removal, host engraftment, and remuscularization, the importance of having a comprehensive list of identifying markers for each cell type is once again apparent. Furthermore, the requirement for markers that reside on the surface of the cell in order to contribute to functional characterization and feed into a therapeutic pipeline is highlighted. This situation also emphasizes the importance of not only the identification of markers, but the benefits of having a combination or panel of markers in order to specify a homogenous cell type. The prerequisite of generation a large number of cells with a high purity stems from the potential generation of teratomas after transplantation of undifferentiated PSCs (Zhang et al., 2011). This implies that it is essential to purify PSC-derived CPCs prior to

83 transplantation which requires knowledge of a panel of markers capable of doing so. Additionally, the usage of mouse and other non-human model systems remain invaluable towards the collection of functional data. Many of the studies previously mentioned have been pioneered in animal models prior to application in the hPSC system and initiation of clinical trials. Further studies are required to identify relevant markers for the CPCs and careful characterization of each identified sub-population would provide further evidence to tackle concerns related to cell heterogeneity and to ultimately best generate functional connections between seemingly disparate CPC populations. While questions remain regarding on what the appropriate CPCs or non-cardiac cell types for transplantation would be, further development of improved marker sets for better generation and identification of cell types would only be an asset to the field of regenerative medicine.

CONCLUSIONS AND FUTURE WORK

4.1 Thesis Summary

Comprehensive knowledge of cell surface markers present on stem cells and stem cell progenitors are essential towards understanding the biological properties and behaviour of cell populations as they interact with each other during development. Elucidation of the surface markers involved at critical phases in development can identify the signaling events involved and contribute towards our understanding of the fate decisions that occur. In the cardiac system, there is a distinct lack of identified surface markers on CPCs and to address this, I have turned to an integrated cell surface mass spectrometry and microarray-based approach to increase the candidate pool of surface membrane proteins in CPCs in a stage-specific manner. Subsequent bioinformatics analysis yielded a list of candidate surface markers, which were phenotypically validated using qPCR and flow cytometry, and through differentiation assays with sorted cell populations. After filtration based on protein-gene correlation, antibody availability, and literature review, I focused on the FZD4 receptor, a member of the WNT signaling pathway, for further analysis.

In this analysis, we have first, identified and characterized FZD4 as a new marker for LPM. Where previous studies have primarily used intracellular markers to identify LPM, FZD4 can be used cell sorting assays for further characterization. Second, we utilized FZD4 as a marker, in conjunction with FLK1 and PDGFRA, and demonstrated an increase in CPC purity and a subsequent increase in CM enrichment. Third, we have also shown FZD4 is expressed in the hPSC system and allows for an enrichment in CMs of similar magnitude. This finding highlights the utility of FZD4 in identifying and sorting for CPCs that can potentially contribute to human cell transplantation studies. Fourth, we showed that NORRIN can be presented to the FZD4 receptor to induce WNT signaling-mediated proliferation which resulted in an increase in CM output from CPCs. This demonstrates the value in knowing the set of surface markers present on a cell at a specific stage of development and the potential to leverage that knowledge into more efficient differentiation protocols. Finally, this study has greatly increased the potential pool of surface markers present not only on CPCs, but also present on mPCSs and each stage of development in between. Once validated and characterized, these markers can also contribute to our understanding of the biology of mouse and human development.

In addition to the panel of surface markers available for further characterization, I have generated microarray data using exon tiling arrays, enabling an additional layer of analysis based on

86 alternative splicing. I have identified 21 surface proteins as candidates for AS regulation which further studies can be done to validate and functionally evaluate these findings. Moreover, in combination with the surface proteomic data, I have extended the MS analysis to encompass network analysis by generating a cell-cell interaction network to investigate the dynamics of CPC specification into CM. Upon further validation, these networks can provide insight on signaling between cells within the CPC population and increase our understanding of cardiac development.

4.2 Impact and Future Studies

The findings presented in this thesis provide a new surface marker for early cardiac mesoderm and provide an initial systems biology approach to investigating the cell type hierarchy of cardiac progenitor cells during development. The data provided in this thesis has identified and characterized FZD4 as a new marker for LPM. Additional, surface markers such as FLK1, PDGFRA, and CD34 have also been used to denote lateral plate mesoderm phenotypes (Tan et al., 2013b). However, these markers are generally non-specific and show a high degree of overlap between different cell types. We have provided evidence that FZD4 can be used to further segregate a FLK1+PDGFRA+ population and thus increase CPC purity and the specificity of cell type identification. These results highlight the value of having a blueprint of the specific sets of surface markers present on a cell at specific stages of development and the potential to leverage that knowledge into more efficient differentiation protocols and better understanding of the CPC sub-population segregation within the early stages of cardiac progenitor development. Further studies would carefully investigate the expression timing and spatial localization of each of these markers individually and in combination to better determine the specific boundaries and capabilities of each marker or marker sets to uniquely identify a given cell type.

In terms of a proposed mechanism, the data presented herein also suggested a proliferative role of canonical WNT signaling through FZD4 during early cardiac progenitor differentiation. NORRIN was presented to the FZD4 receptor to induce WNT signaling-mediated proliferation which resulted in an increase in CM output from CPCs. Combined with previous studies that state non- canonical WNT signaling occurs at the CPC stage, these data may indicate both a canonical and non-canonical role of FZD4 signaling in a context-dependent manner. The induction of the non- canonical signaling pathway may be, as previous studies suggest, to promote specification of CPCs into CM, while the canonical signaling pathway induces proliferation of CPCs resulting in a

87 downstream increase of overall CM. As a next step, experiments to carefully tease apart both aspects of the WNT signaling pathway and timing would provide additional insight as to the specific application and duration of FZD4 signaling during the generation of cardiac mesoderm. Moreover, generation of a response surface model for a combination of WNT ligands on the FZD4 receptor at various stages of development may provide important insights as to the impact of FZD4 signaling in the context of other progenitor cell types.

To further provide insight into the understanding of FZD4-NORRIN signaling biology, we have provided a first-step analysis on the cell-cell communication network during cardiac progenitor differentiation. During development, such communication is critical as the development and function is ultimately controlled by neighboring signals (Eichmann et al., 1997; Gale et al., 1996). Using an integrated dataset from surface mass spectrometry and microarray, we compiled a CCC network between the CPC and CM stages. Preliminary analysis suggests higher number of autocrine factors relative to paracrine factors indicating a higher degree of self-regulation from early CPCs. Further studies into the network topology and a completed CPC interactome may yield insight into how different cell sub-types communication with each other in order to initiate differentiation, proliferation, and cell motility. Understanding of these interactions can contribute towards general cardiac developmental biology and be critical in the design of efficient protocols and effective therapies.

Additional interrogation of the data is possible based on the established surface dataset from mass spectrometry and the exon-containing microarray data, allowing for alternative splicing analysis. (Black, 2003; Chen and Manley, 2009). AS events are reported to be extremely common in mammalian systems (Pan et al., 2008) and are involved in the regulation of cardiac development in both mouse and human systems (Cooper, 2005; Emig et al., 2010). Preliminary analysis indicated candidate surface proteins related to cardiogenesis that demonstrated alternative splicing between CPCs and CMs. Future studies can validate these target candidates and further functional analysis may yield key mechanisms impacting cell signaling during early cardiac development.

We have also shown FZD4 is expressed in the hPSC system and allows for a similar enrichment in CM. This finding highlights the utility of FZD4 in identifying and sorting for CPCs that can potentially contribute to human cell transplantation studies. Stem cell therapy for heart repair and regeneration overall has been motivated by the treatment of myocardial infarction and congestive

88 heart failure for the past 10-20 years. PSC-derived CPCs, such as the ones described in this thesis, have also progressed to clinical trials (Menasché et al., 2015), and while the preliminary results seem promising, additional results are required in order to make an appropriate conclusion. Identification of FZD4 as a potential marker to isolate pure CPCs from hPSC-derived cultures can accelerate development of therapeutic protocols and be instrumental in initiating clinical trials. While questions remain regarding on what the appropriate CPC type for transplantation would be, further development into improved marker sets for better generation and identification of cell types would only be an asset to the field of regenerative medicine.

Finally, we have demonstrated a systems biology approach to investigating of the cardiac progenitor cell type hierarchy during cardiac development. From general omics data, I have drilled down and discovered a key surface receptor, FZD4, that is relevant to cardiac development. Such high throughput screening methods enable non-hypothesis driven research leading to increased identification of novel factors associated with cardiac development. The high-volume influx of similar high throughput datasets can then be combined and incorporated into a coherent process where relevant questions regarding cardiac development can be addressed in a holistic manner. This study has vastly expanded the pool of candidate surface proteins expressed during CPC generation. Future studies can validate and characterize these markers, and ultimately using a systems biology approach, further our understanding of mouse and human cardiac development.

References

Abdul-Ghani, M., Dufort, D., Stiles, R., De Repentigny, Y., Kothary, R., and Megeney, L. a (2011). Wnt11 promotes cardiomyocyte development by caspase-mediated suppression of canonical Wnt signals. Mol. Cell. Biol. 31, 163–178.

Abu-issa, R., Waldo, K., and Kirby, M.L. (2004). Heart fields: One, two or more? Dev. Biol. 272, 281–285.

Adams, M.D., Celniker, S.E., Holt, R.A., Evans, C.A., Gocayne, J.D., Amanatides, P.G., Scherer, S.E., Li, P.W., Hoskins, R.A., Galle, R.F., et al. (2000). The genome sequence of Drosophila melanogaster. Science 287, 2185–2195.

Ai, D., Fu, X., Wang, J., Lu, M.-F., Chen, L., Baldini, A., Klein, W.H., and Martin, J.F. (2007). Canonical Wnt signaling functions in second heart field to promote right ventricular growth. Proc. Natl. Acad. Sci. 104, 9319–9324.

Alizadeh, A.A., Eisen, M.B., Davis, R.E., Ma, C., Lossos, I.S., Rosenwald, A., Boldrick, J.C., Sabet, H., Tran, T., Yu, X., et al. (2000). Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403, 503–511.

Amit, M., Carpenter, M.K., Inokuma, M.S., Chiu, C.P., Harris, C.P., Waknitz, M. a, Itskovitz- Eldor, J., and Thomson, J. a (2000). Clonally derived human embryonic stem cell lines maintain pluripotency and proliferative potential for prolonged periods of culture. Dev Biol 227, 271–278.

Amit, M., Chebath, J., Margulets, V., Laevsky, I., Miropolsky, Y., Shariki, K., Peri, M., Blais, I., Slutsky, G., Revel, M., et al. (2010). Suspension culture of undifferentiated human embryonic and induced pluripotent stem cells. Stem Cell Rev. 6, 248–259.

Arai, A., Yamamoto, K., and Toyama, J. (1997). Murine cardiac progenitor cells require visceral embryonic endoderm and primitive streak for terminal differentiation. Dev. Dyn. Off. Publ. Am. Assoc. Anat. 210, 344–353.

Arbeitman, M.N., Furlong, E.E.M., Imam, F., Johnson, E., Null, B.H., Baker, B.S., Krasnow, M.A., Scott, M.P., Davis, R.W., and White, K.P. (2002). Gene expression during the life cycle of Drosophila melanogaster. Science 297, 2270–2275.

Arjunan, S., Reinartz, M., Emde, B., Zanger, K., and Schrader, J. (2009). Limitations of the colloidal silica method in mapping the endothelial plasma membrane proteome of the mouse heart. Cell Biochem. Biophys. 53, 135–143.

Arnold, S.J., Hofmann, U.K., Bikoff, E.K., and Robertson, E.J. (2008). Pivotal roles for eomesodermin during axis formation, epithelium-to-mesenchyme transition and endoderm specification in the mouse. Dev. Camb. Engl. 135, 501–511.

Astrof, S., Crowley, D., and Hynes, R.O. (2007). Multiple cardiovascular defects caused by the absence of alternatively spliced segments of fibronectin. Dev. Biol. 311, 11–24.

Attisano, L., Cárcamo, J., Ventura, F., Weis, F.M., Massagué, J., and Wrana, J.L. (1993). Identification of human activin and TGF beta type I receptors that form heteromeric kinase complexes with type II receptors. Cell 75, 671–680.

Attisano, L., Wrana, J.L., Montalvo, E., and Massagué, J. (1996). Activation of signalling by the activin receptor complex. Mol. Cell. Biol. 16, 1066–1073.

Bantscheff, M., Schirle, M., Sweetman, G., Rick, J., and Kuster, B. (2007). Quantitative mass spectrometry in proteomics: a critical review. Anal. Bioanal. Chem. 389, 1017–1031.

Bao, Z.Z., Bruneau, B.G., Seidman, J.G., Seidman, C.E., and Cepko, C.L. (1999). Regulation of chamber-specific gene expression in the developing heart by Irx4. Science 283, 1161–1164.

Baptista, R.P., Fluri, D.A., and Zandstra, P.W. (2013). High density continuous production of murine pluripotent cells in an acoustic perfused bioreactor at different oxygen concentrations. Biotechnol. Bioeng. 110, 648–655.

Barker, N., van Es, J.H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., Haegebarth, A., Korving, J., Begthel, H., Peters, P.J., et al. (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449, 1003–1007.

Bausch-Fluck, D., Hofmann, A., Bock, T., Frei, A.P., Cerciello, F., Jacobs, A., Moest, H., Omasits, U., Gundry, R.L., Yoon, C., et al. (2015a). A Mass Spectrometric-Derived Cell Surface Protein Atlas. PLOS ONE 10, e0121314.

Bausch-Fluck, D., Hofmann, A., Bock, T., Frei, A.P., Cerciello, F., Jacobs, A., Moest, H., Omasits, U., Gundry, R.L., Yoon, C., et al. (2015b). A mass spectrometric-derived cell surface protein atlas. PLoS ONE 10.

Bauwens, C., Yin, T., Dang, S., Peerani, R., and Zandstra, P. (2005). Development of a perfusion fed bioreactor for embryonic stem cell-derived cardiomyocyte generation: Oxygen-mediated enhancement of cardiomyocyte output. Biotechnol. Bioeng. 90, 452–461.

Beck, M., Schmidt, A., Malmstroem, J., Claassen, M., Ori, A., Szymborska, A., Herzog, F., Rinner, O., Ellenberg, J., and Aebersold, R. (2011). The quantitative proteome of a human cell line. Mol. Syst. Biol. 7, 549.

Behfar, A., Crespo-Diaz, R., Terzic, A., and Gersh, B.J. (2014). Cell therapy for cardiac repair-- lessons from clinical trials. Nat. Rev. Cardiol. 11, 232–246.

Bianchi-Smiraglia, A., Kunnev, D., Limoge, M., Lee, A., Beckerle, M.C., and Bakin, A. V (2013). Integrin-β5 and zyxin mediate formation of ventral stress fibers in response to transforming growth factor β. Cell Cycle 12, 3377–3389.

Biben, C., and Harvey, R.P. (1997). Homeodomain factor Nkx2-5 controls left/right asymmetric expression of bHLH gene eHand during murine heart development. Genes Dev. 11, 1357–1369.

Black, D.L. (2003). Mechanisms of alternative pre-messenger RNA splicing. Annu. Rev. Biochem. 72, 291–336.

Boheler, K.R., Czyz, J., Tweedie, D., Yang, H.T., Anisimov, S. V., and Wobus, A.M. (2002). Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circ. Res. 91, 189–201.

Boheler, K.R., Bhattacharya, S., Kropp, E.M., Chuppa, S., Riordon, D.R., Bausch-Fluck, D., Burridge, P.W., Wu, J.C., Wersto, R.P., Chan, G.C.F., et al. (2014a). A Human Pluripotent Stem Cell Surface N-Glycoproteome Resource Reveals Markers, Extracellular Epitopes, and Drug Targets. Stem Cell Rep. 3, 185–203.

Boheler, K.R., Bhattacharya, S., Kropp, E.M., Chuppa, S., Riordon, D.R., Bausch-Fluck, D., Burridge, P.W., Wu, J.C., Wersto, R.P., Chan, G.C.F., et al. (2014b). A Human Pluripotent Stem Cell Surface N-Glycoproteome Resource Reveals Markers, Extracellular Epitopes, and Drug Targets. Stem Cell Rep. 3, 185–203.

Bolli, R., Chugh, A.R., D’Amario, D., Loughran, J.H., Stoddard, M.F., Ikram, S., Beache, G.M., Wagner, S.G., Leri, A., Hosoda, T., et al. (2011). Cardiac stem cells in patients with ischaemic cardiomyopathy (SCIPIO): initial results of a randomised phase 1 trial. Lancet Lond. Engl. 378, 1847–1857.

Bondue, A., Lapouge, G., Paulissen, C., Semeraro, C., Iacovino, M., Kyba, M., and Blanpain, C. (2008). Mesp1 acts as a master regulator of multipotent cardiovascular progenitor specification. Cell Stem Cell 3, 69–84.

Bondue, A., Tännler, S., Chiapparo, G., Chabab, S., Ramialison, M., Paulissen, C., Beck, B., Harvey, R., and Blanpain, C. (2011). Defining the earliest step of cardiovascular progenitor specification during embryonic stem cell differentiation. J. Cell Biol. 192, 751–765. van den Bos, E.J., Wagner, A., Mahrholdt, H., Thompson, R.B., Morimoto, Y., Sutton, B.S., Judd, R.M., and Taylor, D.A. (2003). Improved efficacy of stem cell labeling for magnetic resonance imaging studies by the use of cationic liposomes. Cell Transplant. 12, 743–756.

Brade, T., Männer, J., and Kühl, M. (2006). The role of Wnt signalling in cardiac development and tissue remodelling in the mature heart. Cardiovasc. Res. 72, 198–209.

Brand, T. (2003). Heart development: Molecular insights into cardiac specification and early morphogenesis. Dev. Biol. 258, 1–19.

Brauchle, E., Knopf, A., Bauer, H., Shen, N., Linder, S., Monaghan, M.G., Ellwanger, K., Layland, S.L., Brucker, S.Y., Nsair, A., et al. (2016). Non-invasive Chamber-Specific Identification of Cardiomyocytes in Differentiating Pluripotent Stem Cells. Stem Cell Rep. 6, 188–199.

Brem, R.B., Yvert, G., Clinton, R., and Kruglyak, L. (2002). Genetic dissection of transcriptional regulation in budding yeast. Science 296, 752–755.

Bruneau, B.G., Logan, M., Davis, N., Levi, T., Tabin, C.J., Seidman, J.G., and Seidman, C.E. (1999). Chamber-specific cardiac expression of Tbx5 and heart defects in Holt-Oram syndrome. Dev. Biol. 211, 100–108.

Bruneau, B.G., Nemer, G., Schmitt, J.P., Charron, F., Robitaille, L., Caron, S., Conner, D.A., Gessler, M., Nemer, M., Seidman, C.E., et al. (2001). A murine model of Holt-Oram syndrome

92 defines roles of the T-box transcription factor Tbx5 in cardiogenesis and disease. Cell 106, 709– 721.

Bubis, J.A., Levitsky, L.I., Ivanov, M.V., Tarasova, I.A., and Gorshkov, M.V. (2017). Comparative evaluation of label-free quantification methods for shotgun proteomics. Rapid Commun. Mass Spectrom. 31, 606–612.

Buckingham, M., Meilhac, S., and Zaffran, S. (2005). Building the mammalian heart from two sources of myocardial cells. Nat. Rev. Genet. 6, 826–837.

Budnik, B., Levy, E., and Slavov, N. (2017). Mass-spectrometry of single mammalian cells quantifies proteome heterogeneity during cell differentiation. BioRxiv 102681.

Burridge, P.W., Thompson, S., Millrod, M.A., Weinberg, S., Yuan, X., Peters, A., Mahairaki, V., Koliatsos, V.E., Tung, L., and Zambidis, E.T. (2011). A universal system for highly efficient cardiac differentiation of human induced pluripotent stem cells that eliminates interline variability. PloS One 6, e18293.

Cadigan, K.M., and Nusse, R. (1997). Wnt signaling: a common theme in animal development. Genes Dev. 11, 3286–3305.

Cai, C.L., Liang, X., Shi, Y., Chu, P.H., Pfaff, S.L., Chen, J., and Evans, S. (2003). Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. Dev. Cell 5, 877–889.

