The Role of the X-Chromosomal Porcupine Homolog Gene in Mouse Development

The Role of the X-Chromosomal Porcupine homolog Gene in Mouse Development

Steffen Biechele

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

The Role of the X-Chromosomal Porcupine homolog Gene in Mouse Embryonic Development

Steffen Biechele

Doctor of Philosophy

Department of Molecular Genetics University of Toronto

2013 Abstract

WNT ligands are secreted proteins that act as signals between cells. WNTs activate several interconnected signaling pathways that are required for embryonic development as well as tissue homeostasis in adults. The X-chromosomal Porcn gene encodes a membrane-bound O-acyl transferase that is required for the acylation of all 19 WNT ligands encoded in the mammalian genome. Non-acylated WNTs fail to be secreted from the producing cell and thus do not activate downstream signaling targets. In my thesis research, I have investigated the function of Porcn in mouse embryonic development. In vitro, I have shown that Porcn is required for canonical WNT signaling in ES cells and further, for their differentiation into endodermal and mesodermal derivatives. Taking advantage of a mouse line carrying a conditional (floxed) Porcn allele that I have generated, I have focused my studies on the early embryonic roles of Porcn using Cre recombinase-mediated and X chromosome inactivation-based ablation of Porcn function in vivo.

I have found that the earliest requirement for Porcn in mouse development is the induction of gastrulation. In contrast to findings from in vitro studies, I have provided evidence that Porcn is not required for pre-implantation development in vivo. Dissecting embryonic and extra- embryonic roles of Porcn, I have been able to show that Porcn is required in the extra-embryonic chorion in order to mediate chorio-allantoic fusion, whereas ablation in the extra-embryonic ii visceral endoderm had no apparent effects. The extra-embryonic requirement for Porcn results in a parent-of-origin effect in Porcn heterozygous females due to X chromosome inactivation. In contrast to the placentation defect causing embryonic lethality of maternal allele mutants, deletion of the paternal allele caused variable fetal defects resulting in perinatal lethality with only rare survivors to adulthood. Both fetuses and adults represent a mouse model for Focal

Dermal Hypoplasia (FDH), the syndrome caused by mutations in the human PORCN gene. My studies highlight the importance of PORCN-mediated WNT signaling for gastrulation, placentation, and fetal development, but suggest that endogenous WNT secretion does not play an essential role in either implantation or blastocyst lineage specification.

iii

Acknowledgments

I would like to thank my supervisor, Dr. Janet Rossant for her guidance over the course of my Ph.D, in particular, throughout the ups and downs of the Porcn project. I would also like to thank my supervisory committee Dr. Helen McNeill, Dr. Ian Scott, and Dr. Gordon Keller for seeing me through this endeavour.

I am incredibly indebted to Dr. Brian Cox, who has been a brilliant mentor, friend, and organizer of coffee breaks. Another huge thank-you to Jorge Cabezas, who always went above and beyond to help with mouse work. I want to thank Valerie Prideaux for protecting me from bureaucracy and keeping us organized, Andres Nieto for putting up with all my unfiltered thoughts, Dr. Amy Wong for teaching me how to run, Dr. Fredrik Lanner for great advice and for dissociating blastocysts, Dr. Oliver Tam for having an answer for everything, Angela McDonald for being my flow-cytometry princess, Katie Cockburn for being the queen of both blastocysts and reality TV, and Jodi Garner for countless hours of fun in- and outside of tissue culture. Thank you, also, to several unforgettable former lab members: Dr. Amy Ralston, Dr. Cheryle Seguin, Dr. Yojiro Yamanaka, Dr. Peter Rugg-Gunn and Dr. Jon Draper. I cannot express my gratitude enough for everything you all have done.

The vast majority of my project involved mice and all imaginable aspects of mouse generation and breeding, which would have been impossible without the expert help of Sue MacMaster, Sandra Tondat, Linda Wei, Dr. Monica Perreira and Marina Gertsenstein from the Transgenic Core. I would like to acknowledge the Toronto Centre for Phenogenomics, specifically, Emi Yano, Betty-Jo Edgell, and Shella Paje. As well, I would like to thank the members of the Centre for Modeling Human Disease; Dr. Ann Flenniken, Igor Vukobradovic, Lois Kelsey, Zorana Berberovic, Celeste Owen, Lily Morikawa, and Dr. Susan Newbigging. A special thank-you to Dr. Hibret Adissu who was invaluable in analyzing, interpreting and discussing phenotypes.

Last, but not least, I want to thank my friends and family. Thank you for sharing joy and pain – even when you did not know what I was talking about. Finally, thank you to my family for their endless encouragement, support and faith in me. Danke.

Table of Contents

Acknowledgments ...... iv

Table of Contents ...... v

List of Tables ...... x

List of Figures ...... xi

List of Abbreviations ...... xiv

1 Introduction ...... 1

1.1 The WNT family of secreted signaling molecules ...... 1

1.2 WNT ligand biogenesis ...... 2

1.2.1 Post-translational lipid-modification ...... 4

1.2.2 Post-translational glycosylation ...... 5

1.3 The importance of PORCN for WNT ligand biogenesis ...... 5

1.4 Specialized mechanisms for WNT secretion ...... 7

1.5 Receptors and pathways activated by WNT ligands ...... 8

1.5.1 The FZD receptor family ...... 10

1.5.1.1 Heterodimeric FZD/LRP complexes activate the canonical, Beta- Catenin-mediated WNT signaling pathway ...... 10

1.5.1.2 Non-canonical use of FZD receptors ...... 11

1.5.1.3 FZD receptor regulation by R-spondin signaling ...... 13

1.5.2 ROR receptor family ...... 13

1.5.3 RYK receptor ...... 14

1.6 WNT ligands - conserved secretion & complex response ...... 14

1.7 WNT signaling in mouse embryogenesis ...... 16

1.7.1 WNT signaling in pre-implantation development ...... 19

1.7.2 WNT signaling in post-implantation development ...... 21

1.8 WNT and PORCN mutations in human disease ...... 24 v

1.9 Thesis Research ...... 24

2 Porcupine homolog is required for canonical WNT signaling and gastrulation in mouse embryos ...... 26

2.1 Contributions ...... 26

2.2 Abstract ...... 26

2.3 Introduction ...... 26

2.4 Results ...... 27

2.4.1 Porcn expression analysis in the peri-gastrulation mouse embryo ...... 27

2.4.2 Porcn genetrap ES cells exhibit defects in canonical WNT signaling ...... 29

2.4.3 Normal NODAL secretion and signaling in Porcn genetrap ES cells ...... 32

2.4.4 Porcn null epiblasts fail to differentiate and establish anterior-posterior identity ...... 33

2.4.5 Porcn null ES cells fail to differentiate into mesoderm and endoderm derivatives in vitro ...... 38

2.5 Discussion ...... 43

2.6 Materials and Methods ...... 46

2.6.1 ES cell culture ...... 46

2.6.2 RT-PCR: ...... 46

2.6.3 Constructs ...... 46

2.6.4 Canonical WNT Activity Assay ...... 46

2.6.5 Nodal Activity Assay ...... 47

2.6.6 Embryo generation ...... 47

2.6.7 Whole-mount in situ hybridization ...... 48

2.6.8 In vitro differentiation and flow cytometric analysis ...... 48

2.6.9 Quantitative Real-time PCR analysis ...... 49

2.7 Acknowledgements ...... 50

3 Porcn deletion reveals non-essential role for WNT signaling prior to mouse gastrulation ..... 51

3.1 Contributions ...... 51

3.2 Abstract ...... 51

3.3 Introduction ...... 52

3.4 Results ...... 53

3.4.1 Generation of a Porcn floxed allele ...... 53

3.4.2 Zygotic Porcn deletion causes gastrulation defects in hemizygous male embryos ...... 55

3.4.3 Extra-embryonic deletion of Porcn produces a chorio-allantoic fusion defect and phenocopies Wnt7b mutants ...... 58

3.4.4 Porcn is not required in the visceral endoderm ...... 60

3.4.5 Porcn mediated WNT signaling is not required prior to gastrulation ...... 61

3.4.6 Normal cell fate establishment in Porcn mutant blastocysts ...... 64

3.4.7 Porcn function in ES cell maintenance ...... 69

3.4.8 PORCN-mediated WNT signaling is not required in diapause embryos ...... 72

3.5 Discussion ...... 73

3.6 Materials and Methods ...... 76

3.6.1 Generation of Porcn floxed allele ...... 76

3.6.2 Mouse alleles and genetic backgrounds ...... 76

3.6.3 Post-implantation embryo collection, staining and imaging ...... 77

3.6.4 Pre-implantation embryo collection and imaging ...... 77

3.6.5 Single cell gene expression analysis ...... 77

3.6.6 Sex-separated pregnancies ...... 81

3.6.7 Diapause induction ...... 81

3.6.8 RT-PCR ...... 81

3.6.9 Autocrine Tcf/Lef-Luciferase Assay ...... 81

3.6.10 Flow-cytometric analysis of ES cells ...... 81 vii

3.7 Acknowledgements ...... 82

4 Zygotic Porcn paternal allele deletion in mice to model human FDH ...... 83

4.1 Contributions ...... 83

4.2 Abstract ...... 83

4.3 Introduction ...... 84

4.4 Results ...... 86

4.4.1 Embryonic defects cause perinatal lethality ...... 86

4.4.2 Rare adult Porcn+/del females as a model for human FDH ...... 91

4.4.3 Skeletal defects in Porcn+/del females ...... 93

4.4.4 Novel observations in the FDH mouse model ...... 95

4.5 Discussion ...... 98

4.6 Materials and Methods ...... 100

4.6.1 Mouse alleles and genetic background ...... 100

4.6.2 Genotyping of mice and fetuses ...... 100

4.6.3 Staging and Imaging ...... 100

4.6.4 Modified SHIRPA ...... 101

4.6.5 Hematology and blood biochemistry ...... 101

4.6.6 Urinalysis ...... 101

4.6.7 Bone mineral density analysis ...... 101

4.6.8 Faxitron analysis ...... 101

4.6.9 Necropsy and histology ...... 102

4.6.10 Immunohistochemistry ...... 102

4.7 Acknowledgements ...... 102

5 Conclusions and future directions ...... 103

5.1 Summary of thesis research ...... 103

5.2 Porcn in pre-implantation development and implantation ...... 104 viii

5.3 Porcn in post-implantation development and embryo patterning ...... 105

5.4 WNT redundancy in development ...... 107

5.5 PORCN functions at the cellular level ...... 108

5.6 Implications for human disease ...... 109

5.6.1 Focal Dermal Hypoplasia ...... 109

5.6.2 PORCN as a therapeutic target ...... 111

5.7 The future of Porcn research ...... 112

References ...... 114

List of Tables

Table 1-1: Mouse WNT ligands and mutant phenotypes ...... 17

Table 1-2: Compound WNT mutant mouse phenotypes ...... 18

Table 2-1: Primers used for quantitative RT-PCR analysis of embryoid bodies ...... 50

Table 3-1: TaqMan GeneExpression Assays (Applied Biosystem) used for single cell gene expression analysis ...... 80

List of Figures

Figure 1-1: WNT ligand biogenesis and secretion ...... 3

Figure 1-2: Signaling pathways activated by WNT ligands ...... 9

Figure 1-3: Overview of mouse embryonic development up to E8.5 ...... 20

Figure 1-4: WNT signaling in the induction of gastrulation ...... 23

Figure 2-1: Porcn expression pattern in peri-gastrulation embryos ...... 28

Figure 2-2: Validation of Porcn genetrap ES cell line and canonical Wnt signaling defects ...... 31

Figure 2-3: Normal NODAL signaling in Porcn genetrap ES cells ...... 32

Figure 2-4: Normal initiation of gastrulation in Porcn genetrap aggregation embryos ...... 35

Figure 2-5: Analysis of posterior marker gene expression in Porcn genetrap aggregation embryos ...... 36

Figure 2-6: Analysis of anterior marker gene expression in Porcn genetrap aggregation embryos ...... 37

Figure 2-7: Porcn mutant embryoid bodies (EBs) fail to generate FLK1+ mesoderm and CXCR4+ endoderm in vitro ...... 39

Figure 2-8: Activation of the canonical WNT signaling cascade by exogenous WNT3A protein fails to induce mesoderm differentiation in Porcn mutant embryoid bodies (EBs) in vitro ...... 40

Figure 2-9: Gene expression analysis of Porcn mutant embryoid bodies (EBs) confirms failure in generation of endodermal and mesodermal derivatives in vitro ...... 42

Figure 3-1: Generation and characterization of a conditional Porcn allele ...... 54

Figure 3-2: In situ gene expression analysis of hemizygous zygotic Porcn mutants ...... 57

Figure 3-3: Chorio-allantoic fusion defect in Porcndel/+ female embryos ...... 59

Figure 3-4: Visceral endoderm specific deletion of Porcn ...... 60

Figure 3-5: Gastrulation defect in Porcn maternal zygotic mutants ...... 63

Figure 3-6: Pre-implantation development is unperturbed in Porcn mutants ...... 65

Figure 3-7: Pre-implantation development is unperturbed in embryos with ectopic canonical WNT signaling ...... 66

Figure 3-8: Genetic manipulation of WNT signaling has no detectable effect on lineage marker gene expression ...... 67

Figure 3-9: Genetic manipulation of Porcn and Ctnnb1 has no significant effects on direct canonical WNT signaling target gene expression at E4.5 ...... 68

Figure 3-10: Porcn is not required for maintenance of pluripotency of ES cells in vitro ...... 70

Figure 3-11: Porcn mutant ES cells maintain high levels of CD81 in vitro ...... 71

Figure 3-12: Porcn is not required for diapause in vivo ...... 72

Figure 3-13: Genotyping of Porcn mutants by sex-specific gene expression and cell numbers and fates obtained for BioMark gene expression analysis ...... 79

Figure 4-1: Schematic outlining the genetic strategy to generate Porcn+/del females ...... 86

Figure 4-2: Analysis of survival of Porcn+/del fetuses and neonates ...... 87

Figure 4-3: Gross morphological abnormalities in Porcn+/del fetuses ...... 89

Figure 4-4: Histological analysis of Porcn+/del females at E18.5 ...... 90

Figure 4-5: Adult Porcn+/del females as a model for FDH ...... 92

Figure 4-6: Body composition and skeletal phenotypes in adult Porcn+/del females ...... 94

Figure 4-7: Novel observations in Porcn+/del FDH mouse model ...... 97

Figure 5-1: Graphical summary of thesis research ...... 103

xii

xiii

List of Abbreviations

+ wildtype

AA amino acid(s)

Apc adenomatosis polyposis coli

APC Allophycocyanin

Apcmin adenomatosis polyposis coli; multiple intestinal neoplasia allele arm armadillo

AVE anterior visceral endoderm

Bcl9 B cell CLL/lymphoma 9 beta-TrCP beta-transducin repeat containing protein

BMC bone mineral content

BMD bone mineral density

Bmp4 bone morphogenetic protein 4 bp base pair(s)

Brg ATPase subunit of BAF chromatin remodeling complexes

C57MG mouse mammary epithelial cell line

C77 Cysteine residue 77 of the mouse Wnt3a gene

C93A Cysteine residue 93 of the fly wg gene mutated to Alanine

Cadherins Calcium dependent adhesion molecules

CaMK calcium/calmodulin-dependent kinase

xiv

CCAC Canadian Council for Animal Care

CD cluster of differentiation

CD31 cluster of differentiation 31, (a.k.a. PECAM1) cDNA complementary deoxyribonucleic acid

Cdx2 caudal type homeobox 2

CE convergent extension

Cer1 cerberus 1 homolog (Xenopus laevis)

Chr chromosome

CKI Casein Kinase I

CMHD Centre for Modeling Human Disease

CMTX1 Charcot-Marie-Tooth Neuropathy X Type 1

CMV Cytomegalovirus

CO2 Carbon dioxide

CRD Cysteine-rich domain

Cre cyclization recombinase (Enterobacteria Phage P1)

CSD256 mouse ES cell line carrying a genetrap insertion into the Porcn locus

Ct Cycle threshold

Ctnnb1 catenin (cadherin associated protein), beta 1, (a.k.a. beta-catenin)

Cxcr4 chemokine (C-X-C motif) receptor 4

Cys cysteine

DAB 3,3’-diaminobenizidine

DE definitive endoderm del deleted

DICOM Digital Imaging and Communications in Medicine

Dkk1 dickkopf 1

DMEM Dulbecco’s modified eagle medium

DNA deoxyribonucleic acid

DVE distal visceral endoderm

Dvl dishevelled, dsh homolog (Drosophila)

E14Tg2a.4 wildtype ES cell line

E embryonic day (e.g. E7.5)

EB embryoid body

EDTA Ethylenediaminetetraacetic acid

EGFP Enhanced green fluorescent protein

EMT epithelial to mesenchymal transition

EPI epiblast

EpiSC epiblast stem cell(s)

ER endoplasmic reticulum

ES cells embryonic stem cell(s)

ExE extra-embryonic ectoderm

xvi

EYFP enhanced yellow fluorescent protein

Fbx15 F-box protein 15

FDH Focal Dermal Hypoplasia (a.k.a. Goltz Syndrome)

Fgf fibroblast growth factor

FITC fluorescein isothiocyanate

Flk1 fetal liver kinase 1 (a.k.a VEGFR2)

Fzd frizzled homolog (Drosophila)

Gapdh glyceraldehyde-3-phosphate dehydrogenase

Gata6 GATA binding protein 6

Gbx2 gastrulation brain homeobox 2

GFP green fluorescent protein

GPCR G-protein coupled receptor

GPI glycolphosphatidylinositol

Gsk glycogen Synthase Kinase

GTPase guanosine triphosphate hydrolase

H&E Hematoxylin and eosin

H341 histidine residue 341 of mouse/human Porcn gene

HEK293T human embryonic kidney cell line 293

Hesx1 homeobox gene expressed in ES cells

Hoxb1 homeobox B1

xvii

Hspa5 heat shock protein 5 (a.k.a. Grp78, BiP)

HSPG heparin sulfate proteoglycan

ICM inner cell mass

ICR outbred albino mouse strain (named after Institute for Cancer Research)

IgG Immunoglobulin G

IHC immunohistochemistry iPSC(s) induced pluripotent stem cell(s)

IWP-2 inhibitor of Wnt production 2 (Porcn inhibitor)

JNK c-JUN N-terminal kinase (a.k.a. Mapk8)

JUN jun oncogene kDa kiloDalton

Klf4 Kruppel-like factor 4 (gut)

Lef lymphoid enhancer binding factor

Lefty1 left right determination factor 1

Lgr leucine-rich repeat-containing G protein coupled receptor

Lhx1 LIM homeobox protein 1

LIF Leukemia inhibiting factor

Lrp lipoprotein receptor-related protein

MBOAT Membrane bound O-acyl transferase

MEF mouse embryonic fibroblast

xviii

Mesd1 mesoderm development 1

MGC mammalian gene collection

Mixl1 Mix1 homeobox-like 1 (Xenopus laevis)

MMTV mouse mammary tumor virus mRNA messenger ribonucleic acid

Myc myelocytomatosis oncogene

Nanog Nanog homeobox

NLS nuclear localization signal

Oct3/4 POU domain, class 5, transcription factor 1 (Pou5f1)

OMIM Online mendelian inheritance in man

ORF open reading frame

OST oligosaccharyl transferase

Otx2 orthodenticle homolog 2 (Drosophila)

P0 Postnatal day 0 p24 endomembrane protein precursor of 24 kDa (a.k.a EMP24/GP25L/Erp)

PAS Periodic Acid-Schiff

Pax6 paired box gene 6

PBS Phosphate buffered saline

PCP planar cell polarity

PCR polymerase chain reaction

xix

PE Phycoerythrin (fluorophore for flow cytometry)

PE primitive endoderm

PECAM1 platelet/endothelial cell adhesion molecule 1 (a.k.a. CD31)

Pgap1 post-GPI-attachment-to-proteins 1 (a.k.a. Oto)

Pgfra platelet derived growth factor receptor, alpha polypeptide

PMID Pubmed identification por Drosophila gene: porcupine porcn Zebrafish gene: porcupine homolog

Porcn Mouse gene: Porcupine homolog (Drosophila)

PORCN Human gene: Porcupine homolog (Drosophila)

PORCN Porcupine homolog protein (all species)

PS primitive streak

PVE posterior visceral endoderm qPCR quantitative real-time polymerase chain reaction

Rac1 RAS-related C3 botulinum substrate 1

RBC red blood cell(s)

RhoA ras homolog gene family, member A

RNA ribonucleic acid

Ror receptor tyrosine kinase-like orphan receptor

Rspo r-spondin

RT-PCR reverse transcriptase polymerase chain reaction

Ryk receptor-like tyrosine kinase

S209 serine residue 209 of the mouse Wnt3a protein

S239A serine residue 239 of the fly wg protein mutated to alanine

SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis

Ser serine

Sfrp2 secreted frizzled related protein 2

Shh sonic hedgehog

SHIRPA SmithKline Beecham Pharmaceuticals; Harwell, MRC Mouse Genome Centre and Mammalian Genetics Unit; Imperial College School of Medicine at St Mary’s; Royal London Hospital, St Bartholomew’s and the Royal London School of Medicine; Phenotype Assessment

Sox SRY-box containing gene

SSEA-1 stage-specific embryonic antigen 1

T brachyury

Tbx6 T-box 6

TCF transcription factor 7, T cell specific

TCP Toronto Centre for Phenogenomics

TE trophectoderm tg transgene or transgenic

Tmed5 transmembrane emp24 protein transport domain containing 5

Vangl2 van gogh like 2 xxi

VE visceral endoderm

Vegfr2 vascular endothelial growth factor receptor 2 (a.k.a Flk1)

WBC white blood cell(s) wg wingless

Wls wntless (a.k.a. Gpr177, spt or evi)

Wnt wingless-related MMTV integration site wntD wnt inhibitor of dorsal (a.k.a. Wnt8, DWnt8, DWnt-8) wt wild type

X-Gal 5-bromo-4-chloro-3-indolyl-beta-D-galacto-pyranoside

XCI X chromosome inactivation

XEGFP X-linked enhanced green fluorescent protein

ZFN zinc-finger nuclease

xxii 1

1 Introduction

The development of all multicellular organisms is dependent on communication between cells in order to generate a functional organism with all required cell types in appropriate numbers in the correct locations. One of the conserved signaling cascades frequently employed for cell communication is the WNT signaling pathway, which uses a family of secreted lipid-modified glycoproteins (WNT ligands) as secreted messengers between cells. In this thesis, I report the generation and use of a tool to genetically ablate secretion of all WNT ligands by deletion of the Porcupine homolog (Porcn) gene in the mouse. While my studies have focused on embryonic roles of Porcn and thus WNT ligand secretion, this tool can be used widely throughout the embryonic or adult life of the mouse to investigate the sources and functions of WNT ligands, and the associated effects on the WNT signaling pathways in development and disease.

1.1 The WNT family of secreted signaling molecules

In 1982, the mouse Wnt1 gene was identified and opened up the exploration of this signaling pathway in mammals (Nusse and Varmus, 1982). Originally named Int-1 due to its discovery as an integration site of a mouse mammary tumor virus, it was recognized that this gene is homologous to the fly Wingless (wg) gene (Nüsslein-Volhard and Wieschaus, 1980), and the fusion term ‘Wnt’ was established several years later (Nusse et al., 1991). The fly wg gene was discovered in a screen for embryonic lethal mutants and was shown to control segment polarity during larval development (Nüsslein-Volhard and Wieschaus, 1980). Subsequent screens and epistasis experiments in the fly lead to the identification of numerous pathway members that were later identified in mammals as well. The two genes most relevant for this work are armadillo (arm) (Noordermeer et al., 1994; Siegfried et al., 1994) and its mammalian homolog beta-catenin (Ctnnb1), as well as porcupine (por) (Perrimon et al., 1989) and Porcupine homolog (Porcn) (Tanaka et al., 2002). While Beta-Catenin is a major downstream effector in signal receiving cells, PORCN is a key molecule required for the WNT biogenesis in signal producing cells (Figure 1-1).

WNT ligands are secreted lipid-modified glycoproteins and are conserved from the sea anemone Nematostella to humans. To date, 7 WNT family members have been identified in flies, 5 in worms, 15 in zebrafish, and 19 in humans and mice (Table 1-1). The identification of the

2 nucleating members Wnt1 and wg in very different contexts, oncogenesis in the mouse and embryonic development in the fly, already highlights the importance of this signaling pathway for numerous processes. The impact of this pathway on human health was first discovered in 1991, when two groups identified the Adenomatous polyposis coli (Apc) gene in a hereditary cancer syndrome (Kinzler et al., 1991; Nishisho et al., 1991). The APC protein was later shown to interact with Beta-Catenin (Rubinfeld et al., 1993; Su et al., 1993), the downstream effector of what is known today as the canonical WNT signaling cascade.

1.2 WNT ligand biogenesis

WNT proteins contain N-terminal signal peptides that direct them to the endoplasmic reticulum (ER) to enter the secretory pathway (Nielsen et al., 1997). WNT proteins are approximately 40 kiloDalton (kDa) in size and lack any conserved domains found in other protein families. One characterizing feature of WNTs is 23-26 conserved cysteine residues (Table 1-1), which are thought to form disulfide bonds to stabilize the three-dimensional structure of the protein (Tanaka et al., 2002). The ER-resident Heat shock protein 5 (HSPA5/GRP78/BiP) acts as a chaperone and is required for proper folding of WNT ligands (Burrus and McMahon, 1995; Verras et al., 2008). Further, WNT proteins are also glycosylated and lipid-modified in the ER (Figure 1-1).

Exosome Lipoprotein Direct Release particle

Endosome + H Golgi

Nucleus

Legend: RNA Polymerase Ribosome Wnt precursor protein with signal peptide Wnt precursor protein acylated Wnt protein Signal Peptidase OST Complex Porcn Hspa5

P24 receptor Pgap1

Wntless (Wls) V-ATPase

Retromer Complex Lipoprotein particle

Figure 1-1: WNT ligand biogenesis and secretion

1.2.1 Post-translational lipid-modification

While the amino acid composition of WNTs suggests a polar, water-soluble behaviour, their actual hydrophobic behaviour has been attributed to two lipid-modifications; palmitoylation of the N-terminal cysteine (C77) and acylation of a conserved serine residue (S209) with monounsaturated palmitoleic acid (Galli et al., 2007; Kurayoshi et al., 2007; Takada et al., 2006; Willert et al., 2003; Zhai et al., 2004). The residue designations are based on their position in mouse WNT3A, but both residues are conserved in all mammalian WNTs and most of the invertebrate WNT homologs.