Cai, W., Albini, S., Wei, K., Willems, E., Guzzo, R.M., Tsuda, M., Giordani, L., Spiering, S., Kurian, L., Yeo, G.W., et al. (2013). Coordinate Nodal and BMP inhibition directs Baf60c- dependent cardiomyocyte commitment. Genes Dev. 27, 2332–2344.

Cameron, C.M., Hu, W.S., and Kaufman, D.S. (2006a). Improved development of human embryonic stem cell-derived embryoid bodies by stirred vessel cultivation. Biotechnol. Bioeng. 94, 938–948.

Cameron, C.M., Hu, W.S., and Kaufman, D.S. (2006b). Improved development of human embryonic stem cell-derived embryoid bodies by stirred vessel cultivation. Biotechnol Bioeng 94, 938–948.

Carpenter, A.E., Jones, T.R., Lamprecht, M.R., Clarke, C., Kang, I.H., Friman, O., Guertin, D.A., Chang, J.H., Lindquist, R.A., Moffat, J., et al. (2006). CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biol. 7, R100.

Castle, J.D. (2003). Purification of organelles from mammalian cells. Curr. Protoc. Immunol. Chapter 8, Unit 8.1B.

Chan, E.T., Quon, G.T., Chua, G., Babak, T., Trochesset, M., Zirngibl, R.A., Aubin, J., Ratcliffe, M.J.H., Wilde, A., Brudno, M., et al. (2009). Conservation of core gene expression in vertebrate tissues. J. Biol. 8, 33.

Chaney, L.K., and Jacobson, B.S. (1983). Coating cells with colloidal silica for high yield isolation of plasma membrane sheets and identification of transmembrane proteins. J. Biol. Chem. 258, 10062–10072.

Chang, H., Brown, C.W., and Matzuk, M.M. (2002). Genetic analysis of the mammalian transforming growth factor-beta superfamily. Endocr. Rev. 23, 787–823.

Chapman, J.D., Goodlett, D.R., and Masselon, C.D. (2014). Multiplexed and data‐independent tandem mass spectrometry for global proteome profiling. Mass Spectrom. Rev. 33, 452–470.

Chen, M., and Manley, J.L. (2009). Mechanisms of alternative splicing regulation: insights from molecular and genomics approaches. Nat. Rev. Mol. Cell Biol. 10, 741–754.

Chen, B., Dodge, M.E., Tang, W., Lu, J., Ma, Z., Fan, C.-W., Wei, S., Hao, W., Kilgore, J., Williams, N.S., et al. (2009). Small molecule-mediated disruption of Wnt-dependent signaling in tissue regeneration and cancer. Nat. Chem. Biol. 5, 100–107.

Chen, H., Shi, S., Acosta, L., Li, W., Lu, J., Bao, S., Chen, Z., Yang, Z., Schneider, M.D., Chien, K.R., et al. (2004). BMP10 is essential for maintaining cardiac growth during murine cardiogenesis. Dev. Camb. Engl. 131, 2219–2231.

Chen, V.C., Couture, S.M., Ye, J., Lin, Z., Hua, G., Huang, H.-I.P., Wu, J., Hsu, D., Carpenter, M.K., and Couture, L.A. (2012). Scalable GMP compliant suspension culture system for human ES cells. Stem Cell Res. 8, 388–402.

Chen, V.C., Ye, J., Shukla, P., Hua, G., Chen, D., Lin, Z., Liu, J., Chai, J., Gold, J., Wu, J., et al. (2015). Development of a scalable suspension culture for cardiac differentiation from human pluripotent stem cells. Stem Cell Res. 15, 365–375.

Cheung, C., Bernardo, A.S., Trotter, M.W.B., Pedersen, R.A., and Sinha, S. (2012). Generation of human vascular smooth muscle subtypes provides insight into embryological origin-dependent disease susceptibility. Nat. Biotechnol. 30, 165–173.

Chiriac, A., Terzic, A., Park, S., Ikeda, Y., Faustino, R., and Nelson, T.J. (2010). SDF-1-Enhanced Cardiogenesis Requires CXCR4 Induction in Pluripotent Stem Cells. J Cardiovasc. Transl. Res. 3, 674–682.

Cho, H.-J., Lee, H.-J., Youn, S.-W., Koh, S.-J., Won, J.-Y., Chung, Y.-J., Cho, H.-J., Yoon, C.- H., Lee, S.-W., Lee, E.J., et al. (2012). Secondary sphere formation enhances the functionality of cardiac progenitor cells. Mol. Ther. J. Am. Soc. Gene Ther. 20, 1750–1766.

Clevers, H. (2004). Wnt Signaling: Ig-Norrin the Dogma. Curr. Biol. 14, R436–R437.

Cohen, E.D., Tian, Y., and Morrisey, E.E. (2008a). Wnt signaling: an essential regulator of cardiovascular differentiation, morphogenesis and progenitor self-renewal. Development 135, 789–798.

Cohen, E.D., Tian, Y., and Morrisey, E.E. (2008b). Wnt signaling: an essential regulator of cardiovascular differentiation, morphogenesis and progenitor self-renewal. Dev. Camb. Engl. 135, 789–798.

Cohen, E.D., Miller, M.F., Wang, Z., Moon, R.T., and Morrisey, E.E. (2012). Wnt5a and Wnt11 are essential for second heart field progenitor development. Dev. Camb. Engl. 139, 1931–1940.

Collier, A.J., Panula, S.P., Schell, J.P., Chovanec, P., Reyes, A.P., Petropoulos, S., Corcoran, A.E., Walker, R., Douagi, I., Lanner, F., et al. (2017). Comprehensive Cell Surface Protein Profiling Identifies Specific Markers of Human Naive and Primed Pluripotent States. Cell Stem Cell 20, 874–890.e7.

Conlon, F.L., Lyons, K.M., Takaesu, N., Barth, K.S., Kispert, A., Herrmann, B., and Robertson, E.J. (1994). A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Dev. Camb. Engl. 120, 1919–1928.

Cooper, T.A. (2005). Alternative Splicing Regulation Impacts Heart Development. Cell 120, 1–2.

Costello, I., Pimeisl, I.-M., Dräger, S., Bikoff, E.K., Robertson, E.J., and Arnold, S.J. (2011). The T-box transcription factor Eomesodermin acts upstream of Mesp1 to specify cardiac mesoderm during mouse gastrulation. Nat. Cell Biol. 13, 1084–1091.

Crossley, P.H., and Martin, G.R. (1995). The mouse Fgf8 gene encodes a family of polypeptides and is expressed in regions that direct outgrowth and patterning in the developing embryo. Dev. Camb. Engl. 121, 439–451.

Csősz, É., Emri, G., Kalló, G., Tsaprailis, G., and Tőzsér, J. (2015). Highly abundant defense proteins in human sweat as revealed by targeted proteomics and label-free quantification mass spectrometry. J. Eur. Acad. Dermatol. Venereol. JEADV 29, 2024–2031.

D’Amour, K.A., Agulnick, A.D., Eliazer, S., Kelly, O.G., Kroon, E., and Baetge, E.E. (2005). Efficient differentiation of human embryonic stem cells to definitive endoderm. Nat. Biotechnol. 23, 1534–1541.

David, R., Brenner, C., Stieber, J., Schwarz, F., Brunner, S., Vollmer, M., Mentele, E., Müller- Höcker, J., Kitajima, S., Lickert, H., et al. (2008). MesP1 drives vertebrate cardiovascular differentiation through Dkk-1-mediated blockade of Wnt-signalling. Nat. Cell Biol. 10, 338–345.

DeRossi, C., Laiosa, M.D., Silverstone, A.E., and Holdener, B.C. (2000). Mouse fzd4 maps within a region of chromosome 7 important for thymus and cardiac development. Genes. N. Y. N 2000 27, 64–75.

Desiderio, D.M., and Kai, M. (1983). Preparation of stable isotope-incorporated peptide internal standards for field desorption mass spectrometry quantification of peptides in biologic tissue. Biol. Mass Spectrom. 10, 471–479.

DeVeale, B., Bausch-Fluck, D., Seaberg, R., Runciman, S., Akbarian, V., Karpowicz, P., Yoon, C., Song, H., Leeder, R., Zandstra, P.W., et al. (2014). Surfaceome profiling reveals regulators of neural stem cell function. Stem Cells Dayt. Ohio 32, 258–68.

Diehl, S., Bruno, R., Wilkinson, G.A., Loose, D.A., Wilting, J., Schweigerer, L., and Klein, R. (2005). Altered expression patterns of EphrinB2 and EphB2 in human umbilical vessels and congenital venous malformations. Pediatr. Res. 57, 537–544. ten Dijke, P., Yamashita, H., Ichijo, H., Franzén, P., Laiho, M., Miyazono, K., and Heldin, C.H. (1994). Characterization of type I receptors for transforming growth factor-beta and activin. Science 264, 101–104.

Distler, U., Kuharev, J., Navarro, P., and Tenzer, S. (2016). Label-free quantification in ion mobility-enhanced data-independent acquisition proteomics. Nat. Protoc. 11, 795–812.

Dubois, N.C., Craft, A.M., Sharma, P., Elliott, D. a, Stanley, E.G., Elefanty, A.G., Gramolini, A., and Keller, G. (2011). SIRPA is a specific cell-surface marker for isolating cardiomyocytes derived from human pluripotent stem cells. Nat. Biotechnol. 29, 1011–8.

Dunn, N.R., Vincent, S.D., Oxburgh, L., Robertson, E.J., and Bikoff, E.K. (2004). Combinatorial activities of Smad2 and Smad3 regulate mesoderm formation and patterning in the mouse embryo. Dev. Camb. Engl. 131, 1717–1728.

Dunty, W.C., Biris, K.K., Chalamalasetty, R.B., Taketo, M.M., Lewandoski, M., and Yamaguchi, T.P. (2008). Wnt3a/beta-catenin signaling controls posterior body development by coordinating mesoderm formation and segmentation. Dev. Camb. Engl. 135, 85–94.

Durr, E., Yu, J., Krasinska, K.M., Carver, L.A., Yates, J.R., Testa, J.E., Oh, P., and Schnitzer, J.E. (2004). Direct proteomic mapping of the lung microvascular endothelial cell surface in vivo and in cell culture. Nat. Biotechnol. 22, 985–992.

Eaker, S., Armant, M., Brandwein, H., Burger, S., Campbell, A., Carpenito, C., Clarke, D., Fong, T., Karnieli, O., Niss, K., et al. (2013). Concise Review: Guidance in Developing Commercializable Autologous/Patient-Specific Cell Therapy Manufacturing. Stem Cells Transl. Med. 2, 871–883.

Edgar, R., Domrachev, M., and Lash, A.E. (2002). Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 30, 207–210.

Eichmann, A., Corbel, C., Nataf, V., Vaigot, P., Bréant, C., and Le Douarin, N.M. (1997). Ligand- dependent development of the endothelial and hemopoietic lineages from embryonic mesodermal cells expressing vascular endothelial growth factor receptor 2. Proc. Natl. Acad. Sci. U. S. A. 94, 5141–5146.

Eisenberg, L.M., and Eisenberg, C.A. (2006). Wnt signal transduction and the formation of the myocardium. Dev. Biol. 293, 305–315.

Elia, G. (2008). Biotinylation reagents for the study of cell surface proteins. Proteomics 8, 4012– 4024.

Elliott, D.A., Braam, S.R., Koutsis, K., Ng, E.S., Jenny, R., Lagerqvist, E.L., Biben, C., Hatzistavrou, T., Hirst, C.E., Yu, Q.C., et al. (2011). NKX2-5eGFP/w hPSCs for isolation of human cardiac progenitors and cardiomyocytes. Nat. Methods 8, 1037–1040.

Elortza, F., Nü Hse, T.S., Foster, L.J., Stensballe, A., Peck, S.C., and Jensen, O.N. (2003). Proteomic Analysis of Glycosylphosphatidylinositol-anchored Membrane Proteins*. Mol. Cell. Proteomics 2, 1261–1270.

Ema, M., Takahashi, S., and Rossant, J. (2006). Deletion of the selection cassette, but not cis- acting elements, in targeted Flk1-lacZ allele reveals Flk1 expression in multipotent mesodermal progenitors. Blood 107, 111–117.

Emig, D., Salomonis, N., Baumbach, J., Lengauer, T., Conklin, B.R., and Albrecht, M. (2010). AltAnalyze and DomainGraph: analyzing and visualizing exon expression data. Nucleic Acids Res. 38, W755–62.

Eng, J., McCormack, A., and Yates, J. (1994a). An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrom. 5, 976–989.

Eng, J.K., McCormack, A.L., and Yates, J.R. (1994b). An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrom. 5, 976–989.

Engleka, K.A., Manderfield, L.J., Brust, R.D., Li, L., Cohen, A., Dymecki, S.M., and Epstein, J.A. (2012). Islet1 Derivatives in the Heart Are of Both Neural Crest and Second Heart Field OriginNovelty and Significance. Circ. Res. 110, 922–926.

Fabre, B., Lambour, T., Bouyssié, D., Menneteau, T., Monsarrat, B., Burlet-Schiltz, O., and Bousquet-Dubouch, M.-P. (2014). Comparison of label-free quantification methods for the determination of protein complexes subunits stoichiometry. EuPA Open Proteomics 4, 82–86.

Fare, T.L., Coffey, E.M., Dai, H., He, Y.D., Kessler, D.A., Kilian, K.A., Koch, J.E., LeProust, E., Marton, M.J., Meyer, M.R., et al. (2003). Effects of atmospheric ozone on microarray data quality. Anal. Chem. 75, 4672–4675.

Faustino, R.S., and Terzic, A. (2008). Interactome of a cardiopoietic precursor. J Cardiovasc. Transl. Res. 1, 120–6.

Faustino, R., Behfar, A., Perez-Terzic, C., and Terzic, A. (2008). Genomic chart guiding embryonic stem cell cardiopoiesis. Genome Biol. 9, R6.

Fazzari, P., Penachioni, J., Gianola, S., Rossi, F., Eickholt, B.J., Maina, F., Alexopoulou, L., Sottile, A., Comoglio, P.M., Flavell, R.A., et al. (2007). Plexin-B1 plays a redundant role during mouse development and in tumour angiogenesis. BMC Dev. Biol. 7, 55.

Fehling, H.J. (2003). Tracking mesoderm induction and its specification to the hemangioblast during embryonic stem cell differentiation. Development 130, 4217–4227.

Feng, Y., Valley, M.T., Lazar, J., Yang, A.L., Bronson, R.T., Firestein, S., Coetzee, W.A., and Manley, J.L. (2009). SRp38 regulates alternative splicing and is required for Ca2+ handling in the embryonic heart. Dev. Cell 16, 528–538.

Fenn, J.B., Mann, M., Meng, C.K., Wong, S.F., and Whitehouse, C.M. (1989). Electrospray ionization for mass spectrometry of large biomolecules. Science 246, 926.

Fernandes, S., Chong, J.J.H., Paige, S.L., Iwata, M., Torok-Storb, B., Keller, G., Reinecke, H., and Murry, C.E. (2015). Comparison of Human Embryonic Stem Cell-Derived Cardiomyocytes, Cardiovascular Progenitors, and Bone Marrow Mononuclear Cells for Cardiac Repair. Stem Cell Rep. 5, 753–762.

Fischer, A., and Gessler, M. (2003). Hey genes in cardiovascular development. Trends Cardiovasc. Med. 13, 221–226.

Fisher, S.A., Doree, C., Mathur, A., and Martin-Rendon, E. (2015). Meta-Analysis of Cell Therapy Trials for Patients With Heart FailureNovelty and Significance. Circ. Res. 116, 1361–1377.

Fletcher, R.B., Baker, J.C., and Harland, R.M. (2006). FGF8 spliceforms mediate early mesoderm and posterior neural tissue formation in Xenopus. Dev. Camb. Engl. 133, 1703–1714.

Foley, A.C., Korol, O., Timmer, A.M., and Mercola, M. (2007). Multiple functions of Cerberus cooperate to induce heart downstream of Nodal. Dev. Biol. 303, 57–65.

Frank, J.A., Miller, B.R., Arbab, A.S., Zywicke, H.A., Jordan, E.K., Lewis, B.K., Bryant, L.H., and Bulte, J.W.M. (2003). Clinically applicable labeling of mammalian and stem cells by combining superparamagnetic iron oxides and transfection agents. Radiology 228, 480–487.

Gale, N.W., Holland, S.J., Valenzuela, D.M., Flenniken, A., Pan, L., Ryan, T.E., Henkemeyer, M., Strebhardt, K., Hirai, H., Wilkinson, D.G., et al. (1996). Eph receptors and ligands comprise two major specificity subclasses and are reciprocally compartmentalized during embryogenesis. Neuron 17, 9–19.

Gardina, P.J., Clark, T.A., Shimada, B., Staples, M.K., Yang, Q., Veitch, J., Schweitzer, A., Awad, T., Sugnet, C., Dee, S., et al. (2006). Alternative splicing and differential gene expression in colon cancer detected by a whole genome exon array. BMC Genomics 7, 325.

Garot, J., Unterseeh, T., Teiger, E., Champagne, S., Chazaud, B., Gherardi, R., Hittinger, L., Guéret, P., and Rahmouni, A. (2003). Magnetic resonance imaging of targeted catheter-based implantation of myogenic precursor cells into infarcted left ventricular myocardium. J. Am. Coll. Cardiol. 41, 1841–1846.

Gassmann, M., Casagranda, F., Orioli, D., Simon, H., Lai, C., Klein, R., and Lemke, G. (1995). Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor. Nature 378, 390–394.

Gauthier, D.J., Gibbs, B.F., Rabah, N., and Lazure, C. (2004). Utilization of a new biotinylation reagent in the development of a nondiscriminatory investigative approach for the study of cell surface proteins. Proteomics 4, 3783–3790.

Gemel, J., Gorry, M., Ehrlich, G.D., and MacArthur, C.A. (1996). Structure and sequence of human FGF8. Genomics 35, 253–257.

Gerber, S.A., Rush, J., Stemman, O., Kirschner, M.W., and Gygi, S.P. (2003). Absolute quantification of proteins and phosphoproteins from cell lysates by tandem MS. Proc. Natl. Acad. Sci. 100, 6940–6945.

Gerecht-Nir, S., Cohen, S., and Itskovitz-Eldor, J. (2004a). Bioreactor cultivation enhances the efficiency of human embryoid body (hEB) formation and differentiation. Biotechnol. Bioeng. 86, 493–502.

Gerecht-Nir, S., Cohen, S., and Itskovitz-Eldor, J. (2004b). Bioreactor cultivation enhances the efficiency of human embryoid body (hEB) formation and differentiation. Biotechnol Bioeng 86, 493–502.

Gessert, S., and Kühl, M. (2010a). The multiple phases and faces of Wnt signaling during cardiac differentiation and development. Circ. Res. 107, 186–199.

Gessert, S., and Kühl, M. (2010b). The Multiple Phases and Faces of Wnt Signaling During Cardiac Differentiation and Development. Circ. Res. 107, 186–199.

Giles, W.R., and Imaizumi, Y. (1988). Comparison of potassium currents in rabbit atrial and ventricular cells. J. Physiol. 405, 123–145.

Glinka, A., Wu, W., Delius, H., Monaghan, A.P., Blumenstock, C., and Niehrs, C. (1998). Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391, 357–362. de Godoy, L.M.F., Olsen, J.V., Cox, J., Nielsen, M.L., Hubner, N.C., Fröhlich, F., Walther, T.C., and Mann, M. (2008). Comprehensive mass-spectrometry-based proteome quantification of haploid versus diploid yeast. Nature 455, 1251–1254.

Golpanian, S., Wolf, A., Hatzistergos, K.E., and Hare, J.M. (2016). Rebuilding the Damaged Heart: Mesenchymal Stem Cells, Cell-Based Therapy, and Engineered Heart Tissue. Physiol. Rev. 96, 1127–1168.

Golub, T.R., Slonim, D.K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J.P., Coller, H., Loh, M.L., Downing, J.R., Caligiuri, M.A., et al. (1999). Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286, 531–537.

Gordon, M.D., and Nusse, R. (2006). Wnt Signaling: Multiple Pathways, Multiple Receptors, and Multiple Transcription Factors. J. Biol. Chem. 281, 22429–22433.

Goumans, M.J., and Mummery, C. (2000). Functional analysis of the TGFbeta receptor/Smad pathway through gene ablation in mice. Int. J. Dev. Biol. 44, 253–265.

Grabherr, M.G., Haas, B.J., Yassour, M., Levin, J.Z., Thompson, D.A., Amit, I., Adiconis, X., Fan, L., Raychowdhury, R., Zeng, Q., et al. (2011). Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 29, 644–652.

Green, J.B., New, H.V., and Smith, J.C. (1992). Responses of embryonic Xenopus cells to activin and FGF are separated by multiple dose thresholds and correspond to distinct axes of the mesoderm. Cell 71, 731–739.

Griffin, N.M., Yu, J., Long, F., Oh, P., Shore, S., Li, Y., Koziol, J.A., and Schnitzer, J.E. (2010). Label-free, normalized quantification of complex mass spectrometry data for proteomic analysis. Nat. Biotechnol. 28, 83–89.

Grigoropoulos, N.F., and Mathur, A. (2006). Stem cells in cardiac repair. Curr. Opin. Pharmacol. 6, 169–175.

Gundry, R.L., Raginski, K., Tarasova, Y., Tchernyshyov, I., Bausch-Fluck, D., Elliott, S.T., Boheler, K.R., Eyk, J.E.V., and Wollscheid, B. (2009). The Mouse C2C12 Myoblast Cell Surface N-Linked Glycoproteome IDENTIFICATION, GLYCOSITE OCCUPANCY, AND MEMBRANE ORIENTATION. Mol. Cell. Proteomics 8, 2555–2569.

Guo, Q., and Li, J.Y.H. (2007). Distinct functions of the major Fgf8 spliceform, Fgf8b, before and during mouse gastrulation. Dev. Camb. Engl. 134, 2251–2260.

Gygi, S.P., Rist, B., Gerber, S.A., Turecek, F., Gelb, M.H., and Aebersold, R. (1999). Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 17, 994– 999.

Gyöngyösi, M., Wojakowski, W., Lemarchand, P., Lunde, K., Tendera, M., Bartunek, J., Marban, E., Assmus, B., Henry, T.D., Traverse, J.H., et al. (2015). Meta-Analysis of Cell-based CaRdiac stUdiEs (ACCRUE) in Patients With Acute Myocardial Infarction Based on Individual Patient DataNovelty and Significance. Circ. Res. 116, 1346–1360.

Gyöngyösi, M., Wojakowski, W., Navarese, E.P., and Moye, L.À. (2016). Meta-Analyses of Human Cell-Based Cardiac Regeneration TherapiesResponse to Gyöngyösi, Wojakowski, Navarese, Moyé, and the ACCRUE Investigators: Controversies in Meta-Analyses Results on Cardiac Cell-Based Regenerative Studies. Circ. Res. 118, 1254–1263.

Hartogh, S.C. den, Wolstencroft, K., Mummery, C.L., and Passier, R. (2016). A comprehensive gene expression analysis at sequential stages of in vitro cardiac differentiation from isolated MESP1-expressing-mesoderm progenitors. Sci. Rep. 6, 19386.

Harvey, R. (2002). Organogenesis: Patterning the vertebrate heart. Nat. Rev. Genet. 3, 544–556.

Hatzistergos, K.E., and Vedenko, A. (2017). Cardiac Cell Therapy 3.0: The Beginning of the End or the End of the Beginning? Circ. Res. 121, 95–97.

He, A., Kong, S.W., Ma, Q., and Pu, W.T. (2011). Co-occupancy by multiple cardiac transcription factors identifies transcriptional enhancers active in heart. Proc. Natl. Acad. Sci. 108, 5632–5637.

He, J.-Q., Ma, Y., Lee, Y., Thomson, J.A., and Kamp, T.J. (2003). Human embryonic stem cells develop into multiple types of cardiac myocytes: action potential characterization. Circ. Res. 93, 32–39.

100

He, X., Semenov, M., Tamai, K., and Zeng, X. (2004). LDL receptor-related proteins 5 and 6 in Wnt/beta-catenin signaling: arrows point the way. Dev. Camb. Engl. 131, 1663–1677.

Hebert, A.S., Richards, A.L., Bailey, D.J., Ulbrich, A., Coughlin, E.E., Westphall, M.S., and Coon, J.J. (2014). The One Hour Yeast Proteome. Mol. Cell. Proteomics 13, 339–347.

Held, G.A., Grinstein, G., and Tu, Y. (2006). Relationship between gene expression and observed intensities in DNA microarrays--a modeling study. Nucleic Acids Res. 34, e70.

Hendrickx, M., and Leyns, L. (2008). Non-conventional Frizzled ligands and Wnt receptors. Dev. Growth Differ. 50, 229–243.

Henzel, W.J., Billeci, T.M., Stults, J.T., Wong, S.C., Grimley, C., and Watanabe, C. (1993). Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc. Natl. Acad. Sci. U. S. A. 90, 5011–5015.

Hillenkamp, F., Karas, M., Beavis, R.C., and Chait, B.T. (1991). Matrix-assisted laser desorption/ionization mass spectrometry of biopolymers. Anal. Chem. 63, 1193A–1203A.

Hoang, B.H., Thomas, J.T., Abdul-Karim, F.W., Correia, K.M., Conlon, R.A., Luyten, F.P., and Ballock, R.T. (1998). Expression pattern of two Frizzled-related genes, Frzb-1 and Sfrp-1, during mouse embryogenesis suggests a role for modulating action of Wnt family members. Dev. Dyn. Off. Publ. Am. Assoc. Anat. 212, 364–372.

Hu, T., Yamagishi, H., Maeda, J., McAnally, J., Yamagishi, C., and Srivastava, D. (2004). Tbx1 regulates fibroblast growth factors in the anterior heart field through a reinforcing autoregulatory loop involving forkhead transcription factors. Dev. Camb. Engl. 131, 5491–5502.

Huang, Q., Yang, L., Luo, J., Guo, L., Wang, Z., Yang, X., Jin, W., Fang, Y., Ye, J., Shan, B., et al. (2015). SWATH enables precise label-free quantification on proteome scale. Proteomics 15, 1215–1223.

Huang, S.-M.A., Mishina, Y.M., Liu, S., Cheung, A., Stegmeier, F., Michaud, G.A., Charlat, O., Wiellette, E., Zhang, Y., Wiessner, S., et al. (2009). Tankyrase inhibition stabilizes axin and antagonizes Wnt signalling. Nature 461, 614–620.

Huber, L.A., Pfaller, K., and Vietor, I. (2003). Organelle Proteomics. Circ. Res. 92, 962–968.

Huelsken, J., Vogel, R., Brinkmann, V., Erdmann, B., Birchmeier, C., and Birchmeier, W. (2000). Requirement for beta-catenin in anterior-posterior axis formation in mice. J. Cell Biol. 148, 567– 578.

Hunt, D.F., Henderson, R.A., Shabanowitz, J., Sakaguchi, K., Michel, H., Sevilir, N., Cox, A.L., Appella, E., and Engelhard, V.H. (1992). Characterization of peptides bound to the class I MHC molecule HLA-A2.1 by mass spectrometry. Science 255, 1261–1263.

Hussain, W., Moens, N., Veraitch, F.S., Hernandez, D., Mason, C., and Lye, G.J. (2013). Reproducible culture and differentiation of mouse embryonic stem cells using an automated microwell platform. Biochem. Eng. J. 77, 246–257.

101

Ieda, M., Tsuchihashi, T., Ivey, K.N., Ross, R.S., Hong, T.-T., Shaw, R.M., and Srivastava, D. (2009). Cardiac fibroblasts regulate myocardial proliferation through beta1 integrin signaling. Dev. Cell 16, 233–244.

Irizarry, R.A., Warren, D., Spencer, F., Kim, I.F., Biswal, S., Frank, B.C., Gabrielson, E., Garcia, J.G.N., Geoghegan, J., Germino, G., et al. (2005). Multiple-laboratory comparison of microarray platforms. Nat. Methods 2, 345–350.

Ishida, H., Saba, R., Kokkinopoulos, I., Hashimoto, M., Yamaguchi, O., Nowotschin, S., Shiraishi, M., Ruchaya, P., Miller, D., Harmer, S., et al. (2016). GFRA2 Identifies Cardiac Progenitors and Mediates Cardiomyocyte Differentiation in a RET-Independent Signaling Pathway. Cell Rep. 16, 1026–1038.

Ito, M., Liu, Y., Yang, Z., Nguyen, J., Liang, F., Morris, R.J., and Cotsarelis, G. (2005). Stem cells in the hair follicle bulge contribute to wound repair but not to homeostasis of the epidermis. Nat. Med. 11, 1351–1354.

Iyer, D., Gambardella, L., Bernard, W.G., Serrano, F., Mascetti, V.L., Pedersen, R.A., Talasila, A., and Sinha, S. (2015). Robust derivation of epicardium and its differentiated smooth muscle cell progeny from human pluripotent stem cells. Dev. Camb. Engl. 142, 1528–1541.

James, P., Quadroni, M., Carafoli, E., and Gonnet, G. (1993). Protein identification by mass profile fingerprinting. Biochem. Biophys. Res. Commun. 195, 58–64.

Jerome, L.A., and Papaioannou, V.E. (2001). DiGeorge syndrome phenotype in mice mutant for the T-box gene, Tbx1. Nat. Genet. 27, 286–291.

Josic, D., and Clifton, J.G. (2007). Mammalian plasma membrane proteomics. Proteomics 7, 3010–3029.

Kai, T., Williams, D., and Spradling, A.C. (2005). The expression profile of purified Drosophila germline stem cells. Dev. Biol. 283, 486–502.

Kaji, H., Yamauchi, Y., Takahashi, N., and Isobe, T. (2006). Mass spectrometric identification of N-linked glycopeptides using lectin-mediated affinity capture and glycosylation site-specific stable isotope tagging. Nat. Protoc. 1, 3019–3027.

Kalxdorf, M., Gade, S., Eberl, H.C., and Bantscheff, M. (2017). Monitoring Cell-surface N- Glycoproteome Dynamics by Quantitative Proteomics Reveals Mechanistic Insights into Macrophage Differentiation. Mol. Cell. Proteomics MCP 16, 770–785.

Kataoka, H., Takakura, N., Nishikawa, S., Tsuchida, K., Kodama, H., Kunisada, T., Risau, W., Kita, T., and Nishikawa, S.-I. (1997). Expressions of PDGF receptor alpha, c-Kit and Flk1 genes clustering in mouse chromosome 5 define distinct subsets of nascent mesodermal cells. Dev. Growth Differ. 39, 729–740.

Kattman, S., Huber, T., and Keller, G. (2006). Multipotent Flk-1+ Cardiovascular Progenitor Cells Give Rise to the Cardiomyocyte, Endothelial, and Vascular Smooth Muscle Lineages. Dev. Cell 11, 723–732.

102

Kattman, S.J., Witty, A.D., Gagliardi, M., Dubois, N.C., Niapour, M., Hotta, A., Ellis, J., and Keller, G. (2011). Stage-specific optimization of activin/nodal and BMP signaling promotes cardiac differentiation of mouse and human pluripotent stem cell lines. Cell Stem Cell 8, 228–240.

Kehoe, D.E., Lock, L.T., Parikh, A., and Tzanakakis, E.S. (2008). Propagation of embryonic stem cells in stirred suspension without serum. Biotechnol. Prog. 24, 1342–1352.

Keller, A., Eng, J., Zhang, N., Li, X., and Aebersold, R. (2005). A uniform proteomics MS/MS analysis platform utilizing open XML file formats. Mol. Syst. Biol. 1, 2005.0017.

Kelly, R.G., and Buckingham, M.E. (2002). The anterior heart-forming field: Voyage to the arterial pole of the heart. Trends Genet. 18, 210–216.

Kinder, S.J., Tsang, T.E., Quinlan, G.A., Hadjantonakis, A.K., Nagy, A., and Tam, P.P. (1999). The orderly allocation of mesodermal cells to the extraembryonic structures and the anteroposterior axis during gastrulation of the mouse embryo. Development 126, 4691–4701.

Kislinger, T., Gramolini, A.O., MacLennan, D.H., and Emili, A. (2005). Multidimensional Protein Identification Technology (MudPIT): Technical Overview of a Profiling Method Optimized for the Comprehensive Proteomic Investigation of Normal and Diseased Heart Tissue. J. Am. Soc. Mass Spectrom. 16, 1207–1220.

Klaus, A., Saga, Y., Taketo, M.M., Tzahor, E., and Birchmeier, W. (2007). Distinct roles of Wnt/β- catenin and Bmp signaling during early cardiogenesis. Proc. Natl. Acad. Sci. 104, 18531–18536.

Kohn, A.D., and Moon, R.T. (2005). Wnt and calcium signaling: β-Catenin-independent pathways. Cell Calcium 38, 439–446.

Kokubo, H., Miyagawa-Tomita, S., and Johnson, R.L. (2005). Hesr, a mediator of the Notch signaling, functions in heart and vessel development. Trends Cardiovasc. Med. 15, 190–194.

Kokubo, H., Tomita-Miyagawa, S., Hamada, Y., and Saga, Y. (2007). Hesr1 and Hesr2 regulate atrioventricular boundary formation in the developing heart through the repression of Tbx2. Dev. Camb. Engl. 134, 747–755.

Kowalski, M.P., Yoder, A., Liu, L., and Pajak, L. (2012). Controlling embryonic stem cell growth and differentiation by automation: enhanced and more reliable differentiation for drug discovery. J. Biomol. Screen. 17, 1171–1179.

Krawetz, R., Taiani, J.T., Liu, S., Meng, G., Li, X., Kallos, M.S., and Rancourt, D.E. (2010). Large-scale expansion of pluripotent human embryonic stem cells in stirred-suspension bioreactors. Tissue Eng. Part C Methods 16, 573–582.

Kropp, C., Kempf, H., Halloin, C., Robles-Diaz, D., Franke, A., Scheper, T., Kinast, K., Knorpp, T., Joos, T.O., Haverich, A., et al. (2016). Impact of Feeding Strategies on the Scalable Expansion of Human Pluripotent Stem Cells in Single-Use Stirred Tank Bioreactors. Stem Cells Transl. Med. 5, 1289–1301.

103

Kubalak, S.W., Miller-Hance, W.C., O’Brien, T.X., Dyson, E., and Chien, K.R. (1994). Chamber specification of atrial myosin light chain-2 expression precedes septation during murine cardiogenesis. J. Biol. Chem. 269, 16961–16970.

Kulak, N.A., Pichler, G., Paron, I., Nagaraj, N., and Mann, M. (2014). Minimal, encapsulated proteomic-sample processing applied to copy-number estimation in eukaryotic cells. Nat. Methods 11, 319–324.

Kuo, C.T., Morrisey, E.E., Anandappa, R., Sigrist, K., Lu, M.M., Parmacek, M.S., Soudais, C., and Leiden, J.M. (1997). GATA4 transcription factor is required for ventral morphogenesis and heart tube formation. Genes Dev. 11, 1048–1060.

Kwon, C., Arnold, J., Hsiao, E.C., Taketo, M.M., Conklin, B.R., and Srivastava, D. (2007). Canonical Wnt signaling is a positive regulator of mammalian cardiac progenitors. Proc. Natl. Acad. Sci. 104, 10894–10899. van Laake, L.W., Passier, R., Monshouwer-Kloots, J., Verkleij, A.J., Lips, D.J., Freund, C., den Ouden, K., Ward-van Oostwaard, D., Korving, J., Tertoolen, L.G., et al. (2007). Human embryonic stem cell-derived cardiomyocytes survive and mature in the mouse heart and transiently improve function after myocardial infarction. Stem Cell Res. 1, 9–24.

Laflamme, M.A., Chen, K.Y., Naumova, A.V., Muskheli, V., Fugate, J.A., Dupras, S.K., Reinecke, H., Xu, C., Hassanipour, M., Police, S., et al. (2007). Cardiomyocytes derived from human embryonic stem cells in pro-survival factors enhance function of infarcted rat hearts. Nat. Biotechnol. 25, 1015–1024.

Lange, V., Picotti, P., Domon, B., and Aebersold, R. (2008). Selected reaction monitoring for quantitative proteomics: a tutorial. Mol. Syst. Biol. 4, 222.

Lavery, D.L., Martin, J., Turnbull, Y.D., and Hoppler, S. (2008). Wnt6 signaling regulates heart muscle development during organogenesis. Dev. Biol. 323, 177–188.

Lee, H.H., and Frasch, M. (2000). Wingless effects mesoderm patterning and ectoderm segmentation events via induction of its downstream target sloppy paired. Dev. Camb. Engl. 127, 5497–5508.

Lee, J.H., Protze, S.I., Laksman, Z., Backx, P.H., and Keller, G.M. (2017). Human Pluripotent Stem Cell-Derived Atrial and Ventricular Cardiomyocytes Develop from Distinct Mesoderm Populations. Cell Stem Cell 21, 179–194.e4.

Lekven, A.C., Thorpe, C.J., Waxman, J.S., and Moon, R.T. (2001). Zebrafish wnt8 encodes two wnt8 proteins on a bicistronic transcript and is required for mesoderm and neurectoderm patterning. Dev. Cell 1, 103–114.

Lewandrowski, U. (2005). Elucidation of N-Glycosylation Sites on Human Platelet Proteins: A Glycoproteomic Approach. Mol. Cell. Proteomics 5, 226–233.

Li, T.-S., Cheng, K., Malliaras, K., Smith, R.R., Zhang, Y., Sun, B., Matsushita, N., Blusztajn, A., Terrovitis, J., Kusuoka, H., et al. (2012). Direct comparison of different stem cell types and

104 subpopulations reveals superior paracrine potency and myocardial repair efficacy with cardiosphere-derived cells. J. Am. Coll. Cardiol. 59, 942–953.

Lian, X., Zhang, J., Azarin, S.M., Zhu, K., Hazeltine, L.B., Bao, X., Hsiao, C., Kamp, T.J., and Palecek, S.P. (2013). Directed cardiomyocyte differentiation from human pluripotent stem cells by modulating Wnt/β-catenin signaling under fully defined conditions. Nat. Protoc. 8, 162–175.

Lickert, H., Kutsch, S., Kanzler, B., Tamai, Y., Taketo, M.M., and Kemler, R. (2002). Formation of multiple hearts in mice following deletion of beta-catenin in the embryonic endoderm. Dev. Cell 3, 171–181.

Lindsley, R.C., Gill, J.G., Kyba, M., Murphy, T.L., and Murphy, K.M. (2006). Canonical Wnt signaling is required for development of embryonic stem cell-derived mesoderm. Development 133, 3787–3796.

Lindsley, R.C., Gill, J.G., Murphy, T.L., Langer, E.M., Cai, M., Mashayekhi, M., Wang, W., Niwa, N., Nerbonne, J.M., Kyba, M., et al. (2008). Mesp1 coordinately regulates cardiovascular fate restriction and epithelial-mesenchymal transition in differentiating PSCs. Cell Stem Cell 3, 55–68.

Liu, F., Ventura, F., Doody, J., and Massagué, J. (1995). Human type II receptor for bone morphogenic proteins (BMPs): extension of the two-kinase receptor model to the BMPs. Mol. Cell. Biol. 15, 3479–3486.

Liu, H., Sadygov, R.G., and Yates, J.R. (2004a). A Model for Random Sampling and Estimation of Relative Protein Abundance in Shotgun Proteomics. Anal. Chem. 76, 4193–4201.

Liu, P., Wakamiya, M., Shea, M.J., Albrecht, U., Behringer, R.R., and Bradley, A. (1999). Requirement for Wnt3 in vertebrate axis formation. Nat. Genet. 22, 361–365.

Liu, W., Selever, J., Wang, D., Lu, M.-F., Moses, K.A., Schwartz, R.J., and Martin, J.F. (2004b). Bmp4 signaling is required for outflow-tract septation and branchial-arch artery remodeling. Proc. Natl. Acad. Sci. U. S. A. 101, 4489–4494.