Biochemical studies in mammalian systems have shown that mutation of the cysteine (C77A) leads to a reduction in WNT function, whereas mutation of the serine (S209A) causes complete ablation of WNT function, potentially due to retention in the ER (Doubravska et al., 2011; Takada et al., 2006). In contrast to mammals, mutations in these corresponding residues in fly wg have the opposite consequences: WG C93A (corresponding to WNT3A C77A) accumulates in the ER in vivo, whereas WG S239A (corresponding to WNT3A S209A) can be secreted, but causes a strong reduction in signaling activity in imaginal discs (Franch-Marro et al., 2008; Herr and Basler, 2012). While some in vitro studies show that S209A mutants are retained in the ER in mammalian cells (Takada et al., 2006), other studies show that this mutant variant can still be secreted (Doubravska et al., 2011; Galli and Burrus, 2011). Despite some conflicting data, all studies find that WNTs carrying S209A, or S209A in combination with C77A, are completely non-functional in assays for canonical WNT function (Biechele et al., 2011; Doubravska et al., 2011; Galli and Burrus, 2011; Galli et al., 2007; Takada et al., 2006). Intriguingly, one study suggests that S209 palmitoylation is not required for secretion from HEK293T cells and that non-palmitoylated WNT1 can activate a non-defined pathway in vitro (Galli and Burrus, 2011). Such an effect however, has not been observed to date in vivo.

While these data show that WNT palmitoylation is required for classical WNT function in both vertebrates and invertebrates, the molecular mechanisms leading to this effect have been a source for speculation. Currently, S209 palmitoylation has been shown to be necessary for both the binding to the cargo receptor Wntless (WLS) (Coombs et al., 2010; Herr and Basler, 2012; Najdi et al., 2012) and binding of WNT ligands to Frizzled (FZD) receptors on signal receiving cells (Janda et al., 2012).

A third lipid-modification of WNTs has been described; the ER-resident post-GPI-attachment-to- proteins 1 (PGAP1/OTO) protein promotes the modification of WNTs with a glycolphosphatidyl-inositol (GPI) anchor that results in WNT retention in the ER (Figure 1-1) (Zoltewicz et al., 2009). The residue(s) modified and the exact mode of action remains undetermined to date.

1.2.2 Post-translational glycosylation Based on sequence analysis, all WNTs contain 1-4 N-glycosylation sites (Table 1-1). Biochemical studies using WG, WNT1, WNT3A, and WNT5A have shown that N-glycosylation is mediated by the oligosaccharyl transferase (OST) complex (Figure 1-1) (Tanaka et al., 2002), but there is conflicting data regarding the function of N-glycosylation in WNT post-translational modification. While some studies show that N-glycosylation is required for efficient secretion and function of WNT ligands (Komekado et al., 2007; Kurayoshi et al., 2007), others show that N-glycosylation is not strictly required (Herr and Basler, 2012; Mason et al., 1992). Similarly, there is no consensus in what order these modifications occur with respect to the lipid- modifications (Komekado et al., 2007; Tanaka et al., 2002). The discrepancies between these studies are most likely specific to the cell types and assays used, and might reflect differential requirements of individual WNTs and/or cell types. In contrast to these apparently variable requirements for glycosylation, all 19 WNTs required lipid-modification in order to interact with cargo-receptor WLS and/or to be functional (Najdi et al., 2012).

1.3 The importance of PORCN for WNT ligand biogenesis

The PORCN protein is strictly required for the lipid-modification of WNT and thus the secretion and function of all mammalian as well as most fly WNT ligands (Figure 1-1). PORCN is a member of the membrane bound O-acyl transferase (MBOAT) superfamily (Hofmann, 2000) and is localized in the ER (Dodge et al., 2012; Takada et al., 2006; Tanaka et al., 2000). Based on bioinformatic analyses, PORCN has been proposed to contain 8 transmembrane domains and a catalytic histidine (H341) residue (Hofmann, 2000; Kadowaki et al., 1996). Chemical inhibition (Chen et al., 2009; Dodge et al., 2012), as well as knock-down approaches show that PORCN is required for the acylation of S209 of WNT ligands (Galli and Burrus, 2011; Takada et al., 2006). Evidence for the catalytic activity in cell-free systems is lacking, but transgenes mutant for the proposed catalytic residue (H341A) fail to rescue WNT palmitoylation in human

PORCN mutant cell lines (Proffitt and Virshup, 2012), suggesting that PORCN catalyzes S209 acylation of WNT ligands.

S209 acylation is a pre-requisite for palmitoylation of C77 (Doubravska et al., 2011). Whether PORCN or an additional enzyme catalyzes this second palmitoylation event is currently not clear; while PORCN binds the conserved domain surrounding C77 in flies (Tanaka et al., 2000; Tanaka et al., 2002), data from mouse cells suggests that C77 palmitoylation is not catalyzed by PORCN (Galli and Burrus, 2011). Another notable exception in the fly is WNTD/WNT8, which lacks the conserved serine residue and is PORCN-independent (Ching et al., 2008). In zebrafish, morpholino-mediated knock-down of porcn appears to affect only a subset of WNT ligands despite conservation of the relevant serine residue (Chen et al., 2012b). This effect might be due to incomplete ablation of PORCN function by the morpholino and/or maternal PORCN protein deposition in the egg. Studies using a mutant zebrafish porcn allele have not been reported to date.

Addressing the dependence of all 19 human WNT ligands on PORCN function systematically (Najdi et al., 2012), the activity of each WNT ligand was assessed in a PORCN mutant fibrosarcoma HT1080 cell line. Depending on the ligand, the signaling response was tested using a canonical WNT reporter construct (14/19) or Dishevelled 2 (DVL2) phosphorylation (4/19). All 18 WNTs tested were inactive when expressed in PORCN mutant cells. The activity of remaining WNT5B could not be tested using either of these assays, but WNT5B was unable to interact with the WNT cargo-receptor WLS in PORCN mutant cells, suggesting that WNT5B is also dependent on PORCN function. In summary, this study showed that all human WNT ligands are dependent on PORCN. Whether the discrepancies observed between mammals (Najdi et al., 2012) and zebrafish (Chen et al., 2012b) reflect different technical limitations (genetic vs. morpholino) or functional differences between the two systems has not been conclusively resolved.

Flies, mice and humans carry one por/Porcn/PORCN gene on their X chromosome. In contrast to flies, where this gene gives rise to only one isoform, four different isoforms (A-D) are generated by alternative splicing of exon 7 (18bp) and exon 8 (15bp) in mice and humans (Caricasole et al., 2002; Tanaka et al., 2000). While studies in differentiating embryocarcinoma cells suggest that overexpression of individual isoforms has different effects on differentiation (Tanaka et al.,

2003), the exact functions have not been described to date. As mammals have more Wnt ligand genes than flies (19 vs. 7 respectively), it is tempting to speculate that individual isoforms may have different substrate specificities. While this hypothesis has not been addressed exhaustively, data from the Virshup lab (Proffitt and Virshup, 2012), as well as my published results (Biechele et al., 2011), show that each isoform can palmitoylate all WNT ligands tested to date in overexpression studies (10/19). Despite this ‘promiscuity’ within the family of WNT ligands, no effects of PORCN on other cell communication pathways, such as Sonic hedgehog (Shh) and Notch signaling have been observed to date (Chen et al., 2009). WNT signaling, however, has recently been shown to be extremely sensitive to PORCN expression levels over at least three orders of magnitude (Proffitt and Virshup, 2012), suggesting that PORCN-mediated palmitoylation is a rate-limiting step in WNT biogenesis. While expression pattern analysis suggests that Porcn expression is regulated in a tissue-specific fashion (Barrott et al., 2011; Biechele et al., 2011; Diez-Roux et al., 2011; Liu et al., 2012), the functional details of Porcn regulation at level of gene expression or protein level remains elusive.

Apart from its function in WNT biogenesis, Covey et al. showed that PORCN has at least one additional function (Covey et al., 2012); Knock-down of PORCN in human cancer cells leads to a proliferation defect that was WNT independent. Intriguingly this proliferation defect is not dependent on the catalytic activity of PORCN. The mechanism causing this defect remains elusive.

1.4 Specialized mechanisms for WNT secretion

Cumulatively, studies in multiple cell types and organisms show that post-translationally modified WNT ligands follow a specialized exocytic pathway (Figure 1-1). In the fly, several members of the p24 family of cargo-adapter proteins shuttle WNT ligands from the ER to the Golgi network (Buechling et al., 2011; Port et al., 2011). This function appears to be conserved, as at least one of the mammalian p24 homologs (TMED5), is also required for WNT shuttling in human cells (Buechling et al., 2011).

Once in the Golgi, WNTs interact with WLS, a cargo receptor that is required for the transport of WNT ligands from the Golgi network to the cell surface. This interaction is S209 palmitoylation dependent and likely mediated by a Lipocalin-like domain in WLS (Coombs et al., 2010; Herr and Basler, 2012). The release of WNTs from WLS requires acidification of the exocytic

8 vesicles, but acidification alone is not sufficient for WNT release (Coombs et al., 2010), suggesting that additional mechanisms or proteins are involved in this process. Upon release of the WNT ligands, WLS is endocytosed and recycled by the retromer complex in a retrograde transport that shuttles WLS from endosomes to the Golgi network (Belenkaya et al., 2008; Port et al., 2008). WLS protein can thus participate in multiple rounds of WNT ligand secretion. In the fly, all WNTs except the atypical WNTD/WNT8 require WLS for their secretion (Ching et al., 2008; Herr and Basler, 2012). Consistent with these findings, mouse Wls mutants fail to initiate gastrulation, phenocopying the earliest single Wnt mutant (Wnt3) (Fu et al., 2009b; Liu et al., 1999). Similar to PORCN, WLS appears to be required for the secretion of all mammalian WNTs, as all evaluated WNTs (11/19) are WLS-dependent (Najdi et al., 2012).

In what form WNTs are secreted from the cell has remained elusive for a long time, but several recent studies have shed light on this question. A portion of WNT ligands is secreted on exosomes in mammals and flies (Gross et al., 2012). This process is dependent on cargo receptor WLS and in some, but not all cases, WLS is also a component of these exosomes (Gross et al., 2012; Korkut et al., 2009). Distinct from exosomes, WNTs have also been shown to be part of extracellular lipoprotein complexes in vitro and in vivo (Neumann et al., 2009; Panáková et al., 2005). In summary, these studies show that WNT ligands are secreted in multiple forms, including exosomes, lipoprotein particles and potentially also direct release from the plasma membrane (Figure 1-1).

1.5 Receptors and pathways activated by WNT ligands

In the three decades since the discovery of the first Wnt ligand gene, numerous receptors and downstream pathways have been identified (Angers and Moon, 2009). While it was initially thought that WNT ligands are inherently canonical (i.e. acting through Beta-Catenin) or non- canonical (i.e. all other effectors), it has become increasingly clear that the response to a particular WNT is highly dependent on the receptor(s) expressed on the signaling receiving cell. Here, I will briefly review the cell surface receptors known to bind WNT ligands as well as the major pathways downstream of them (Figure 1-2).

Fzd Ror2 Ryk Fzd/Lrp Fzd/G protein

Vangl

β γ Dvl Dvl Dvl Lrp5/6 Dvl Dvl Gα

RhoA Rac1 ++ Axin Apc Ca ++ Gsk3 PKC Ca β-Catenin CkIα Ca++

Rock JNK

CamKII Ryk fragment β-Catenin JUN Tcf/Lef

Transcription of target genes Cytoskeleton

Nucleus

Legend:

acylated Ryk intracellular Wnt ligand Fzd Ror2 Ryk fragment

Figure 1-2: Signaling pathways activated by WNT ligands

1.5.1 The FZD receptor family

The prototypical WNT receptors of the Frizzled family (Fzd) were identified as pathway components in Drosophila and vertebrates (Bhanot et al., 1996; Yang-Snyder et al., 1996). In contrast to Drosophila’s 4 Fz receptors, the human and mouse genomes encode 10 Fzd receptors (van Amerongen and Nusse, 2009). Biochemically, Fzd receptors have a conserved structure of 7 transmembrane domains and an extra-cellular cysteine-rich domain (CRD) that is required for WNT binding. While FZD receptors share similarities with G-protein coupled receptors (GPCRs), they are not prototypical GPCRs as they lack other key hallmarks, such as a G-protein binding motif in the second intracellular loop. Only recently, the structure of the WNT/FZD complex has been resolved for Xenopus XWNT8 bound to the CRD of mouse FZD8 (Janda et al., 2012). Highlighting and explaining the importance of Porcn for WNT signaling, the crystal structure revealed that the PORCN-dependent palmitoleate adduct makes multiple contacts within a hydrophobic groove of the FZD CRD (Janda et al., 2012).

1.5.1.1 Heterodimeric FZD/LRP complexes activate the canonical, Beta- Catenin-mediated WNT signaling pathway

Lipoprotein receptor-related protein 5 and 6 (LRP5/6) are single-pass transmembrane proteins of the low-density lipoprotein receptor-related family that form heterodimeric complexes with FZD (Tamai et al., 2000). This complex acts as a receptor for the activation of the canonical WNT signaling cascade (Figure 1-2). This pathway is also referred to as the WNT/Beta-Catenin pathway because its effects are mediated by transcriptional activator Beta-Catenin (reviewed in (Angers and Moon, 2009; Niehrs, 2012)).

Central to pathway activity is the post-transcriptional control of Beta-Catenin stability. In the absence of WNT ligands, cellular availability of Beta-Catenin is tightly controlled by a cytoplasmic destruction complex. This complex is composed of AXIN, APC, Casein Kinase I (CKI) and Glycogen Synthase Kinase 3 Alpha/Beta (GSK3A/B) and binds Beta-Catenin (Figure 1-2). This binding allows for the phosphorylation of N-terminal serine and threonine residues of Beta-Catenin by CKI and GSK3. This phosphorylated motif is recognized by a component of an E3 ubiquitin ligase complex, beta-TrCP, and ubiquitinated. Ubiquitinated Beta-Catenin is then rapidly degraded by the proteasome.

Binding of WNT ligands to the FZD/LRP5/6 receptor complex activates cytoplasmic Dishevelled (DVL) phosphoproteins and causes re-localization of the destruction complex. Due to these events, phosphorylated Beta-Catenin is no longer removed from the destrucation complex and degraded by the proteasome (Li et al., 2012). Activation of the pathway thus results in accumulation of cytoplasmic Beta-Catenin and subsequent translocation to the nucleus. In the nucleus, Beta-Catenin replaces members of the Groucho family of transcriptional repressors from DNA-binding proteins of the T-cell-factor/lymphoid enhancer factor (TCF/LEF) family (Figure 1-2). Beta-Catenin further recruits chromatin modifiers and transcriptional coactivators, such as BRG1, BCL9, and PYGOPUS, ultimately activating the transcription of genes containing TCF/LEF binding motifs in their regulatory sequences.

Besides its role as transcriptional activator, Beta-Catenin has a second function in adherens junctions (Peifer et al., 1992), where it mediates the interaction between Calcium dependent adhesion molecules (Cadherins) and the cytoskeleton. Only recently has it become possible to separate these two functions genetically by specific mutations in the fly arm and mouse Ctnnb1 genes (Valenta et al., 2011).

1.5.1.2 Non-canonical use of FZD receptors

Canonical WNT signaling underlies the ability of WNTs to transform the mouse mammary epithelial cell line C57MG in vitro (Brown et al., 1986; Shimizu et al., 1997) and induce a secondary axis in Xenopus embryos in vivo (McMahon and Moon, 1989). While the canonical WNT signaling pathway is defined by its Beta-Catenin-mediated activity, all other effects mediated by WNT ligands have been classified as ‘alternative’ or ‘non-canonical’ (van Amerongen, 2012). In contrast to canonical signaling that affects cell fate and proliferation, alternative WNT pathways are generally involved in the control of cell migration and polarity. These pathways remain poorly defined and it is unclear whether there are multiple completely separate pathways.

One of the FZD mediated non-canonical pathways is planar cell polarity (PCP). In contrast to apical-basal polarity, PCP defines the polarity of cells within the plane of an epithelium. Initially discovered in the fly, the best know examples for PCP are the distally pointing hairs on wing cells (Taylor et al., 1998) and the orientation of ommatidia within the compound eye of flies (Jenny et al., 2005). FZD receptors are key components of PCP in both flies and mice (Feiguin et

12 al., 2001; Guo et al., 2004), suggesting that WNT ligands that bind FZD receptors could have an impact on this process. While there is currently no evidence that WNT ligands are involved in PCP in the fly (Chen et al., 2008; Lawrence et al., 2002), at least WNT7A plays a role in PCP in vertebrates, as was shown for the orientation of stereocilia bundles in cells of the inner ear of the mouse (Dabdoub et al., 2003). While PCP defects have also been observed in Fzd6 mutant mice that show randomized directions of hair follicle orientation (Guo et al., 2004), the WNT ligand(s) potentially regulating this process, have not been identified. In zebrafish, both pipetail/wnt5b and silberblick/wnt11 mutants exhibit PCP/CE defects (Heisenberg et al., 2000; Kilian et al., 2003), substantiating that WNT ligands are involved in PCP in vertebrates.

Similar to PCP, convergent extension (CE) movements during embryogenesis of the mouse use several key components of WNT signaling and PCP, such as FZD, DVL and Van Gogh like 2 (VANGL2) (Torban et al., 2004; Wang et al., 2006). CE is a process that causes a tissue to elongate and narrow by directional intercalation of the cells in the tissue. While it is typically studied in Xenopus and zebrafish, CE has been shown to be important in mammals as well. In mammalian CE mutants, neural tube closure fails because the distance between the neural folds is increased due to a wider neural plate. Mammalian CE defects were first identified in Looptail mutants, which carry a hypomorphic allele of Vangl2 and display the neural tube closure defect craniorachischisis (Kibar et al., 2001; Torban et al., 2004).

At the molecular level, activation of the PCP/CE signaling cascade through FZD involves the activation of the small GTPases RAC1 and RHOA, which can activate JUN N-terminal kinase (JNK). Downstream effects of this pathway include cytoskeletal changes and activation of JNK- dependent transcription factors (Figure 1-2). Also independent of Beta-Catenin is the WNT/Ca++ pathway (Figure 1-2), which is mediated by FZD and acts through heterotrimeric G-proteins and/or DVL to induce Ca++ fluxes and activate calcium/calmodulin-dependent protein kinase (CaMK) II and Protein Kinase C (PKC) (Sheldahl et al., 2003; Slusarski et al., 1997a; Slusarski et al., 1997b). Similar to all non-canonical WNT pathways, the detailed mechanisms and whether these mechanisms are species- or even tissue-specific remain poorly defined.

1.5.1.3 FZD receptor regulation by R-spondin signaling

R-spondins (RSPO) are secreted WNT agonists that can promote both canonical WNT signaling and PCP signaling, and bind to several receptors. Binding of RSPO3 to Syndecan, a heparan sulfate proteoglycan (HSPG), has been shown to induce PCP signaling in Xenopus development (Ohkawara et al., 2011). R-spondins also act as stem cell growth factors in the intestine and the skin (Schuijers and Clevers, 2012). In these rapidly proliferating epithelia, R-spondins bind to Leucine-rich repeat-containing G protein-coupled receptors (LGR4, LGR5, LGR6) and can promote both PCP and Beta-Catenin-mediated signaling (Carmon et al., 2011; de Lau et al., 2011; Glinka et al., 2011). RSPO/LGR acts on the WNT signaling pathway by protecting FZD receptors from internalization (Hao et al., 2012), thereby sensitizing canonical WNT signaling. Consistent with these findings, the canonical WNT signaling response induced by RSPO/LGR signaling is abolished, when WNT ligand secretion is ablated (Kim et al., 2008).

1.5.2 ROR receptor family

The receptor tyrosine kinase-like orphan receptor (ROR) family of cell surface receptor kinases has two members (ROR1 and ROR2) in mammals and has been shown to contain an extracellular cysteine-rich WNT-binding domain similar to FZD receptors (Saldanha et al., 1998). WNT5A signaling through ROR2 is able to antagonize Beta-Catenin-mediated WNT signaling (Mikels and Nusse, 2006) and in Xenopus, WNT5A has been shown to activate JNK through ROR2 to coordinate cell polarity during morphogenetic movements (Figure 1-2) (Schambony and Wedlich, 2007; Unterseher et al., 2004). Surprisingly, studies using Ror1/2 double mutant mouse embryonic fibroblasts (MEF) show that WNT5A-activated ROR receptors are required for DVL phosphorylation but not activation of JNK or inhibition of canonical WNT signaling (Ho et al., 2012).

Consistent with a WNT5A-activated ROR2 signaling pathway, Ror2 mutant mice present with defects in limb outgrowth similar to Wnt5a mutants (Schwabe et al., 2004; Yamaguchi et al., 1999a). Linking WNT ligands and ROR2 to CE in vivo, it has recently been shown that ROR2 and VANGL2 form a complex that is activated by WNT5A and required for limb outgrowth in the mouse (Gao et al., 2011). While it is clear that WNT5A bind and activate ROR2 in vitro and in vivo, the detailed mechanisms acting intracellularly downstream of this event remain to be elucidated.

1.5.3 RYK receptor

Similar to ROR receptors, RYK is a receptor tyrosine kinase. RYK was identified as a WNT- binding protein due to the homology of its extra-cellular domain to the WNT binding domain of the secreted WNT inhibitory factor protein (WIF) (Patthy, 2000). Mammalian RYK appears to form a complex with WNT and FZD and interacts with DVL intracellularly, leading to activation of WNT/Beta-Catenin signaling (Lu et al., 2004). Interestingly, it was also shown that RYK can be cleaved and that the carboxy-terminal fragment is translocated to the nucleus upon WNT stimulation during neuronal differentiation (Lyu et al., 2008). In addition to its role in canonical WNT signaling, RYK has also recently been linked with PCP, as Ryk interacts genetically with Vangl2 and regulates its stability (Andre et al., 2012). While many questions about the underlying mechanisms remain unanswered, these results clearly show that RYK is involved in canonical and non-canonical WNT signaling (Figure 1-2).

1.6 WNT ligands - conserved secretion & complex response

As outlined above, many studies have elucidated details of the various effects of WNT signaling. The separation of WNT-activated pathways is understandable in a historic context, but a dramatic simplification. Based on the numerous WNT signaling components that participate in more than one ‘pathway’, the view of WNT signaling shifts more towards an integrated, but complex signaling network (van Amerongen and Nusse, 2009). Both WNT ligands and FZD receptors belong to multi-gene families and the complexity of intracellular WNT signaling is in part caused by the high number of ligand/receptor/co-receptor combinations (Niehrs, 2012; van Amerongen, 2012), and further complicated but the fact that single WNTs, such as WNT5A, can activate non-canonical WNT signaling, and inhibit or activate canonical WNT signaling (van Amerongen et al., 2012).

In contrast to the high number of receptors, co-receptors, and intra-cellular effectors of WNT signaling (Figure 1-2) (Niehrs, 2012), the secretion of all WNT ligands is highly conserved and dependent on the irreplaceable function of PORCN (Figure 1-1). In the absence of PORCN, WNT ligands cannot be transported to the cell surface by WLS (Najdi et al., 2012). More importantly, the PORCN-mediated lipid-modification is involved in binding of WNT ligands to FZD receptors (Janda et al., 2012). PORCN thus represents a unique bottleneck in WNT ligand

15 biogenesis and an ideal tool to ablate activation of all WNT ligand dependent signaling pathways.

1.7 WNT signaling in mouse embryogenesis

Many studies elucidating the molecular details of WNT signaling have been investigated in cell lines in vitro. The functional significance of WNT signaling however is most striking in vivo. Consistent with the discovery of Wnt1 (Int-1) in mouse mammary tumors and wg in embryonic development of the fly, numerous studies have shown that de-regulation of WNT signaling has dramatic effects on the development and life of an organism. Due to the relative ease of genetic screens, the fly has been the key organism for the discovery of Wnt genes. While not easy to manipulate genetically, the easily accessible embryos of Xenopus and zebrafish have also provided great models to test hypotheses and identify more players involved in WNT signaling. While embryos of the mouse are not as easily accessible, the mouse has proven to be an excellent model organism, as targeted mutations can be generated. Being a mammal, the mouse further represents the genetically tractable model organism evolutionarily closest to humans. Over the last two decades, mouse mutants for all major players in WNT signaling have been generated (van Amerongen and Berns, 2006).

Relevant to this thesis, I will focus my review on Wnt ligand mutants and their involvement in mouse development. The phenotypes of all zygotic Wnt ligand mutants are summarized in table 1. Highlighting the importance of WNT ligands for development, 10 out of the 19 single Wnt ligand mutants exhibit defects in embryonic and fetal development that are incompatible with postnatal survival (Table 1-1). The perinatal lethality observed in 7 out of these 10 mutants is due to a variety of defects, affecting the development of the brain, heart, lung, placenta, kidney, urogenital tract, bone and joints. Further, the morphogenesis of outgrowing structures, such as the limbs and the face are affected by mutations in Wnt ligand genes. Three single Wnt mutants cause defects that are not compatible with survival to term: Wnt3, Wnt3a and Wnt7b (Table 1-1). While Wnt3 mutant mice display the earliest phenotype in development and die due to gastrulation defects around embryonic day 7.5 (E7.5) (Liu et al., 1999), Wnt3a mutants die between E10.5 and E12.5 due to defects in mesoderm formation (Takada et al., 1994). In contrast to the requirements for Wnt3 and Wnt3a in the embryo proper, Wnt7b is required in the extra- embryonic region of the conceptus (Parr et al., 2001). WNT7B secreted from the chorion is necessary for successful fusion of the chorion with the allantois. In the absence of this fusion, placenta and embryo remain unconnected and the embryo fails to thrive due to the absence of maternal nutrients and oxygen (Parr et al., 2001).