Liu, Y., Asakura, M., Inoue, H., Nakamura, T., Sano, M., Niu, Z., Chen, M., Schwartz, R.J., and Schneider, M.D. (2007). Sox17 is essential for the specification of cardiac mesoderm in embryonic stem cells. Proc. Natl. Acad. Sci. U. S. A. 104, 3859–3864.

Lombard-Banek, C., Reddy, S., Moody, S.A., and Nemes, P. (2016). Label-free Quantification of Proteins in Single Embryonic Cells with Neural Fate in the Cleavage-Stage Frog (Xenopus laevis) Embryo using CE-ESI-HRMS. Mol. Cell. Proteomics mcp.M115.057760.

Lorbeer, R.-A., Heidrich, M., Lorbeer, C., Ojeda, D.F.R., Bicker, G., Meyer, H., and Heisterkamp, A. (2011). Highly efficient 3D fluorescence microscopy with a scanning laser optical tomograph. Opt. Express 19, 5419–5430.

Lough, J., Barron, M., Brogley, M., Sugi, Y., Bolender, D.L., and Zhu, X. (1996). Combined BMP- 2 and FGF-4, but neither factor alone, induces cardiogenesis in non-precardiac embryonic mesoderm. Dev. Biol. 178, 198–202.

105

Luo, J., Schumacher, M., Scherer, A., Sanoudou, D., Megherbi, D., Davison, T., Shi, T., Tong, W., Shi, L., Hong, H., et al. (2010). A comparison of batch effect removal methods for enhancement of prediction performance using MAQC-II microarray gene expression data. Pharmacogenomics J. 10, 278–291.

Lyons, I., Parsons, L.M., Hartley, L., Li, R., Andrews, J.E., Robb, L., and Harvey, R.P. (1995). Myogenic and morphogenetic defects in the heart tubes of murine embryos lacking the homeo box gene Nkx2-5. Genes Dev. 9, 1654–1666.

Macher, B.A., and Yen, T.-Y. (2007). Proteins at membrane surfaces—a review of approaches. Mol. Biosyst. 3, 705–713.

Mackay, J., Mensah, G. A., Mendis, S., & Greenlund, K. (2004). The Atlas of Heart Disease and Stroke. World Health Organization.

Maesner, C.C., Almada, A.E., and Wagers, A.J. (2016). Established cell surface markers efficiently isolate highly overlapping populations of skeletal muscle satellite cells by fluorescence- activated cell sorting. Skelet. Muscle 6, 35.

Malliaras, K., Makkar, R.R., Smith, R.R., Cheng, K., Wu, E., Bonow, R.O., Marbán, L., Mendizabal, A., Cingolani, E., Johnston, P.V., et al. (2014). Intracoronary cardiosphere-derived cells after myocardial infarction: evidence of therapeutic regeneration in the final 1-year results of the CADUCEUS trial (CArdiosphere-Derived aUtologous stem CElls to reverse ventricUlar dySfunction). J. Am. Coll. Cardiol. 63, 110–122.

Maltsev, V.A., Rohwedel, J., Hescheler, J., and Wobus, A.M. (1993). Embryonic stem cells differentiate in vitro into cardiomyocytes representing sinusnodal, atrial and ventricular cell types. Mech. Dev. 44, 41–50.

Mann, M., Højrup, P., and Roepstorff, P. (1993). Use of mass spectrometric molecular weight information to identify proteins in sequence databases. Biol. Mass Spectrom. 22, 338–345.

MAQC Consortium, Shi, L., Reid, L.H., Jones, W.D., Shippy, R., Warrington, J.A., Baker, S.C., Collins, P.J., de Longueville, F., Kawasaki, E.S., et al. (2006). The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat. Biotechnol. 24, 1151–1161.

Mardis, E.R. (2008). Next-generation DNA sequencing methods. Annu. Rev. Genomics Hum. Genet. 9, 387–402.

Marvin, M.J., Rocco, G.D., Gardiner, A., Bush, S.M., and Lassar, A.B. (2001). Inhibition of Wnt activity induces heart formation from posterior mesoderm. Genes Dev. 15, 316–327.

Mauritz, C., Martens, A., Rojas, S.V., Schnick, T., Rathert, C., Schecker, N., Menke, S., Glage, S., Zweigerdt, R., Haverich, A., et al. (2011). Induced pluripotent stem cell (iPSC)-derived Flk-1 progenitor cells engraft, differentiate, and improve heart function in a mouse model of acute myocardial infarction. Eur. Heart J. 32, 2634–2641.

106

Mazzotta, S., Neves, C., Bonner, R.J., Bernardo, A.S., Docherty, K., and Hoppler, S. (2016). Distinctive Roles of Canonical and Noncanonical Wnt Signaling in Human Embryonic Cardiomyocyte Development. Stem Cell Rep. 7, 764–776.

Melo, L.G., Pachori, A.S., Kong, D., Gnecchi, M., Wang, K., Pratt, R.E., and Dzau, V.J. (2004). Gene and cell-based therapies for heart disease. FASEB J. Off. Publ. Fed. Am. Soc. Exp. Biol. 18, 648–663.

Menasché, P., Vanneaux, V., Hagège, A., Bel, A., Cholley, B., Cacciapuoti, I., Parouchev, A., Benhamouda, N., Tachdjian, G., Tosca, L., et al. (2015). Human embryonic stem cell-derived cardiac progenitors for severe heart failure treatment: first clinical case report. Eur. Heart J. 36, 2011–2017.

Mesnard, D., Guzman-Ayala, M., and Constam, D.B. (2006). Nodal specifies embryonic visceral endoderm and sustains pluripotent cells in the epiblast before overt axial patterning. Dev. Camb. Engl. 133, 2497–2505.

Messina, E., Angelis, L.D., Frati, G., Morrone, S., Chimenti, S., Fiordaliso, F., Salio, M., Battaglia, M., Latronico, M.V.G., Coletta, M., et al. (2004). Isolation and Expansion of Adult Cardiac Stem Cells From Human and Murine Heart. Circ. Res. 95, 911–921.

Mishina, Y., Suzuki, A., Ueno, N., and Behringer, R.R. (1995). Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis. Genes Dev. 9, 3027–3037.

Molkentin, J.D., Lin, Q., Duncan, S.A., and Olson, E.N. (1997). Requirement of the transcription factor GATA4 for heart tube formation and ventral morphogenesis. Genes Dev. 11, 1061–1072.

Moon, R.T., and Gough, N.R. (2016). Beyond canonical: The Wnt and β-catenin story. Sci Signal 9, eg5-eg5.

Moon, R.T., Kohn, A.D., Ferrari, G.V.D., and Kaykas, A. (2004). WNT and β-catenin signalling: diseases and therapies. Nat. Rev. Genet. 5, 691–701.

Moretti, A., Caron, L., Nakano, A., Lam, J.T., Bernshausen, A., Chen, Y., Qyang, Y., Bu, L., Sasaki, M., Martin-Puig, S., et al. (2006). Multipotent Embryonic Isl1+ Progenitor Cells Lead to Cardiac, Smooth Muscle, and Endothelial Cell Diversification. Cell 127, 1151–1165.

Mortazavi, A., Williams, B.A., McCue, K., Schaeffer, L., and Wold, B. (2008). Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat. Methods 5, 621–628.

Motoike, T., Markham, D.W., Rossant, J., and Sato, T.N. (2003). Evidence for novel fate of Flk1+ progenitor: contribution to muscle lineage. Genes. N. Y. N 2000 35, 153–159.

Mummery, C.L. (2005). Cardiology: solace for the broken-hearted? Nature 433, 585–587.

Munoz, J., Low, T.Y., Kok, Y.J., Chin, A., Frese, C.K., Ding, V., Choo, A., and Heck, A.J.R. (2011). The quantitative proteomes of human‐induced pluripotent stem cells and embryonic stem cells. Mol. Syst. Biol. 7, 550.

107

Nagalakshmi, U., Wang, Z., Waern, K., Shou, C., Raha, D., Gerstein, M., and Snyder, M. (2008). The transcriptional landscape of the yeast genome defined by RNA sequencing. Science 320, 1344–1349.

Nagano, K., Taoka, M., Yamauchi, Y., Itagaki, C., Shinkawa, T., Nunomura, K., Okamura, N., Takahashi, N., Izumi, T., and Isobe, T. (2005). Large-scale identification of proteins expressed in mouse embryonic stem cells. Proteomics 5, 1346–1361.

Nagaraj, N., Wisniewski, J.R., Geiger, T., Cox, J., Kircher, M., Kelso, J., Pääbo, S., and Mann, M. (2011). Deep proteome and transcriptome mapping of a human cancer cell line. Mol. Syst. Biol. 7, 548.

Naito, A.T., Shiojima, I., Akazawa, H., Hidaka, K., Morisaki, T., Kikuchi, A., and Komuro, I. (2006). Developmental stage-specific biphasic roles of Wnt/β-catenin signaling in cardiomyogenesis and hematopoiesis. Proc. Natl. Acad. Sci. 103, 19812–19817.

Nakagawa, O., Nakagawa, M., Richardson, J.A., Olson, E.N., and Srivastava, D. (1999). HRT1, HRT2, and HRT3: a new subclass of bHLH transcription factors marking specific cardiac, somitic, and pharyngeal arch segments. Dev. Biol. 216, 72–84.

Nelson, T., Faustino, R., Chiriac, A., Crespo-Diaz, R., Behfar, A., and Terzic, A. (2008). CXCR4+/FLK-1+ Biomarkers Select a Cardiopoietic Lineage from Embryonic Stem Cells. Stem Cells 26, 1464–1473.

Niebruegge, S., Nehring, A., Bär, H., Schroeder, M., Zweigerdt, R., and Lehmann, J. (2008). Cardiomyocyte production in mass suspension culture: embryonic stem cells as a source for great amounts of functional cardiomyocytes. Tissue Eng. Part A 14, 1591–1601.

Nijmeijer, R.M., Leeuwis, J.W., DeLisio, A., Mummery, C.L., and Chuva de Sousa Lopes, S.M. (2009). Visceral endoderm induces specification of cardiomyocytes in mice. Stem Cell Res. 3, 170–178.

Nishitoh, H., Ichijo, H., Kimura, M., Matsumoto, T., Makishima, F., Yamaguchi, A., Yamashita, H., Enomoto, S., and Miyazono, K. (1996). Identification of type I and type II serine/threonine kinase receptors for growth/differentiation factor-5. J. Biol. Chem. 271, 21345–21352.

Nunomura, K. (2005b). Cell Surface Labeling and Mass Spectrometry Reveal Diversity of Cell Surface Markers and Signaling Molecules Expressed in Undifferentiated Mouse Embryonic Stem Cells. Mol. Cell. Proteomics 4, 1968–1976.

Nunomura, K. (2005a). Cell Surface Labeling and Mass Spectrometry Reveal Diversity of Cell Surface Markers and Signaling Molecules Expressed in Undifferentiated Mouse Embryonic Stem Cells. Mol. Cell. Proteomics 4, 1968–1976.

O’Brien, T.X., Lee, K.J., and Chien, K.R. (1993). Positional specification of ventricular myosin light chain 2 expression in the primitive murine heart tube. Proc. Natl. Acad. Sci. U. S. A. 90, 5157–5161.

108

Ohlmann, A., Scholz, M., Goldwich, A., Chauhan, B.K., Hudl, K., Ohlmann, A.V., Zrenner, E., Berger, W., Cvekl, A., Seeliger, M.W., et al. (2005). Ectopic norrin induces growth of ocular capillaries and restores normal retinal angiogenesis in Norrie disease mutant mice. J. Neurosci. Off. J. Soc. Neurosci. 25, 1701–1710.

Olsen, S.K., Li, J.Y.H., Bromleigh, C., Eliseenkova, A.V., Ibrahimi, O.A., Lao, Z., Zhang, F., Linhardt, R.J., Joyner, A.L., and Mohammadi, M. (2006). Structural basis by which alternative splicing modulates the organizer activity of FGF8 in the brain. Genes Dev. 20, 185–198.

Ong, S.-E., and Mann, M. (2005). Mass spectrometry–based proteomics turns quantitative. Nat. Chem. Biol. 1, 252–262.

Ong, S.-E., Blagoev, B., Kratchmarova, I., Kristensen, D.B., Steen, H., Pandey, A., and Mann, M. (2002). Stable Isotope Labeling by Amino Acids in Cell Culture, SILAC, as a Simple and Accurate Approach to Expression Proteomics. Mol. Cell. Proteomics 1, 376–386.

Ong, S.-G., Huber, B.C., Lee, W.H., Kodo, K., Ebert, A.D., Ma, Y., Nguyen, P.K., Diecke, S., Chen, W.-Y., and Wu, J.C. (2015). Microfluidic Single-Cell Analysis of Transplanted Human Induced Pluripotent Stem Cell–Derived Cardiomyocytes After Acute Myocardial Infarction. Circulation 132, 762–771.

Ono, M., Shitashige, M., Honda, K., Isobe, T., Kuwabara, H., Matsuzuki, H., Hirohashi, S., and Yamada, T. (2006). Label-free Quantitative Proteomics Using Large Peptide Data Sets Generated by Nanoflow Liquid Chromatography and Mass Spectrometry. Mol. Cell. Proteomics 5, 1338– 1347.

Orlic, D., Hill, J.M., and Arai, A.E. (2002). Stem cells for myocardial regeneration. Circ. Res. 91, 1092–1102.

Oskouei, B.N., Lamirault, G., Joseph, C., Treuer, A.V., Landa, S., Da Silva, J., Hatzistergos, K., Dauer, M., Balkan, W., McNiece, I., et al. (2012). Increased Potency of Cardiac Stem Cells Compared with Bone Marrow Mesenchymal Stem Cells in Cardiac Repair. STEM CELLS Transl. Med. 1, 116–124.

Pampaloni, F., Reynaud, E.G., and Stelzer, E.H.K. (2007). The third dimension bridges the gap between cell culture and live tissue. Nat. Rev. Mol. Cell Biol. 8, 839–845.

Pan, Q., Shai, O., Lee, L.J., Frey, B.J., and Blencowe, B.J. (2008). Deep surveying of alternative splicing complexity in the human transcriptome by high-throughput sequencing. Nat. Genet. 40, 1413–1415.

Pandur, P., Läsche, M., Eisenberg, L.M., and Kühl, M. (2002). Wnt-11 activation of a non- canonical Wnt signalling pathway is required for cardiogenesis. Nature 418, 636–641.

Passier, R., Oostwaard, D.W., Snapper, J., Kloots, J., Hassink, R.J., Kuijk, E., Roelen, B., de la Riviere, A.B., and Mummery, C. (2005a). Increased cardiomyocyte differentiation from human embryonic stem cells in serum-free cultures. Stem Cells 23, 772–780.

109

Passier, R., Oostwaard, D.W., Snapper, J., Kloots, J., Hassink, R.J., Kuijk, E., Roelen, B., de la Riviere, A.B., and Mummery, C. (2005b). Increased cardiomyocyte differentiation from human embryonic stem cells in serum-free cultures. Stem Cells 23, 772–780.

Patel, V.J., Thalassinos, K., Slade, S.E., Connolly, J.B., Crombie, A., Murrell, J.C., and Scrivens, J.H. (2009). A Comparison of Labeling and Label-Free Mass Spectrometry-Based Proteomics Approaches. J. Proteome Res. 8, 3752–3759.

Paxton, C.N., Bleyl, S.B., Chapman, S.C., and Schoenwolf, G.C. (2010). Identification of differentially expressed genes in early inner ear development. Gene Expr. Patterns GEP 10, 31.

Pease, A.C., Solas, D., Sullivan, E.J., Cronin, M.T., Holmes, C.P., and Fodor, S.P. (1994). Light- generated oligonucleotide arrays for rapid DNA sequence analysis. Proc. Natl. Acad. Sci. U. S. A. 91, 5022–5026.

Peng, G., Suo, S., Chen, J., Chen, W., Liu, C., Yu, F., Wang, R., Chen, S., Sun, N., Cui, G., et al. (2016). Spatial Transcriptome for the Molecular Annotation of Lineage Fates and Cell Identity in Mid-gastrula Mouse Embryo. Dev. Cell 36, 681–697.

Perea-Gomez, A., Vella, F.D.J., Shawlot, W., Oulad-Abdelghani, M., Chazaud, C., Meno, C., Pfister, V., Chen, L., Robertson, E., Hamada, H., et al. (2002). Nodal antagonists in the anterior visceral endoderm prevent the formation of multiple primitive streaks. Dev. Cell 3, 745–756.

Pontén, A., Walsh, S., Malan, D., Xian, X., Schéele, S., Tarnawski, L., Fleischmann, B.K., and Jovinge, S. (2013). FACS-Based Isolation, Propagation and Characterization of Mouse Embryonic Cardiomyocytes Based on VCAM-1 Surface Marker Expression. PLoS ONE 8.

Prall, O.W.J., Menon, M.K., Solloway, M.J., Watanabe, Y., Zaffran, S., Bajolle, F., Biben, C., McBride, J.J., Robertson, B.R., Chaulet, H., et al. (2007). An Nkx2-5/Bmp2/Smad1 negative feedback loop controls heart progenitor specification and proliferation. Cell 128, 947–959.

Pulina, M.V., Hou, S.-Y., Mittal, A., Julich, D., Whittaker, C.A., Holley, S.A., Hynes, R.O., and Astrof, S. (2011). Essential roles of fibronectin in the development of the left-right embryonic body plan. Dev. Biol. 354, 208–220.

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.

Rana, M.S., Christoffels, V.M., and Moorman, A.F.M. (2013). A molecular and genetic outline of cardiac morphogenesis. Acta Physiol. 207, 588–615.

Re’em-Kalma, Y., Lamb, T., and Frank, D. (1995). Competition between noggin and bone morphogenetic protein 4 activities may regulate dorsalization during Xenopus development. Proc. Natl. Acad. Sci. U. S. A. 92, 12141–12145.

Reimers, M. (2010). Making Informed Choices about Microarray Data Analysis. PLOS Comput. Biol. 6, e1000786.

110

Ren, Y., Lee, M.Y., Schliffke, S., Paavola, J., Amos, P.J., Ge, X., Ye, M., Zhu, S., Senyei, G., Lum, L., et al. (2011). Small molecule Wnt inhibitors enhance the efficiency of BMP-4-directed cardiac differentiation of human pluripotent stem cells. J. Mol. Cell. Cardiol. 51, 280–287.

Revil, T., Gaffney, D., Dias, C., Majewski, J., and Jerome-Majewska, L.A. (2010). Alternative splicing is frequent during early embryonic development in mouse. BMC Genomics 11, 399.

Richardson, L., Venkataraman, S., Stevenson, P., Yang, Y., Moss, J., Graham, L., Burton, N., Hill, B., Rao, J., Baldock, R.A., et al. (2014). EMAGE mouse embryo spatial gene expression database: 2014 update. Nucleic Acids Res. 42, D835–D844.

Ringwald, M., Eppig, J.T., Begley, D.A., Corradi, J.P., McCright, I.J., Hayamizu, T.F., Hill, D.P., Kadin, J.A., and Richardson, J.E. (2001). The Mouse Gene Expression Database (GXD). Nucleic Acids Res. 29, 98–101.

Rivera-Pérez, J.A., and Magnuson, T. (2005). Primitive streak formation in mice is preceded by localized activation of Brachyury and Wnt3. Dev. Biol. 288, 363–371.

Robitaille, J., MacDonald, M.L.E., Kaykas, A., Sheldahl, L.C., Zeisler, J., Dubé, M.-P., Zhang, L.-H., Singaraja, R.R., Guernsey, D.L., Zheng, B., et al. (2002). Mutant frizzled-4 disrupts retinal angiogenesis in familial exudative vitreoretinopathy. Nat. Genet. 32, 326–330.

Rodrigues, C.A.V., Fernandes, T.G., Diogo, M.M., da Silva, C.L., and Cabral, J.M.S. (2011). Stem cell cultivation in bioreactors. Biotechnol. Adv. 29, 815–829.