Table 1-1: Mouse WNT ligands and mutant phenotypes

Gene Chr Length Weight # of N- Porcn Timepoint Process affected in Reference(s) for (AA) (kDa) Cys glyc target of lethality zygotic mouse mutants knockout sites Ser Wnt1 15 370 41 23 4 S224 Perinatal Brain Development (McMahon and Bradley, 1990; Thomas and Capecchi, 1990) Wnt2 6 360 40 24 1 S212 Perinatal Lung and placental (Goss et al., 2011; development Monkley et al., 1996) Wnt2b 3 389 44 26 1 S241 Viable Olfactory bulb (Goss et al., 2009; Tsukiyama and Yamaguchi, 2012) Wnt3 11 355 40 24 2 S212 E7.5 Primitive Streak (Liu et al., 1999) Formation Wnt3a 11 352 39 25 2 S209 E12.5 Mesoderm formation (Takada et al., 1994) and dysmorphology of CNS Wnt4 4 351 39 25 2 S212 Perinatal Kidney and female (Stark et al., 1994) gonad formation, (Später et al., 2006; Chondrocyte Vainio et al., 1999) maturation Wnt5a 14 379 42 24 4 S244 Perinatal Morphogenesis of (Tai et al., 2009; outgrowing structures, Yamaguchi et al., 1999a) anorectal development Wnt5b 6 359 40 24 3 S236 Viable None detected (Agalliu et al., 2009) D. Agalliu, personal communication Wnt6 1 364 40 25 2 S227 Viable Developmental (Potok et al., 2008) retardation, Stromal A. Kispert, personal cell proliferation in communication decidua Wnt7a 6 349 39 25 3 S206 Viable Sterile, limb patterning (Parr and McMahon, and female 1995) (Miller and reproductive tract Sassoon, 1998; Parr and McMahon, 1998) Wnt7b 15 349 39 25 3 S206 (1) E10.5 (1) Chorio-allantoic (1) (Parr et al., 2001) fusion (2) perinatal (2) Lung vascular (2) (Shu et al., 2002) development Wnt8a 18 354 40 23 1 S186 Viable None detected (Vendrell et al., 2012)

Wnt8b 19 350 39 24 2 S185 Viable Changes in expression (Fotaki et al., 2010) of Wnt genes in telencephalon Wnt9a 11 365 40 24 1 S221 Perinatal Cause of death (Später et al., 2006) unknown, defects in long bone and joint development Wnt9b 11 359 39 25 1 S218 Perinatal Urogenital and facial (Carroll et al., 2005) development Wnt10a 1 417 47 24 2 S268 Viable unknown A. McMahon, personal communication Wnt10b 15 389 43 24 2 S253 Viable Bone mass and (Stevens et al., 2010) mesenchymal progenitor cell defects Wnt11 7 354 39 26 5 S215 Perinatal Ureteric branching, (Majumdar et al., 2003; ventricular Nagy et al., 2010) myocardium development Wnt16 6 364 41 26 1 S226 Viable Bone cortical thickness (Zheng et al., 2012) and bone strength

The first double Wnt ligand mutant mouse model (Wnt1; Wnt3a) was published in 1997 (Ikeya et al., 1997). Not surprising, the phenotype of the double mutant was more severe than either single mutant (Ikeya and Takada, 1998; Ikeya et al., 1997). Even though this phenotype highlighted that there is redundancy in WNT signaling, only 10 more zygotic or epiblast-specific double mutant mouse models have been published (Table 1-2). In total these 11 mouse models only represent 6.4% of conceivable double mutants (11 out of 171). Further, only one triple Wnt mutant mouse model has been published (Wnt4; Wnt5a; Wnt5b), but only the neural tube defects were described (Agalliu et al., 2009). Similar to numerous double mutants, the triple mutant phenotype was more severe than all other allelic combinations of the three genes (D. Agalliu, personal communication). These results show that WNT ligands have redundant functions in vivo that have not been addressed systematically.

Table 1-2: Compound WNT mutant mouse phenotypes

Compound mutant Lethality Phenotype Reference Wnt1; Wnt3a Lethal between Defects in expansion of neural crest and CNS (Ikeya and Takada, 1998; Ikeya et al., E10.5 and E18.5 progenitors, Patterning defect in dermomyotome 1997)

Wnt1; Wnt4 perinatal Decreased thymocyte number E15-E16 (Mulroy et al., 2002) S. Jyoti, personal communication Wnt1; Wnt8a unknown No phenotype observed up to E18.5 (Vendrell et al., 2012) T. Schimann, personal communication Wnt2; Wnt2b Between E14.5 Lung agenesis, cardiovascular defects (Goss et al., 2009) and P0 E. Morrisey personal communication Wnt2b; Wnt7b Lethal by E13.5 Lung branching defects (Miller et al., 2012)

Wnt4; Wnt5a Lethal after Dysmorphology and mild hypoplasia of pituitary, (Potok et al., 2008) E13.5 abnormal motor neuron number and morphology (Agalliu et al., 2009) at E13.5 Wnt4; Wnt5b unknown Abnormal motor neuron number and morphology (Agalliu et al., 2009) at E13.5 Wnt4; Wnt5a; Wnt5b Lethal before unknown (Agalliu et al., 2009) E13.5 D. Agalliu, personal communication Wnt4; Wnt9b unknown Joint integrity defects at E15.5 Spaeter et al., 2006

Wnt5a; Wnt5b Lethal after At E13.5 More severe limb truncations than (Agalliu et al., 2009) E13.5 single mutants, shortened AP body axis, D. Agalliu, personal communication duplication of lumbar neural tube, motor neuron number and morphology at E13.5 Wnt5a; Wnt11 E11.5 Cardiac defect due to insufficient second heart (Cohen et al., 2012) field progenitor cells Wnt7a; Wnt7b E12.5 Severe CNS-specific hemorrhaging and (Stenman et al., 2008) (epiblast-specific) disorganization of neural tissue

1.7.1 WNT signaling in pre-implantation development

At the beginning of mouse embryonic development, the fertilized zygote undergoes cleavage divisions and morphological changes to form the blastocyst by E3.5 (Stephenson et al., 2012). The blastocyst is a vesicle of trophectoderm (TE) enclosing the fluid-filled blastocoel cavity and a compact group of inner cells called the inner cell mass (ICM). By the time of implantation (E4.5), cells of the ICM are committed to become primitive endoderm (PE) or early epiblast (EPI) (Figure 1-3). While the TE and PE will give rise to extra-embryonic tissues of the conceptus, the EPI cells will give rise to the embryo proper. Each cell lineage can be identified by the expression of unique transcription factors, such as Cdx2 (TE), Gata6 (PE) and Nanog (EPI). The cell fate decisions involved in the generation of these three cell types are mediated by Hippo signaling in the case of TE versus ICM (Nishioka et al., 2009), and FGF signaling in the case of PE versus EPI (Yamanaka et al., 2010).

Despite numerous WNT ligands and pathway components being expressed at the blastocyst stage (Kemp et al., 2005), there is no in vivo evidence to date that WNT signaling is required for pre- implantation development. No zygotic single or compound Wnt ligand mutant exhibits a pre- implantation phenotype (Table 1-1, 1-2). Further, genetic ablation of WNT cargo receptor Wls or canonical WNT ligand co-receptors Lrp5;Lrp6 has no effect on blastocyst development (Fu et al., 2009b; Kelly et al., 2004). Perhaps most strikingly, even when both maternally deposited and zygotic Ctnnb1 gene products are ablated, embryos form blastocysts and develop past implantation (De Vries et al., 2004; Rudloff and Kemler, 2012). Canonical WNT signaling has also been silenced in blastocysts by ectopic expression of canonical WNT signaling inhibitor Dickkopf1 (DKK1) (Xie et al., 2008). While implantation into the uterus was affected, the development of blastocysts was normal. Similar to ablation approaches, ectopic activation of canonical WNT signaling, which is achieved by removing the degradation motif from the Ctnnb1 locus (Harada et al., 1999), also had no measurable effects on pre-implantation development (Kemler et al., 2004). In summary, these studies show that both ablation or activation at least canonical WNT signaling has no effect on blastocyst development.

In contrast to these in vivo studies, it has been shown that activation of canonical WNT signaling contributes to the maintenance of pluripotent mouse embryonic stem (ES) cells in vitro. This effect has been observed by small molecule-mediated inhibition of GSK3 (Ying et al., 2008), as

20 well as addition of WNT3A protein to ES cell cultures (Berge et al., 2011). The main mode of action appears to be the de-repression of TCF3 target genes by Beta-Catenin, since ablation of TCF3 repressor function recapitulates most effects of GSK3 inhibition (Wray et al., 2011). As ES cells are derived from epiblast cells of the blastocyst (Figure 1-3), it has been suggested that WNT ligands are also required for early epiblast maintenance in vivo (Berge et al., 2011), but in vivo evidence is lacking.

E0 E4.5 E5.5 E6.5-E7.5 E8.5 Zygote Blastocyst Egg Cylinder Gastrula Pre-Implantation Post-Implantation Extraembryonic

Embryonic Epiblast Embryonic Stem Cells Stem Cells Anterior Posterior

Trophectoderm Extraembryonic Ectoderm Primitive Endoderm Visceral Endoderm Epiblast Anterior Visceral Endoderm Embryonic Ectoderm Mesoderm Definitive Endoderm

Figure 1-3: Overview of mouse embryonic development up to E8.5

1.7.2 WNT signaling in post-implantation development

After implantation of the embryo into the uterine tissue (E4.5), the epiblast and overlying TE grow into the blastocoel where they form a cylindrical embryo covered by the visceral endoderm (VE), which is a derivative of the PE (Figure 1-3). Epithelialization of the epiblast and cavitation lead to the formation of the pro-amniotic cavity in the center of the cylindrical embryo (E5.5). The inside of the cylinder is the future dorsal side of the embryo. By E6.0, a specialized group of VE cells, established at the distal tip of the cylinder, has followed the migration of distal visceral endoderm (DVE) cells to the anterior side of the embryo, and forms the anterior visceral endoderm (AVE) (Takaoka et al., 2011). The AVE acts as a signaling center by secreting inhibitors of WNT and TGF-beta signaling (Tam and Loebel, 2007). On the posterior side of the embryo, the primitive streak indicates the initiation of gastrulation at E6.5 (Figure 1-3). Gastrulation establishes the three germ layers of the embryo: ectoderm, mesoderm and endoderm. At this point, dorsal-ventral and anterior-posterior axes, as well as the three germ layers that will give rise to all embryonic tissues have been established (Figure 1-3).

Initiating gastrulation, NODAL precursor protein induces Bmp4 expression in the extra- embryonic ectoderm (Figure 1-4) (Ben-Haim et al., 2006). BMP4 initially induces the expression of Wnt3 in the proximal epiblast and later in the posterior epiblast and visceral endoderm (Ben- Haim et al., 2006; Winnier et al., 1995). WNT3 promotes its own expression in positive feedback-loops mediated by Beta-Catenin (Tortelote et al., 2012), as well as through the indirect maintenance of Bmp4 expression by induction of Nodal (Ben-Haim et al., 2006). Importantly, WNT3 induces the expression of the key transcription factor Brachyury (T) (Liu et al., 1999; Rashbass et al., 1994; Tortelote et al., 2012; Yamaguchi et al., 1999b). T marks the morphological site of gastrulation in the posterior region of the embryo: the primitive streak. The anterior region of the epiblast is protected from WNT and NODAL signaling due to the secretion of antagonists (DKK, LEFTY1, CER1) from the AVE (Tam and Loebel, 2007).

Upon expression of Brachyury, mediated by FGF4 and FGF8 (Ciruna and Rossant, 2001), epiblast cells are recruited to the primitive streak, undergo epithelial to mesenchymal transition (EMT) and migrate between epiblast and VE in a lateral/anterior direction to form mesoderm. Definitive embryonic endoderm (DE) is established by the integration of cells exiting the primitive streak into the VE epithelium by intercalation in a scattered fashion (Kwon et al.,

2008). While this process leads to a proximal displacement of VE cells (Figure 1-3), some VE cells remain in the DE layer and can contribute to the embryonic gut tube (Kwon et al., 2008).

WNT signaling is critically involved in the establishment and maintenance of the primitive streak and thus gastrulation. Wnt3 mutant embryos fail to express Brachyury and do not establish a primitive streak (Liu et al., 1999). Based on marker gene expression patterns, these embryos appear to remain in an epiblast-like state, with high expression levels of Oct3/4 (Pou5f1) and Otx2. An identical phenotype is observed in Wls mutant mice (Fu et al., 2009b), which fail to secrete WNT ligands. Similarly, mutants at the canonical WNT receptor level phenocopy this defect; Lrp5; Lrp6 compound mutants (Kelly et al., 2004), as well as LRP5/6 chaperone Mesd1 (Hsieh et al., 2003; Lighthouse et al., 2011). Similar in phenotype, Ctnnb1 mutants fail to gastrulate (Huelsken et al., 2000; Rudloff and Kemler, 2012; Valenta et al., 2011). In contrast to all other WNT signaling mutants described however, Ctnnb1 mutants also fail to establish the AVE signaling centre. This effect was attributed to WNT-independent functions of Beta-Catenin (Morkel et al., 2003).

Ectopic activation of the canonical signaling cascade, as exemplified by mis-expression of chicken Wnt8C, can result in duplication of the primitive streak (Pöpperl et al., 1997). Similarly, ablation of Axin, a negative regulator of canonical WNT signaling, can cause axial duplications (Tilghman, 1996; Zeng et al., 1997). Perhaps most strikingly, stabilization of Beta-Catenin causes premature expression of Brachyury in the epiblast and EMT, similar to the primitive streak (Kemler et al., 2004). Similar but more severely affected, Apcmin mutant embryos display defects in proximal-distal axis development and express Brachyury throughout the epiblast by E5.75 (Chazaud and Rossant, 2006).

During the course of my studies, two Porcn floxed alleles were published (Barrott et al., 2011; Liu et al., 2012). Consistent with a role for PORCN in at least WNT3 secretion, epiblast-specific Porcn mutants fail to express primitive streak marker Brachyury, and die around E7.5 (Barrott et al., 2011). Zygotic deletion has been reported to be lethal by E7.5 as well (Liu et al., 2012), but a detailed phenotypic analysis was lacking in both studies.

ExE VE

Nodal precursor protein Embryo E6.5

Extraembryonic Ectoderm

Visceral Endoderm Bmp4

Proamniotic Cavity

Epiblast Nodal Wnt3 Wnt3

Anterior Posterior Brachyury

Gastrulation

Epi

Figure 1-4: WNT signaling in the induction of gastrulation

Legend: Black arrows indicate inductive relationships. White arrows indicate PORCN-mediated signaling.

1.8 WNT and PORCN mutations in human disease

Numerous mutations in WNT pathway components, including 6 WNT ligand genes, have been associated with human diseases and developmental defects, and have been reviewed recently (Clevers and Nusse, 2012).

Similar to mice, the PORCN gene is also on the X chromosome in humans. Mutations in PORCN cause Focal Dermal Hypoplasia (FDH, OMIM#305600), also known as Goltz Syndrome (Grzeschik et al., 2007; Wang et al., 2007). The majority of patients are heterozygous females and all males observed show post-zygotic mutations, suggesting that PORCN is required for embryonic/fetal development in humans. The only exception published to date is a male Klinefelter/FDH patient (47,XXY) without detectable mosaicism for the PORCN mutation (Alkindi et al., 2012). The survival of this patient past embryonic development is most likely due to the presence of a second X chromosome carrying a wildtype PORCN allele. These observations in humans, along with my published mouse data showing that Porcn is required in the embryo proper for gastrulation (Barrott et al., 2011; Biechele et al., 2011), have substantiated the suggestion that human hemizygous PORCN mutants are embryonic lethal.

Heterozygous females however, exhibit a variable phenotype, most likely due to differences in individual XCI patterns and frequencies. Phenotypically, FDH is characterized by patchy, hypoplastic skin (Goltz, 1992). Other features of the syndrome include digital abnormalities, microphthalmia, hypodontia, kidney abnormalities, abdominal wall defects, skeletal abnormalities, and reduced bone density (Bornholdt et al., 2009c). While a number of these defects have been recapitulated in mouse chimeras and fetuses with embryo-specific deletion of the maternal Porcn allele (Barrott et al., 2011; Liu et al., 2012), no zygotic heterozygous adult females have been described to date, limiting the comparison of human and mouse phenotypes.

1.9 Thesis Research

Numerous studies have shown that ectopic activation or inactivation of WNT signaling has detrimental effects in mammals, ranging from embryonic malformations and lethality to bone maintenance problems and cancer in adults. The investigation of WNT signaling in vivo is complicated by the fact that there are 19 WNT ligands encoded in the mammalian genome, with redundant but also distinct functions. These ligands can activate multiple, sometimes opposing

25 signaling pathways with different outcomes. As the scarcity of compound Wnt mutant phenotypes published shows (Table 1-2), generation of compound mutant mice for Wnt loci is laborious. The dependence of all 19 WNT ligands on PORCN for their function however, allows functional ablation of all WNT ligands. I have investigated the role of Porcn in embryonic development in vivo. These studies aimed to identify the earliest roles for WNT ligand sources in embryonic and extra-embryonic development.

During my studies, I have generated a mouse line carrying a floxed allele for Porcn as a tool to ablate WNT secretion in vivo. Using several tissue-specific cyclization recombinase (Cre) deletions and genetic approaches, I show that WNTs are first required for the initiation of gastrulation, where ligand secretion from epiblast, but not the visceral endoderm, is required. Furthermore, I identify the first extra-embryonic requirement: chorionic PORCN is required for WNT7B-mediated chorio-allantoic fusion. Investigating the earliest hypothesized roles for WNT signaling during pre-implantation development, I demonstrate that implantation and pre- implantation development do not require secretion of embryonic WNT ligands. My studies highlight the importance of Wnt3 and Wnt7b for embryonic development and define the earliest WNT ligand requirements in mouse development in both embryonic and extra-embryonic tissues of the conceptus.

Furthermore, I have generated heterozygous female embryos and mice as a model for human FDH. I identify several frequent defects that cause perinatal lethality. Rare surviving females recapitulate numerous defects seen in human patients and thus represent a mouse model for human FDH patients carrying mutations in PORCN.

2 Porcupine homolog is required for canonical WNT signaling and gastrulation in mouse embryos 2.1 Contributions

The data in this chapter was published in Developmental Biology in July 2011 (PMID: 21554866). All experiments were designed, performed and analyzed by Steffen Biechele. Dr. Brian Cox and Dr. Janet Rossant guided experimental design and interpretation of results. Generation of ES cell – embryo aggregates and transfer into surrogates was performed by the Transgenic Core at the Toronto Centre for Phenogenomics.

2.2 Abstract

In this study, Porcn mutant mouse ES cells were used to analyze the role of PORCN in mammalian embryonic development. In vitro, I show an exclusive requirement for PORCN in WNT secreting cells and further, that any of the four PORCN isoforms is sufficient to allow for the secretion of functional WNT3A. Embryos generated by aggregation of Porcn mutant ES cells with wildtype embryos fail to complete gastrulation in vivo, but remain in an epiblast-like state, similar to Wnt3 and Wls mutants. Consistent with this phenotype, in vitro differentiated mutant ES cells fail to generate endoderm and mesoderm derivatives. Taken together, these data highlight the importance of PORCN for WNT secretion and germ layer formation in vitro and in vivo.

2.3 Introduction

In a screen for X-chromosomal embryonic lethal genes in the mouse, members of the Rossant lab identified Porcn (Cox et al., 2010), providing evidence that Porcn is required for mammalian embryonic development. While embryonic requirement for Porcn had been hypothesized based on its critical role in WNT signaling (Takada et al., 2006) and based on the absence of human males carrying zygotic mutations of X-chromosomal PORCN (Grzeschik et al., 2007; Wang et al., 2007), no vertebrate model lacking Porcn had been published. Taking advantage of the available Porcn null genetrap ES cell line used in the initial screen, I performed a detailed analysis of the Porcn mutant phenotype in pluripotent and differentiating ES cells in vitro, as well as in embryos lacking PORCN function in the epiblast in vivo.

2.4 Results

2.4.1 Porcn expression analysis in the peri-gastrulation mouse embryo

The 19 WNT ligands encoded in the mouse genome are expressed dynamically both temporally and spatially throughout embryonic development. In order to determine whether Porcn, which is required for the acylation of all mammalian WNT ligands (Najdi et al., 2012), would have dynamic or ubiquitous expression during embryo development, I analyzed the expression of Porcn in the gastrulating mouse embryo using RNA in situ hybridization.

At E6.5, Porcn is expressed in epiblast cells undergoing gastrulation movements and forming the primitive streak on the posterior side of the embryo, as well as the AVE (Fig. 2-1 A). One day later, Porcn is no longer expressed in the AVE but in the lateral proximal region of the endoderm layer surrounding the egg cylinder, the lateral regions of the epiblast and at lower levels in the primitive streak region and the migrating mesoderm (Fig. 2-1 B, E). Porcn is not detectable in the anterior most region of the embryo (Fig. 2-1 E). At E8.25, Porcn expression is restricted to the embryo proper with the exception of the region of the cardiogenic plate and foregut pocket (Fig. 2-1 C). After turning (E9.0), Porcn is expressed strongly on the dorsal side of the embryo in the neural tube where it forms a gradient with highest expression levels on the dorsal side (Fig. 2-1 D, G). Modest Porcn expression was detected throughout the cranial region and in the optic vesicles (Fig. 2-1D, F) with a marked absence of expression in the surface ectoderm (Fig. 2-1 F). Porcn was also undetectable in the developing heart (Fig. 2-1 D, G).

Although not ubiquitous, the expression pattern of Porcn at E9.0 is broad enough at these stages to encompass the published expression domains of the Wnt ligand genes (Witte et al., 2009). At gastrulation, Porcn expression in the posterior epiblast of the gastrulating embryo is consistent with the expression pattern of several WNT proteins, but its expression in the AVE was unexpected as this is a source of WNT inhibitors (Kemp et al., 2005). Wnt3, however, has been reported to be expressed in the visceral endoderm, proximal to the AVE (Rivera-Pérez and Magnuson, 2005). A functional relevance for this expression has not been reported to date. While it is possible that Porcn expression in the AVE has no function, it opens the possibility that this expression is relevant for potential non-WNT substrates.

A E6.5 B E7.5 C E8.25 D E9.0

F G AVE PS E

E Meso F G NT

V PS

Endo O

Figure 2-1: Porcn expression pattern in peri-gastrulation embryos

Wildtype mouse embryos after whole mount in situ hybridization for Porcn (A-D, F) and sections thereof (E, G). Gastrulation stage embryos and sections are oriented with their anterior to the left (A-C, E). (A) At 6.5, Porcn expression can be detected in the primitive streak (PS), as well as the anterior visceral endoderm (AVE). (B, E) Porcn expression remains restricted to the embryo proper at E7.5; expression can be detected at moderate levels in the primitive streak (PS) and migrating mesoderm (Meso) and at higher levels in the lateral regions of the epiblast and the endodermal cell layer (Endo). At E8.5 (C), Porcn expression is restricted to the embryo proper with slightly higher levels on the posterior side. After turning (D), Porcn is highly expressed in the neural tube (NT) and at moderate levels throughout the cranial region. Porcn expression cannot be detected in the surface ectoderm (arrow) of the head (F) overlying the optic vesicle (O). Porcn is also not expressed in the developing cardiac ventricle (V, Figures D, G).

2.4.2 Porcn genetrap ES cells exhibit defects in canonical WNT signaling

I obtained a mouse male embryonic stem cell line (CSD256) from Baygenomics, carrying a genetrap insertion in the second intron of Porcn, 3’ to the open reading frame start site (Fig. 2-2 A). This integration site is predicted to terminate Porcn transcripts after 136 bases of coding sequence, affecting all 4 isoforms that are generated by alternative splicing of exon 7 (18 bp) and exon 8 (15 bp) (Fig. 2-2 A) (Tanaka et al., 2000). In contrast to cell lines carrying heterozygous genetrap insertions in autosomal genes, this cell line is a functional Porcn null cell line due to male hemizygosity. Absence of Porcn transcripts was confirmed by RT-PCR using exon-junction spanning primers that allow detection of all 4 isoforms (Figure 2-2B).

As PORCN has been shown to be involved in the lipid-modification of WNT ligands in both mammals and flies, I tested whether Porcn was required for canonical WNT signaling using Tcf/Lef-luciferase assays (Veeman et al., 2003) in ES cells (Fig. 2-2 C). Mock-transfected Porcn null cells (CSD256) showed reduced luciferase expression compared to the parental wildtype cell line (E14Tg2a.4, Fig. 2-2 C). Upon transfection with a Wnt3a overexpression plasmid, luciferase activity in wildtype cells increased 39-fold, but Porcn mutant cells did not show any upregulation and luciferase expression remained at basal levels, indicating defects in canonical WNT signaling mediated by WNT3A protein (Fig. 2-2 C). Co-transfection of Wnt3a with a mixture of all four Porcn isoforms rescued this defect of mutant cells and caused a 25-fold increase in luciferase activity, but had no significant effect on the WNT3A-mediated induction in wildtype cells (Fig. 2-2 C). Transfection with the Porcn mixture alone had no effect on luciferase activity in wildtype or mutant cells (Fig. 2-2 C).

I next tested whether WNT3A palmitoylation is catalyzed by a specific PORCN isoform; Wnt3a and individual PORCN isoforms (A-D) were over-expressed in Porcn null ES cells and canonical WNT activity was measured using the Tcf/Lef-luciferase assay (Fig. 2-2 D). All of the PORCN isoforms were able to rescue the Porcn null defect in ES cells and no significant differences could be detected between cells transfected with the mixture of isoforms or individual isoforms (Fig. 2-2 D). This does not preclude some isoform-specific effect on other WNT ligands, but suggests considerable overlap in function of PORCN isoforms.

In order to confirm that lack of WNT3A palmitoylation causes the same effects in our Tcf/Lef- luciferase assays, I over-expressed WNT3A or point-mutant variants that cannot be lipid- modified (C77A, S209 and double mutant) in wildtype and Porcn mutant ES cells (Fig. 2-2 E). As expected, over-expression of wildtype WNT3A led to a strong increase (70-fold) in luciferase activity (Fig. 2-2 E). Consistent with results obtained in other systems (Doubravska et al., 2011), the S209A and double-mutant WNT3A showed no upregulation of reporter activity at all, whereas WNT3A C77A retained some activity and showed a 6-fold increase in WNT reporter activity compared to mock-transfected cells (Fig. 2-2 E). In Porcn mutant ES cells, none of the Wnt3a constructs showed any activity over basal levels (Fig. 2-2 E). These results are consistent with a previous report showing that S209 is not acylated upon knock-down of Porcn (Takada et al., 2006). Whether PORCN is also required for palmitoylation of C77 cannot be addressed in this system, as S209 acylation has been reported to be a pre-requisite for C77 palmitoylation (Doubravska et al., 2011).