Ross, P.L., Huang, Y.N., Marchese, J.N., Williamson, B., Parker, K., Hattan, S., Khainovski, N., Pillai, S., Dey, S., Daniels, S., et al. (2004). Multiplexed Protein Quantitation in Saccharomyces cerevisiae Using Amine-reactive Isobaric Tagging Reagents. Mol. Cell. Proteomics 3, 1154–1169.

Rossini, A., Frati, C., Lagrasta, C., Graiani, G., Scopece, A., Cavalli, S., Musso, E., Baccarin, M., Di Segni, M., Fagnoni, F., et al. (2011). Human cardiac and bone marrow stromal cells exhibit distinctive properties related to their origin. Cardiovasc. Res. 89, 650–660.

Rugg-Gunn, P.J., Cox, B.J., Lanner, F., Sharma, P., Ignatchenko, V., McDonald, A.C.H., Garner, J., Gramolini, A.O., Rossant, J., and Kislinger, T. (2012). Cell-Surface Proteomics Identifies Lineage-Specific Markers of Embryo-Derived Stem Cells. Dev. Cell 22, 887–901.

Rybak, J.-N., Ettorre, A., Kaissling, B., Giavazzi, R., Neri, D., and Elia, G. (2005). In vivo protein biotinylation for identification of organ-specific antigens accessible from the vasculature. Nat Meth 291–298.

Sa, S., Wong, L., and McCloskey, K.E. (2014). Combinatorial Fibronectin and Laminin Signaling Promote Highly Efficient Cardiac Differentiation of Human Embryonic Stem Cells. BioResearch Open Access 3, 150–161.

Sachinidis, A., Fleischmann, B.K., Kolossov, E., Wartenberg, M., Sauer, H., and Hescheler, J. (2003). Cardiac specific differentiation of mouse embryonic stem cells. Cardiovasc. Res. 58, 278– 291.

111

Saga, Y., Miyagawa-Tomita, S., Takagi, A., Kitajima, S., Miyazaki, J. i, and Inoue, T. (1999). MesP1 is expressed in the heart precursor cells and required for the formation of a single heart tube. Development 126, 3437–3447. dos Santos, F.F., Andrade, P.Z., da Silva, C.L., and Cabral, J.M.S. (2013). Bioreactor design for clinical-grade expansion of stem cells. Biotechnol. J. 8, 644–654.

Schena, M., Shalon, D., Davis, R.W., and Brown, P.O. (1995). Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270, 467–470.

Schiess, R., Mueller, L.N., Schmidt, A., Mueller, M., Wollscheid, B., and Aebersold, R. (2008). Analysis of cell surface proteome changes via label-free, quantitative mass spectrometry. Mol Cell Proteomics 8, 624–638.

Schlange, T., Andrée, B., Arnold, H.-H., and Brand, T. (2000). BMP2 is required for early heart development during a distinct time period. Mech. Dev. 91, 259–270.

Schleiffarth, J.R., Person, A.D., Martinsen, B.J., Sukovich, D.J., Neumann, A., Baker, C.V.H., Lohr, J.L., Cornfield, D.N., Ekker, S.C., and Petryk, A. (2007). Wnt5a is required for cardiac outflow tract septation in mice. Pediatr. Res. 61, 386–391.

Schneider, V.A., and Mercola, M. (2001). Wnt antagonism initiates cardiogenesis in Xenopus laevis. Genes Dev. 15, 304–315.

Schoenwolf, G.C., and Garcia-Martinez, V. (1995). Primitive-streak origin and state of commitment of cells of the cardiovascular system in avian and mammalian embryos. Cell. Mol. Biol. Res. 41, 233–240.

Schram, G., Pourrier, M., Melnyk, P., and Nattel, S. (2002). Differential distribution of cardiac ion channel expression as a basis for regional specialization in electrical function. Circ. Res. 90, 939– 950.

Schultheiss, T.M., Burch, J.B., and Lassar, A.B. (1997). A role for bone morphogenetic proteins in the induction of cardiac myogenesis. Genes Dev. 11, 451–462.

Scialdone, A., Tanaka, Y., Jawaid, W., Moignard, V., Wilson, N.K., Macaulay, I.C., Marioni, J.C., and Göttgens, B. (2016). Resolving early mesoderm diversification through single-cell expression profiling. Nature 535, 289–293.

Seifert, J.R.K., and Mlodzik, M. (2007). Frizzled/PCP signalling: a conserved mechanism regulating cell polarity and directed motility. Nat. Rev. Genet. 8, 126–138.

Serra, M., Brito, C., Sousa, M.F.Q., Jensen, J., Tostões, R., Clemente, J., Strehl, R., Hyllner, J., Carrondo, M.J.T., and Alves, P.M. (2010). Improving expansion of pluripotent human embryonic stem cells in perfused bioreactors through oxygen control. J. Biotechnol. 148, 208–215.

Sharma, P., Abbasi, C., Lazic, S., Teng, A.C.T., Wang, D., Dubois, N., Ignatchenko, V., Wong, V., Liu, J., Araki, T., et al. (2015). Evolutionarily conserved intercalated disc protein Tmem65 regulates cardiac conduction and connexin 43 function. Nat. Commun. 6, 8391.

112

Sheng, J.-J., and Jin, J.-P. (2014). Gene regulation, alternative splicing, and posttranslational modification of troponin subunits in cardiac development and adaptation: a focused review. Front. Physiol. 5.

Shiba, Y., Gomibuchi, T., Seto, T., Wada, Y., Ichimura, H., Tanaka, Y., Ogasawara, T., Okada, K., Shiba, N., Sakamoto, K., et al. (2016). Allogeneic transplantation of iPS cell-derived cardiomyocytes regenerates primate hearts. Nature 538, 388–391.

Shin, C., and Manley, J.L. (2002). The SR protein SRp38 represses splicing in M phase cells. Cell 111, 407–417.

Shin, B.K., Wang, H., Yim, A.M., Le Naour, F., Brichory, F., Jang, J.H., Zhao, R., Puravs, E., Tra, J., Michael, C.W., et al. (2003). Global profiling of the cell surface proteome of cancer cells uncovers an abundance of proteins with chaperone function. J. Biol. Chem. 278, 7607–7616.

Shiratori, H., Sakuma, R., Watanabe, M., Hashiguchi, H., Mochida, K., Sakai, Y., Nishino, J., Saijoh, Y., Whitman, M., and Hamada, H. (2001). Two-step regulation of left-right asymmetric expression of Pitx2: initiation by nodal signaling and maintenance by Nkx2. Mol. Cell 7, 137–149.

Singh, H., Mok, P., Balakrishnan, T., Rahmat, S.N.B., and Zweigerdt, R. (2010). Up-scaling single cell-inoculated suspension culture of human embryonic stem cells. Stem Cell Res. 4, 165–179.

Sizarov, A., Ya, J., Boer, B.A. de, Lamers, W.H., Christoffels, V.M., and Moorman, A.F.M. (2011). Formation of the Building Plan of the Human Heart: Morphogenesis, Growth, and Differentiation. Circulation 123, 1125–1135.

Später, D., Abramczuk, M.K., Buac, K., Zangi, L., Stachel, M.W., Clarke, J., Sahara, M., Ludwig, A., and Chien, K.R. (2013). A HCN4+ cardiomyogenic progenitor derived from the first heart field and human pluripotent stem cells. Nat. Cell Biol. 15, 1098–1106.

Spellman, P.T., Sherlock, G., Zhang, M.Q., Iyer, V.R., Anders, K., Eisen, M.B., Brown, P.O., Botstein, D., and Futcher, B. (1998). Comprehensive identification of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol. Biol. Cell 9, 3273–3297.

Srinivasan, K., Shiue, L., Hayes, J.D., Centers, R., Fitzwater, S., Loewen, R., Edmondson, L.R., Bryant, J., Smith, M., Rommelfanger, C., et al. (2005). Detection and measurement of alternative splicing using splicing-sensitive microarrays. Methods San Diego Calif 37, 345–359.

Srivastava, D., and Olson, E.N. (2000). A genetic blueprint for cardiac development. Nature 407, 221–226.

Srivastava, D., Cserjesi, P., and Olson, E.N. (1995). A Subclass of bHLH Proteins Required for Cardiac Morphogenesis. Science 270, 1995–1999.

Sumi, T., Tsuneyoshi, N., Nakatsuji, N., and Suemori, H. (2008). Defining early lineage specification of human embryonic stem cells by the orchestrated balance of canonical Wnt/Â - catenin, Activin/Nodal and BMP signaling. Development 135, 2969–2979.

113

Sweetman, D., Wagstaff, L., Cooper, O., Weijer, C., and Münsterberg, A. (2008a). The migration of paraxial and lateral plate mesoderm cells emerging from the late primitive streak is controlled by different Wnt signals. BMC Dev. Biol. 8, 63.

Sweetman, D., Wagstaff, L., Cooper, O., Weijer, C., and Münsterberg, A. (2008b). The migration of paraxial and lateral plate mesoderm cells emerging from the late primitive streak is controlled by different Wnt signals. BMC Dev. Biol. 8, 63.

Tada, S., Era, T., Furusawa, C., Sakurai, H., Nishikawa, S., Kinoshita, M., Nakao, K., Chiba, T., and Nishikawa, S.-I. (2005). Characterization of mesendoderm: a diverging point of the definitive endoderm and mesoderm in embryonic stem cell differentiation culture. Dev. Camb. Engl. 132, 4363–4374.

Takeuchi, J.K., Ohgi, M., Koshiba-Takeuchi, K., Shiratori, H., Sakaki, I., Ogura, K., Saijoh, Y., and Ogura, T. (2003). Tbx5 specifies the left/right ventricles and ventricular septum position during cardiogenesis. Development 130, 5953–5964.

Tam, P.P.L., and Loebel, D.A.F. (2007). Gene function in mouse embryogenesis: get set for gastrulation. Nat. Rev. Genet. 8, 368–381.

Tam, P.P., Parameswaran, M., Kinder, S.J., and Weinberger, R.P. (1997). The allocation of epiblast cells to the embryonic heart and other mesodermal lineages: the role of ingression and tissue movement during gastrulation. Development 124, 1631–1642.

Tan, J.Y., Sriram, G., Rufaihah, A.J., Neoh, K.G., and Cao, T. (2013a). Efficient derivation of lateral plate and paraxial mesoderm subtypes from human embryonic stem cells through GSKi- mediated differentiation. Stem Cells Dev. 22, 1893–1906.

Tan, J.Y., Sriram, G., Rufaihah, A.J., Neoh, K.G., and Cao, T. (2013b). Efficient Derivation of Lateral Plate and Paraxial Mesoderm Subtypes from Human Embryonic Stem Cells Through GSKi-Mediated Differentiation. Stem Cells Dev. 22, 1893–1906.

Tan, P.K., Downey, T.J., Spitznagel, E.L., Jr, Xu, P., Fu, D., Dimitrov, D.S., Lempicki, R.A., Raaka, B.M., and Cam, M.C. (2003). Evaluation of gene expression measurements from commercial microarray platforms. Nucleic Acids Res. 31, 5676.

Tarca, A.L., Romero, R., and Draghici, S. (2006). Analysis of microarray experiments of gene expression profiling. Am. J. Obstet. Gynecol. 195, 373–388.

Tata, P.R., Mou, H., Pardo-Saganta, A., Zhao, R., Prabhu, M., Law, B.M., Vinarsky, V., Cho, J.L., Breton, S., Sahay, A., et al. (2013). Dedifferentiation of committed epithelial cells into stem cells in vivo. Nature 503, 218–223.

Telford, W.G., Babin, S.A., Khorev, S.V., and Rowe, S.H. (2009). Green fiber lasers: An alternative to traditional DPSS green lasers for flow cytometry. Cytom. Part J. Int. Soc. Anal. Cytol. 75, 1031–1039.

114

Telonis-Scott, M., Kopp, A., Wayne, M.L., Nuzhdin, S.V., and McIntyre, L.M. (2009). Sex- Specific Splicing in Drosophila: Widespread Occurrence, Tissue Specificity and Evolutionary Conservation. Genetics 181, 421–434.

Thomas, R.J., Anderson, D., Chandra, A., Smith, N.M., Young, L.E., Williams, D., and Denning, C. (2009). Automated, scalable culture of human embryonic stem cells in feeder-free conditions. Biotechnol. Bioeng. 102, 1636–1644.

Thomas, T., Yamagishi, H., Overbeek, P.A., Olson, E.N., and Srivastava, D. (1998). The bHLH factors, dHAND and eHAND, specify pulmonary and systemic cardiac ventricles independent of left-right sidedness. Dev. Biol. 196, 228–236.

Timmerman, L.A., Grego-Bessa, J., Raya, A., Bertrán, E., Pérez-Pomares, J.M., Díez, J., Aranda, S., Palomo, S., McCormick, F., Izpisúa-Belmonte, J.C., et al. (2004). Notch promotes epithelial- mesenchymal transition during cardiac development and oncogenic transformation. Genes Dev. 18, 99–115.

Tirosh-Finkel, L., Elhanany, H., Rinon, A., and Tzahor, E. (2006). Mesoderm progenitor cells of common origin contribute to the head musculature and the cardiac outflow tract. Development 133, 1943–1953.

Trinh, L.A., and Stainier, D.Y.R. (2004). Fibronectin regulates epithelial organization during myocardial migration in zebrafish. Dev. Cell 6, 371–382.

Trounson, A., and DeWitt, N.D. (2016). Pluripotent stem cells progressing to the clinic. Nat. Rev. Mol. Cell Biol. 17, 194–200.

Tzahor, E., and Lassar, A.B. (2001). Wnt signals from the neural tube block ectopic cardiogenesis. Genes Dev. 15, 255–260.

Ueno, S., Weidinger, G., Osugi, T., Kohn, A.D., Golob, J.L., Pabon, L., Reinecke, H., Moon, R.T., and Murry, C.E. (2007a). Biphasic role for Wnt/β-catenin signaling in cardiac specification in zebrafish and embryonic stem cells. Proc. Natl. Acad. Sci. 104, 9685–9690.

Ueno, S., Weidinger, G., Osugi, T., Kohn, A.D., Golob, J.L., Pabon, L., Reinecke, H., Moon, R.T., and Murry, C.E. (2007b). Biphasic role for Wnt/β-catenin signaling in cardiac specification in zebrafish and embryonic stem cells. Proc. Natl. Acad. Sci. 104, 9685–9690.

Vincent, S.D., Dunn, N.R., Hayashi, S., Norris, D.P., and Robertson, E.J. (2003). Cell fate decisions within the mouse organizer are governed by graded Nodal signals. Genes Dev. 17, 1646– 1662.

Vitelli, F., Taddei, I., Morishima, M., Meyers, E.N., Lindsay, E.A., and Baldini, A. (2002). A genetic link between Tbx1 and fibroblast growth factor signaling. Dev. Camb. Engl. 129, 4605– 4611.

Vizcaíno, J.A., Deutsch, E.W., Wang, R., Csordas, A., Reisinger, F., Ríos, D., Dianes, J.A., Sun, Z., Farrah, T., Bandeira, N., et al. (2014). ProteomeXchange provides globally coordinated proteomics data submission and dissemination. Nat. Biotechnol. 32, 223–226.

115

Wallingford, J.B., and Habas, R. (2005). The developmental biology of Dishevelled: an enigmatic protein governing cell fate and cell polarity. Dev. Camb. Engl. 132, 4421–4436.

Wang, F., Hou, J., Han, B., Nie, Y., Cong, X., Hu, S., and Chen, X. (2012). Developmental changes in lysophospholipid receptor expression in rodent heart from near-term fetus to adult. Mol. Biol. Rep. 39, 9075–9084.

Wang, Y., Cheng, L., and Gerecht, S. (2014). Efficient and scalable expansion of human pluripotent stem cells under clinically compliant settings: a view in 2013. Ann. Biomed. Eng. 42, 1357–1372.

Washburn, M.P., Wolters, D., and Yates, J.R. (2001). Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 19, 242–247.

Watarai, H., Hinohara, A., Nagafune, J., Nakayama, T., Taniguchi, M., and Yamaguchi, Y. (2005). Plasma membrane-focused proteomics: Dramatic changes in surface expression during the maturation of human dendritic cells. Proteomics 5, 4001–4011.

Whitman, C.O. (1887). A contribution to the history of germ layers in Clepsine. J. Morphol. 1, 105–182.

Wilkinson, D.G., Bhatt, S., and Herrmann, B.G. (1990). Expression pattern of the mouse T gene and its role in mesoderm formation. Nature 657–659.

Wilson, E.B. (1892). The Cell-lineage of Nereis. A contribution to the cytogeny of the Annelid body. J. Morphol. 386.

Winnier, G., Blessing, M., Labosky, P.A., and Hogan, B.L. (1995). Bone morphogenetic protein- 4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 9, 2105–2116.

Wolf-Yadlin, A., Hautaniemi, S., Lauffenburger, D., and White, F. (2007). Multiple reaction monitoring for robust quantitative proteomic analysis of cellular signaling networks. Proc. Natl. Acad. Sci. U. S. A. 104, 5860–5865.

Wollscheid, B., Bausch-Fluck, D., Henderson, C., O’Brien, R., Bibel, M., Schiess, R., Aebersold, R., and Watts, J.D. (2009a). Mass-spectrometric identification and relative quantification of N- linked cell surface glycoproteins. Nat. Biotechnol. 27, 378–386.

Wollscheid, B., Bausch-Fluck, D., Henderson, C., O’Brien, R., Bibel, M., Schiess, R., Aebersold, R., and Watts, J.D. (2009b). Mass-spectrometric identification and relative quantification of N- linked cell surface glycoproteins. Nat. Biotechnol. 27, 378–386.

Wu, C.C., and Yates, J.R. (2003). The application of mass spectrometry to membrane proteomics. Nat. Biotechnol. 21, 262–267.

Wu, C.C., MacCoss, M.J., Howell, K.E., and Yates 3rd, J.R. (2003). A method for the comprehensive proteomic analysis of membrane proteins. Nat Biotechnol 21, 532–538.

116

Wu, S.M., Fujiwara, Y., Cibulsky, S.M., Clapham, D.E., Lien, C., Schultheiss, T.M., and Orkin, S.H. (2006a). Developmental Origin of a Bipotential Myocardial and Smooth Muscle Cell Precursor in the Mammalian Heart. Cell 127, 1137–1150.

Wu, S.M., Fujiwara, Y., Cibulsky, S.M., Clapham, D.E., Lien, C. ling, Schultheiss, T.M., and Orkin, S.H. (2006b). Developmental Origin of a Bipotential Myocardial and Smooth Muscle Cell Precursor in the Mammalian Heart. Cell 127, 1137–1150.

Xiong, Q., Ye, L., Zhang, P., Lepley, M., Tian, J., Li, J., Zhang, L., Swingen, C., Vaughan, J.T., Kaufman, D.S., et al. (2013). Functional Consequences of Human Induced Pluripotent Stem Cell TherapyClinical Perspective: Myocardial ATP Turnover Rate in the In Vivo Swine Heart With Postinfarction Remodeling. Circulation 127, 997–1008.

Xu, Q., Wang, Y., Dabdoub, A., Smallwood, P.M., Williams, J., Woods, C., Kelley, M.W., Jiang, L., Tasman, W., Zhang, K., et al. (2004). Vascular development in the retina and inner ear: control by Norrin and Frizzled-4, a high-affinity ligand-receptor pair. Cell 116, 883–895.

Xu, X.Q., Soo, S.Y., Sun, W., and Zweigerdt, R. (2009). Global expression profile of highly enriched cardiomyocytes derived from human embryonic stem cells. Stem Cells Dayt. Ohio 27, 2163–74.

Yamaguchi, T.P. (2001). Heads or tails: Wnts and anterior-posterior patterning. Curr. Biol. CB 11, R713-724.

Yang, Y. (2012). Wnt signaling in development and disease. Cell Biosci. 2, 14.

Yang, L., Soonpaa, M.H., Adler, E.D., Roepke, T.K., Kattman, S.J., Kennedy, M., Henckaerts, E., Bonham, K., Abbott, G.W., Linden, R.M., et al. (2008a). Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population. Nature 453, 524–528.