I next investigated whether PORCN function is required cell autonomously (Fig. 2-2 F). The secreting cell population was transfected with EYFP or Wnt3a expression constructs and mixed with cells transfected with the canonical WNT reporter construct. Expression of WNT3A in wildtype cells led to significant upregulation of luciferase activity in both wildtype and mutant recipient cells, as compared to the EYFP control (Fig. 2-2 F). In contrast, Porcn mutant cells transfected with Wnt3a were unable to induce a response in either wildtype or mutant recipient cells (Fig. 2-2 F). Thus PORCN function is dispensable in WNT3A signal receiving cells and only required in WNT3A secreting cells, supporting the role of PORCN in WNT processing and secretion.

genetrap alternative integration splice region A B 5’ 3’ RT-PCR primers E14 (wt) CSD256 PorcnA Porcn PorcnB PorcnC PorcnD H2afz mutant (CSD256) C D Topflash Fopflash

E F

C77A C77A C77A C77A S209AS209A S209A S209A

Figure 2-2: Validation of Porcn genetrap ES cell line and canonical Wnt signaling defects

(A) Schematic representation of Porcn locus with genetrap integration site and Porcn transcripts. Untranslated regions of are indicated in dark grey, translated regions in light grey. Primers for RT-PCR are indicated by arrows and allow amplification of exons 4-9. Wildtype Porcn transcripts (PorcnA-D) as well as mutant transcript in genetrap cell line CSD256 are indicated in black. (B) RT-PCR of Porcn ES cells showing absence of Porcn transcript in CSD256 Porcn genetrap ES cell line as compared to parental wildtype E14 ES cell line. RT-PCR for H2afz was used as housekeeping control. (C) Tcf/Lef-Luciferase assay in Porcn wt (E14) and mutant (CSD256) ES cells. Porcn mutant ES cells show an absence of canonical Wnt signaling reponse upon transfection with Wnt3a expression plasmid. Co-transfection of Wnt3a expression plasmid with a mixture of PorcnA-D expression plasmids rescues the defect in canonical Wnt signaling. (D) Tcf/Lef- Luciferase assay of Porcn mutant ES cells transfected with all 4 or individual Porcn splice variants shows that any of the 4 Porcn isoforms (PorcnA-D) can rescue the canonical Wnt signaling defect in Porcn mutant ES cells. (E) Paracrine Tcf/Lef-Luciferase assay shows that Porcn is only required in Wnt signal secreting cells. Expression of Wnt3a in wt ES cells (Sender) leads to significant upregulation of Tcf/Lef-Luciferase activity, whereas expression of Wnt3a in mutant cells (Sender) had no effect on canonical Wnt reporter activity as compared to mock (EYFP) transfection, independent of the genotype of Tcf/Lef-Luciferase transfected recipient ES cells. ** p<0.01, * p<0.05, ns=not significant, error bars display standard error, wt = wildtype, mut = mutant

2.4.3 Normal NODAL secretion and signaling in Porcn genetrap ES cells

While it has been shown that PORCN is required for the acylation of WNT ligands, potential roles in other signaling pathways have not been investigated extensively. As NODAL signaling is another essential pathway for gastrulation, I investigated whether Porcn genetrap ES cells exhibit defects in NODAL secretion and signaling in an autocrine/paracrine assay using the

NODAL-responsive (n2)7-luciferase construct (Saijoh et al., 2000). Both wildtype and mutant cells showed a moderate increase in luciferase activity upon Nodal overexpression and no statistically significant differences could be detected between the two cell lines (Fig. 2-3). Similarly, no significant differences between the cell lines were found upon stimulation of the pathway with ACTIVIN A protein or inhibition using SB431542 (Fig. 2-3). These results suggest that Porcn is not involved in the secretion of NODAL ligand or the reception of NODAL signaling.

Figure 2-3: Normal NODAL signaling in Porcn genetrap ES cells

NODAL responsive (n2)7-Luciferase assay in Porcn wt (E14Tg2a.4) and mut (CSD256) ES cells. No significant differences could be detected between the two cell lines upon over-expression of Nodal, inhibition of the pathway using 10 mM SB431542 or activation of the pathway by addition of 100 ng/ml ACTIVIN A protein.

2.4.4 Porcn null epiblasts fail to differentiate and establish anterior- posterior identity

In mouse embryogenesis, canonical WNT signaling is required for gastrulation and the establishment of the anterior-posterior (A-P) axis (Huelsken et al., 2000; Liu et al., 1999). Multiple WNT ligands are expressed in posterior domains of the embryo and the primitive streak (Wnt2b, Wnt3, Wnt3a, Wnt5a, Wnt8a, Wnt11) and WNT antagonists such as Dkk and Sfrp2 are expressed by the anterior visceral endoderm (AVE) (Kemp et al., 2005). Consistent with this concept, mutations in the earliest acting WNT gene, Wnt3, causes a lack of primitive streak and the associated ingression of mesodermal cells in the posterior region (Liu et al., 1999).

We have previously shown that Porcn null embryos exhibit prolonged expression of the pluripotency marker Oct4 (Cox et al., 2010), suggesting that Porcn null embryos phenocopy Wnt3 mutant embryos (Liu et al., 1999). In order to determine the phenotype of Porcn null embryos in more detail, I analyzed expression patterns of several marker genes at gastrulation stages. As males generated from Porcn null ES cells die in utero and hence cannot be used to generate a viable mouse line, Porcn null embryos were generated by aggregation of mutant ES cells with wildtype GFP-marked host embryos. Only those embryos in which the entire epiblast was ES-derived (40% of dissected aggregation embryos), as judged by absence of GFP expression, were used for this analysis.

As Wnt3 mutant embryos display a gastrulation phenotype, I analyzed the expression of early gastrulation markers at E6.5 by in situ hybridization, to determine whether gastrulation was initiated in Porcn null embryos. Wnt3 expression on the posterior side of the embryo (Fig. 2-4 A) precedes primitive streak formation and appears normal in Porcn aggregation embryos (n=3/4, Fig. 2-4 B). Further, all embryos analyzed showed expression of Cer1 (n=4, Fig. 2-4 C, D) and Lhx1 (n=5, Fig. 2-4, E-F’) in the AVE, confirming the proper formation of this anterior signaling centre required for gastrulation. Several embryos also showed posterior embryonic Lhx1 expression (n=3/5, Fig. 2-4 F’). Although this expression region was smaller than in control embryos (Fig. 2-4 E), its presence suggests the initiation of gastrulation.

At later stages, Wnt3 marks the primitive streak (Fig. 2-5 A) and is expressed in ectopic patches throughout the Porcn mutant embryo proper at both E7.5 (n=4) and E8.5 (n=5, Fig. 2-5 B, D). As Porcn is required for the lipid modification of WNT3, this expression is not expected to result

34 in secretion of functional WNT3 ligand. In order to determine whether the canonical WNT signaling cascade was active in Porcn null embryos, we used Axin2, a direct target and negative feedback regulator of canonical WNT signaling as a read-out (Jho et al., 2002). Axin2 expression was detected at lower levels and in an ectopic, patchy pattern compared to wildtype control embryos (n=16), suggesting that canonical WNT signaling was considerably reduced in Porcn mutant embryos (Fig. 2-5, E-H).

Brachyury (T) is also a direct target of canonical WNT signaling (Yamaguchi et al., 1999b) and required for gastrulation in the mouse embryo where it marks the primitive streak at E7.5 (Fig. 2- 5 I) and axial mesoderm and the tailbud at E8.5 (Fig. 2-5 K). In the majority of Porcn null aggregation embryos (20/24), T was only detectable in 1-3 small ectopic patches at E7.5 (n=16) and E8.5 (n=8, Fig. 2-5 J, L). The remaining embryos showed either a normal looking primitive streak (2/24) or complete absence of T (2/24). These observations, in combination with the observed patches of mesodermal Lhx1 expression at E6.5 (Fig. 2-4 F’), suggest that primitive streak formation is initiated, but a fully developed streak is not formed. Expression of HoxB1 was undetectable in all Porcn null embryos analyzed (n=7, Fig. 2-5 M-P), confirming the absence of posterior cell fates normally induced by posteriorizing WNT signals.

As the posterior markers T and Hoxb1 were markedly reduced or absent, I next examined the expression of the anterior markers Hesx1 and Otx2 in order to determine if anterior neuro- ectodermal cell fate was established properly or possibly expanded due to a lack of posteriorizing signals. Porcn null embryos lacked Hesx1 expression entirely or showed a strongly reduced expression (n=9, Fig. 2-6 B). In contrast, strong Otx2 expression was detectable throughout the embryo proper (n=9, Fig. 2-6 D, F). In wildtype embryos, Otx2 is dynamically expressed, marking the epiblast (E5.0), anterior visceral endoderm (AVE, E6.5) or anterior neuroectoderm (E7.5, E8.5, Fig. 2-6 C, E). As previously described (Cox et al., 2010), Porcn null embryos maintain a strong expression of Oct3/4 (Pou5f1) in the embryonic ectoderm.

While the expression patterns of Wnt3, Axin2 and T suggest that the primitive streak may be initiated at E7.5 but not maintained, Porcn null embryos fail to establish posterior cell fates based on HoxB1 expression. The strong expression of Oct3/4 (Cox et al., 2010) and Otx2 is likely a continuance of the early epiblast expression from E5.0 rather than an expansion of

35 anterior fates. This conclusion is supported by the observation of the reduction in Hesx1 expression.

A Wnt3 B Wnt3

E6.5 wt E6.5 mut C Cer1 D Cer1

E6.5 wt E6.5 mut E Lhx1 F Lhx1 F’ Lhx1

E6.5 wt E6.5 mut E6.5 mut

Figure 2-4: Normal initiation of gastrulation in Porcn genetrap aggregation embryos

Comparison of expression patterns of Wnt3 (A, B), Cer1 (C, D) and Lhx1 (E, F, F’) in wildtype and Porcn genetrap aggregation embryos at E6.5. Expression of genes was visualized by whole mount RNA in situ hybridization. All embryos are oriented with their anterior to the left. At 6.5, Wnt3 is marks the posterior side of the embryo (A) where the primitive streak is initiated, whereas Cer1 marks the anterior visceral endoderm (AVE) (C). Lhx1 marks both the AVE and mesoderm on the posterior side of the embryo proper (E). Porcn mutants display highly similar expression patterns for all three marker genes at E6.5 (B, D, F, F’). In the case of Lhx1, 40% of Porcn genetrap aggregation embryos showed only AVE expression (F, n=2/5), whereas 60% showed both AVE and initiation of mesodermal expression (F’, n=3/5).

E7.5, wt E7.5, mut E8.5, wt E8.5, mut A Wnt3 B Wnt3 C Wnt3 D Wnt3

E Axin2 F Axin2 G Axin2 H Axin2

I T J T K T L T

M Hoxb1 N Hoxb1 O Hoxb1 P Hoxb1

Figure 2-5: Analysis of posterior marker gene expression in Porcn genetrap aggregation embryos

Examples of wildtype and Porcn genetrap aggregation embryos after in situ hybridization for Wnt3 (A-D), Axin2 (E-H), Brachyury (T) (I-L) and Hoxb1 (M-P). Wildtype embryos are oriented with their anterior to the left. Mutant embryos cannot be oriented, due to a lack of discernible markers or structures. At E7.5, Wnt3 (A), canonical Wnt signaling target Axin2 (E) and T (I) are expressed in the primitive streak, whereas HoxB1 (M) marks posterior cell fates in wildtype embryos. Porcn mutants display a reduced and ectopic expression of Wnt3 (B) as well as Axin2 (F). Expression of T (J) is strongly reduced and Hoxb1 (N) is entirely absent. At E8.5, Wnt3 expression is restricted to the developing neural tissues and the tailbud in wildtype embryos (C), but expressed ectopically in mutant embryos (D). Axin2 expression at E8.5 is ectopic and strongly reduced in Porcn mutant embryos (H) as compared to wildtype embryos (G). Brachyury (T) marks the axial mesoderm and tailbud in E8.5 wildtype embryos (K), but is strongly reduced in Porcn mutants (L). Posterior cell fate marker Hoxb1 (O) is undetectable in Porcn mutant embryos.

A Hesx1 B Hesx1

E8.5 wt E8.5 mut C Otx2 D Otx2

E7.5 wt E7.5 mut E Otx2 F Otx2

E8.5 wt E8.5 mut

Figure 2-6: Analysis of anterior marker gene expression in Porcn genetrap aggregation embryos

Examples of wildtype and Porcn genetrap aggregation embryos after in situ hybridization for Hesx1 (A, B) and Otx2 (C-F). Wildtype embryos are oriented with their anterior to the left. Mutant embryos cannot be oriented, due to a lack of discernible markers or structures. Hesx1 marks the anterior neuroectoderm in wildtype embryos (A) and is strongly reduced (B) or entirely absent in Porcn mutant embryos. Otx2 expression in wildtype embryos is restricted to the AVE at E7.5 (C) and the anterior neuroectoderm at E8.5 (E), but is expressed throughout the entire embryo proper of Porcn mutant embryos at both E7.5 and E8.5 (D,F).

2.4.5 Porcn null ES cells fail to differentiate into mesoderm and endoderm derivatives in vitro

In order to assess the differentiation properties of Porcn mutant cells in more detail, I studied the differentiation of mutant ES cells as embryoid bodies in vitro (Gadue et al., 2005). It has previously been shown that induction of Flk1 (VEGFR2), a marker expressed in the lateral plate mesoderm, is dependent on canonical WNT signaling (Nostro et al., 2008). To test whether Porcn genetrap ES cells can generate Flk1-expressing mesoderm, I differentiated Porcn wildtype and null ES cells as embryoid bodies (EBs) in vitro in serum-free conditions. Mesoderm formation was induced by addition of BMP4 (0.5 ng/ml) and ACTIVIN A (5 ng/ml) after 48 hrs and EBs were cultured for an additional 48 hrs before flow-cytometric analysis (Fig. 2-7 A-H).

Using SSEA-1 as a marker for ES and epiblast cells and FLK1 as a marker for mesoderm, I could observe that approximately 40% of wildtype cells downregulated SSEA-1 and expressed FLK1 (Fig. 2-7 A), indicating successful mesoderm formation. In contrast, FLK1+ cells were absent entirely in Porcn mutant cells and SSEA-1 expression remained high (Fig. 2-7 B).

Absence of FLK1+ cells was also observed in wildtype EBs upon addition of canonical WNT inhibitor DKK-1 (300 ng/ml, Fig. 2-7 G, H) or PORCN inhibitor IWP-2 (1 uM, Fig. 2-7 C, D), indicating that the observed defect is in fact dependent on canonical WNT signaling and the specific function of PORCN. In order to circumvent the block in differentiation of Porcn mutant cells, I added the GSK3 inhibitor CHIR99014 (3 uM) to differentiating EBs 6 hours after induction with BMP4 and ACTIVIN A. Inhibition of GSK3 stabilizes free Beta-Catenin, which is then translocated to the nucleus where it can bind TCF/LEF transcription factors and activate gene transcription. EBs treated with GSK3 inhibitor showed downregulation of SSEA-1 and a significant proportion of cells expressing FLK1 were observed in both wildtype (22.60%, Fig. 2- 7 E) and mutant cells (32.91%, Fig. 2-7 F). The higher amount of FLK1+ cells in mutant EBs may be due to different kinetics of FLK1 induction between these two cells lines or the failure of Porcn mutant cells to secrete a negative modulator of FLK1 induction. Likely candidates would be non-canonical WNTs that may require PORCN for their secretion.

In contrast to activation of the canonical WNT signaling cascade at the level of GSK3, addition of WNT3A (up to 100 ng/ml) protein had no effect on mutant EB differentiation (Fig. 2-8 B, D, F). These observations suggest a positive feedback loop, in which a trigger amount of exogenous

WNT3A results in the transcription of multiple other canonical WNTs that are required for gastrulation events. Similar observations have been made for WNT3 in an embryocarcinoma model (Marikawa et al., 2009). This feedback loop has recently been confirmed in vivo (Tortelote et al., 2012).

I further tested whether EBs could be induced to form CXCR4+ endoderm upon induction with ACTIVIN A alone at high concentrations (100 ng/ml, Fig. 2-7 I, J) (Morrison et al., 2008; Yasunaga et al., 2005). While wildtype EBs showed downregulation of SSEA-1 and up to 30% of cells became CXCR4+, no CXCR4+ cells were observed in Porcn null embryoid bodies (Fig. 2-7 I, J). These results indicate absence of both endoderm and mesoderm in Porcn null EBs in vitro.

standard Porcn inhibitor Gsk3 inhibitor Wnt inhibitor Activin A conditions IWP-2 (1 uM) CHIR99021 (3 uM) Dkk-1 (300 ng/ml) (100 ng/ml) A C E G I 58.23%58.23% 40.22% 99.50%99.50% 0.10% 61.09%61.09% 22.60% 95.50%95.50% 2.12% 66.86% 29.92%

E14Tg2a.4 (wt)

1.45% 0.45% 16.37% 2.38% 1.97% 1.26%

B D F H J SSEA1 98.95%98.95% 0.09% 98.89%98.89% 0.24% 56.13%56.13% 32.91% 98.68%98.68% 0.13% SSEA1 97.51% 0.02%

CSD256 (Porcn mutant)

0.98% 0.81% 10.92% 1.19% 2.47% 0.00%

Flk1 CXCR4

Figure 2-7: Porcn mutant embryoid bodies (EBs) fail to generate FLK1+ mesoderm and CXCR4+ endoderm in vitro

Porcn wildtype (E14Tg2a.4) and mutant (CSD256) ES cells were differentiated in vitro as embryoid bodies in serum-free media with 0.5 ng/ml BMP4 and 5 ng/ml ACTIVIN A. After 4 days of differentiation, EBs were dissociated and analyzed for the expression of cell surface markers SSEA-1 (ES/epiblast cells), FLK1 (VEGFR2, lateral plate mesoderm) and CXCR4 (endoderm) by flow-cytometry. While wildtype ES cells generated FLK1+ SSEA1low mesoderm (A), this population was entirely absent in Porcn mutant cells and SSEA1 expression remained high (B). This defect could also be induced by addition of PORCN inhibitor IWP-2 to wildtype (C) or mutant (D) cells. Activation of the canonical WNT signaling pathway using GSK3 inhibitor CHIR99021 was able to rescue the defect and allowed differentiation into FLK1+ SSEA1low mesoderm of both wildtype (E) and mutant (F) cells. Addition of extra-cellular canonical WNT signaling inhibitor DKK-1 (G, H) caused a phenotype highly similar to Porcn mutation (B) or inhibition (C, D). Differentiation of cells without BMP4 and with high levels of ACTIVIN A (100ng/ml) directs differentiation of wildtype ES cells towards CXCR4+ endoderm (I). Porcn mutant cells fail to generate CXCR4+ endoderm (J).

A 20 ng/ml Wnt3a C 100 ng/ml Wnt3a E 200 ng/ml Wnt3a

9.38% 17.57% 17.71% E14Tg2a.4 (wt)

B D F Count

0.31% 0.22% 0.25% CSD256 (Porcn mutant)

Flk1

Figure 2-8: Activation of the canonical WNT signaling cascade by exogenous WNT3A protein fails to induce mesoderm differentiation in Porcn mutant embryoid bodies (EBs) in vitro

Flow-cytometric analysis of FLK1 expression in embryoid bodies (EBs) cultured with WNT3A protein. Porcn wildtype (E14Tg2a.4) and mutant (CSD256) ES cells were differentiated in vitro as embryoid bodies in serum- free media. At day 2 of differentiation, 20-200 ng/ml WNT3A protein were added to differentiating EBs. After 4 days of differentiation, EBs were dissociated and analyzed for the expression of lateral plate mesoderm marker FLK1 (VEGFR2) by flow-cytometry. While FLK1+ mesoderm could be induced in wildtype cells by addition 20 ng/ml WNT3A (A), Porcn mutant cells failed to generate a FLK1+ cell population. Increasing amounts of WNT3A (C, E) increase the FLK1+ cell population generated by wildtype cells, but did not induce FLK1+ mesoderm in Porcn mutant cells.

Expanding the analysis of in vitro differentiation, I performed quantitative real-time PCR analysis of ES cells (EB day 0, Fig. 2-9 A), EBs before induction with BMP4 and ACTIVIN A (EB day 2, Fig. 2-9 B) and at the time of flow-cytometric analysis (EB day 4, Fig. 2-9 C). Using a panel of marker genes for pluripotency (Fbx15, Klf4, Oct4), epiblast (E-cadherin, Fgf5), neuroectoderm (Gbx2, Sox1, Pax6, Nestin), mesoderm (T, Tbx6, Pdgfra, Mixl), endoderm (Sox17, Cer1, Cxcr4) as well as WNT pathway members (Wnt5a, Axin2, c-Myc), I assessed the phenotype of Porcn mutant ES cells and EBs on a molecular level (Fig. 2-9 A-C).

At EB day 0 and EB day 2 (Fig 2-9 A, B), only minor changes in gene expression levels were detected that were not statistically significant for the majority of genes assessed (Student’s t-test, p>0.05), suggesting that Porcn mutant ES cells had normal expression levels compared to wildtype ES cells and successfully adopted an epiblast-like state at EB day 2. Upon induction with BMP4 and ACTIVIN A (EB day 4, Fig. 2-9 C), the difference between Porcn wildtype and mutant cells in pluripotency, epiblast and neuroectoderm marker gene expression remained statistically not significant with the exception of the downregulation of Klf4 and Gbx2. In contrast, all mesodermal and endodermal marker genes were significantly downregulated with Mixl being the most extreme. Further, the primitive streak marker Wnt5a, as well as the direct canonical WNT signaling targets Axin2 and c-Myc were also significantly downregulated. Visual assessment of differentiating EBs as well as the proportion of dead cells detected during flow- cytometric analysis by propidium iodide staining suggests no major defects in cell proliferation or apoptosis under these conditions.

Taken together, my in vitro data suggest that Porcn mutant ES cells fail to differentiate into endodermal and mesodermal derivatives, consistent with the requirement for canonical WNT signaling in the induction of these germ layers.

EB day 0 A 100 10 Pluripotency

1 Epiblast Neuroectoderm 0.1 Mesoderm 0.01 Endoderm Wnt pathway genes 0.001

0.0001 * * * 5 4 4 d 2 n T 6 a l 7 1 4 a 2 c lf t ti x fr ix r K c bx M in bx1 O -ca Fgf5 Sox1Pax6 Tb dg ox1 Cer xc F E G Nes P S C Wnt5 Ax c-My EB day 2 B 100 10

0.1

0.01

0.001

0.0001 * 5 4 4 d 2 n T 6 a l 7 1 4 a 2 c lf t ti x fr ix r K c bx M in bx1 O -ca Fgf5 Sox1Pax6 Tb dg ox1 Cer xc F E G Nes P S C Wnt5 Ax c-My EB day 4 C 100 10

0.1

0.01

0.001

0.0001 * * * * * * * * * * * * 5 4 4 d 2 n T 6 a l 7 1 4 a 2 c lf t ti x fr ix r K c bx M in bx1 O -ca Fgf5 Sox1Pax6 Tb dg ox1 Cer xc F E G Nes P S C Wnt5 Ax c-My

Figure 2-9: Gene expression analysis of Porcn mutant embryoid bodies (EBs) confirms failure in generation of endodermal and mesodermal derivatives in vitro

Quantitative real-time PCR analysis of in Porcn wildtype (E14Tg2a.4) and mutant (CSD256) ES cells differentiated in vitro as embryoid bodies in serum-free media. RNA was harvested from ES cells at the time of EB set-up (EB day 0, A), before addition of growth factors BMP4 and ACTIVIN A (EB day 2, B) and at the time of flow-cytometric analysis (EB day 4, C). Bars indicate deviation of gene expression level in Porcn mutant cells compared to wildtype cells at the respective time point. Error bars indicate standard errors. Statistical significance indicated by * (Student’s t-test, p<0.05). The selected marker genes show only minor differences in expression levels at EB day 0 (A) and EB day 2 (B), suggesting that Porcn mutant ES cells successful transition into an epiblast-like state (B). At EB day 4 (C), changes in pluripotency, epiblast and neuroectoderm marker gene expression are statistically not significant, with the exception of Klf4 and Gbx2. In contrast, all markers for mesodermal and endodermal derivatives were significantly reduced as compared to wildtype EBs, indicating absence or strong reduction in those cell types. Canonical WNT signaling targets Axin2 and c-Myc were also significantly reduced.

2.5 Discussion

We have recently discovered Porcn in a screen for X chromosome linked embryonic lethal mutations in the mouse (Cox et al., 2010). Here, I have expanded the analysis of Porcn mutant embryos derived from aggregations of Porcn null ES cells with wildtype donor embryos. While it has previously been established in lower organisms and in vitro systems that PORCN is required for the lipid-modification of WNT ligands, I presented the first analysis showing that mammalian embryos and ES cells require PORCN for the secretion of functional ligands to induce canonical WNT signaling responses.

At the time of publication of these data, it had been established in vitro using siRNA and chemical inhibition that PORCN is required for acylation and secretion of WNT3A (Takada et al., 2006) and WNT5A ligands (Chen et al., 2009) and for activity of WNT1 and WNT2 (Chen et al., 2009), suggesting that PORCN function might be required for all WNT ligands. More recent data has now confirmed, that PORCN is required for the secretion and function of all human WNT ligands (Najdi et al., 2012). Sequence conservation between mouse and humans suggests that this is also true for mouse WNT ligands. In keeping with these observations, I have been able to show that genetic ablation of Porcn leads to a failure to generate functional secreted WNTs as assessed in Tcf/Lef-luciferase assays in mouse ES cells. By using appropriate mixtures of wildtype and mutant cells, I was able to definitively demonstrate an exclusive requirement for Porcn in WNT producing cells, whereas Porcn is dispensable in WNT receiving cells.

Consistent with previously established roles for canonical WNT signaling in establishing endodermal and mesodermal cell fates, Porcn mutant ES cells failed to generate these two germ layers after differentiation in embryoid bodies in vitro. Treatment of wild type cells with a chemical inhibitor of PORCN (IWP-2) or an extracellular WNT inhibitor (DKK) phenocopied the Porcn mutant differentiation defect. However the defect could be rescued by treatment of mutant cells with a GSK3 inhibitor (CHIR99014), which would stabilize Beta-Catenin and activate the canonical WNT signaling pathway in mutant cells. These results are all consistent with a specific role for PORCN in promoting the correct processing and secretion of WNTs.

These in vitro studies on Porcn mutant ES cells strongly suggest that Porcn will be required for the earliest WNT dependent developmental events in vivo. This was confirmed by examining the phenotype of Porcn mutant ES-derived embryos. Mutant embryos showed major gastrulation

44 defects. Porcn mutant epiblast cells fail to specify anterior and posterior tissue fates, but instead remain in an “epiblast-like” state with high levels of Oct4 (Cox et al., 2010) and Otx2 transcription. The epiblast continues to proliferate, leading to folds of embryonic ectoderm within the visceral endoderm (Cox et al., 2010). These observations support the role for PORCN in WNT signaling, as Wnt3-/- embryos show a similar phenotype (Barrow et al., 2003; Liu et al., 1999). Further, Wls, encoding a protein involved in trafficking WNT ligands from the Golgi to the cell surface, has been knocked out in the mouse and also shows a highly similar phenotype (Fu et al., 2009a). Based on the similarity of these three mutants, I can conclude that Porcn and Wls are involved in the generation and secretion of functional WNT3 ligand in vivo.

In the ES-derived Porcn mutant embryos, expression of canonical WNT targets, such as Axin2 and T, was reduced but not absent, suggesting that there might not be a complete block in WNT function in these embryos. In contrast to zygotic mutant embryos, the extraembryonic tissues derived from trophectoderm and primitive endoderm in the analyzed aggregation embryos are derived from the wild type donor embryos. Hence, there could be some residual WNT secretion from these tissues. Porcn was not expressed strongly in the extraembryonic tissues of the peri- gastrulation mouse embryo but it was expressed in some regions of the visceral endoderm, such as the AVE at E6.5. The AVE, however, has not been described as a source for WNT ligands so far, but has been shown to secrete WNT inhibitors DKK and SFRP. Transient expression of Wnt3 in the region proximal to the AVE has been reported (Rivera-Pérez and Magnuson, 2005). While it is possible that PORCN has no functional role in the AVE, it opens the possibility that PORCN may have WNT-independent functions or substrates. More recent data shows that WNT3 ligands secreted from the posterior visceral endoderm are sufficient to induce gastrulation in Wnt3 mutant epiblast cells, but are insufficient to support the completion of gastrulation (Tortelote et al., 2012). These data are consistent with the phenotype observed in the Porcn null aggregation embryos presented in this study. However, I was not able to detect Porcn transcript in wildtype posterior visceral endoderm, the proposed source of the gastrulation-inducing WNT3 ligand. While it is possible that these discrepancies are due to experimental limitations, such as staging or the detection limit of the Porcn in situ hybridization probe, it is also possible, that lack of WNT ligands from the Wnt3 mutant epiblast induces Porcn expression in the visceral endoderm, thereby allowing WNT ligands to be secreted from the posterior visceral endoderm.

Due to its significant role in gastrulation, I investigated whether PORCN could be involved in Nodal signaling, but was unable to detect defects in NODAL secretion or downstream signaling in the ES cell system. Using a chemical PORCN inhibitor, another study has excluded effects of PORCN on Notch and Sonic Hedgehog (Shh) signaling (Chen et al., 2009). While these data support a model where Porcn has no effects on other secreted signaling molecules, a recent study has shown that PORCN has effects that are not mediated by WNT ligands (Covey et al., 2012). Interestingly, these effect appear to be independent of PORCN’s catalytic O-acyl transferase function, since the observed phenotype could be rescued by catalytically inactive PORCN (Covey et al., 2012).

The obvious limitation of my experimental approach was the fact that Porcn null ES cells only allow for the investigation of the earliest roles in ES cells and aggregation embryos. Due to the embryonic lethality of aggregation embryos, no Porcn mutant mouse line can be established from these ES cells, making it impossible to investigate the defects of zygotic Porcn mutants. To allow for detailed, time- or tissue-specific analyses of Porcn’s role throughout mouse embryonic development and adult tissue homeostasis, a conditional Porcn allele avoiding embryonic lethality of founder males was required.

2.6 Materials and Methods

2.6.1 ES cell culture

E14Tg2a.4 (parental wildtype) and CSD256 (Porcn genetrap) cells were obtained from Baygenomics and cultured in feeder-free, serum-free conditions containing 1000 U/ml Leukemia Inhibitory Factor (LIF), 1 mM PD0325901 (MEK inhibitor, Stemgent) and 3 mM CHIR99021 (GSK3 inhibitor, Stemgent) as previously described (Nichols and Ying, 2006). Tissue culture dishes were coated with gelatin and ES-qualified fetal calf serum (Hyclone) immediately before plating of ES cells.

2.6.2 RT-PCR:

RNA was extracted from E14Tg2a.4 and CSD256 ES cells using Trizol according to manufacturer’s instructions. RNA was transcribed into cDNA using SuperScript™ III Reverse Transcriptase (Invitrogen) with oligo(dT) primers. Gene specific PCR was performed using Taq DNA Polymerase (Roche) with the following Porcn-Fwd (GTGAGATGCACATGGTGGAC) and Porcn-Rev (ACTGTCAGGTCCCATTCCAG) primers, spanning exons 4 to 9. Detection of H2afz transcript was used as control (Mamo et al., 2007).

2.6.3 Constructs

For over-expression experiments, pCX-EYFP (Hadjantonakis et al., 2002) was used as a control plasmid for mock transfections. The promoter/enhancer elements in this vector drive strong, widespread transgene expression in ES cells and live mice (Niwa et al., 1991). The PorcnC ORF was amplified by PCR from a Mammalian Gene Collection (MGC) clone (accession: BC032284) whereas Wnt3a and Porcn isoforms A, B and D from previously published vectors (gift from T. Kadowaki) (Tanaka et al., 2000). Wnt3a point mutants C77A, S209A, and C77AS209A were generated by site-directed mutagenesis of the wildtype Wnt3a ORF. Amplicons were ligated into the EcoRI site of pCX-EYFP to replace the EYFP ORF. All plasmids were verified by sequencing.

2.6.4 Canonical WNT Activity Assay

For the initial autocrine assay, ES cells were plated at 200,000 cells/well in 24 well plates in serum-free ES media containing LIF, but no PD0325901 or CHIR99021. Cells were transfected

47 in triplicate at time of plating with 20 ng pCMV-RenillaLuciferase (pCMV-RL), 400 ng Super8XTOPFlash (or Super8XFOPFlash) (Veeman et al., 2003), as well as pCX-EYFP (500ng/1000ng), pCX-Wnt3a (500 ng), pCX-PorcnA-D (total amount: 500ng) using FugeneHD (Roche). Luciferase activity was determined 24h after transfection using Dual-Luciferase® Reporter Assay System (Promega) according to manufacturer’s protocol.

The autocrine Wnt3a point-mutant assay was performed as above, with the following modification; 500 ng of pCX-Wnt3a (wt or point-mutants) or 500 ng of pCX-EYFP were transfected along with 20 ng of pCMV-RL and 400 ng Super8XTOPFlash (or Super8XFOPFlash).

For paracrine assays, cells were transfected with pCX-Wnt3a or pCX-EYFP using FugeneHD in serum-free media without inhibitors. Recipient cells were transfected with pCMV-RL and Super8XTOPFlash (or Super8XFOPFlash). 24h after transfection cells were dissociated using Trypsin and 75,000 producer cells and 75,000 recipient cells were plated per well of a 24 well plate in triplicate. Luciferase activity was determined 24h after cell mixing using Dual- Luciferase® Reporter Assay System (Promega) according to manufacturer’s protocol.

All Luciferase assays were performed at least three times with three technical replicates. Results were analyzed using Student’s t-test.

2.6.5 Nodal Activity Assay

Assay was performed and analyzed as described for the autocrine canonical Wnt signaling assay with the following modifications: Inhibitors PD0325901 and CHIR99021 were included in the media. Cells were transfected with 500 ng pBOS-Nodal (Yamamoto et al., 2003) (or pCX-EYFP

(Hadjantonakis et al., 2002)), 300 ng (n2)7-luc (Saijoh et al., 2000) (gift from H. Hamada) and 20 ng pCMV-RL. For inhibition and activation of the pathway, 10 mM SB431542 (Sigma) and 100 ng/ml ACTIVIN A (R&D Systems) were used respectively.

2.6.6 Embryo generation

Porcn null embryos were generated by aggregation of Porcn genetrap ES cells (CSD256, Baygenomics) with 8-cell stage embryos as previously described (Cox et al., 2010). Briefly, male mice homozygous for a ubiquitously expressed eGFP transgene (B5/EGFP) [Tg(CAG-

EGFP)B5Nagy] (Hadjantonakis et al., 1998b) were mated with superovulated ICR (Harlan) female mice and embryos were collected at day E1.5. After over-night culture until the uncompacted eight-cell stage, each embryo was aggregated with a clump of 8 to 15 ES cells in depression wells and cultured one more night, before morulae and blastocysts were transferred into the uteri of pseudopregnant females at E2.5. Embryos were dissected five (E7.5) or six days (E8.5) after embryo transfer and fixed in 4% PFA in PBS overnight.

All aggregation experiments were performed as a service by the Transgenic Core at the Toronto Centre for Phenogenomics. All procedures were performed according to animal use protocols approved by the institutional Animal Care Committee in accordance with guidelines by the Canadian Council for Animal Care (CCAC).

2.6.7 Whole-mount in situ hybridization

Whole-mount embryo in situ hybridization was performed as previously described (Yamanaka et al., 2007). The probes RNA probes used were: Axin2 (HindIII fragment, F. Costantini), Brachyury (Herrmann, 1991), Cer1 (Shawlot et al., 1998), Hesx1 (Thomas and Beddington, 1996), Hoxb1 (Wilkinson et al., 1989), Lhx1 (Shawlot and Behringer, 1995), Otx2 (Ang et al., 1994), Wnt3 (Roelink et al., 1990).

For in situ hybridization of Porcn, two probes were used simultaneously. The templates for Porcn probes were amplified from cDNA clones by PCR using primers Fwd1: CTGGAATGGGACCTGACAGT, Rev1: GCATCCAAAAGTGACCCAGT, Fwd2: GTGAGATGCACATGGTGGAC, Rev2: ACTGTCAGGTCCCATTCCAG and cloned into pBSII for transcription with sp6/T7 polymerases.

2.6.8 In vitro differentiation and flow cytometric analysis

Serum-free in vitro differentiation as embryoid bodies (EB) was performed as previously described (Gadue et al., 2006) with minor modifications; ES cells were cultured in differentiation media at 75,000 cells/ml on low attachment plates (Costar). For standard differentiation, 5 ng/ml ACTIVIN A (R&D), 0.5 ng/ml rhBMP4 (R&D) were added after 2 days and EBs were cultured for another 2 days without dissociation/re-aggregation. Other factors used: 1 uM IWP-2 (PORCN inhibitor, gift from L. Lum); 3 uM CHIR99021 (Stemgent), 150 ng/ml rmDKK (R&D), 20-200 ng/ml rmWNT3A (R&D).

For endoderm differentiation, 100 ng/ml Activin A (R&D) was added at day 2 without addition of BMP4.

For flow-cytometric analysis, EBs were dissociated by incubation with Trypsin/EDTA and stained using the following antibodies: Phycoerythrin (PE) Rat anti-mouse FLK-1 (BD Pharmingen), Allophycocyanin (APC) Rat anti-Mouse CD184 (CXCR4, BD Pharmingen) and fluorescein isothiocyanate (FITC) Mouse anti-SSEA-1 (BD Pharmingen). The cell suspension was analyzed on a LSRII or FACSCanto flow cytometer (BD Biosciences) and dead cells were excluded from the analysis based on propidium iodide staining. Data analysis was performed using FlowJo Software (Tree Star Inc.).

2.6.9 Quantitative Real-time PCR analysis

RNA was isolated (Trizol Reagent, GIBCO BRL) at day 0, 2 and 4 of EB differentiation according to the manufacturer’s instructions. Two micrograms of RNA were reverse transcribed using QuantiTect RT Kit (Qiagen) according to manufacturer’s protocol. Generated cDNA was used for gene expression analysis by real-time PCR on a Roche LightCycler 480 using 2X SYBR master mix (Roche). Primers used are listed in table 2-1. Melt curve analysis was used to determine primer specificity and standard curves were generated to control for primer efficiency. Gene expression was determined by relative quantification with values corrected for input using GAPDH and normalized relative to wildtype EBs at the respective time-point. Data presented in Figure 6 was obtained in three independent experiments and analyzed using Student’s t-test (p<0.05).

Table 2-1: Primers used for quantitative RT-PCR analysis of embryoid bodies

Gene Forward Primer Reverse Primer Axin2 GGGGGAAAACACAGCTTACA TTGACTGGGTCGCTTCTCTT Cer1 GTCATCCTGCCCATCAAAAG ATTTGCCAAAGCAAAGGTTG c-Myc TGCCCGCGATCAGCTCTCCT GGGGCATCGTCGTGGCTGTC Cxcr4 TCCAACAAGGAACCCTGCTTC TTGCCGACTATGCCAGTCAAG E-cad TCTACCAAAGTGACGCTGAAGTCC GGTACACGCTGGGAAACATGAG Fgf5 GCTGTGTCTCAGGGGATTGT CACTCTCGGCCTGTCTTTTC Fbx15 CATCTGTCACGAAGCAGCAT GGTCACCGCATCCAAGTAAG Gapdh CTCGTCCCGTAGACAAAA TGAATTTGCCGTGAGTGG Gbx2 AGACGGCAAAGCCTTCTTGG TCATCTTCCACCTTTGACTCGTCT Klf4 ACACTTGTGACTATGCAGGCTGTG TCCCAGTCACAGTGGTAAGGTTTC Mixl GCTGCTACCCGAGTCCAGGAT GCCTTGAGGATAAGGGCTGAAA Nestin AACTCTCGCTTGCAGACACCTG AGGTGCTGGTCCTCTGGTATCC Oct4 AGCTGCTGAAGCAGAAGAGG AGATGGTGGTCTGGCTGAAC Pax6 GGACTTCAGTACCAGGGCAACC GCATCTGAGCTTCATCCGAGTC Pdgfra TCCATGCTAGACTCAGAAAGTCAA TCCCGGTGGACACAATTTTTC Sox1 TTCCCCAGGACTCCGAGGCG GCTGTGTGCCTCCTCTGCGG Sox17 TATGGTGTGGGCCAAAGACGAA AACGCCTTCCAAGACTTGCCTA T TCCCGAGACCCAGTTCATAG TTCTTTGGCATCAAGGAAGG Tbx6 TGGAGAACCAGGAACTGTGGAA ATACTCGGCAAGCAGGGAACAT Wnt5a GAGAAAGGGAACGAATCCACGCTAA GGAGCCAGACACTCCATGACACT 2.7 Acknowledgements

I wish to thank Dionne White for flow cytometry and Ken Harpal for sectioning; Dr. Tatsuhiko Kadowaki and Dr. Hiroshi Hamada for DNA constructs and Dr. Lawrence Lum for sharing IWP- 2. Supported in part by Research Grant No. 6-FY08-315 from the March of Dimes Foundation.

3 Porcn deletion reveals non-essential role for WNT signaling prior to mouse gastrulation 3.1 Contributions

The data in this chapter is currently (April 2013) under review at Development for publication. Steffen Biechele designed experiments, and performed and analyzed experiments involving post- implantation embryos and ES cells. Katie Jean Cockburn performed cell fate analysis in immunostained pre-implantation embryos. Dr. Fredrik Lanner dissociated blastocysts and performed single cell gene expression analysis. Dr. Brian Cox performed statistical analyses on single cell gene expression data, guided experimental design and interpretation of results. Dr. Janet Rossant guided experimental design and interpretation of results.

3.2 Abstract

In this study, I have generated a mouse line carrying a floxed allele for Porcn as a tool to ablate WNT secretion from specific tissues and used zygotic, oocyte-specific and visceral endoderm- specific deletions to investigate embryonic and extra-embryonic requirements for WNT ligand secretion. I show that there is no requirement for WNT ligands during preimplantation development of the mouse embryo. WNTs are first required for the initiation of gastrulation, where PORCN-mediated acylation of WNTs is required in the epiblast, but not the visceral endoderm. Heterozygous female embryos, which are functionally mutant in both trophoblast and visceral endoderm due to imprinted X chromosome inactivation, complete gastrulation but display chorio-allantoic fusion defects similar to Wnt7b mutants. My studies highlight the importance of WNT3 and WNT7B for embryonic development but suggest that endogenous WNT secretion does not play an essential role in either implantation or blastocyst lineage specification.

3.3 Introduction

In my initial studies (Chapter 2), I was able to show that embryos lacking Porcn specifically in the epiblast fail to gastrulate (also (Barrott et al., 2011; Biechele et al., 2011)). While these data established the role in the embryo proper, the experimental strategy did not allow for analysis of extra-embryonic and pre-implantation functions of Porcn. Even though zygotic deletion of Porcn was reported during the course of my studies (Liu et al., 2012), the phenotype of zygotic mutants had not been investigated in detail. Early embryonic Porcn functions are of particular interest, as PORCN-mediated WNT signaling has been reported to be necessary for the maintenance of pluripotent mouse embryonic stem (ES) cells in vitro (Berge et al., 2011). These cell lines are derived from the inner cell mass (ICM) of pre-implantation mouse blastocysts.

I have generated a mouse line carrying a floxed allele for Porcn as a tool to ablate all WNT ligand secretion and used zygotic, oocyte-specific and visceral endoderm-specific deletions to investigate embryonic and extra-embryonic requirements for WNT signaling in early mouse development. I demonstrate that PORCN-mediated embryonic WNT signals are not required in pre-implantation development and implantation itself. Consistent with numerous WNT pathway mutants, my study identifies gastrulation as the first PORCN/WNT-dependent event in embryonic development. Taking advantage of extra-embryonic imprinted XCI in heterozygous females, I show that early extra-embryonic requirements are limited to the chorion, where Porcn is required for chorio-allantoic fusion.

3.4 Results

3.4.1 Generation of a Porcn floxed allele

I have previously shown that mice generated by aggregation of Porcn mutant ES cells with tetraploid embryos die at gastrulation stages, making it impossible to generate a viable mouse line (Biechele et al., 2011; Cox et al., 2010). In order to circumvent this problem and analyze Porcn function during embryonic development in more detail, I have generated a conditional Porcn allele carrying loxP sites flanking exon 3 (Figure 3-1A) by homologous recombination in G4 mouse ES cells (George et al., 2007). After excision of the FRT-flanked Neomycin resistance lox/Y cassette (Figure 3-1A), Porcn floxed (Porcn ) ES cells were used to generate chimeric F0 founder males. Germline-transmission of the allele was observed by coat color and PCR genotyping, and heterozygous floxed female F1 offspring were used to establish a breeding colony. Both hemizygous and homozygous allele carriers showed no defects in embryogenesis or adult life and have a normal fertility on outbred ICR and inbred C57BL/6J backgrounds (>F10).

On a molecular level, deletion of exon 3 is predicted to cause a frameshift and premature stop- codons leading to nonsense-mediated decay of the mutant transcript. In order to confirm this prediction, I generated Porcn deleted ES cells (Porcndel/Y) by transient expression of Cre recombinase. Unexpectedly, Porcn transcript was still detectable by RT-PCR in Porcn mutant ES cells (Fig. 3-1B) and sequencing of RT-PCR products revealed that deletion of exon 3 caused aberrant splicing between exons 2 and 4, resulting in the inclusion of the majority of the fusion intron in the transcript. This mutant transcript contains 10 stop-codons before exon 4 and should thus not lead to a functional protein product. In order to confirm that Porcndel/Y ES cells are functionally mutant, I performed an autocrine WNT secretion assay based on the canonical WNT reporter Tcf/Lef-Luciferase (Veeman et al., 2003). Upon over-expression of Wnt3a, Luciferase activity increased 8-fold in Porcnlox/Y ES cells (Fig. 3-1C). In contrast, Porcndel/Y ES cells showed no upregulation of Luciferase activity. This defect could be rescued by co-transfection of Wnt3a with a mixture of all four Porcn isoforms (Fig. 3-1C). Control transfections with EYFP or Porcn in the absence of Wnt3a expression plasmid had no effect on Luciferase activity. These results show that deletion of Porcn exon 3 results in a functionally mutant allele which phenocopies the Porcn genetrap allele (CSD256, Baygenomics) and an independent Porcn floxed allele (Barrott et al., 2011) in vitro.

alternative A splice region

Porcn wt 5’ 3’

FRT loxP

Porcn neo 5’ NeoR 3’

loxP

Porcn lox 5’ 3’ F1 R1 R3

Porcn del 5’ 3’

RT-F2 RT-R2 Porcn del transcript RT-F1 RT-R1 B C lox/Y del/Y -RT water Topflash

Porcn (Ex2-4) Fopflash

Porcn (Ex5-9) Fold Change Avg

H2afz Genotype pCX-EYFP pCX-Wnt3a pCX-PorcnA-D

Figure 3-1: Generation and characterization of a conditional Porcn allele

(A) LoxP sites flanking exon 3 were introduced into the Porcn locus through homologous recombination in ES cells (Porcn neo). The Neomycin resistance cassette was excised in vitro (Porcn lox) before aggregation and chimera generation. Removal of exon 3 by Cre-mediated excision (Porcn del) results in production of a mutant transcript that exhibits aberrant splicing and inclusion of intron 2 and 3 (Porcn del transcript). (B) RT-PCR reveals an aberrantly spliced transcript in Porcndel/Y cells containing all exons downstream of exon 3 (Ex5-9), but with intronic sequences between exon 2 & 4 in place of the excised exon 3 (Ex2-4). H2afz was used as control. (C) Porcndel/Y ES cells fail to express Tcf/Lef-Luciferase in response to Wnt3a transfection, compared to 8-fold upregulation in Porcnlox/Y ES cells. This is rescued with co-transfection of Wnt3a and a mixture of all 4 Porcn isoforms, resulting in similar (11x) upregulation of Tcf/Lef-Luciferase activity in both Porcnlox/Y and Porcndel/Y ES cells. Control transfections with EYFP or Porcn expression plasmids alone had no effect. Primers F1, R1 and R3 were used for genotyping. Primers RT-F1/R1, RT-F2/R2 were used for RT-PCR.

3.4.2 Zygotic Porcn deletion causes gastrulation defects in hemizygous male embryos

I and others have previously shown that embryos lacking Porcn specifically in the epiblast fail to gastrulate (Barrott et al., 2011; Biechele et al., 2011). As these embryos have wildtype extra- embryonic tissues that can act as sources of WNT signaling (Tortelote et al., 2012), I aimed to investigate whether zygotic and epiblast-specific mutants differ in phenotype. To this end, I generated zygotic Porcn mutant embryos by deletion with a ubiquitously expressed Cre recombinase (pCX-NLS-Cre) (Belteki et al., 2005). Hemizygous Porcn mutant embryos (Porcndel/Y) could be recovered up to E7.5 and were detected at the expected Mendelian frequencies, but exhibited an abnormal morphology (Figure 3-2). Mutant embryos were smaller than wildtype littermates and lacked the amnion. There was no folding of the epiblast as in embryos generated by aggregation of ES cells (Biechele et al., 2011), but the morphology was very similar to the published epiblast-specific Porcn mutants (Barrott et al., 2011).

Consistent with complete ablation of WNT signaling secretion, Porcn mutant embryos lack canonical WNT signaling response based on a Tcf/Lef-LacZ reporter allele (Mohamed et al., 2004) at E6.5 (n=2, Fig 3-2A’) and E7.5 (n=4, Fig. 3-2E’), while Beta-Galactosidase activity was readily detectable in the primitive streak at both E6.5 (n=5, Fig. 3-2A) and E7.5 (n=5, Fig. 3-2E) in wildtype embryos. To extend these observations, I assessed the expression of endogenous targets of canonical WNT signaling in the primitive streak. Wls/Gpr177 (Fu et al., 2009b) and Brachyury (Arnold et al., 2000; Yamaguchi et al., 1999b) were both undetectable in mutant embryos (Fig. 3-2I’, 3-2G’, n=6 and n=12 respectively). Consistent with previous suggestions of gastrulation defects (Biechele et al., 2011), I was unable to detect the migrating mesoderm marker Lhx1 (Shawlot et al., 1999)(n=6, Fig. 3-2H, H’), and the posterior cell fate marker HoxB1 (n=4, Fig. 3-2H, H’) in Porcn mutant embryos. In contrast to the absence of primitive streak and posterior marker genes, the early epiblast marker Otx2 (n=12, Fig. 3-2K, K’) and the pluripotency marker Oct3/4 (n=6/10, Fig. 3-2L, L’) were highly expressed throughout the epiblast of zygotic Porcndel/Y embryos, suggesting that these embryos remain in an ‘epiblast- like’ state, similar to Wnt3 mutant embryos (Liu et al., 1999).

It has been shown that the primitive streak, and thus gastrulation, is induced by WNT3 (Liu et al., 1999; Tortelote et al., 2012). While Wnt3 expression is induced and maintained by BMP4 secreted from the extra-embryonic ectoderm (Ben-Haim et al., 2006; Miura et al., 2010), it has

56 recently been shown that it is also regulated by canonical WNT signaling in an autoregulatory feedback loop (Tortelote et al., 2012). Consistent with auto-regulation in a canonical WNT positive feedback loop, I was unable to detect Wnt3 transcript at E7.5 in Porn mutant embryos (n=6, Fig. 3-2F, F’). The expression of WNT3-inducing Bmp4 at E6.5 however, was normal in both wildtype (n=3/4, Fig. 3-2D) and mutant (n=3/4, Fig. 3-2D’).

As canonical WNT signaling has been implicated in anterior posterior (AP) axis development in the mouse embryo (Huelsken et al., 2000; Morkel et al., 2003), I tested AP axis establishment in Porcn mutant embryos by observing the anterior localization of the anterior visceral endoderm (AVE) as visualized by the Hhex-eGFP transgene (Rodriguez et al., 2001). At E6.5, the majority of mutant embryos (n=7/12, Fig. 3-2B, B’) showed anterior localization of GFP+ cells, suggesting that PORCN-mediated canonical WNT signaling is not required for the anterior localization of the AVE. The remaining embryos showed distal localization (n=2/12) or strong reduction (n=3/12) of GFP expression potentially indicating a developmental delay in Porcn mutant embryos. Proper AVE localization was further confirmed by in situ hybridization for Cer1(Shawlot et al., 1998). Similar to wildtype embryos (n=2/3, Fig. 3-2C), Cer1 transcripts were detected asymmetrically on one side of the embryo at E6.5 (n=3/4, Fig. 3-2C’), further substantiating that PORCN-mediated WNT signaling is not required for distal visceral endoderm (DVE) to AVE transition.

In summary, this analysis shows that Porcn is required for canonical WNT signaling and gastrulation in vivo. This phenotype appears identical to zygotic Wnt3 mutants (Liu et al., 1999), which also fail to initiate gastrulation. In contrast, epiblast-specific Wnt3 mutants and Porcn null aggregation embryos initiate gastrulation, but fail to maintain it (Biechele et al., 2011; Tortelote et al., 2012). These observations suggest that VE-secreted WNT3 is sufficient to induce the initial phases of gastrulation in Porcn or Wnt3 mutant epiblast, as the extra-embryonic tissues are wildtype in both settings. Whether VE-secreted WNT3 is also required to induce gastrulation cannot be answered conclusively based on these mutants.

lox/Y del/Y lox/Y del/Y lox/Y del/Y lox/Y del/Y A A’ B B’ C C’ D D’

E6.5

Tcf/Lef-LacZ Tcf/Lef-LacZ Hhex-EGFP Hhex-EGFP Cer1 Cer1 Bmp4 Bmp4 E E’ F F’ G G’ H H’

E7.5

Tcf/Lef-LacZ Tcf/Lef-LacZ Wnt3 Wnt3 T T Hoxb1 Hoxb1 I I’ J J’ K K’ L L’

E7.5

Wls Wls Lhx1 Lhx1 Otx2 Otx2 Oct3/4 Oct3/4

Figure 3-2: In situ gene expression analysis of hemizygous zygotic Porcn mutants

Representative images of Porcnlox/Y (A-L) and Porcndel/Y (A’-L’) embryos analyzed by in situ hybridization for marker genes or reporter gene expression. (A, A’) At E6.5 Porcn mutant embryos fail to activate the canonical Wnt signaling reporter Tcf/Lef-LacZ, but show normal expression of the AVE markers, Hhex- EGFP (B, B’) and Cer1 (C,C’). Bmp4 is expressed normally in the extraembryonic ectoderm (D, D’). At E7.5, canonical Wnt signaling response is undetectable based on the absence of the Tcf/Lef-LacZ reporter (E, E’) and Wnt signaling targets’ expression, such as Wnt3 (F, F’), Brachyury (G, G’) and Wls (I, I’). Posterior cell fate marker, Hoxb1 (H, H’), and migrating mesoderm marker, Lhx1 (J, J’), are also undetectable in zygotic hemizygous Porcn mutants. In contrast, Otx2 (K, K’) and Oct3/4 (L, L’) are strongly expressed throughout the epiblast of mutant embryos.