Yang, L., Soonpaa, M.H., Adler, E.D., Roepke, T.K., Kattman, S.J., Kennedy, M., Henckaerts, E., Bonham, K., Abbott, G.W., Linden, R.M., et al. (2008b). Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population. Nature 453, 524–528.

Yang, L., Soonpaa, M.H., Adler, E.D., Roepke, T.K., Kattman, S.J., Kennedy, M., Henckaerts, E., Bonham, K., Abbott, G.W., Linden, R.M., et al. (2008c). Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population. Nature 453, 524–528.

Yang-Snyder, J., Miller, J.R., Brown, J.D., Lai, C.J., and Moon, R.T. (1996). A frizzled homolog functions in a vertebrate Wnt signaling pathway. Curr. Biol. CB 6, 1302–1306.

Yasunaga, M., Tada, S., Torikai-Nishikawa, S., Nakano, Y., Okada, M., Jakt, L.M., Nishikawa, S., Chiba, T., Era, T., and Nishikawa, S.-I. (2005). Induction and monitoring of definitive and visceral endoderm differentiation of mouse ES cells. Nat. Biotechnol. 23, 1542–1550.

Yates, J.R., Speicher, S., Griffin, P.R., and Hunkapiller, T. (1993). Peptide mass maps: a highly informative approach to protein identification. Anal. Biochem. 214, 397–408.

117

Ye, J., and Yeghiazarians, Y. (2015). Cardiac stem cell therapy: Have we put too much hype in which cell type to use? Heart Fail. Rev. 20, 613–619.

Yoshida, T., Vivatbutsiri, P., Morriss-Kay, G., Saga, Y., and Iseki, S. (2008). Cell lineage in mammalian craniofacial mesenchyme. Mech. Dev. 125, 797–808.

Yuasa, S., Itabashi, Y., Koshimizu, U., Tanaka, T., Sugimura, K., Kinoshita, M., Hattori, F., Fukami, S., Shimazaki, T., Ogawa, S., et al. (2005). Transient inhibition of BMP signaling by Noggin induces cardiomyocyte differentiation of mouse embryonic stem cells. Nat. Biotechnol. 23, 607–611.

Zaffran, S., Kelly, R.G., Meilhac, S.M., Buckingham, M.E., and Brown, N.A. (2004). Right Ventricular Myocardium Derives From the Anterior Heart Field. Circ. Res. 95, 261–268.

Zhang, J., Wilson, G.F., Soerens, A.G., Koonce, C.H., Yu, J., Palecek, S.P., Thomson, J.A., and Kamp, T.J. (2009). Functional cardiomyocytes derived from human induced pluripotent stem cells. Circ. Res. 104, e30-41.

Zhang, Y., Sturgill, D., Parisi, M., Kumar, S., and Oliver, B. (2007). Constraint and turnover in sex-biased gene expression in the genus Drosophila. Nature 450, 233–237.

Zhang, Y., Wang, D., Chen, M., Yang, B., Zhang, F., and Cao, K. (2011). Intramyocardial transplantation of undifferentiated rat induced pluripotent stem cells causes tumorigenesis in the heart. PloS One 6, e19012.

Zhao, S., Fung-Leung, W.-P., Bittner, A., Ngo, K., and Liu, X. (2014). Comparison of RNA-Seq and Microarray in Transcriptome Profiling of Activated T Cells. PLOS ONE 9, e78644.

Zhou, B., von Gise, A., Ma, Q., Rivera-Feliciano, J., and Pu, W.T. (2008). Nkx2-5- and Isl1- expressing cardiac progenitors contribute to proepicardium. Biochem. Biophys. Res. Commun. 375, 450–453.

Zimmerman, L.B., De Jesús-Escobar, J.M., and Harland, R.M. (1996). The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. Cell 86, 599–606.

Zweigerdt, R. (2009). Large scale production of stem cells and their derivatives. Adv. Biochem. Eng. Biotechnol. 114, 201–235.

Appendices

Appendix I: Protein Quantification and Clustering.

Table of protein intensities determined by mass spectrometry and clustered using k-means clustering. Protein GeneID Description d0 d2 d3 d3.75 cluster 0610007L01RIK 71667 UPF0458 protein C7orf42 homolog -0.98779 -0.6261 0.400761 1.213124 1 Adipocyte plasma membrane-associated 2310001A20RIK 71881 -1.43661 0.11024 0.51422 0.812148 1 protein 2310010M20RIK 69576 hypothetical protein LOC69576 -1.3994 0.467571 0.024947 0.906882 1 2610524H06RIK 330173 hypothetical protein LOC330173 0.016576 0.232555 -1.33175 1.082623 4 Isoform 1 of Signal peptide peptidase- 3110056O03RIK 73218 -0.19126 0.95046 0.561637 -1.32084 5 like 2B Killer cell lectin-like receptor subfamily 4930431A04RIK 100043861 -0.33236 0.665717 0.926665 -1.26002 5 B member 1G Glycosyltransferase 54 domain- 4933434I20RIK 67555 -0.0138 0.3511 1.018711 -1.35601 5 containing protein Isoform 2 of Uncharacterized protein 8430419L09RIK 74525 0.349483 1.137445 -1.23265 -0.25428 3 KIAA1467 ATP-binding cassette sub-family A ABCA1 11303 0.74959 0.468884 -1.46828 0.24981 3 member 1 ATP-binding cassette, sub-family A ABCA12 74591 -0.46076 -0.83617 -0.14131 1.438244 4 (ABC1), member 12 ATP-binding cassette, sub-family A ABCA15 320631 -1.40553 0.769029 -0.02123 0.657738 1 (ABC1), member 15 ABCB1A 18671 Multidrug resistance protein 3 -1.21076 -0.23725 0.26065 1.187363 1 ABCC4 239273 Putative uncharacterized protein -0.83494 -0.67292 0.156751 1.351106 1 Abhydrolase domain-containing protein ABHD14A 68644 -1.19073 -0.34303 1.146538 0.387222 1 14A ACVR1 11477 Activin receptor type-1 0.018586 0.685516 0.716089 -1.42019 5 Isoform ActR-IIB1 of Activin receptor ACVR2B 11481 0.407467 0.742687 -1.47534 0.325189 3 type-2B Disintegrin and metalloproteinase ADAM10 11487 -0.07948 1.146901 0.210803 -1.27823 5 domain-containing protein 10 ADAM17 11491 Putative uncharacterized protein -1.29182 0.740478 -0.28598 0.837324 1 ADCY3 104111 adenylate cyclase type 3 isoform 1 1.100171 0.246734 -1.31442 -0.03249 3 ADCY9 11515 Adenylate cyclase type 9 1.275724 -0.72763 -0.86264 0.314544 3 ADORA2B 11541 Adenosine receptor A2b 1.443593 -0.09842 -0.64425 -0.70093 3 Isoform 1 of Cytosolic carboxypeptidase AGBL3 76223 0.361382 -1.10004 -0.46329 1.201947 4 3 AGPAT6 102247 Glycerol-3-phosphate acyltransferase 4 -0.86001 -0.52667 -0.02135 1.408028 1 AGPAT9 231510 Glycerol-3-phosphate acyltransferase 3 -0.86001 -0.52667 -0.02135 1.408028 1 AGTR1B 11608 Type-1B angiotensin II receptor 0.741895 0.943413 -1.10669 -0.57861 3 Uncharacterized family 31 glucosidase AI464131 329828 0.511113 1.149028 -0.68783 -0.97231 3 KIAA1161 ALCAM 11658 CD166 antigen -0.84257 -0.67998 0.183075 1.339473 1 ALPI 76768 intestinal alkaline phosphatase -1.08763 -0.4847 0.370219 1.202111 1

118

119

Alkaline phosphatase, tissue-nonspecific ALPL 11647 0.599998 0.64472 -1.47373 0.229016 3 isozyme Isoform 1 of E3 ubiquitin-protein ligase AMFR 23802 -0.30414 -0.4008 -0.76521 1.470143 4 AMFR ANK1 11733 Ank1 protein -0.25135 0.982669 -1.28864 0.557314 3 ANO6 105722 Isoform 2 of Anoctamin-6 0.828568 0.46215 0.150542 -1.44126 5 ANO9 71345 Anoctamin-9 1.455546 -0.42653 -0.80608 -0.22294 3 ANPEP 16790 Aminopeptidase N 0.104612 -0.10906 -1.21785 1.222296 4 ANTXR1 69538 Isoform 2 of Anthrax toxin receptor 1 -1.25969 -0.26863 0.469923 1.058395 1 APLP2 11804 CDE1-binding protein CDEBP -0.57735 1.02275 0.649497 -1.0949 5 Cyclic AMP-dependent transcription ATF6B 12915 -1.08592 -0.60148 0.71073 0.976665 1 factor ATF-6 beta ATP11A 50770 Putative uncharacterized protein -0.22329 0.830896 -1.32242 0.714813 4 Isoform 1 of Probable cation- ATP13A3 224088 0.352917 1.092623 -0.15849 -1.28705 5 transporting ATPase 13A3 Sodium/potassium-transporting ATPase ATP1B1 11931 0.075119 0.693763 0.668522 -1.4374 5 subunit beta-1 Sodium/potassium-transporting ATPase ATP1B2 11932 -0.60937 -0.79617 -0.0068 1.412348 1 subunit beta-2 Sodium/potassium-transporting ATPase ATP1B3 11933 -1.39929 0.286624 0.969147 0.143516 2 subunit beta-3 Isoform A1-II of V-type proton ATPase ATP6V0A1 11975 -1.16061 -0.43996 0.501536 1.099036 1 116 kDa subunit a isoform 1 Isoform 1 of V-type proton ATPase 116 ATP6V0A2 21871 -0.67258 -0.7203 -0.03114 1.424021 1 kDa subunit a isoform 2 Probable phospholipid-transporting ATP8A2 50769 -1.11458 0.471581 -0.49355 1.136546 1 ATPase IB ATRN 11990 Attractin -0.51238 0.142991 1.335519 -0.96613 2 AU023871 106722 G6b protein 0.028834 0.443452 0.923813 -1.3961 5 Late secretory pathway protein AVL9 AVL9 78937 0.753462 0.914646 -0.51435 -1.15376 3 homolog BAI3 210933 Brain-specific angiogenesis inhibitor 3 0.211767 -0.29882 -1.15273 1.239786 4 BIRC6 12211 baculoviral IAP repeat-containing 6 -0.95502 0.898745 0.827703 -0.77142 2 BMP1 12153 Bone morphogenetic protein 1 0.549414 -0.44863 -1.16659 1.065807 4 Bone morphogenetic protein receptor BMPR2 12168 -0.34608 -0.87538 -0.21629 1.437759 4 type-2 BSG 12215 Isoform 2 of Basigin -0.71949 0.133998 -0.77799 1.363488 4 BST2 69550 Bone marrow stromal antigen 2 -0.81682 -0.01452 -0.58108 1.412424 4 Isoform 1 of Uncharacterized protein C230096C10RIK 230866 -1.11528 -0.57158 0.901189 0.785679 1 KIAA0090 Isoform 1 of VWFA and cache domain- CACHD1 320508 -1.0892 -0.60833 0.811453 0.886074 1 containing protein 1 Voltage-dependent P/Q-type calcium CACNA1A 12286 0.035623 -0.28553 -1.07629 1.326199 4 channel subunit alpha-1A Isoform 2B of Voltage-dependent CACNA2D1 12293 -1.09802 -0.42064 0.277519 1.241133 1 calcium channel subunit alpha-2/delta-1 CADM1 54725 Isoform 1 of Cell adhesion molecule 1 -1.32729 -0.1402 0.467402 1.00009 1 CADM4 260299 Cell adhesion molecule 4 0.284336 0.063661 -1.36868 1.020687 4 Isoform 1 of Calcium/calmodulin- CAMK2G 12325 dependent protein kinase type II subunit -0.0442 -0.17058 -1.10626 1.321038 4 gamma

120

Isoform 5 of Peripheral plasma CASK 12361 0.262319 -1.27384 1.137002 -0.12548 2 membrane protein CASK CD151 12476 CD151 antigen -0.90214 0.758199 0.966654 -0.82271 2 CD164 53599 Sialomucin core protein 24 0.758949 0.839703 -0.32972 -1.26893 5 CD22 12483 Isoform 1 of B-cell receptor CD22 -0.85501 -0.69785 0.239631 1.313222 1 CD276 102657 CD276 antigen -1.25994 0.053569 0.019228 1.187144 1 CD300LB 217304 CMRF35-like molecule 7 -0.03726 -1.32984 0.299079 1.068025 4 CD320 54219 Isoform 1 of CD320 antigen 0.864563 -1.388 0.569133 -0.0457 4 CD38 12494 ADP-ribosyl cyclase 1 0.12453 0.941259 0.341912 -1.4077 5 Isoform 2 of Leukocyte surface antigen CD47 16423 -1.18054 0.571037 1.044516 -0.43502 2 CD47 CD80 12519 Putative uncharacterized protein 0.679755 1.031185 -0.78154 -0.9294 3 CD97 26364 CD97 antigen isoform 2 0.97938 0.671689 -0.48891 -1.16216 3 CDCP1 109332 CUB domain-containing protein 1 -1.20865 0.771268 0.870096 -0.43271 2 CDH2 12558 Cadherin-2 -0.71406 -0.47212 -0.29114 1.477317 4 CDH4 12561 Cadherin-4 -1.01981 -0.31534 -0.02857 1.36372 1 CDH6 12563 cadherin-6 precursor -1.30169 -0.13024 0.353031 1.078896 1 Isoform Long of Carcinoembryonic CEACAM1 26365 0.79149 0.878723 -1.1565 -0.51372 3 antigen-related cell adhesion molecule 1 Cadherin EGF LAG seven-pass G-type CELSR1 12614 -1.37299 0.185318 0.160674 1.027001 1 receptor 1 Isoform 1 of Cadherin EGF LAG seven- CELSR2 53883 -1.31739 -0.05842 0.291285 1.084534 1 pass G-type receptor 2 Isoform 2 of Chromodomain-helicase- CHD9 109151 -0.12572 -0.73604 -0.58674 1.448505 4 DNA-binding protein 9 CHRM4 12672 Muscarinic acetylcholine receptor M4 0.518758 -1.4992 0.526778 0.45366 4 neuronal acetylcholine receptor subunit CHRNA3 110834 -0.1681 -0.88297 -0.37993 1.430999 4 alpha-3 neuronal acetylcholine receptor subunit CHRNB2 11444 -0.12296 -0.77146 -0.55022 1.444641 4 beta-2 precursor CLCN5 12728 H(+)/Cl(-) exchange transporter 5 -1.04146 -0.42507 0.159096 1.307429 1 CLDND1 224250 Claudin domain-containing protein 1 -1.49971 0.527537 0.485254 0.486923 2 Lectin-like transmembrane protein CLEC2J 677440 -1.02111 0.109716 -0.41893 1.330331 1 (Fragment) CLEC4A1 269799 Dendritic cell inhibitory receptor 4 -0.92115 -0.67268 0.33385 1.259975 1 Cleft lip and palate transmembrane CLPTM1 56457 -1.44899 0.117356 0.670935 0.660693 1 protein 1 homolog CLU 12759 Clusterin -1.24044 -0.1934 0.271984 1.161862 1 CNGA1 12788 cGMP-gated cation channel alpha-1 -0.06826 0.637454 0.816995 -1.38619 5 CNNM2 94219 Isoform 2 of Metal transporter CNNM2 0.554376 1.071899 -1.14579 -0.48048 3 CNNM4 94220 Metal transporter CNNM4 -1.20707 -0.22526 0.233228 1.199102 1 Ciliary neurotrophic factor receptor CNTFR 12804 0.654413 -0.15976 0.853556 -1.34821 5 subunit alpha CNTN3 18488 Contactin-3 -1.48099 0.620454 0.269918 0.590622 1 Isoform 1 of Contactin-associated CNTNAP2 66797 1.440995 -0.45633 -0.83183 -0.15284 3 protein-like 2 CNTNAP5A 636808 Contactin-associated protein like 5-1 0.212608 1.203798 -0.20692 -1.20948 5 COLEC12 140792 Collectin-12 -0.98304 -0.6805 0.52609 1.137448 1

121

COMP 12845 Cartilage oligomeric matrix protein -0.62921 -0.23743 -0.60884 1.475479 4 CPD 12874 Carboxypeptidase D -0.48189 -0.40466 -0.60816 1.494714 4 CPM 70574 Isoform 1 of Carboxypeptidase M -1.16932 -0.3684 0.372204 1.165519 1 Carnitine O-palmitoyltransferase 1, CPT1C 78070 0.448835 -0.13957 -1.32323 1.013967 4 brain isoform Isoform 1 of Complement regulatory CR1L 12946 -0.3722 -0.95341 -0.06995 1.395561 4 protein Crry CRB2 241324 Crumbs homolog 2 -1.1739 -0.38167 0.412836 1.142736 1 CRYBA1 12957 Beta-crystallin A1 0.962655 0.4323 -0.02166 -1.3733 5 Macrophage colony-stimulating factor 1 CSF1R 12978 0.035623 -0.28553 -1.07629 1.326199 4 receptor Isoform 2 of CUB and sushi domain- CSMD1 94109 -1.455 0.416138 0.227708 0.81115 1 containing protein 1 CTNS 83429 Cystinosin -0.79319 0.732606 0.987451 -0.92686 2 CTSC 13032 Uncharacterized protein -1.30916 -0.08161 1.090561 0.300208 2 Isoform 1 of Coxsackievirus and CXADR 13052 -1.16313 0.090707 -0.19312 1.265545 1 adenovirus receptor homolog Isoform 1 of Cysteine and histidine-rich CYHR1 54151 -0.02317 0.722413 0.707053 -1.40629 5 protein 1 CYP2A12 13085 MCG133379, isoform CRA_a 0.96692 0.536939 -0.18361 -1.32025 5 CYP2C38 13097 Cytochrome P450 2C38 0.982188 0.610093 -0.3515 -1.24078 3 CYP2C54 404195 Cytochrome P450 2C54 1.039394 0.638053 -0.6089 -1.06854 3 CYP2U1 71519 Isoform 1 of Cytochrome P450 2U1 -0.99362 -0.14795 1.389089 -0.24752 2 DAG1 13138 Dystroglycan -0.31426 1.105039 0.43149 -1.22227 5 Isoform 1 of Discoidin, CUB and LCCL DCBLD1 66686 0.029641 0.324521 -1.36685 1.012686 4 domain-containing protein 1 Discoidin, CUB and LCCL domain- DCBLD2 73379 -0.86132 -0.39889 -0.17765 1.437862 1 containing protein 2 DCHS1 233651 dachsous 1 -0.37101 -0.11655 -0.92454 1.412091 4 Isoform 2 of ATP-dependent RNA DHX9 13211 -0.6254 -0.48891 -0.37799 1.492301 4 helicase A Isoform 1 of Protein dispatched DISP1 68897 -1.12312 -0.40038 0.304718 1.218785 1 homolog 1 deleted in lung and esophageal cancer DLEC1 320256 0.469324 0.965524 -0.07785 -1.357 5 protein 1 homolog DPP4 13482 Dipeptidyl peptidase 4 0.570314 1.095694 -1.04085 -0.62515 3 Dipeptidyl aminopeptidase-like protein DPP6 13483 -0.68608 -0.03113 1.424234 -0.70702 2 6 DSC3 13507 Isoform 3A of Desmocollin-3 0.773173 0.403788 0.290877 -1.46784 5 DSG1B 225256 Desmoglein-1-beta -0.72499 0.045218 1.398494 -0.71872 2 DSG2 13511 Desmoglein-2 0.620938 0.645627 0.202034 -1.4686 5 Isoform B of Endothelin-converting ECE1 230857 -0.93107 -0.79811 0.841153 0.888027 1 enzyme 1 EFNA3 13638 Ephrin-A3 -0.79578 -0.7538 0.228022 1.32156 1 EFNA5 13640 Isoform Long of Ephrin-A5 -0.64941 -0.15761 -0.6519 1.458921 4 EFNB3 13643 Ephrin-B3 -0.48393 -0.57692 -0.4366 1.49745 4 Elongation of very long chain fatty acids ELOVL4 83603 -0.84797 -0.70089 0.231176 1.317676 1 protein 4 EMB 13723 Embigin -1.30011 -0.05695 0.239394 1.117663 1