3.4.3 Extra-embryonic deletion of Porcn produces a chorio-allantoic fusion defect and phenocopies Wnt7b mutants

To test the extra-embryonic requirement for Porcn more directly, I made use of female embryos with imprinted XCI as well as VE-specific Cre-mediated deletion of Porcn. As Porcn is a X- chromosomal gene, it is subject to X chromosome inactivation (XCI) in females (Barakat and Gribnau, 2012). While XCI is random in the embryo proper, XCI is imprinted in the extra- embryonic trophoblast and primitive endoderm lineages, resulting in paternal specific silencing (Barakat and Gribnau, 2012). Heterozygous females carrying a mutant maternal Porcn allele (Porcndel/+) are therefore mosaic in the embryo proper but have functionally mutant extra- embryonic tissues, allowing us to assess extra-embryonic requirements for Porcn.

Porcnlox/lox females were crossed to males carrying a ubiquitously expressed pCX-NLS-Cre transgene (Belteki et al., 2005). Female embryos derived from this cross thus inherited a paternal wildtype Porcn allele (Xp) and a maternal floxed or zygotically deleted Porcn allele (Xm). Zygotically deleted Porcndel/+ embryos were recovered up to E11.5 (Fig. 3-3A). Females could first be distinguished morphologically at E9.5, as they display a ball of allantoic tissue at the posterior end of the embryo, typical of failed chorio-allantoic fusion (Fig. 3-3B). Porcndel/+ embryos fail to establish a functional umbilical cord and placenta, which are required to provide the embryo with nutrients and oxygen. Consistent with lack of maternal nutrients and embryonic hypoxia, Porcndel/+ embryos showed a failure to thrive compared to Porcnlox/+ littermates (Fig. 3- 3C). This phenotype is highly reminiscent of the Wnt7b (Parr et al., 2001) and Integrin alpha 4 (Itga4) (Yang et al., 1995) mutant phenotypes. In contrast to a report using an independent floxed Porcn and Cre allele to generate Porcndel/+ embryos (Liu et al., 2012), I only rarely observed neural tube closure defects (n=2/33), which we attribute to a general developmental delay associated with the chorio-allantoic fusion defect.

Addressing the mechanism of this phenotype, I hypothesized that Porcndel/+ females fail to palmitoylate and secrete WNT7B from the extra-embryonic chorion, which results in the absence of Integrin alpha 4 (Itga4) expression required for chorio-allantoic fusion (Parr et al., 2001). To test this hypothesis, I performed immunostaining for ITGA4 in the chorion at E8.5 (Fig. 3-3D, E). Consistent with the proposed mechanism, ITGA4 protein is undetectable in the chorion of Porcndel/+females (Fig. 3-3E, n=2), while it is readily detectable in Porcnlox/+females (Fig. 3-3D, n=3).

In these maternally inherited Porcndel/+ mutants, both trophoblast and primitive endoderm are functionally mutant due to imprinted X-inactivation. The later chorio-allantoic phenotype suggests that WNT signaling does play a key role in trophoblast but not earlier in initiating gastrulation from the VE.

B E9.5

C E10.5

D E

Figure 3-3: Chorio-allantoic fusion defect in Porcndel/+ female embryos

(A) Porcndel/+ embryos can be detected at Mendelian ratios (25%) at E9.5 and can be recovered up to E11.5. At E9.5 (B) and E10.5 (C), Porcndel/+ embryos are distinguishable from Porcnlox/+ littermates, and exhibit failure in chorio-allantoic fusion as indicated by a ball of allantoic tissue at the posterior end of the embryo (arrow) and a failure to thrive. At E8.5, Porcndel/+ females (E) fail to express Integrin alpha-4 in the chorion (arrow), while it is readily detectable in Porcnlox/+ littermates (D). Allantoides are indicated by arrowhead.

3.4.4 Porcn is not required in the visceral endoderm

The phenotype of the Porcndel/+ embryos strongly suggests that the earliest extra-embryonic requirement for PORCN-mediated WNT secretion is in the chorion at E8.5, leading to embryonic lethality by E11.5. However, several lines of evidence indicate that Porcn could play an earlier extra-embryonic role in the VE: (1) it has recently been shown that Wnt3 expression from the posterior VE is sufficient to induce gastrulation (Tortelote et al., 2012) and (2) Porcn is expressed at E6.5 in the AVE (Biechele et al., 2011; Gonçalves et al., 2011).

I thus investigated a potential requirement for Porcn in the VE by a direct approach not dependent on XCI, using the previously described VE-specific Ttr::Cre allele (Kwon and Hadjantonakis, 2009). Successful Porcn deletion by the Ttr::Cre allele was confirmed by PCR genotyping of embryos at E7.5 (Fig. 3-4A). Despite successful deletion, Porcnlox/Y; Ttr::Cre+/tg males and Porcnlox/+; Ttr::Cre+/tg females were observed at the expected ratios at weaning age (Fig. 3-4B) and were indistinguishable from littermates.

These results show that Porcn is not required in the Ttr::Cre-expressing visceral endoderm and its derivatives for normal embryonic development and is consistent with the phenotype of zygotic Porcndel/+ females. In combination with the results from zygotic and epiblast-specific Porcn and Wnt3 mutants (Biechele et al., 2011; Tortelote et al., 2012), these results show that WNT3 secreted from wildtype VE is sufficient to induce gastrulation in mutant epiblast, but not required for the induction of gastrulation in wildtype epiblast.

A B

Embryo Embryo1 Embryo 2 wildtype3 positive DNAwater Ctrl DNA Sry

Porcn lox Porcn wt

Porcn del Cre

Figure 3-4: Visceral endoderm specific deletion of Porcn

Porcnlox/lox females were crossed to Ttr::Cretg/+ males to generate embryos with VE-specific Porcn deletion. (A) PCR analysis shows Cre-mediated deletion in E7.5 embryos carrying Cre recombinase (Embryos 1 and 3). Sry-specific primers were used to determine sex of embryos. (B) All possible genotypes were observed at the expected Mendelian ratios.

3.4.5 Porcn mediated WNT signaling is not required prior to gastrulation

All of my studies of Porcn up to this point showed that Porcn function is not required prior to gastrulation. Although several WNT ligands are expressed in the pre-implantation blastocyst (Kemp et al., 2005), no WNT pathway mutants described to date exhibit defects prior to the egg cylinder stage (van Amerongen and Berns, 2006). However, recent data shows that PORCN- mediated WNT signaling is necessary for the maintenance of pluripotent ES cells in vitro (Berge et al., 2011). As ES cells are considered the in vitro equivalent of the epiblast progenitors in the blastocyst, I aimed to determine whether embryo-secreted WNTs are also required for development to the blastocyst stage and early postimplantation stages in vivo.

Several studies using different methods have failed to detect Porcn transcript in oocytes (Macfarlan et al., 2012; Posfai et al., 2012), suggesting that rescue by maternal protein is not likely to explain the absence of any early embryo defects in Porcn zygotic mutants. In order to test this more directly, I generated Porcnlox/lox; ZP3-Cre+/tg females that delete Porcn in the developing oocytes (Lewandoski et al., 1997) and crossed them to wildtype males carrying an X- linked EGFP transgene (Hadjantonakis et al., 1998a). The transgene allowed me to sex and thus genotype the resulting embryos. Porcndel/+ females are fluorescently labeled by the X-linked GFP transgene, whereas maternal zygotic Porcndel/Y (mzPorcndel/Y) male embryos are not.

In order to functionally assess whether maternal transcript is required for development prior to gastrulation stages, I dissected pregnant females at E6.5 and E7.5. Hemizygous mutant male and heterozygous female embryos implanted successfully with no apparent deviation from the expected ratio (Fig. 3-5E). The recovered mutant male embryos failed to gastrulate and morphologically resembled zygotic Porcn mutants. This observation was confirmed by in situ hybridization for Bmp4, Cer1, and Otx2 (Fig. 3-5A-C’). Furthermore, expression of the canonical WNT signaling target Axin2 (Jho et al., 2002) was undetectable in mzPorcndel/Y embryos (Fig. 3-5D, n=5), but apparent in Porcndel/+ littermates at E7.5 (Fig. 3-5D’, n=4). These results support that maternal PORCN expression and early WNT secretion are not critical for early development.

Blastocyst-secreted WNT ligands have been proposed to play a role in implantation, since WNT- coated beads are sufficient to activate the canonical WNT signaling pathway in the luminal epithelium of the uterus (Mohamed et al., 2005). Further, WNT pathway inhibition by SFRP2

62 reduces implantation rates (Mohamed et al., 2005), suggesting that WNT signaling from the blastocyst might be required for implantation. The fact that mzPorcndel/Y embryos successfully implanted and developed to E7.5 would tend to argue against a key role for blastocyst-derived WNTs in promoting implantation. However, we could not rule out that implantation was rescued by WNT signals produced by heterozygous littermates present in the same uterus.

To test this possibility, we separated blastocysts from the above cross at E3.5 by genotype based on sex and transferred them separately into surrogate mothers. Dissection at E7.5 revealed that male mutant embryos and female heterozygous embryos implanted successfully at similar frequencies of 53% (n=17) and 60% (n=10) respectively (Fig. 3-5F). These results substantiate our previous finding that WNT ligands secreted from the embryo are not required for implantation.

Bmp4 Cer1 Otx2 Axin2 E6.5 E6.5 E7.5 E7.5 A B C D mutant

A’ B’ C’ D’ control

E n=142 n=63 100

75 Implantation Rates F

no embryo no embryo after sex separation 50 mzPorcn del/Y n=17 25 del/Y Porcn

% of implantations del/+ n=10 del/+ 0 0 25 50 75 100% Maternal Genotype (Control) Zp3-Cre +/tg Zp3-Cre +/tg Porcn lox/lox

Figure 3-5: Gastrulation defect in Porcn maternal zygotic mutants

Representative images of maternal zygotic Porcn mutant (A-D) and control (A’-D’) embryos analyzed by in situ hybridization. At E6.5, Bmp4 (A, A’) and Cer1 (B, B’) expression are normal in mutant embryos, whereas Otx2 is expressed ectopically throughout the epiblast of Porcn mutants (C). Canonical WNT signaling target, Axin2, is undetectable in maternal zygotic mutants (D) compared to wildtype (D’). (E) Dissection of litters from Porcnlox/lox; Zp3-Cre+/tg females at E7.5 revealed the expected 1:1 ratio of the two embryonic genotypes. The presence of empty deciduas (50% of implantations) in the mutant litters (E) was also observed in control crosses using Zp3-Cre+/tg control females (E), suggesting that this lethality is due to the Zp3-Cre transgene. (F) After sex-separation and transfer into wildtype surrogates at E3.5, both embryonic genotypes (Porcndel/Y and Porcndel/+) showed similar implantation rates atE7.5.

3.4.6 Normal cell fate establishment in Porcn mutant blastocysts

As maternal zygotic Porcn mutant embryos develop successfully to pre-gastrulation stages, pre- implantation development cannot be severely affected by loss of WNT signaling, consistent with previously published studies manipulating canonical WNT signaling in pre-implantation development (Kemler et al., 2004; Xie et al., 2008). However, mild effects on lineage allocation would not necessarily be incompatible with normal implantation.

In order to determine whether there was any observed defect in blastocyst cell lineage segregation, we immuno-stained maternal/zygotic Porcn mutant blastocysts for lineage specific markers NANOG (epiblast), GATA6 (primitive endoderm) and CDX2 (trophoblast). We also compared results with embryos carrying a stabilized allele of Beta-Catenin (Harada et al., 1999), which activates canonical WNT signaling. All embryos were recovered at E3.5 and cultured to E4.5. All three lineages were present and appropriately located in embryos of all genotypes (Fig. 3-6, A-D). Quantification of cell numbers for each lineage revealed that both Porcndel/Y and Porcndel/+ had normal cell numbers and cell fate distributions compared to control embryos (Fig. 3-6E, Chi square test, p=0.5). Activating the downstream canonical WNT pathway by stabilization of Beta-Catenin did cause an increase in the total number of ICM cells approaching statistical significance (Fig. 3-7E, Chi square test, p=0.05118). However, the ratio of NANOG+ epiblast to GATA6+ PE cells remained similar to wildtype or Porcn mutant embryos (Fig. 3-7E). The number of outer trophectoderm cells remained similar to control embryos (Fig. 3-7F). These in vivo findings show that WNT signaling is not necessary for ICM maintenance, but sufficient to increase the number of ICM cells.

We also assessed whether ablation or activation of WNT signaling resulted in a molecular phenotype at the transcriptional level. To this end, we dissociated E4.5 blastocysts and performed quantitative gene expression analysis using the BioMark Fluidigm System (Rugg- Gunn et al., 2012). Gene expression in single cells was analyzed for a panel of cell fate marker genes, as well as direct canonical WNT signaling targets (Table 3-1). As suggested by our immunostaining for single marker genes, unsupervised clustering of gene expression levels for all genes analyzed revealed that all three cell fates of the blastcyst are established in the Porcn mutant as well as Beta-Catenin stabilized embryos and show highly similar gene expression profiles compared to wildtype control embryos (Figure 3-8). Importantly, the main driver of

65 clustering is cell fate and not genotype (Figure 3-8). The canonical WNT target genes assessed showed lineage-specific expression patterns (Figure 3-9), but no significant changes in expression levels upon genetic activation or inactivation of the canonical WNT pathway, suggesting that the canonical WNT pathway is not fully functional at this stage of development. Together these data confirm that the three cell fates of the blastocyst are established independent of canonical WNT pathway manipulation.

Nanog A B C D Gata6 Cdx2 mzPorcn Porcn Zp3-Cre Porcn del/Y del/+ tg/+ lox/lox E F PE EPI TE uncl F average % of ICM average cell number

Porcn del/+ Porcn del/+ Zp3-CrePorcn tg/+ lox/lox Zp3-CrePorcn tg/+ lox/lox mzPorcn del/Y mzPorcn del/Y

Figure 3-6: Pre-implantation development is unperturbed in Porcn mutants

The Porcn floxed allele was backcrossed to C57BL/6J background (>F5) and deleted using the Tg(Zp3- cre)93Knw/J transgene, avoiding the artefactual lethality of the Tg(Zp3-Cre)3Mrt transgene observed in previous experiments (Fig. 4E). (A-D) Representative confocal sections of E4.5 blastocysts immuno-stained for NANOG, GATA6 and CDX2. (E) No significant differences could be observed in the quantification of cell fates between maternal zygotic Porcn mutants and control genotypes (Chi square, p=0.5). Graph displays average cell number/embryo for each cell fate (n=5 embryos/genotype). (F) Normalized cell fate distributions within inner cell mass (ICM). Key – TE = Trophectoderm (blue), PE = primitive endoderm (green), EPI = epiblast (red), uncl = unclassified cells (grey). Error bars indicate standard error of themean.

66 Nanog A B C Gata6 Cdx2 Ctnnb1 pCX-NLS-Cre Ctnnb1 +/del ex3 tg/+ +/lox ex3

E F PE EPI TE uncl F average % of ICM average cell number

Ctnnb1 +/del ex3 Ctnnb1 +/lox ex3 Ctnnb1 +/del ex3 Ctnnb1 +/lox ex3 pCX-NLS-Cre tg/+ pCX-NLS-Cre tg/+ G average % of cells

Porcn del/+ Zp3-CrePorcn tg/+ lox/lox mzPorcn del/Y Ctnnb1 +/del ex3Ctnnb1 +/lox ex3 pCX-NLS-Cre tg/+

Figure 3-7: Pre-implantation development is unperturbed in embryos with ectopic canonical WNT signaling

(A-C) Representative confocal sections of E4.5 blastocysts immuno-stained for NANOG, GATA6 and CDX2. (D) Differences approaching significance can be observed in the quantification of cell fates between embryos carrying an allele encoding stabilized Beta-Catenin (Ctnnb1+/del ex3) and control embryos (Chi square, p=0.05118). Graph displays average cell number/embryo for each cell fate (n=5 embryos/genotype). (E) Normalized cell fate distributions within inner cell mass (ICM). Key – TE = Trophectoderm (blue), PE = primitive endoderm (green), EPI = epiblast (red), uncl = unclassified cells (grey). (F) Normalized distribution of inner cells (ICM, yellow) and outer cells (TE, blue) of all genotypes assessed at E4.5. Error bars indicate standard error of the mean.

Normalized Ct value

0 5 10 Rex1 Sox17 Sox7 Stella Pecam1 Sox2 Gbx2 Otx2 Porcn Fgf5 Brachyury Gata4 Pdgfra Gata6 Cripto Fgf4 Nanog Klf2 Klf4 Eomes Fgfr2 Gata3 Gapdh Dnmt3b Oct3/4 Actb wildtypeTE wildtype PE wildtype Epi Porcn del/+ TE Porcn del/+ PE Porcn del/Y TE Porcn del/Y PE Porcn del/+ Epi Porcn del/Y Epi Ctnnb1 +/stab TE Ctnnb1 +/stab PE Ctnnb1 +/stab Epi

Figure 3-8: Genetic manipulation of WNT signaling has no detectable effect on lineage marker gene expression

Heat map of cell fate marker gene expression in E4.5 blastocysts. Single cell gene expression levels of Porcndel/Y, Porcndel/+, Porcn+/+ and Ctnnb1+/del ex3 embryos were determined using the BioMark System and assigned to one of the three lineages (TE, PE, Epi) based on marker genes. I was unable to detect major differences between genotypes within each cell lineage as shown by unsupervised hierarchical clustering of the samples. Actb and Gapdh served as housekeeping controls.

Normalized Ct value

0 5 10 Actb Gapdh CyclinD1 Tcf1 c-Myc Axin2 Gbx2 Brachyury Lef1 Wls Ctnnb1 wildtypeTE wildtype PE wildtype Epi Porcn del/+ TE Porcn del/+ PE Porcn del/Y TE Porcn del/Y PE Porcn del/+ Epi Porcn del/Y Epi Ctnnb1 +/stab TE Ctnnb1 +/stab PE Ctnnb1 +/stab Epi

Figure 3-9: Genetic manipulation of Porcn and Ctnnb1 has no significant effects on direct canonical WNT signaling target gene expression at E4.5

Heat map of direct canonical Wnt target gene expression in E4.5 blastocysts. I was unable to detect significant differences in WNT target gene expression levels between Porcn mutant, wildtype and Ctnnb1 stabilized embryos within each cell lineage as shown by unsupervised hierarchical clustering of the samples. Transcript levels of Ctnnb1 are unaffected across all genotypes. Actb and Gapdh are not WNT signaling targets, but served as housekeeping controls.

3.4.7 Porcn function in ES cell maintenance

To complement these in vivo studies and investigate whether pluripotency in vitro is dependent on PORCN-mediated canonical WNT signaling, I cultured both Porcnlox/Y and Porcndel/Y ES cells in feeder-free conditions in chemically defined media including LIF and MEK inhibitor PD0325901 (Nichols and Ying, 2006). Both cell lines were cultured for three passages in the presence of GSK3 inhibitor CHIR99021, DMSO, or PORCN inhibitor IWP2 and the expression of cell surface markers of ESCs and EpiSCs was analyzed by flow-cytometry (Rugg-Gunn et al., 2012). Cells in all conditions maintained similarly high levels of ESC markers CD31 (Figure 3- 10) and CD81 (Figure 3-11). When assessed for EpiSC marker CD40, cells of all conditions showed expression levels typical for ESCs, which are distinctly lower than EpiSCs (Figure 3- 10). Similar to my in vivo results, these data support that PORCN-mediated WNT signaling is not required for the maintenance of pluripotency in ESCs.

A 3uM CHIR99021 B DMSO C 1uM IWP2

5 0.088% 0.062% 5 0.100% 3.26% 5 0.043% 4.07% 1 0 1 0 1 0

4 4 4 1 0 1 0 1 0 Porcn 3 3 3 lox/Y 1 0 1 0 1 0

2 2 2 1 0 1 0 1 0

0 0 0 0.071% 99.8% 1.54% 95.1% 0.675% 95.2% 2 3 4 5 2 3 4 5 2 3 4 5 0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 D E F

5 0.00% 0.021% 5 0.043% 5.93% 5 0.032% 7.43% 1 0 1 0 1 0

4 4 4 1 0 1 0 1 0 Porcn 3 3 3 del/Y 1 0 1 0 1 0

2 2 2 1 0 1 0 1 0

0 0 0 0.034% 99.9% 0.369% 93.7% 0.345% 92.2%

2 3 4 5 2 3 4 5 2 3 4 5 0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 CD31 (PE-Cy7) G unstained ESC + EpiSC H EpiSC CD40 (APC) 1.60% 0.123% 59.3% 9.07% 5 5 1 0 1 0

4 4 1 0 1 0

3 3 Controls 1 0 1 0

2 2 1 0 1 0

0 0 98.1% 0.212% 30.3% 1.33% 2 3 4 5 2 3 4 5 0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0

CD31 (PE-Cy7)

Figure 3-10: Porcn is not required for maintenance of pluripotency of ES cells in vitro

Expression levels of ES cell marker, CD31, and EpiSC marker, CD40, were analyzed by flow-cytometry in Porcn wildtype (lox/Y, A-C) and mutant (del/Y, D-F) ES cells after three passages in the absence of serum and feeder cells, and in the presence of FGF inhibitor and LIF. WNT signaling levels were manipulated by GSK3 inhibition (CHIR99021, A, D) or PORCN inhibition (IWP2, C, F). DMSO treatment (B, E) served as control. PORCN inhibitor treated (C) and Porcn mutant cells (D-F) maintained high levels of CD31 and low levels of CD40, similar to wildtype ES cells under control conditions (B) or GSK3 inhibition (A).

unstained ESC + EpiSC EpiSC Porcn lox/Y - 3uM CHIR99021 t Porcn del/Y - 3uM CHIR99021 Porcn lox/Y - DMSO Coun Porcn del/Y - DMSO Porcn lox/Y - 1uM IWP Porcn del/Y - 1uM IWP

0 102 103 104 105 CD81 (FITC)

Figure 3-11: Porcn mutant ES cells maintain high levels of CD81 in vitro

Expression levels of ES cell marker CD81 was analyzed by flow-cytometry in Porcn wildtype (lox/Y) and mutant (del/Y) ES cells cultured for three passages in the absence of serum and feeder, and in the presence of FGF inhibitor and LIF. WNT signaling levels were manipulated by GSK3 inhibition (CHIR99021) or PORCN inhibition (IWP2). DMSO treatment served as control. Consistent with CD31 expression (Fig. 3-10), CD81 expression levels remained similarly high in all conditions tested when compared to EpiSC.

3.4.8 PORCN-mediated WNT signaling is not required in diapause embryos

In contrast to cultured ES cells in vitro, pluripotency in the embryo is a transient state from E3.5 to E5.5. It is thus possible, that WNT signaling in vivo is only required for prolonged maintenance of epiblast progenitor cells, similar to components of the LIF receptor complex (Nichols et al., 2001). To test this possibility, I delayed implantation of Porcn mutant embryos in vivo for 6 days by chemically inducing and maintaining diapause to EDG10 (equivalent days of gestation) (Hunter, 1999). Mutant pre-implantation embryos flushed at EDG10 had normal morphology (Figure 3-12) and were transferred into surrogates where they successfully implanted (n=9/10) and displayed the characteristic gastrulation phenotype at E7.5 (according to surrogate pregnancy). These data substantiate that embryonic Porcn deletion has no functional effects on pre-implantation development in vivo.

A Brightfield B GFP

Figure 3-12: Porcn is not required for diapause in vivo

Representative brightfield (A) and GFP (B) image of diapause blastocysts recovered at EDG10 from Porcnlox/lox, Zp3-Cre+/tg females crossed to XEGFPtg/Y males. The GFP+ve embryo is heterozygous for Porcn, whereas the GFP-ve embryo is a maternal and zygotic Porcn mutant (Porcndel/Y). When transferred into surrogates, 90% of Porcndel/Y embryos implanted and developed to gastrulation stages (n=10).

3.5 Discussion

In this study, I have used zygotic and tissue-specific deletion of Porcn to ablate WNT ligand secretion in embryonic development. Using this approach, I have determined the earliest requirements for embryonic and extra-embryonic WNT ligand secretion.

Although others and I have reported epiblast-specific Porcn mutants (Barrott et al., 2011; Biechele et al., 2011), the phenotype of zygotic inactivation of Porcn has not been described in any detail (Liu et al., 2012). I show in this study that zygotic Porcn mutants fail to gastrulate and remain in an OCT4+ OTX2+ epiblast-like state, similar to Porcn epiblast-specific mutants. This phenotype not only phenocopies Wnt3 mutants (Liu et al., 1999), but also a group of ‘canonical WNT null’ phenotypes, such as Wls (Fu et al., 2011), Mesd1 (Hsieh et al., 2003) and Lrp5/6 compound mutants (Kelly et al., 2004). Strikingly, the phenotype of the downstream effector Ctnnb1 differs slightly (Huelsken et al., 2000). While all ‘canonical WNT null’ mutants show proper DVE to AVE transition, Ctnnb1 mutants fail to establish the AVE signaling center indicative of anterior-posterior axis formation. My data supports the notion that this phenotypic discrepancy reflects a WNT independent function of Beta-Catenin (Morkel et al., 2003).

In contrast to zygotic Porcn mutants, data from epiblast-specific mutants show failure to induce primitive streak marker Brachyury at E6.5 (Barrott et al., 2011), but some residual WNT signaling response and delayed induction of the primitive streak marker Brachyury at E7.5 (Chapter 2.4.4). This phenotypic discrepancy suggests that the VE is a source of WNT ligands at E7.5. This finding is supported by a recent study showing that WNT3 secreted from the visceral endoderm is sufficient to induce, but not maintain, gastrulation in epiblast-specific Wnt3 mutants (Tortelote et al., 2012). In order to determine whether VE-secreted WNT3 is also necessary, I generated embryos with functionally Porcn mutant extra-embryonic tissues based on imprinted XCI. Surprisingly, these embryos were embryonic lethal due to a defect in chorio-allantoic fusion, similar to Wnt7b mutants (Parr et al., 2001). As WNT7B-mediated chorio-allantoic fusion occurs approximately one day after WNT3-induced gastrulation, I conclude that (1) PORCN- mediated WNT3 secretion from the VE is not necessary for gastrulation, and (2) WNT7B is the first WNT required from an extra-embryonic source. In contrast to mice, human FDH patients can inherit a mutant X-chromosomal PORCN allele from either parent (Grzeschik et al., 2007). This discrepancy in phenotypes is most likely due to the lack of stringency in imprinted extra-

74 embryonic XCI in humans (Zeng and Yankowitz, 2003), but could also indicate that WNT ligands are not required for chorio-allantoic fusion in humans.