122

EMP1 13730 Epithelial membrane protein 1 1.006784 0.711877 -0.82245 -0.89621 3 ENAM 13801 Enamelin 0.518758 -1.4992 0.526778 0.45366 4 ENG 13805 endoglin isoform 1 -0.36618 1.255233 -1.11352 0.224461 3 Isoform 2 of Ectonucleotide ENPP1 18605 pyrophosphatase/phosphodiesterase -0.64211 0.931486 0.771325 -1.0607 2 family member 1 Ectonucleotide ENPP3 209558 pyrophosphatase/phosphodiesterase 0.438532 0.674927 0.373934 -1.48739 5 family member 3 Isoform 1 of Ectonucleotide ENPP4 224794 pyrophosphatase/phosphodiesterase 0.843006 0.863685 -0.64479 -1.0619 3 family member 4 Isoform Long of Ectonucleoside ENTPD2 12496 -1.30983 -0.10516 0.338582 1.076407 1 triphosphate diphosphohydrolase 2 Isoform 1 of Ectonucleoside ENTPD7 93685 -1.03769 -0.60608 0.501834 1.141938 1 triphosphate diphosphohydrolase 7 EPCAM 17075 Epithelial cell adhesion molecule -1.12218 0.794092 0.890785 -0.56269 2 EPHA2 13836 Ephrin type-A receptor 2 0.827142 0.305305 0.322991 -1.45544 5 EPHB2 13844 Isoform 1 of Ephrin type-B receptor 2 -0.31345 -0.15607 -0.94149 1.41102 4 EPHB4 13846 ephrin type-B receptor 4 isoform b 0.293829 0.509353 0.678148 -1.48133 5 EXTL3 54616 Exostosin-like 3 -0.5532 -1.04886 0.406968 1.195091 1 junctional adhesion molecule A F11R 16456 -1.45627 0.485735 0.778842 0.191692 2 precursor F2R 14062 Proteinase-activated receptor 1 -0.70694 0.000209 -0.70741 1.414144 4 F2RL1 14063 Proteinase-activated receptor 2 -1.35325 -0.1588 0.7695 0.742549 1 F3 14066 Tissue factor 0.262121 0.666681 0.549377 -1.47818 5 FAM38A 234839 Protein PIEZO1 -0.87042 0.203259 -0.66103 1.328192 4 Isoform 1 of Basic fibroblast growth FGFR1 14182 -1.28535 0.299089 1.114446 -0.12818 2 factor receptor 1 Vascular endothelial growth factor FLT4 14257 -0.93509 0.712282 -0.78119 1.003995 4 receptor 3 Formyl peptide receptor-related FPR-RS4 14291 0.096903 -0.93922 1.357569 -0.51526 2 sequence 4 FRRS1 20321 Ferric-chelate reductase 1 0.874773 0.806312 -0.54762 -1.13347 3 FZD4 14366 Frizzled-4 -0.68432 -0.72551 -0.00627 1.41609 1 Gamma-aminobutyric acid receptor GABRA4 14397 -0.32901 -0.91007 -0.18541 1.424485 4 subunit alpha-4 GAK 231580 Isoform 1 of Cyclin-G-associated kinase -1.23995 0.669654 -0.36734 0.937633 1 Putative polypeptide N- GALNTL5 67909 acetylgalactosaminyltransferase-like 0.080039 0.064677 -1.2928 1.14808 4 protein 5 Glial cell line derived neurotrophic GFRA1 14585 factor family receptor alpha 1, isoform -0.62941 -0.23715 -0.60886 1.475427 4 CRA_a GGT1 14598 Gamma-glutamyltranspeptidase 1 -1.15198 0.629524 1.016264 -0.4938 2 N-acetyllactosaminide alpha-1,3- GGTA1 14594 1.317688 0.224371 -0.88144 -0.66062 3 galactosyltransferase isoform 1 GLG1 20340 Golgi apparatus protein 1 -0.20398 -0.89048 -0.33778 1.432247 4 GLP2R 93896 Glucagon-like peptide 2 receptor 0.374345 1.193226 -1.10584 -0.46173 3 GLRB 14658 Glycine receptor subunit beta -1.21473 0.635588 0.979772 -0.40063 2

123

GM106 226866 RPE-spondin 0.340708 1.258824 -0.89984 -0.69969 3 GM10933 100169866 Putative uncharacterized protein 0.829986 0.855207 -1.12527 -0.55992 3 GM13941 668802 hypothetical protein -0.88373 1.031545 -0.83022 0.682403 4 complement factor H-related protein C GM4788 214403 1.467398 -0.26689 -0.76594 -0.43457 3 isoform 3 GM725 277899 similar to putative taste receptor T2R22 0.755507 0.353846 0.3642 -1.47355 5 Isoform 2 of N-acetylglucosamine-1- GNPTAB 432486 -1.01778 -0.51647 0.25702 1.277231 1 phosphotransferase subunits alpha/beta GOLM1 105348 Golgi membrane protein 1 -0.8855 -0.64323 0.201499 1.327231 1 GPC1 14733 Glypican-1 -1.18375 -0.45089 0.644997 0.989649 1 GPC3 14734 Glypican-3 -0.35724 1.12108 0.431664 -1.1955 5 Isoform 2 of Probable G-protein GPR112 236798 -1.26895 -0.24684 0.457863 1.057924 1 coupled receptor 112 GPR125 70693 Putative uncharacterized protein -1.38728 0.427322 0.015167 0.944789 1 GPR126 215798 G-protein coupled receptor 126 0.410548 -0.55032 -1.05235 1.192123 4 GPR85 64450 Probable G-protein coupled receptor 85 0.238095 -0.69729 -0.85476 1.313956 4 GRID1 14803 Glutamate receptor delta-1 subunit 0.707287 -0.64541 -1.05105 0.989179 4 GRIK1 14805 Glutamate receptor, ionotropic kainate 1 -0.55872 -0.67862 1.473598 -0.23626 2 glutamate receptor, ionotropic, kainate 3 GRIK3 14807 -0.98051 0.559876 1.114937 -0.69431 2 precursor Glutamate [NMDA] receptor subunit GRIN2A 14811 -1.08257 -0.24441 -0.00278 1.32977 1 epsilon-1 Glutamate [NMDA] receptor subunit GRIN2B 14812 -0.19218 0.319678 1.130558 -1.25806 5 epsilon-2 GUCY2G 73707 Guanylate cyclase 2G -1.30851 1.042459 -0.15479 0.420844 2 H2-K1 14972 42 kDa protein -0.30564 1.376946 -0.06834 -1.00296 5 H-2 class I histocompatibility antigen, H2-L 14980 -0.87479 -0.6205 0.142756 1.352533 1 L-D alpha chain H60A 15101 Putative uncharacterized protein 0.040759 0.090138 -1.28644 1.155544 4 HAS1 15116 Hyaluronan synthase 1 0.938786 0.654655 -1.2553 -0.33814 3 Hepatitis A virus cellular receptor 2 HAVCR2 171285 -0.96192 1.395794 -0.34411 -0.08976 2 homolog HEATR7B2 223825 HEAT repeat-containing protein 7B2 -0.70676 4.34E-05 -0.70748 1.414199 4 Erythroid cell-specific and testis- HEMT1 15202 -0.12838 -0.66454 -0.65889 1.451815 4 specific protein 2 HIGD1B 75689 HIG1 domain family member 1B -1.19765 -0.09431 0.04507 1.246884 1 HMCN1 545370 Uncharacterized protein -1.02111 0.109716 -0.41893 1.330331 1 HS2ST1 23908 Heparan sulfate 2-O-sulfotransferase 1 -1.01819 -0.08532 -0.26896 1.372469 1 HTR7 15566 5-hydroxytryptamine receptor 7 0.969044 0.383269 -1.38315 0.030838 3 HTT 15194 huntingtin 0.136196 -0.15987 -1.20381 1.227488 4 HYOU1 12282 Hypoxia up-regulated protein 1 -1.05135 -0.65414 0.93153 0.773961 1 Isoform 1 of Intercellular adhesion ICAM1 15894 0.973706 0.407471 -0.00793 -1.37325 5 molecule 1 ICOSL 50723 Isoform 2 of ICOS ligand -1.1584 1.284381 -0.04645 -0.07954 2 IFNAR1 15975 Interferon alpha/beta receptor 1 -1.21265 0.090393 -0.10655 1.228805 1 IFNGR2 15980 Putative uncharacterized protein -1.13266 -0.3395 0.227054 1.245101 1 Isoform 1 of Immunoglobulin IGDCC3 19289 -0.91292 -0.64636 0.263348 1.295933 1 superfamily DCC subclass member 3

124

IGF1R 16001 Insulin-like growth factor 1 receptor -0.70026 -0.51096 -0.26475 1.475964 4 Cation-independent mannose-6- IGF2R 16004 -1.27044 -0.24505 0.45977 1.055718 1 phosphate receptor IGSF8 140559 Immunoglobulin superfamily member 8 0.753038 0.82899 -0.29387 -1.28816 5 IGSF9 93842 Isoform 1 of Protein turtle homolog A -1.14909 0.727978 0.938903 -0.51779 2 IL17RA 16172 Interleukin-17 receptor A -0.31695 -0.23042 -0.88749 1.434856 4 IL17RD 171463 Isoform 1 of Interleukin-17 receptor D 0.603021 0.823699 -0.02775 -1.39897 5 INSR 16337 Insulin receptor -1.08326 -0.35041 0.135421 1.298242 1 ITFG3 106581 Protein ITFG3 0.660628 0.574219 -1.47516 0.240312 3 ITGA1 109700 Integrin alpha-1 -1.04861 -0.43319 0.18642 1.295379 1 ITGA11 319480 Integrin alpha-11 -1.18746 0.076611 1.250791 -0.13994 2 ITGA2B 16399 Integrin alpha-IIb -0.66277 -0.55112 -0.2649 1.478793 4 ITGA3 16400 Isoform Alpha-3A of Integrin alpha-3 -0.83575 -0.0308 1.414376 -0.54783 2 ITGA5 16402 Integrin alpha-5 1.379893 -0.11913 -0.25161 -1.00916 3 Isoform Alpha-6X1A of Integrin alpha- ITGA6 16403 -0.69288 0.948235 0.763435 -1.01879 2 6 ITGA8 241226 Isoform 1 of Integrin alpha-8 -0.817 -0.63609 0.066175 1.386912 1 ITGA9 104099 integrin alpha 9 isoform a -0.2073 -0.7344 -0.5227 1.464399 4 ITGAV 16410 Integrin alpha-V -0.94761 -0.44945 0.018776 1.378286 1 ITGB1 16412 Integrin beta-1 0.184546 0.220261 -1.3931 0.988288 4 ITGB5 16419 Isoform Beta-5B of Integrin beta-5 -0.72889 -0.46026 -0.28571 1.474869 4 Isoform 1 of Inositol 1,4,5-trisphosphate ITPR2 16439 -1.43591 0.848183 0.142117 0.445606 2 receptor type 2 JAG1 16449 Protein jagged-1 0.801429 0.336732 -1.46256 0.324401 3 JAM2 67374 Putative uncharacterized protein -1.22975 0.741477 0.883951 -0.39568 2 Potassium voltage-gated channel KCNB2 98741 1.315599 0.105149 -1.06314 -0.35761 3 subfamily B member 2 Potassium voltage-gated channel KCNH1 16510 -0.12572 -0.73604 -0.58674 1.448505 4 subfamily H member 1 KCNJ4 16520 Inward rectifier potassium channel 4 0.912124 0.570426 -1.35106 -0.13149 3 Vascular endothelial growth factor KDR 16542 -0.40684 -0.62767 -0.45881 1.49332 4 receptor 2 Isoform D1 of Killer cell lectin-like KLRA4 16635 -1.21473 0.635588 0.979772 -0.40063 2 receptor 4 L1CAM 16728 Neural cell adhesion molecule L1 -0.49669 -0.54703 -0.45523 1.498943 4 LAMA1 16772 Laminin subunit alpha-1 0.68594 0.84145 -1.33597 -0.19142 3 LAMP1 16783 Putative uncharacterized protein -1.02944 -0.57363 0.397621 1.205447 1 LDLR 16835 Low-density lipoprotein receptor 0.454333 0.971692 -0.06779 -1.35823 5 Isoform 1 of Leukemia inhibitory factor LIFR 16880 0.091489 -0.18773 -1.16673 1.26298 4 receptor LNPEP 240028 Leucyl-cystinyl aminopeptidase 0.704282 0.969228 -0.55017 -1.12334 3 LPAR4 78134 Lysophosphatidic acid receptor 4 -0.5558 0.570927 1.079868 -1.095 2 LPHN1 330814 Isoform 2 of Latrophilin-1 0.886729 0.828615 -0.69048 -1.02486 3 LPHN2 99633 Lphn2 protein -0.40684 -0.62767 -0.45881 1.49332 4 Prolow-density lipoprotein receptor- LRP1 16971 -0.97596 -0.37655 -0.02769 1.380197 1 related protein 1

125

Low density lipoprotein-related protein LRP1B 94217 -0.55267 1.067766 -1.10069 0.585592 4 1B Low-density lipoprotein receptor-related LRP2 14725 -1.12226 0.120308 -0.28137 1.283316 1 protein 2 Isoform 1 of Low-density lipoprotein LRP4 228357 -1.04709 -0.39259 0.122699 1.316977 1 receptor-related protein 4 Low-density lipoprotein receptor-related LRP5 16973 -1.27024 -0.23392 0.435449 1.068713 1 protein 5 Leucine-rich repeat-containing protein LRRC8C 100604 0.116889 0.572478 0.755542 -1.44491 5 8C LRRN1 16979 Leucine-rich repeat neuronal protein 1 0.439624 0.950397 -0.0104 -1.37963 5 LRRN4 320974 Neuronal leucine rich repeat-4 -1.08981 -0.28701 0.063063 1.313759 1 Tumor necrosis factor receptor LTBR 17000 -0.75127 -0.4292 -0.29142 1.471889 4 superfamily member 3 LY75 17076 Lymphocyte antigen 75 -0.06967 -0.13543 -1.11314 1.318231 4 Cation-dependent mannose-6-phosphate M6PR 17113 -0.4752 -0.10834 -0.84669 1.430231 4 receptor MAN1B1 227619 Mannosidase alpha class 1B member 1 -0.72461 0.856463 0.864858 -0.99671 2 Membrane-bound transcription factor MBTPS1 56453 -1.14018 -0.45155 0.45726 1.134465 1 site-1 protease MCOLN1 94178 Isoform 1 of Mucolipin-1 -0.93897 -0.76491 0.649863 1.05401 1 Multiple epidermal growth factor-like MEGF10 70417 0.351387 -1.01458 -0.57042 1.233606 4 domains protein 10 Major facilitator superfamily domain- MFSD2A 76574 0.120761 0.577942 0.74776 -1.44646 5 containing protein 2A Isoform 3 of Melanoma inhibitory MIA3 338366 0.524612 -1.29256 -0.23215 1.000092 4 activity protein 3 MME 17380 Neprilysin 1.318667 -0.07952 -1.11306 -0.12609 3 MMP15 17388 Matrix metalloproteinase-15 -1.30121 -0.18416 0.45131 1.034054 1 MOXD1 59012 DBH-like monooxygenase protein 1 0.538879 -1.14638 1.081857 -0.47436 5 myelin protein zero-like protein 1 MPZL1 68481 -0.52685 -0.05773 -0.83648 1.421056 4 isoform b mas-related G-protein coupled receptor MRGPRA3 233222 -0.64941 -0.15761 -0.6519 1.458921 4 member A3 MRGPRF 211577 Uncharacterized protein 0.524612 -1.29256 -0.23215 1.000092 4 MUP3 17842 Major urinary protein 3 0.099368 0.162462 1.079344 -1.34117 5 probable E3 ubiquitin-protein ligase MYCBP2 105689 -0.58613 -0.96922 1.281422 0.273921 1 MYCBP2 NBEA 26422 Isoform 3 of Neurobeachin -0.69613 -0.58509 -0.18168 1.462897 4 Isoform 1 of Neural cell adhesion NCAM1 17967 0.691282 0.000791 -1.41436 0.72229 4 molecule 1 NCAN 13004 Neurocan core protein -0.05507 0.685264 0.763978 -1.39417 5 NCSTN 59287 Nicastrin -1.22861 -0.08611 1.21362 0.101104 2 NELL2 54003 protein kinase C-binding protein NELL2 -1.43594 0.407098 0.863054 0.165787 2 NEO1 18007 Isoform 1 of Neogenin 0.439675 0.879293 0.103304 -1.42227 5 NFASC 269116 neurofascin isoform 3 precursor -1.18584 -0.43992 0.614512 1.01125 1 NPC1 18145 Niemann-Pick C1 protein -0.40756 0.843475 0.788877 -1.22479 5 NPNT 114249 Isoform 4 of Nephronectin -1.39723 0.065897 0.385344 0.945994 1 NPR1 18160 Atrial natriuretic peptide receptor 1 0.483844 0.779543 0.192969 -1.45636 5 NPTN 20320 Isoform 1 of Neuroplastin -0.2823 0.206456 -1.16099 1.23684 4

126

NPTXR 73340 Neuronal pentraxin receptor 0.646209 1.04758 -0.68795 -1.00584 3 NRADD 67169 Putative uncharacterized protein 0.487235 -0.7267 -0.93047 1.169932 4 NRP1 18186 Neuropilin-1 -0.84334 -0.12086 -0.46903 1.433239 4 NRP2 18187 Isoform A22 of Neuropilin-2 -0.95544 -0.75717 0.702533 1.010082 1 Isoform GP145-TRKB of BDNF/NT-3 NTRK2 18212 0.615707 0.24899 0.612892 -1.47759 5 growth factors receptor Isoform 3 of NT-3 growth factor NTRK3 18213 -1.30169 -0.13024 0.353031 1.078896 1 receptor ODZ3 23965 Isoform 1 of Teneurin-3 -1.1771 -0.31135 0.291577 1.196867 1 ODZ4 23966 Isoform 3 of Teneurin-4 -1.40352 0.925876 0.41032 0.06732 2 OLFR1079 258402 Olfactory receptor MOR189-1 -0.47011 -0.57059 -0.45737 1.498076 4 OLFR113 258286 Olfactory receptor 113 0.518265 0.599735 0.375547 -1.49355 5 OLFR1131 258652 olfactory receptor 1131 -0.72461 0.856463 0.864858 -0.99671 2 OLFR1392 258462 olfactory receptor 1392 -1.11056 -0.57591 0.763042 0.923437 1 OLFR339 258951 Olfactory receptor Olfr339 0.946613 0.652435 -0.35244 -1.24661 3 OLFR368 258371 Olfr368 protein 0.61847 -0.86148 -0.83918 1.082187 4 OLFR781 258723 Olfactory receptor Olfr781 0.733961 0.522899 0.207641 -1.4645 5 OLFR888 258416 Olfactory receptor Olfr888 0.029641 0.324521 -1.36685 1.012686 4 Isoform 1 of Metalloendopeptidase OMA1 67013 -1.13828 0.058833 -0.20803 1.287474 1 OMA1, mitochondrial OTOF 83762 Isoform 1 of Otoferlin 0.085544 0.519191 -1.42937 0.824636 4 P2RX7 18439 P2X purinoceptor 7 0.08044 1.20826 -0.05136 -1.23734 5 Peptidyl-glycine alpha-amidating PAM 18484 0.602505 0.881541 -0.12612 -1.35793 5 monooxygenase PANX1 55991 Pannexin-1 1.046875 -0.88347 -0.82675 0.663342 4 PCDH1 75599 Protocadherin 1 -0.02267 0.95955 0.437062 -1.37395 5 PCDH19 279653 protocadherin-19 isoform a -0.04274 -0.63078 -0.75262 1.426133 4 PCDH7 54216 protocadherin 7 isoform 2 -1.09119 -0.54718 0.529583 1.108796 1 PCDHA5 12941 Protocadherin alpha 5 1.256388 -0.22361 -1.16376 0.130981 3 PCDHGA12 93724 Protocadherin gamma subfamily A, 11 -0.10229 -1.35467 0.512779 0.944177 1 PCDHGA7 93715 protocadherin gamma subfamily A, 7 0.252121 -0.78546 -0.77671 1.310044 4 PCDHGB5 93702 Protocadherin gamma B5 -0.58529 -0.47473 -0.43701 1.497027 4 PCDHGB7 93704 Pcdhgb7 protein -0.52779 -0.62317 -0.3385 1.489465 4 PCDHGC3 93706 protocadherin gamma subfamily C, 3 -0.8516 -0.78424 0.416813 1.219023 1 Proprotein convertase subtilisin/kexin PCSK9 100102 -0.06826 0.637454 0.816995 -1.38619 5 type 9 PCYOX1 66881 Prenylcysteine oxidase -1.35066 0.245927 1.054876 0.049852 2 Isoform 1 of Alpha-type platelet-derived PDGFRA 18595 -0.89846 -0.41245 -0.10724 1.418151 1 growth factor receptor PDPN 14726 Podoplanin -0.95502 0.898745 0.827703 -0.77142 2 Platelet/endothelial cell adhesion PECAM1 18613 1.044862 0.66522 -0.8972 -0.81288 3 molecule 1, isoform CRA_d PIGU 228812 Putative uncharacterized protein 0.170347 1.341644 -0.63786 -0.87414 3 Isoform 2 of Secretory phospholipase PLA2R1 18779 0.554556 0.911756 -0.10621 -1.36011 5 A2 receptor PLXNA1 18844 Plexin-A1 -1.4195 0.149809 0.355096 0.914595 1

127

PLXNA3 18846 plexin A3 -1.46404 0.18386 0.599269 0.680916 1 PLXNB1 235611 Plexin-B1 -1.2692 -0.06419 0.168669 1.16472 1 PLXNB2 140570 plexin B2 -1.43106 0.22508 0.307853 0.898127 1 Isoform 2 of Lipid phosphate PPAP2A 19012 0.962655 0.4323 -0.02166 -1.3733 5 phosphohydrolase 1 PRNP 19122 Major prion protein -0.40068 -0.56759 -0.52793 1.496193 4 PROM1 19126 Isoform 1 of Prominin-1 -1.41202 0.478849 0.8798 0.053375 2 PROM2 192212 Isoform 1 of Prominin-2 -0.516 1.135067 0.482218 -1.10129 5 PRSS42 235628 Serine protease 42 -1.41327 0.034935 0.849727 0.528604 1 PRTG 235472 Protogenin -0.74019 -0.71147 0.057891 1.393766 1 Prostaglandin F2 receptor negative PTGFRN 19221 1.194035 0.395468 -1.0728 -0.5167 3 regulator PTK7 71461 Tyrosine-protein kinase-like 7 -0.90804 -0.43067 -0.07021 1.408921 1 Isoform 2 of Receptor-type tyrosine- PTPRC 19264 -0.88003 1.17583 0.484248 -0.78005 2 protein phosphatase C Receptor-type tyrosine-protein PTPRF 19268 -1.33168 0.856883 -0.1983 0.673094 1 phosphatase F Receptor-type tyrosine-protein PTPRG 19270 -1.28968 0.036866 0.101723 1.151093 1 phosphatase gamma receptor-type tyrosine-protein PTPRJ 19271 -0.95115 0.944981 -0.77226 0.778425 4 phosphatase eta isoform 2 Receptor-type tyrosine-protein PTPRM 19274 -0.77979 -0.93644 0.712686 1.00354 1 phosphatase mu Isoform 1 of Receptor-type tyrosine- PTPRS 19280 -0.09496 -1.16928 -0.01 1.274236 4 protein phosphatase S PVRL1 58235 Poliovirus receptor-related protein 1 0.400271 1.221291 -0.68136 -0.9402 3 Isoform 1 of Poliovirus receptor-related PVRL3 58998 -0.97437 -0.61327 0.338751 1.248895 1 protein 3 QSOX1 104009 Isoform 1 of Sulfhydryl oxidase 1 -1.14909 0.727978 0.938903 -0.51779 2 Retinoic acid early-inducible protein 1- RAET1A 19368 0.829986 0.855207 -1.12527 -0.55992 3 alpha RHBDF2 217344 Rhomboid family member 2 -0.45193 0.00352 -0.93666 1.38507 4 ROBO4 74144 Uncharacterized protein 0.230062 1.111039 -0.03283 -1.30828 5 Tyrosine-protein kinase transmembrane ROR1 26563 -0.65762 -0.44826 -0.38384 1.48972 4 receptor ROR1 Tyrosine-protein kinase transmembrane ROR2 26564 -0.42273 -0.73344 -0.32052 1.476694 4 receptor ROR2 Proto-oncogene tyrosine-protein kinase ROS1 19886 -1.11887 1.174421 0.40078 -0.45633 2 ROS S1PR2 14739 Sphingosine 1-phosphate receptor 2 0.163826 -0.49594 -0.98981 1.321924 4 SC5D 235293 lathosterol oxidase -1.2412 0.51798 1.043372 -0.32015 2 SCARA3 219151 Scavenger receptor class A member 3 -1.19209 0.023396 -0.08478 1.253474 1 SCARB1 20778 Scavenger receptor class B member 1 -0.68025 -0.69043 -0.06339 1.434069 1 SCARB2 12492 Lysosome membrane protein 2 -0.93117 -0.73393 0.510072 1.155029 1 sodium channel, voltage-gated, type III, SCN3A 20269 -1.10345 1.32474 -0.07203 -0.14926 2 alpha Signal peptide, CUB and EGF-like SCUBE2 56788 -0.2734 -0.19903 -0.9415 1.413935 4 domain-containing protein 2 SECTM1A 209588 Secreted and transmembrane protein 1A -0.23812 0.60401 0.937663 -1.30356 5 SEL1L 20338 Isoform 1 of Protein sel-1 homolog 1 -1.12025 -0.49796 0.502699 1.115513 1

128

SEMA4A 20351 Semaphorin-4A -1.34962 0.886733 0.609 -0.14611 2 SEMA4D 20354 Semaphorin-4D -0.0496 0.991624 0.415136 -1.35716 5 SEMA5B 20357 semaphorin-5B precursor 0.249659 1.120881 -1.29441 -0.07613 3 SEMA7A 20361 Semaphorin-7A -0.34035 -0.31964 -0.8023 1.462295 4 SEZ6L2 233878 Isoform 2 of Seizure 6-like protein 2 -0.83507 -0.40265 -0.21016 1.447881 1 SIAE 22619 Isoform 1 of Sialate O-acetylesterase -0.44486 -0.59097 1.496787 -0.46096 2 Isoform 1 of Tyrosine-protein SIRPA 19261 phosphatase non-receptor type substrate -1.17919 -0.37563 0.417088 1.137737 1 1 SLC12A2 20496 solute carrier family 12 member 2 0.698467 0.718159 -0.00369 -1.41293 5 SLC12A4 20498 Putative uncharacterized protein -0.23164 -0.75709 1.465025 -0.4763 2 Isoform 2 of Solute carrier family 12 SLC12A5 57138 0.270914 1.2522 -1.07673 -0.44639 3 member 5 Isoform 1 of Solute carrier family 12 SLC12A7 20499 0.825613 0.652402 -0.10637 -1.37165 5 member 7 SLC15A4 100561 Solute carrier family 15 member 4 0.022713 -1.07452 -0.27775 1.329564 4 SLC22A21 56517 Solute carrier family 22 member 21 -0.60721 0.342528 1.245657 -0.98097 2 Isoform 1 of Solute carrier family 22 SLC22A23 73102 -0.62965 0.76439 0.935169 -1.06991 2 member 23 SLC26A6 171429 Anion exchanger SLC26A6a 0.886319 0.810451 -0.60539 -1.09138 3 Isoform 1 of Equilibrative nucleoside SLC29A1 63959 0.956932 0.743188 -0.6418 -1.05832 3 transporter 1 Solute carrier family 2, facilitated SLC2A1 20525 -0.06668 1.134881 0.219914 -1.28811 5 glucose transporter member 1 Solute carrier family 2, facilitated SLC2A3 20527 0.982188 0.610093 -0.3515 -1.24078 3 glucose transporter member 3 SLC30A1 22782 Zinc transporter 1 -1.18266 0.254046 1.207815 -0.2792 2 Sodium-dependent phosphate transport SLC34A2 20531 -0.19734 -0.82349 -0.42816 1.448998 4 protein 2B SLC39A14 213053 zinc transporter ZIP14 isoform a 0.384845 1.110675 -0.24776 -1.24776 5 SLC39A6 106957 Zinc transporter ZIP6 -0.5871 -0.51275 -0.39547 1.495326 4 SLC3A2 17254 4F2 cell-surface antigen heavy chain -1.12862 0.649205 1.011015 -0.5316 2 Large neutral amino acids transporter SLC43A1 72401 -0.21225 0.665589 0.87087 -1.32421 5 small subunit 3 Isoform 2 of Choline transporter-like SLC44A2 68682 -0.52483 1.208919 0.375271 -1.05936 5 protein 2 SLC46A1 52466 Proton-coupled folate transporter 1.113897 0.575199 -0.875 -0.81409 3 SLC5A3 53881 Sodium/myo-inositol cotransporter -0.41788 -0.57645 -0.50252 1.496849 4 Sodium- and chloride-dependent taurine SLC6A6 21366 -0.77935 0.804452 0.921568 -0.94667 2 transporter Isoform 1 of Sodium- and chloride- SLC6A8 102857 0.840233 0.580573 -0.02207 -1.39873 5 dependent creatine transporter 1 High affinity cationic amino acid SLC7A1 11987 1.035804 0.3364 -0.02562 -1.34658 5 transporter 1 Isoform 1 of Solute carrier organic SLCO3A1 108116 -0.29016 -0.14183 -0.96772 1.399716 4 anion transporter family member 3A1 SLITRK4 245446 SLIT and NTRK-like protein 4 1.373883 -0.56502 0.078304 -0.88717 3 SMO 319757 Smoothened homolog -1.46785 0.207362 0.566931 0.693554 1 SMOK2A 27263 Sperm motility kinase 2A 0.095999 -0.21176 -1.1544 1.270161 4

129

Acid sphingomyelinase-like SMPDL3B 100340 -0.11715 1.113199 0.292865 -1.28891 5 phosphodiesterase 3b SORL1 20660 Sortilin-related receptor -1.37189 1.02548 0.117049 0.229362 2 SORT1 20661 Isoform 1 of Sortilin -1.47651 0.577706 0.652199 0.246603 2 Suppressor of tumorigenicity 14 protein ST14 19143 -1.42719 0.043282 0.734737 0.64917 1 homolog STIM1 20866 Stromal interaction molecule 1 -1.42758 0.163093 0.892979 0.371503 2 Erythrocyte band 7 integral membrane STOM 13830 0.071568 -1.39874 0.944393 0.382781 1 protein STT3A 16430 Putative uncharacterized protein -1.16924 -0.47904 0.974111 0.674168 1 Dolichyl-diphosphooligosaccharide-- STT3B 68292 protein glycosyltransferase subunit -1.42702 0.066634 0.810308 0.550076 1 STT3B STX1A 20907 Syntaxin-1A -1.28459 0.150391 -0.01772 1.151913 1 SULF1 240725 Extracellular sulfatase Sulf-1 -1.18713 -0.31891 0.331751 1.174292 1 Isoform 2 of SUN domain-containing SUN1 77053 -0.99545 -0.70987 0.692763 1.01255 1 protein 1 Isoform 3 of SUN domain-containing SUN2 223697 -0.99545 -0.70987 0.692763 1.01255 1 protein 2 Isoform 1 of Sushi domain-containing SUSD2 71733 0.813818 0.874685 -0.57305 -1.11546 3 protein 2 SV2A 64051 Synaptic vesicle glycoprotein 2A -1.3552 0.687742 0.817306 -0.14985 2 Isoform 1 of Synaptophysin-like protein SYPL 19027 -1.09651 -0.32803 0.131156 1.293388 1 1 Tumor-associated calcium signal TACSTD2 56753 -0.8077 -0.73445 0.214704 1.327444 1 transducer 2 TAS2R120 387348 Taste receptor type 2 member 120 0.532193 0.684847 0.25957 -1.47661 5 TAS2R137 574417 taste receptor type 2 member 3 0.423243 -1.26593 -0.23563 1.078317 4 TBC1D4 210789 140 kDa protein 1.059525 -1.31921 0.357186 -0.0975 4 TFRC 22042 Transferrin receptor protein 1 -0.60624 1.288069 0.267734 -0.94957 5 Transforming growth factor beta TGFBR3 21814 0.483844 0.779543 0.192969 -1.45636 5 receptor type 3 Trans-Golgi network integral membrane TGOLN1 22134 -1.33341 -0.157 0.536997 0.953414 1 protein 1 THSD7A 330267 Uncharacterized protein -0.6149 -0.50324 -0.37461 1.492753 4 THY1 21838 Thy-1 membrane glycoprotein -0.45175 0.880886 0.767237 -1.19637 5 TLL1 21892 Isoform 1 of Tolloid-like protein 1 -1.21655 -0.41849 0.753558 0.881485 1 TLR2 24088 Toll-like receptor 2 0.480016 0.895085 0.027626 -1.40273 5 TLR4 21898 Toll-like receptor 4 -0.78604 0.667326 1.041643 -0.92293 2 Isoform 1 of TM2 domain-containing TM2D1 94043 -1.08857 0.503156 1.126522 -0.54111 2 protein 1 Isoform 2 of TM2 domain-containing TM2D3 68634 -0.13746 -0.78303 -0.52574 1.446226 4 protein 3 TM7SF2 73166 Putative uncharacterized protein -1.38682 -0.07432 0.687205 0.773931 1 Transmembrane 9 superfamily member TM9SF1 74140 -0.75822 -0.60075 -0.07577 1.434736 1 1 Isoform 2 of Transmembrane channel- TMC5 74424 -1.21655 -0.41849 0.753558 0.881485 1 like protein 5 TMEFF1 230157 Isoform 1 of Tomoregulin-1 -0.28668 -1.28829 0.882019 0.692946 1 TMEM132A 98170 Transmembrane protein 132A -1.05232 -0.34923 0.072889 1.328663 1

130

TMEM150A 232086 Transmembrane protein 150A -0.19275 1.17878 0.244276 -1.23031 5 TMEM181A 77106 Putative uncharacterized protein -1.20602 -0.35217 0.456989 1.101204 1 TMEM196 217951 transmembrane protein 196 0.926349 0.063339 0.412935 -1.40262 5 TMEM2 83921 Transmembrane protein 2 -1.05338 1.048205 0.631732 -0.62656 2 TMEM200A 77220 Transmembrane protein 200A -1.45574 0.545378 0.745866 0.164499 2 TMEM87A 211499 Transmembrane protein 87A -0.70175 -0.67563 -0.05376 1.431146 1 Isoform 1 of Transmembrane protein TMEM87B 72477 -1.39543 -0.02812 0.572571 0.850973 1 87B TMEM9 66241 Transmembrane protein 9 -1.01095 -0.22006 -0.14996 1.380964 1 TMEM9B 56786 Transmembrane protein 9B -0.3422 -0.57945 -0.56933 1.490974 4 Isoform 1 of Transmembrane and TPR TMTC4 70551 -1.06365 -0.55437 0.453827 1.164198 1 repeat-containing protein 4 Tumor necrosis factor receptor TNFRSF10B 21933 0.238712 1.174577 -0.17528 -1.23801 5 superfamily member 10B Isoform 2 of Tumor necrosis factor TNFRSF22 79202 -0.22588 -0.38603 -0.83701 1.448922 4 receptor superfamily member 22 TOR2A 30933 Isoform 1 of Torsin-2A -1.02875 -0.68618 0.870499 0.844432 1 TPBG 21983 Trophoblast glycoprotein -1.45165 0.252209 0.834014 0.365424 2 Isoform 1 of Two pore calcium channel TPCN1 252972 -0.44322 -0.47704 -0.57731 1.497567 4 protein 1 TPST2 22022 protein-tyrosine sulfotransferase 2 -0.57055 0.187664 1.324129 -0.94124 2 Translocating chain-associated TRAM2 170829 -0.99705 -0.23628 -0.15453 1.387868 1 membrane protein 2 Isoform 1 of TPR and ankyrin repeat- TRANK1 320429 0.742063 0.916368 -1.17157 -0.48686 3 containing protein 1 Short transient receptor potential TRPC6 22068 -1.0974 -0.20066 -0.02661 1.324668 1 channel 6 transient receptor potential cation TRPM3 226025 channel, subfamily M, member 3 -0.04696 1.131838 0.208646 -1.29352 5 isoform b Isoform 1 of Transient receptor potential TRPM4 68667 -0.75127 -0.4292 -0.29142 1.471889 4 cation channel subfamily M member 4 TSPAN13 66109 Tetraspanin-13 -0.31288 -0.72035 -0.445 1.478233 4 TSPAN15 70423 tetraspanin 15 0.887977 0.823495 -1.04118 -0.67029 3 TSPAN31 67125 Tetraspanin-31 -0.2497 0.702787 0.855373 -1.30846 5 TSPAN6 56496 Tetraspanin-6 -1.26949 0.037359 1.176384 0.055751 2 TSPAN9 109246 Tetraspanin-9 0.881018 0.381836 -1.43162 0.168764 3 Thioredoxin domain-containing protein TXNDC15 69672 -1.39955 0.078846 0.945772 0.374927 2 15 Isoform 2 of Tyrosine-protein kinase TYRO3 22174 -0.97494 0.453397 -0.66332 1.18487 4 receptor TYRO3 ubiquitin-conjugating enzyme E2 J2 UBE2J2 140499 0.988355 0.502911 -1.31935 -0.17192 3 isoform a UNC5C 22253 Isoform 1 of Netrin receptor UNC5C -1.08153 -0.58352 0.604948 1.060107 1 USH2A 22283 Isoform 1 of Usherin -0.44486 -0.59097 1.496787 -0.46096 2 VANGL2 93840 Vang-like protein 2 0.660628 0.574219 -1.47516 0.240312 3 VASN 246154 Vasorin -0.49118 -0.66521 -0.32944 1.485834 4 Isoform 1 of Vascular cell adhesion VCAM1 22329 -1.18464 -0.22261 0.175935 1.231312 1 protein 1

131

VMN1R201 171255 Vomeronasal receptor V1RI4 0.413808 -0.38637 -1.17111 1.14367 4 VMN2R12 627569 vomeronasal 2, receptor 12 -0.3632 0.733773 0.877963 -1.24854 5 VMN2R16 384220 vomeronasal 2, receptor 16 0.079817 -0.1605 -1.17716 1.257843 4 VMN2R97 627367 vomeronasal receptor Vmn2r97 -0.23885 1.215046 0.215514 -1.19171 5 Isoform 1 of V-set and immunoglobulin VSIG1 78789 -0.34031 -0.92498 -0.15098 1.416267 4 domain-containing protein 1 Vesicle transport through interaction VTI1B 53612 -1.32224 0.11879 1.10839 0.095062 2 with t-SNAREs homolog 1B tRNA (guanine-N(7)-)- WDR4 57773 -0.41405 1.271714 0.220598 -1.07826 5 methyltransferase subunit WDR4 WNT5B 22419 Protein WNT-5b -0.89583 -0.73892 0.420343 1.21441 1 WNT7A 22421 Protein WNT-7a -0.83568 0.95097 0.776216 -0.89151 2 ZFP60 22718 Putative uncharacterized protein 1.092622 0.604228 -0.87522 -0.82163 3 ZMYM3 56364 Zinc finger MYM-type protein 3 0.079817 -0.1605 -1.17716 1.257843 4

Chapter 2

Yoon C, Song H, Yin T, Bausch-Fluck D, Frei AP, Kattman S, Dubois N, Witty AD, Hewel JA, Guo H, Emili A, Wollscheid B, Keller G, Zandstra PW. FZD4 marks lateral plate mesoderm and signals with NORRIN to increase cardiomyocyte induction from pluripotent stem cell-derived cardiac progenitors. Stem Cell Reports 2018 Jan 9; 10(1): 87–100.

132