In order to validate that there is no role for Porcn and PORCN-mediated WNT signaling in the VE, I generated VE-specific Porcn mutants using the Ttr::Cre allele (Kwon and Hadjantonakis, 2009). In keeping with data from Porcndel/+ females, VE-specific deletion had no effect on embryonic development in both males and females. Thus, multiple lines of evidence suggest that PORCN-mediated WNT secretion from the VE and its derivatives, despite being sufficient (Tortelote et al., 2012), is not necessary for the induction of gastrulation or further development to adulthood.

The phenotype of zygotic Porcn mutant embryos shows that WNT secretion is not necessary prior to gastrulation (E6.5). This is in contrast with studies suggesting functions for embryo- secreted WNT ligands in implantation (Mohamed et al., 2005). I have investigated these questions in vivo by oocyte-specific deletion of Porcn (De Vries et al., 2000; Lewandoski et al., 1997), thereby eliminating the possibility of maternal rescue. Using this approach, I was able to show that maternal zygotic Porcn mutant embryos display no implantation defects even in the absence of heterozygous littermates. These data clearly show that PORCN-mediated WNT signaling from the embryo is not required for implantation or pre-implantation development. They further support a model in which an unknown factor secreted from blastocysts is sufficient to induce WNT ligand expression in the uterine epithelium (Mohamed et al., 2005).

Data obtained from in vitro studies in ES cells shows that PORCN-mediated WNT signaling, and WNT3A protein contribute to maintaining ES cells in a pluripotent state (Berge et al., 2011). These data are further supported by the fact that ES cells can be derived and maintained efficiently in the presence of GSK3 inhibitor (Ying et al., 2008), which functionally results in activation of WNT signaling. As ES cells are derived from the epiblast of blastocysts, it has been suggested that WNT signaling is also required in the inner cell mass in vivo. The fact that maternal zygotic Porcn mutants, as well as canonical WNT receptor and Ctnnb1 mutants (Kelly et al., 2004; Rudloff and Kemler, 2012; Valenta et al., 2011), develop to gastrulation stages however shows that canonical WNT signaling is not strictly required for the maintenance of the inner cell mass in vivo. Further, in contrast to LIF signaling (Nichols et al., 2001), I show that

PORCN-mediated WNT signaling is not required for the prolonged maintenance of epiblast during diapause in vivo.

To determine whether there was a more subtle, non-lethal effect of Porcn ablation, similar to heterozygous Fgf4 ablation (Kang et al., 2012), I investigated blastocysts with maternal and zygotic deletion of Porcn, or activated WNT signaling (Ctnnb1 Δexon3). While genetic activation of the canonical WNT signaling pathway was sufficient to increase the number of inner cells of the blastocyst, ablation of Porcn had no effect on cell numbers or cell fate decisions in pre-implantation development. At molecular level, the gene expression profiles for numerous cell fate marker genes remain highly similar between wildtype embryos and embryos with genetic activation or ablation of canonical WNT signaling activity. Our results clearly show that Porcn, and thus WNT ligands, are not required for pre-implantation development.

While several direct canonical WNT signaling target genes exhibit lineage-specific expression patterns in the blastocyst, no target gene was significantly responsive to genetic activation or inactivation of the WNT signaling pathway. These data suggest that the WNT signaling response is inhibited or dampened in pre-implantation development. This dampening might be mediated by the Hippo pathway, which is actively involved in cell fate decisions in the blastocyst (Nishioka et al., 2009) and has recently been shown to be able to inhibit canonical WNT signaling by retaining Beta-Catenin in the cytoplasm (Imajo et al., 2012).

3.6 Materials and Methods

3.6.1 Generation of Porcn floxed allele

G4 ES cells (George et al., 2007) were electroporated with a targeting construct containing loxP sites flanking exon 3 and an FRT-flanked Neomycin resistance cassette. Neomycin-resistant clones were selected and individually expanded. Southern Blot Analysis using a NeomycinR probe identified cell lines with single integrations and fragment lengths indicating homologous recombination. Correct homologous recombination was confirmed in candidate cell lines by long-range PCR (PrimeStar, Takara Bio Inc.) using combinations of genomic and NeoR primers spanning the entire 5’ arm and 3’ arm respectively (~4 kb). PCR amplicons were fully sequenced to confirm correct integration. The Neomycin resistance cassette was excised using transient expression of FlpE recombinase and chimeric founder males were generated by aggregation of ES cells (clone H4D4) with ICR embryos (Cox et al., 2010). Chimeric founder animals were bred to ICR females to identify germline transmission. Offspring were genotyped by PCR using the following primers: PorcnRecF1 5’ctgttaaaccaagacatgaccttca, PorcnRecR1 5’ taactaggacgctttgggataggat, and PorcnRecR3 5’gttctgccttcctaacccatataac.

3.6.2 Mouse alleles and genetic backgrounds

All animal experiments were performed in a specific pathogen free environment at the Toronto Centre for Phenogenomics (TCP) and all procedures were approved by the institutional Animal Care Committee in accordance to guidelines by the Canadian Council for Animal Care (CCAC). Unless indicated otherwise, experiments were performed using outbred ICR mice carrying the Porcn floxed allele and/or the following transgenes; D4/XEGFP (Tg(GFPX)4Nagy) (Hadjantonakis et al., 1998a), pCX-NLS-Cre (Tg(ACTB-cre)1Nagy) (Belteki et al., 2005), Tcf/Lef-LacZ (Mohamed et al., 2004), Hhex-EGFP (Tg(Hhex-EGFP)#Rbe)(Rodriguez et al., 2001), Zp3-cre (Tg(Zp3-Cre)3Mrt) (Lewandoski et al., 1997), Ttr::Cre (Tg(Ttr-cre)1Hadj) (Kwon and Hadjantonakis, 2009), Ctnnb1lox(ex3) (Ctnnb1tm1Mmt) (Harada et al., 1999). For blastocyst studies (Immunostaining, Fluidigm Gene Expression Analysis), the Porcn floxed allele was backcrossed to C57BL/6J for five generations (incipient congenic) and deleted using a Zp3-Cre allele on C57BL/6J background (Tg(Zp3-cre)93Knw/J) (De Vries et al., 2000). Genotyping of mice and embryos was performed using Sigma REDExtract-N-Amp™ Tissue PCR Kit according to manufacturers protocol using primers as indicated in original publications

77 for alleles used. All genotypes mentioned follow the convention and indicate the maternal allele first (i.e. Genemat/pat).

3.6.3 Post-implantation embryo collection, staining and imaging

Embryos were obtained from natural timed mating and dissected in PBS. Whole-mount in situ hybridization was performed as described in chapter 2.6.6. Beta-Galactosidase staining of gastrulating embryos were performed as previously described (Biechele et al., 2011; Cox et al., 2010), and imaged using a LeicaMZ16F microscope. E8.5 embryos were stained for Integrin alpha 4 whole-mount and sectioned subsequently as previously described (Daane et al., 2011).

3.6.4 Pre-implantation embryo collection and imaging

Embryos were generated by natural mating, recovered in M2 (Specialty Media, Chemicon) at E3.5, and cultured in KSOM (Specialty Media, Chemicon) under mineral oil (Sigma) at 37°C and 5% CO2 in air for 24 hours (E4.5). Fixation and immunostaining was performed as previously described (Stephenson et al., 2010) with the following primary antibodies: rabbit anti- NANOG (1:200, Cosmo Bio, REC-RCAB0002), goat anti-GATA6 (1:200, R&D Systems, AF1700), mouse anti-CDX2 (1:200, BioGenex, MU392-UC), donkey anti-rabbit DyLight 549 (1:400, Jackson ImmunoResearch), donkey anti-goat Alexa Fluor 488 (1:400, Molecular Probes), donkey anti-mouse Alexa Fluor 633 (1:400, Molecular Probes). Nuclei were labeled with Hoechst 33342 (Molecular Probes). Images were captured using a Zeiss Axiovert 200 inverted microscope equipped with a Hamamatsu C9100-13 EM-CCD camera, a Quorum spinning disk confocal scan head and Volocity acquisition software (Perkin Elmer). Z-stacks were taken at 2 µm intervals with a 20x water immersion objective (NA=0.75). Images were exported to ImageJ for analysis, and cells were manually scored as epiblast, primitive endoderm or trophectoderm based on position of Hoechst-stained nuclei in the embryo and expression of NANOG, GATA6 and CDX2. Per genotype, cells of 5 embryos were quantified and the data was analyzed using Chi square analysis. Graphs were generated using Prism5 Software (GraphPad Software, Inc.).

3.6.5 Single cell gene expression analysis

For gene expression analysis, cultured E4.5 embryos were treated with acid Tyrode’s solution to remove the zona pellucida, then incubated in 2 mg/ml collagenase IV for 20 min at 37°C, followed by brief trypsinization, and then manually dissociated into single cells using a finely-

78 pulled glass capillary. For maternal zygotic mutants, embryos were dissociated individually and genotype (as X-linked) was determined based on expression of XIST, Uty, and Ddx3y in the subsequent gene expression analysis (Figure 3-13). Control embryos (C57BL6/N and pCX-Nls- Cretg/+; Ctnnb1+/del ex3) were treated as groups.

Gene expression analysis was performed using 48.48 Dynamic Arrays on the BioMark System (Fluidigm). Individual cells were lysed in 5 µL RT-PreAmp Master Mix, containing CellsDirect 2X Reaction Mix (Invitrogen), 0.2X assay pool of recommended TaqMan GeneExpression Assays (20X, Applied Biosystem) and RT/Taq Enzyme (CellsDirect qRT-PCR kit, Invitrogen). Cell lysis, sequence-specific reverse transcription (50°C for 20 min) and sequence-specific amplification (18 cycles of: 95°C for 15 seconds, 60°C for 4 min) were performed immediately. The pre-amplified product was diluted 5-fold before being analyzed on 48.48 Dynamic Arrays on the BioMark System with TaqMan GeneExpression Assays (Applied Biosystem).

Fluidigm data was analyzed using a modified version of the LogEx method as published by Fluidigm. Raw Ct (Cycle threshold) values for all data were assembled in to a table in R. Samples were normalized by determining the lowest observed detection (LOD) for each probe and subtracting the observed Ct values from it. Median and standard deviations for each gene probe were calculated. It was determined that a Ct value ≥5 for control gene probes (Actb, Gapdh) was necessary for the cell sample to pass quality control. Cells that did not pass this QC process were removed. Any Ct values that were less than zero (i.e. less than the LOD) were replaced with zeros. Embryos from Porcn deletion crosses were classified by sex using the gene probes for Xist, Uty and Ddx3y to determine genotypes (female = heterozygous, male = mutant). Cells from each embryo were then classified into each of three lineages using known lineage markers (Table 3-1) by expectation maximization clustering forced to a fit of three populations. Next the median Ct value for each cell lineage and genotype was calculated. These values were then clustered using the default parameters of heatmap.2 method from the R library gplots. Genotyping of Porcn mutants by sex-specific gene expression as well as cell numbers and fates obtained for BioMark gene expression analysis are displayed in supplementary Figure 3-13.

A B

n=35 n=49 n=93 n=180 100 EPI PE XIST TE 80 Uty Ddx3y 60

EmbryoEmbryo 1 Embryo 2 Embryo 3 Embryo 4 Embryo 5 Embryo 6 Embryo 7 Embryo 8 9 EmbryoEmbryo 10 Embryo 11 Embryo 12 13 Normalized Ct value 40

0 5 10 20

Porcn +/+ Porcn del/+ mzPorcn del/Y Ctnnb1 +/del ex3

Figure 3-13: Genotyping of Porcn mutants by sex-specific gene expression and cell numbers and fates obtained for BioMark gene expression analysis

(A) Male Porcn mutant (del/Y) and female heterozygous (del/+) embryos obtained from maternal zygotic deletion crosses were distinguished by gene expression analysis for XIST (female-specific), Uty and Ddx3y (male specific). Hierarchical clustering was used to identify female (embryos 1-6) and male embryos (embryos 7-13). (B) Cell numbers and cell fate distribution of cells analyzed by BioMark (Fluidigm) single cell gene expression analysis.

Table 3-1: TaqMan GeneExpression Assays (Applied Biosystem) used for single cell gene expression analysis Probe Function Lineage Actb Housekeeping gene n/a Axin2 WNT target n/a Brachyury (T) WNT target/lineage marker meso/endoderm c-Myc WNT target n/a CyclinD1 WNT target n.a Ctnnb1 WNT component n/a Cripto Lineage marker epiblast Ddx3y Sexing/genotyping male specific Dnmt3b Lineage marker epiblast Eomes Lineage marker trophectoderm Fgf4 Lineage marker epiblast Fgf5 Lineage marker epiblast Fgfr2 Lineage marker trophectoderm Gapdh Housekeeping gene n/a Gata3 Lineage marker trophectoderm Gata4 Lineage marker primitive endoderm Gata6 Lineage marker primitive endoderm Gbx2 WNT target/lineage marker n/a, neural Klf2 Lineage marker epiblast Klf4 Lineage marker epiblast Lef1 WNT target n/a Nanog Lineage marker epiblast Oct3/4 Lineage marker epiblast Otx2 Lineage marker epiblast/neural Pdgfra Lineage marker primitive endoderm Pecam1 Lineage marker epiblast Porcn WNT component n/a Rex1 Lineage marker epiblast Sox2 Lineage marker epiblast Sox7 Lineage marker primitive endoderm Sox17 Lineage marker primitive endoderm Stella Lineage marker epiblast Tcf1 WNT target n/a Uty Sexing/Genotyping male specific Wls WNT target n/a XIST Sexing/genotyping female specific

3.6.6 Sex-separated pregnancies

Embryos were generated by natural mating of Porcn lox/lox; Zp3-Cre(Mrt)+/tg females to XEGFPtg/Y males. At E3.5 (or EDG10) embryos were harvested, sexed based on EGFP fluorescence (Hadjantonakis et al., 1998a) and transferred separately into E2.5 pseudo-pregnant females. Embryos were recovered from uteri five days later (E7.5).

3.6.7 Diapause induction

Diapause was induced as previously described (Hunter, 1999). Briefly, Porcn lox/lox; Zp3- Cre(Mrt)+/tg females were naturally mated to XEGFPtg/Y males. At E2.5, females were injected subcutaneously with 3 mg Depo-Provera and intraperitoneally with 10 ug of Tamoxifen. Tamoxifen injections were repeated every 3-4 days and embryos were flushed from uteri at indicated stages. As Diapause begins at E4.5, 6 day diapause embryos were harvested at EDG10 (Equivalent Days of Gestation).

3.6.8 RT-PCR

RT-PCR was performed as previously described (Chapter 1.6.9) with the following primers: PorcnEx2-4 Fwd: GGCTGCTTCTTACCATCTGC, Rev: TGTCCACCATGTGCATCTCAC, PorcnEx4-9 Fwd: GTGAGATGCACATGGTGGAC, Rev: ACTGTCAGGTCCCATTCCAG. Detection of H2afz transcript was used as control (Mamo et al., 2007).

3.6.9 Autocrine Tcf/Lef-Luciferase Assay

Autocrine Tcf/Lef-Luciferase assays were performed as previously described (Chapter 2.6.3).

3.6.10 Flow-cytometric analysis of ES cells

For flow-cytometric analysis, a single-cell suspension of Porcnlox/Y (H4D4) or Porcndel/Y (H4D4F9) ESCs was plated onto gelatin/serum-coated plates at 10,000 cells/cm2 in serum-free N2B27 medium supplemented with LIF and 1 uM PD0325901 (Selleck Chemicals) (Nichols and Ying, 2006). In addition, 3 uM CHIR99021 (Selleck Chemicals), DMSO (Sigma), or 1 uM IWP2 (Sigma) were added. Media was changed daily and cells were passaged every 3 days using Trypsin (Gibco) and plated at 10,000 cells/cm2. At every passage, a portion of the cells was triturated to single-cell suspension, stained with anti-CD31–PECy7 (Biolegend, 102418), anti- CD81–FITC (R&D Systems, FAB4865F) antibodies and goat-anti-CD40 (R&D Systems

AF440). Donkey-anti-goat-DyeLight649 (Jackson ImmunoResearch, 705-495-147) was used as a secondary antibody for CD40 detection. Cell suspensions were analysed using a Becton Dickinson LSRII flow cytometer and FACS plots were generated using FlowJo software (Tree Star Inc.). Representative plots and histograms shown in figures 3-9 and 3-10 were obtained after the third passage.

3.7 Acknowledgements

I would like to thank Jorge Cabezas for assistance with mouse husbandry, Malgosia Kownacka and Jodi Garner for assistance with tissue culture, Angela McDonald for flow cytometry, Andres Nieto for Fluidigm assistance and Oliver Tam for discussions. Further, I would like to acknowledge the Toronto Centre for Phenogenomics (TCP) and specifically the TCP Transgenic Core Facility for excellent support in mouse generation and embryo transfers. I would further like to thank Dr. Kat Hadjantonakis for sharing of the Ttr-Cre transgenic mouse line. The author would like to acknowledge the Samuel Lunenfeld Research Institute’s CMHD Pathology Core for Histology for their technical services (www.cmhd.ca). Supported in part by Research Grant No. 6-FY08-315 from the March of Dimes Foundation.

4 Zygotic Porcn paternal allele deletion in mice to model human FDH 4.1 Contributions

The data in this chapter is currently (April 2013) being prepared for submission to PLOS ONE for publication. Steffen Biechele performed all aspects of mouse breeding, dissections and project coordination. The phenotype of adult mice was assessed by the Centre for Modeling Human Disease (CMHD) Mouse Physiology Facility. Histologic processing was performed by the CMHD Pathology Core at the Toronto Centre for Phenogenomics (TCP). Pathologic analysis was performed by Dr. Hibret Adissu. Dr. Brian Cox and Dr. Janet Rossant guided experimental design and interpretation of results.

4.2 Abstract

In humans, mutations in PORCN cause the X-linked dominant syndrome Focal Dermal Hypoplasia (FDH, OMIM#305600). This disorder is characterized by ecto-mesodermal dysplasias and shows a highly variably phenotype, potentially due to individual XCI patterns. In order to establish an adult mouse model for this human disease, I have generated Porcn heterozygous animals by paternal allele deletion in zygotes. Heterozygous female fetuses displayed variable morphological abnormalities during development, but survived to term. However, 95% of these females died perinatally due to body wall closure defects, diaphragmatic hernias and severe kidney defects. Rare adult survivors recapitulated several aspects of human FDH, such as skin and skeletal defects. Further, all mice presented with bronchopneumonia, rhinitis, and otitis media, lesions that are not frequently reported in humans. In contrast to other Porcn mutant mouse models, we have assessed both fetal and adult phenotypes. While all phenotypes might be caused by the expected defects in WNT ligand secretion, we have observed numerous defects that are typically associated with ciliopathies, suggesting that PORCN function may be required for cilia formation or the existence of an unexplained linkage between cilia function and WNT signaling.

4.3 Introduction

Similar to mice, humans carry a single PORCN gene on the X chromosome (Xp11.23). Mutations in human PORCN cause FDH (Goltz Syndrome, OMIM#305600) (Grzeschik et al., 2007; Wang et al., 2007), an X-linked dominant disorder characterized by dysplasias in ecto- mesodermal tissues. Phenotypically, FDH is characterized by patchy, hypoplastic skin (Goltz, 1992), frequently in combination with other defects, such as digital abnormalities, microphthalmia, hypodontia, kidney abnormalities, abdominal wall defects, skeletal abnormalities, and reduced bone density (Bornholdt et al., 2009c). The majority of FDH patients are heterozygous females, which are mosaic for PORCN function due to XCI. Male FDH patients represent approximately 10% of observed cases and carry post-zygotic mutations, leading to functional mosaicism similar to heterozygous females. The only exception published to date is a male Klinefelter/FDH patient (47,XXY) without detectable mosaicism for the PORCN mutation (Alkindi et al., 2012). Similar to females carrying zygotic mutations, the second X chromosome carrying a wildtype PORCN allele leads to functional mosaicism due to XCI. The absence of zygotic mutant male or homozygous female patients, along with mouse data showing that Porcn is required in the embryo proper for gastrulation (Barrott et al., 2011; Biechele et al., 2011), have led to the conclusion that human hemizygous PORCN mutants are embryonic lethal. Heterozygous females however, exhibit a variable phenotype, most likely due to differences in individual XCI patterns and frequencies. Mouse models recapitulating certain aspects of the human phenotype range have been generated by Porcn mutant chimera formation (Liu et al., 2012), as well as tissue-specific deletions of Porcn (Barrott et al., 2011; Liu et al., 2012) or Wls (Chen et al., 2012a; Fu et al., 2011; Zhu et al., 2012a), which is also essential for WNT secretion. Zygotic Porcn heterozygous adult females, however, have not been described to date, limiting the comparison of human and mouse phenotypes.

Taking advantage of the floxed Porcn allele I have generated, I used zygotic deletion of the paternal Porcn allele to generate a mouse model for human FDH. I show that heterozygous female fetuses display variable defects that do not significantly affect survival in the uterus, but lead to perinatal lethality of more than 95% of females. Rare survivors develop to adulthood and display variable skeletal and skin defects, representing an adult zygotic mouse model for human FDH. Although not frequently reported in humans, we consistently observed bronchopneumonia,

85 rhinitis, and otitis media in these animals, suggesting a potential link between Porcn function and ciliopathies.

4.4 Results

4.4.1 Embryonic defects cause perinatal lethality

I have previously shown that zygotic deletion of the maternal Porcn allele in female mouse embryos causes failure in chorio-allantoic fusion and embryonic lethality by E11.5 (Chapter 3.4.3). In order to establish a model for human FDH, I generated heterozygous females by zygotic deletion of the paternal allele (Porcn+/del) using the ubiquitously expressed pCX-NLS- Cre transgene transmitted through the female germline (Figure 4-1). An X-linked EGFP transgene in cis to the mutant allele was used to identify mutant cells in female fetuses and female neonates (Hadjantonakis et al., 2001). In the mice and humans, XCI is random in the embryo proper, but imprinted in extra-embryonic tissues, leading to monoallelic expression from the maternal X chromosome (Barakat and Gribnau, 2012). Paternal allele mutant females, thus have functionally wildtype extra-embryonic tissues.

+/+ lox/Y Porcn Porcn +/tg tg/Y pCX-NLS-Cre X XEGFP

+/del +/Y Porcn Porcn +/tg XEGFP +/tg +/tg pCX-NLS-Cre pCX-NLS-Cre

+/del +/Y Porcn Porcn +/tg XEGFP

Figure 4-1: Schematic outlining the genetic strategy to generate Porcn+/del females

Maternally transmitted, ubiquitously expressed Cre recombinase (pCX-NLS-Cre transgene) is loaded into eggs, leading to the zygotic deletion of a paternally transmitted Porcn floxed allele upon fertilization independent of transgene transmission. A X-linked green fluorescent protein (GFP) transgene in cis to the floxed Porcn allele was used to identify female fetuses and pups. Due to XCI, functionally mutant cells express GFP.

I noted that heterozygous females were underrepresented in newborn litters and a large percentage of these mice were lost due to perinatal mortality (Fig. 4-2). However, by breeding larger numbers of mice rare survivors were identified. Despite observations of occasional embryonic lethality, I was able to observe Porcn+/del female fetuses up to E18.5 with no significant deviations from the expected frequency (50%, Fig. 4-2), suggesting that the majority of malformations observed do not impinge on embryonic survival. Assessment of litters at birth (P0), however, revealed that the majority Porcn+/del females die at birth and are cannibalized by the mothers, as only 7.9% of expected females (22/277) were observed alive (Fig. 4-2). Only 3.6% (10/277) survive to postnatal day 7 (P7), indicating a high rate of perinatal lethality (Fig. 4- 2). No unusual lethality was observed past P7.

Figure 4-2: Analysis of survival of Porcn+/del fetuses and neonates

At fetal stages (E11.5 to E18.5) Porcn+/del females were observed at the expected Mendelian frequency (50%). At birth (P0) only 7.9% of the expected female neonates were observed alive. Lethality in the first week of life further reduced this frequency to 3.6% (10 out of 277 expected) at postnatal day 7 (P7).

Similar to human patients and a previous report (Barrott et al., 2011), I found a wide spectrum of defects during fetal development of Porcn+/del females. Typical defects included posterior truncations (Fig. 4-3 A, B) reminiscent of Cdx1::Cre deletions of Ctnnb1 (Hierholzer and Kemler, 2010), digital abnormalities, body wall closure defects, and defects in tail development (Fig. 4-3 C, D), as well as craniofacial defects. As the craniofacial defects were highly variable between individual female fetuses, frequently unilateral and inconsistent in their occurrence, we did not document these phenotypes in more detail. Digital obnormalities have also been observed in tissue-specific deletions of Porcn or Wls in the limb-ectoderm, suggesting that Wnt ligands secreted from the ectoderm are required for distal limb patterning (Barrott et al., 2011; Zhu et al., 2012b).

In order to investigate the cause of perinatal lethality in more detail, we assessed midsagittal and parasagittal sections of fetuses just prior to birth (E18.5). While some Porcndel/+ fetuses (n=2/9) and Porcn+/Y littermates (n=2/2) showed no obvious abnormalities (Fig. 4-4 A), the majority of fetuses (n=7/9) exhibited defects that could cause perinatal lethality. The individually variable defects can be grouped into four categories; thoracic body wall defects (n=5/9, Fig. 4-4 B), diaphragmatic hernias with abdominal organs protruding into the thoracic cavity (n=5/9, Fig. 4-4 C), kidney defects such as hydronephrosis and hydroureter (n=4/9, Fig. 4-4 E, F), and midline closure defects (Fig. 4-3 D). We also observed a high frequency of fetuses with focal dermal hypoplasia (n=6/9, Fig. 4-4 G, H), which can be excluded as cause of lethality, but is the name- giving feature of the human disease. Thoracic body wall defects and diaphragmatic hernias are likely to impair lung function and compromise postnatal survival. Further, severe kidney defects can also cause lethality within the first days of life. Together with midline closure defects that expose organs to the exterior, these defects can explain the perinatal lethality observed in Porcn+/del females.

A E11.5 B E11.5

+/Y +/del C E18.5 D E18.5

+/Y +/del

Figure 4-3: Gross morphological abnormalities in Porcn+/del fetuses

At E11.5, heterozygous females with posterior truncations could be observed (B), while wildtype Porcn+/Y littermates (A) developed normally. Just prior to birth (E18.5), several Porcn+/del females (D) displayed defects in body wall closure (arrowhead), digital abnormalities (arrow) and lack of tail (open arrowhead). Male littermates never displayed these defects (C).

A B

+/Y +/del C D

+/del +/del E F

+/Y +/del G H

+/Y +/del

Figure 4-4: Histological analysis of Porcn+/del females at E18.5

At E18.5, in contrast to control Porcn+/Y littermates (A), several Porcn+/del females exhibited body wall closure defects (B, n=5/9), diaphragmatic hernias (C, n=5/9), and signs of spina bifida (D, open arrowhead, n=1/9). Arrows indicate the diaphragm. Arrowheads indicate the anterior body wall. Approximately 45% (n=4/9) heterozygous females displayed signs of severe kidney disease, such as hydronephrosis (F), which was never observed in control littermates (E). The skin of the majority of Porcn+/del fetuses (n=6/9) displayed signs of focal dermal hypoplasia (H).

4.4.2 Rare adult Porcn+/del females as a model for human FDH

To establish the relationship of our FDH model mice to human FDH and identify possible novel defects we phenotyped five adult Porcn+/del; XEGFP+/tg; pCX-NLS-Cre+/tg females at nine to ten weeks of age. As the genetic strategy did not generate female control littermates, we used Porcn+/+; XEGFP+/tg; pCX-NLS-Cre+/tg females on the same genetic background as controls. Compared to control females, Porcn+/del females had reduced weight (p=0.07) and significantly reduced locomotor activity (Fig. 4-5 A, B), indicating poor clinical condition. Blood glucose and triglyceride levels were also significantly reduced (Fig. 4-5 C, D), potentially explaining the lethargy observed in locomotor activity testing. We further observed increased blood urea levels (Fig. 4-5 E), pointing towards kidney defects. However, kidney morphology and urinalysis were normal, suggesting a pre-renal cause such as reduced glomerular function due to dehydration.

Similar to my observations at E18.5, some females (2/5) had skin lesions following blaschkoid lines on the torso (Paller, 2007). Consistent with a previous report (Barrott et al., 2011) and the human phenotype, these lesions were characterized by reduction in dermal collagen and adnexal aplasia and/or hypoplasia (Fig. 4-5 F, G). In order to characterize which cells within the lesions have an active mutant X chromosome, I stained sections for EGFP, which labels functionally mutant cells. EGFP expression was equally mosaic in both normal and affected regions of the skin (Fig. 4-5 H-K), suggesting that the requirement for WNT secretion was not confined to a readily identifiable cell source. However, tissue-specific deletions of Porcn and Wls, which similarly affects Wnt ligand secretion, have shown that epidermal WNT ligands are required for the induction of hair follicles and for the proliferation of dermal cells, whereas dermal WNT ligands appear to be dispensable (Barrott et al., 2011; Chen et al., 2012a; Fu and Hsu, 2012; Liu et al., 2012). Intriguingly, Porcn expression has not been observed consistently in surface ectoderm of developing embryos (Biechele et al., 2011; Diez-Roux et al., 2011; Liu et al., 2012). This discrepancy might be due to the detection limit for Porcn in in situ hybridization experiments and suggests that Wnt ligand secretion from the epidermis might be tightly regulated at the level of Porcn expression (Proffitt and Virshup, 2012).

A Weight B Locomotor C Blood D Triglycerides E Urea Activity Glucose

p=0.0700 p=0.0288 p=0.0014 p=0.0017 p=0.0265

+/+ +/tg +/del +/tg Porcn XEGFP Porcn XEGFP F H&E G H&E

500 um 500 um

H GFP IHC I K GFP IHC J

500 um 500 um J K

Figure 4-5: Adult Porcn+/del females as a model for FDH

Adult Porcn+/del exhibit reduced weight and locomotor activity compared to control Porcn+/+ females (A, B). Blood glucose and triglyceride levels were reduced (C, D), whereas urea levels in the blood were increased (E). Heterozygous females further displayed FDH characteristic focal dermal hypoplasia with reduction/absence of dermal collagen and adnexal hypoplasia/aplasia (H&E staining, F, G). Functionally mutant cells are labeled by an X-linked GFP transgene in cis to the deleted allele. Immunohistochemistry for GFP (GFP IHC) did not reveal major differences in GFP expression patterns in the skin of heterozygous or control animals. Figures 5 A-E were analyzed by unpaired student’s t-test.

4.4.3 Skeletal defects in Porcn+/del females

As human patients frequently present with skeletal abnormalities and reduced bone density, we performed body composition analyses, X-ray imaging and necropsies. While we could not detect significant changes in fat and lean mass (Fig. 4-6 A, B), the bone mineral density (BMD) and bone mineral content (BMC) were significantly reduced (Fig 4-6 C, D). X-ray imaging was largely unremarkable (Fig. 4-6 E, F), but necropsies identified one mouse with a thoracic body wall defect; the thorax exhibited a 10 mm wedge-shaped gap in the sternal bone (Fig. 4-6 G). The sternal osseous and cartilaginous structures on either side of the defect were each enveloped by differentiated periosteal and perichondrial tissue consistent with duplication of the sternal skeleton (Fig. 4-6 H). Strikingly, human FDH patients with split sternum have been observed (Leoyklang et al., 2008; Loguercio Leite et al., 2005), highlighting the similarities between mouse and human phenotypes.

A Fat B Lean C BMD D BMC mass mass

p=0.0848 p=0.1283 p=0.0073 p=0.0168

E +/+ F +/del

G H

Figure 4-6: Body composition and skeletal phenotypes in adult Porcn+/del females

Adult Porcn+/del and Porcn+/+ females were subjected to X-ray imaging and body composition analysis. While fat mass (A) and lean mass (B) was not significantly different (t-test), bone mineral density (BMD, C) and bone mineral content (BMC, D) were significantly reduced in heterozygous females. X-ray imaging was unremarkable in both control (E) and heterozygous females (F). One out of the five analyzed heterozygous female exhibited a sternal gap (G, white arrowhead). Consistent with duplication of the sternal skeleton, osseous structures on either side were enveloped by periosteal tissue. Arrow indicates the border between left and right ribcage.

4.4.4 Novel observations in the FDH mouse model

All five Porcn+/del females invariably exhibited otitis media, rhinitis, and bronchopneumonia with bronchiectasis (Fig. 4-7 A-F). Mild bilateral hydrocephalus of the third ventricle was also seen in some females (2/5, Fig. 4-7 G, H). In FDH model mice the pneumonia was characterized by pyogranulomatous inflammation centered on foreign material (hair, food, bedding) and colonies of coccoid bacteria within the bronchioles, consistent with aspiration pneumonia (Fig. 4-7 I, J). Aspiration pneumonia is rare in mice and its presence only in mutants rules out environmental or iatrogenic causes. Consistent with these findings, recurrent pneumonia has been reported in some FDH patients in association with gastroesophageal reflux and nasal regurgitation during feeding (Boothroyd and Hall, 1987; Leoyklang et al., 2008). It is not known if there are lung defects that are associated with this symptom. Mild right ventricular hypertrophy of the heart was observed (Fig. 4-7 K, L); this was likely secondary to pulmonary hypertension associated with pneumonia. Consistent with the chronic, active pulmonary inflammation, we detected significant increases in white blood cell counts, lymphocytes, monocytes and neutrophils (Fig. 4-7 M-P). Further, we detected an increase in red blood cell counts (Fig. 4-7 Q) and significant increases in total hemoglobin concentration (Fig. 4-7 R). Whether these increases are due to dehydration or an adaptive response to poor lung function is not clear.

To investigate the cause of the bronchopneumonia, we performed Periodic Acid-Schiff (PAS) staining and histological analyses on tracheae of adult animals. In contrast to controls (Fig. 4-7 S), the normal ciliated epithelial morphology was disrupted by segmental loss of ciliated epithelial cells (up to 200 um in length, Fig. 4-7 T). In these segments the tracheal epithelium is replaced by single cell layered or disorganized multilayered non-ciliated cuboidal to squamous type epithelium. There is marked goblet cell hyperplasia as evidenced by increased numbers of PAS positive cells within the bronchioles, notably in areas of inflammation (Fig. 4-7 U, V). The PAS stain also indicated excessive mucous within lower airways. These changes in the tracheal epithelium of the mutants could constitute reparative and protective morphological modifications in response to chronic active inflammation. It is however possible that the segmental absence of normal ciliated epithelium might have resulted in suboptimal mucociliary function and subsequent aspiration pneumonia. It remains unclear if similar cilia defects are the cause of the observation of pneumonia in human FDH patients. Nonetheless, the constellation of lesions in

96 the Porcn+/del females, namely: otitis media, rhinitis, aspiration bronchopneumonia, and hydrocephalus strongly suggest defect in motile cilia (ciliopathy). These findings have been linked to ciliopathy in humans (Ware et al., 2011) as well as some mouse models (Vogel et al., 2010).

A +/+ B +/del C +/+ D +/del

E +/+ F +/del G +/+ H +/del

I +/+ J +/del K +/+ L +/del

M N O P Q R

WBC Lymphocytes Monocytes Neutrophils RBC Hemoglobin p=0.0001 p=0.0015 p=0.0061 p=0.0239 p=0.3777 p=0.0008 S +/+ T +/del U +/+ V +/del

Figure 4-7: Novel observations in Porcn+/del FDH mouse model

In contrast to control animals (A, C, E, I), adult Porcn+/del females exhibit otitis media (B), rhinitis (D) and pneumonia obliterans (F, J). Mild hydrocephalus was also observed in 2/5 Porcn+/+ females (G, H). Arrowhead indicates enlarged third ventricle. Bronchioles of heterozygous females contained bedding (arrowhead, J) and large numbers of immune cells. Pneumonia obliterans was accompanied by mild cardiac right ventricular hypertrophy (arrowhead, K, L). Hematology profiles showed significant increases in white blood cells (WBC, M), lymphocytes (N), monocytes (O), and neutrophils (P). Increases were also observed in red blood cell (RBC) counts (Q) and hemoglobin concentration (R). Blood cell counts were analyzed by unpaired student’s t-test. Figures S-V: PAS stain of tracheae and bronchioles showed that tracheal epithelia were slightly disorganized and lacked apical cilia in segments of up to 200 um (T) compared to controls (S). Bronchioles of heterozygous females have increased numbers of goblet cells (V, pink cells, arrowhead) in areas of inflammation indicated by the presence of immune cells (V, arrow) compared to controls (U).

4.5 Discussion

In this chapter, I have used zygotic mutation of the paternal Porcn allele to ablate WNT ligand secretion in embryonic development. Using this approach, I have created a mouse model for the human disease FDH. Several aspects of human FDH have been recapitulated in mice using tissue-specific Porcn or Wls deletions and chimera formation (Barrott et al., 2011; Biechele et al., 2011; Chen et al., 2012a; Liu et al., 2012; Zhu et al., 2012b). The majority of results obtained from these models are consistent and minor deviations are likely attributable to variations in genetic background or specific Cre alleles used. In contrast to these tissue-specific studies, this chapter presents the first heterozygous female mice that have been followed throughout embryonic development and into adulthood.

Similar to a previous report based on the epiblast-specific deletion of the maternal Porcn allele (Barrott et al., 2011), I have encountered variable defects throughout fetal development and a dramatic perinatal lethality of 95% of zygotic paternal allele mutant heterozygous females. The variability of fetal defects can be attributed to XCI patterns that are unique and potentially skewed in each female. The individual phenotypes are thus dependent on which cells or tissues affected functionally. This situation is further complicated by non-cell-autonomy of WNT- related effects, as the actual phenotype may be observed in a functionally wildtype cell due to genetic ablation of the nearby WNT source. While the fetal lethality was not significant, a large majority of heterozygous females died perinatally due to diaphragmatic hernias, body wall closure defects and severe kidney defects. Porcn heterozygous female mice surviving the perinatal period developed fairly normally, but recapitulated typical skin defects and also a more rare occurrence of a split sternum, which has also been observed in human patients (Leoyklang et al., 2008).

Unexpectedly, this FDH mouse model invariably exhibits aspiration pneumonia and several associated effects, such as thickening of the left ventricular wall and hematology profiles consistent with chronic active pulmonary inflammation. In contrast to our observations in the mouse models, such consistent lung phenotypes are only rarely reported in human patients. Thus it remains unclear whether this discrepancy is due to species-specific differences in PORCN and WNT functions, or whether lung phenotypes are not reported adequately in human FDH cases.

Surprisingly, adult Porcn+/del animals display several phenotypes that are frequently associated with ciliary defects: aspiration pneumonia with bronchiectasis, rhinitis, otitis media, kidney defects and mild hydrocephalus. The accumulation of these phenotypes leads us to speculate that PORCN might be involved in ciliary assembly or function. Such an effect could be mediated by WNT ligands: canonical WNT signaling has been shown to regulate ciliogenesis in Kupffer’s vesicle in zebrafish (Caron et al., 2012). The potential effect on cilia function might also depend on WNT/PCP-mediated orientation of cilia, similar to the orientation of stereocilia bundles in the inner ear (Dabdoub et al., 2003). Finally, PORCN might have a WNT-unrelated effect on cilia. WNT-unrelated Porcn functions have been reported previously (Covey et al., 2012), but remain hard to investigate in vivo due to the co-occurrence of severe WNT-related defects. In summary, my analysis of adult Porcn+/del female mice has revealed significant overlap between mouse and human phenotypes, but also highlighted some discrepancies that may represent species-specific features or may have been missed in reported human cases.

100

4.6 Materials and Methods

4.6.1 Mouse alleles and genetic background

All animal experiments were performed in a specific pathogen free environment at the Toronto Centre for Phenogenomics (TCP) and all procedures were approved by the institutional Animal Care Committee in accordance to guidelines by the Canadian Council for Animal Care (CCAC). In order to identify female mice and Porcn mutant cells in heterozygous animals, I established a mouseline carrying my Porcn floxed allele in cis to the X-linked D4/XEGFP transgene (Tg(GFPX)4Nagy) (Hadjantonakis et al., 1998a) on an outbred ICR background. Both hemizygous and heterozygous animals are viable, fertile, and did not display any obvious defects. Porcn+/del female fetuses and adults were generated by crossing Porcnlox/Y; XEFPtg/Y males to pCX-NLS-Cre+/tg females. In this setting, the floxed allele is deleted in all zygotes due to inheritance of the maternal Cre allele or maternal loading of the Cre transcript respectively. Female fetuses and newborns were identified by expression of the GFP transgene. Control females for adult phenotyping were generated by crossing XEFPtg/Y males to pCX-NLS-Cre+/tg females.

4.6.2 Genotyping of mice and fetuses

Genotyping of mice and embryos was performed using Sigma REDExtract-N-Amp™ Tissue PCR Kit according to manufacturer’s protocol using the genotyping primers PorcnRecF1 5’ctgttaaaccaagacatgaccttca, PorcnRecR1 5’ taactaggacgctttgggataggat, and PorcnRecR3 5’ gttctgccttcctaacccatataac. Further, fluorescent (GFP) labeling of females and PCR genotyping for Sry were used to determine the sex of fetuses.

4.6.3 Staging and Imaging

Fetuses were generated by timed mating. The day of finding a vaginal plug was designated embryonic day 0.5 (E0.5) and fetuses were dissected in PBS at the indicated stages. Fetuses older than E15.5 were euthanized by decapitation and imaged on a MZ16F microscope (Leica) equipped with a MicroPublisher 5.0 RTV camera (Qimaging).

101

4.6.4 Modified SHIRPA

The general appearance and behavior screening was performed using a modified SHIRPA protocol (Rogers et al., 1997) with details at www.CMHD.ca. A 20 kHz clickbox (MRC Institute of Hearing, Nottingham, UK) was used to elicit the Preyer reflex indicative of normal hearing. Eyes were scanned for abnormalities using a pen light to reveal opacities and to assess pupillary light reflex. Extended observation and handling was used to detect gait abnormalities and/or limb weakness.

4.6.5 Hematology and blood biochemistry

Blood was collected in 200 ul EDTA-coated capillary tubes prior to euthanasia. Samples were analyzed using a Hemavet Hematology Analyzer (950FS). Biochemical analysis was performed by IDEXX Reference Laboratories (Markham, ON) using a Roche Hitachi 917 Chemistry Analyzer.

4.6.6 Urinalysis

Mouse urine was collected from conscious, restrained mice and analyzed using Chemstrip 4MD urinalysis test strips (Roche Diagnostics, Laval, Quebec).

4.6.7 Bone mineral density analysis

Dual energy X-ray absorptiometry was performed using a PIXImus small animal densitometer (Lunar; GE Medical System, WI). Mice were anaesthetised using 5% isoflurane with 700 mL/min oxygen, and placed in prone position on the specimen tray using 2% isoflurane with 700 mL/min oxygen to maintain anaesthesia. Following whole body scanning, bone mineral content (BMC), bone area and bone mineral density (BMD) were measured, with the skull excluded from results.

4.6.8 Faxitron analysis

A high-resolution digital X-ray was taken at a magnification factor of 1.0 at 26 kVp using a Faxitron model MX-20 Specimen Radiography System with a digital camera attachment (Faxitron X-ray Corporation, IL) to determine bone structure. The images were captured on the Specimen Imaging program in the format of Digital Imaging and Communications in Medicine

102

(DICOM) files for analysis purposes. The images were also save as JPEG files for general viewing purposes.

4.6.9 Necropsy and histology

Adult female mice were euthanized at 9-10 weeks of age by CO2. A standard panel of organs and tissues were collected and fixed by immersion (1:10 volume) in 10% neutral buffered formalin for 48 hours before transfer to 70% Ethanol. Tissues were embedded in paraffin and sectioned at 4µm for routine Hematoxylin and Eosin (H&E) staining. Lung tissues were additionally stained with Periodic acid-Schiff (PAS) stain. E18.5 embryos were removed by cesarean section, euthanized by decapitation, and fixed by immersion in buffered formalin for 48 hours. Fetuses were embedded in paraffin, midsagittal and parasagittal sections were made and routinely stained with H&E.

4.6.10 Immunohistochemistry

Tissue sections were deparaffinized, rehydrated, and antigens were retrieved by Pepsin treatment at room temperature for 10 minutes. Endogenous peroxidase activity was quenched by 3% hydrogen peroxide treatment. After blocking, GFP was detected using anti-GFP rabbit IgG (Invitrogen, A11122), Elite ABC Kit (Vectastain, PK-6101) and DAB Peroxidase Substrate Kit (Vectastain, SK-4100) according to manufacturer’s instructions.

4.7 Acknowledgements

I would like to thank Jorge Cabezas and the Toronto Centre for Phenogenomics (TCP) for assistance with mouse husbandry. The author would like to acknowledge the Samuel Lunenfeld Research Institute's CMHD Pathology Core for histology services, and CMHD Mouse Physiology Facility for their technical screening services (www.cmhd.ca).

103

5 Conclusions and future directions 5.1 Summary of thesis research

In my thesis research, I have investigated the role of the membrane-bound O-acyl transferase PORCN in mouse embryonic development. In vitro, I have shown that PORCN is required for canonical WNT signaling in ES cells and further, for their differentiation into endodermal and mesodermal derivatives. Taking advantage of a mouse line carrying a conditional (floxed) Porcn allele that I have generated, I have focused my studies on the early embryonic roles of Porcn using several Cre-mediated and XCI-mediated inactivation strategies in vivo (graphically summarized in Fig. 5-1).

+/tg +/tg tg/+

lox/Y lox/+ +/lox lox/Y Genetic +/tg Strategy Porcn genetrap ES Embryo Porcn Porcn Porcn Porcn Aggregation pCX-NLS-Cre pCX-NLS-Cre pCX-NLS-Cre Ttr::Cre

E4.5

E5.5

Lethal variable not E8.5 E7.5 E11.5 at age (perinatal) lethal maintenance initiation chorio- none Phenotype of of allantoic variable detected gastrulation fusion

lox/- -/- -/- Human Phenocopy Wnt3 Wnt3 Wnt7b wildtype Sox2-Cre+/tg FDH Legend TE/ExE Porcn mutant tissue PE/VE Porcn wildtype tissue Epiblast Porcn mosaic tissue

Figure 5-1: Graphical summary of thesis research

104

5.2 Porcn in pre-implantation development and implantation

Pre-implantation embryonic development is of particular interest because ES cells are derived from pre-implantation blastocysts. ES cells are pluripotent and thus can differentiate into all tissues of the embryo proper. While mouse ES cells giving rise to germ cells in chimeric mice have been exploited to generate targeted mouse mutants, human ES cells and induced pluripotent stem (iPS) cells show great promise as a potential source of tissues for regenerative medicine in humans.

The signals required to maintain cells in a pluripotent state over prolonged periods in culture have long been under investigation. Relevant to this thesis, activation of canonical WNT signaling pathway has been shown to contribute to the maintenance of pluripotency (Berge et al., 2011; Ying et al., 2008). Based on the similarity of ES cells with the ICM, from which they have been derived, it has been concluded that PORCN-mediated WNT signaling is also required for pluripotency in vivo (Berge et al., 2011; Ying et al., 2008). In this thesis, I have investigated this requirement in vivo using maternal and zygotic ablation of Porcn. In contrast to their function in vitro, PORCN-mediated WNT signals are not required for the maintenance of pluripotency in vivo: Maternal zygotic Porcn mutants implant successfully and develop to gastrulation stages, even when their implantation is artificially delayed (Chapter 3.4.8). Consistent with my data, no WNT pathway mutants have been described that result in a pre-implantation phenotype in mice (Table 1-1, 1-2 and (van Amerongen and Berns, 2006)).

Implantation is strictly required for successful development of all mammals and requires signaling between the embryo and receptive uterine tissues. The nature of this signal is currently not clear, but WNT ligands secreted from the embryo were likely candidates, as WNT can induce the appropriate response in uterine tissues and WNT inhibition decreases implantation rates (Mohamed et al., 2005). As no single or compound WNT ligand mutant mouse exhibits an implantation phenotype (Table 1-1), functional redundancy between WNT ligands was proposed to rescue implantation failures in single WNT mutants. Taking advantage of the fact that PORCN is required for the secretion of all WNT ligands, I have investigated whether maternal zygotic Porcn mutants fail to implant. My results show that PORCN-mediated WNT secretion is not required for successful implantation of mouse embryos. While these studies rule out embryonic WNT ligands, they do not rule out that WNTs secreted from the uterus might play an

105 important function in this process. Uterine-specific deletion of Porcn could be used to investigate this question.

5.3 Porcn in post-implantation development and embryo patterning

Once the developing embryo has implanted (E4.5), the anterior-posterior (AP) axis is established and gastrulation commences with the establishment of its morphological landmark: the primitive streak (PS). The PS is the conduit through which epiblast cells move to give rise to anterior/distally moving mesoderm, as well as definitive endoderm cells that displace the visceral endoderm.

The formation of the PS is dependent on the expression of Wnt3 in the posterior region of the embryo, as zygotic Wnt3 mutants fail to establish a PS (Liu et al., 1999). The fact that several WNT ligands are expressed on the posterior side of the embryo (Kemp et al., 2005), suggested that a compound WNT ligand phenotype could be more severe than the Wnt3 mutant phenotype. I have investigated this question using zygotic (and maternal zygotic) deletion of Porcn to ablate the secretion of all WNT ligands. Based on the expression of numerous marker genes, I was unable to detect any differences between Wnt3 and Porcn mutant phenotypes. My results thus show that other WNT ligands have no additional or compensating functions for the initial patterning of the embryo. Consistent with my results, ablation of WNT secretion chaperone Wls (Fu et al., 2009b), canonical WNT co-receptors Lrp5; Lrp6 (Kelly et al., 2004), or their chaperone Mesd (Hsieh et al., 2003), result in identical phenotypes. While Ctnnb1 mutants also fail to establish the PS, they further display a defect in the establishment of the anterior visceral endoderm (AVE) (Huelsken et al., 2000). As Porcn mutants establish the AVE successfully, my data supports the notion that this AVE defect represents a WNT-independent function of Beta- Catenin (Morkel et al., 2003).

Gastrulation-inducing WNT3 is expressed as early as E5.75 in the posterior visceral endoderm (PVE) and subsequently in the underlying epiblast (Rivera-Pérez and Magnuson, 2005). The initial analysis of epiblast-specific Wnt3 mutants suggested that WNT3 expression in the PVE is not sufficient to induce gastrulation (Barrow et al., 2007). However, a recent analysis of the same mutants showed that WNT3 expressed in the PVE can induce gastrulation albeit slightly delayed (Tortelote et al., 2012). In the absence of epiblast secreted WNT3 however, gastrulation cannot

106 be completed successfully and the mutant embryos die by E9.5 (Tortelote et al., 2012). Consistent with these data, I have been able to show that epiblast-specific Porcn mutants initiate gastrulation, but fail to complete it (Biechele et al., 2011). Together, these data show that PORCN-mediated WNT3 secretion from the PVE is sufficient to induce gastrulation and suggest that the PVE is a PS-inducing signaling center.

In order to confirm this hypothesis, I investigated whether PORCN-mediated WNT secretion from the PVE is also necessary for the induction of gastrulation. Ablation of PORCN function in the VE by imprinted XCI in heterozygous females, as well as VE-specific deletion by Ttr-Cre (Kwon and Hadjantonakis, 2009), had no effect on gastrulation. My data thus suggest that PORCN-mediated WNT secretion from the PVE is sufficient but not necessary for the induction of gastrulation. The data are consistent with a model in which the PVE acts as a secondary/back- up inducer of gastrulation. In this model, PORCN-mediated secretion of WNT3 from the PVE is not normally required to induce gastrulation in wildtype epiblast. However, if gastrulation is genetically inhibited in the epiblast (i.e. epiblast-specific Wnt3 or Porcn mutants), WNT3 is secreted from the PVE and is able to induce, but not maintain, gastrulation (Biechele et al., 2011; Tortelote et al., 2012). Alternatively, WNT3 secretion from the PVE might not depend on PORCN. Such a tissue-specific, alternative secretion could be investigated in embryos with Porcn mutant VE and Wnt3 mutant epiblast. If WNT3 were secreted from the PVE in the absence of PORCN, these mutant embryos would induce gastrulation similar to epiblast-specific Wnt3 mutants. If WNT3 is also dependent on PORCN in the PVE, these embryos will phenocopy zygotic Wnt3 mutants.

Based on the analysis of embryos that are functionally extra-embryonic mutant for Porcn, I was able to identify the earliest extra-embryonic requirement for PORCN-mediated WNT secretion. Consistent with PORCN being required for the function of chorionic WNT7B (Najdi et al., 2012), extra-embryonic Porcn mutants phenocopy the chorio-allantoic fusion defect observed in Wnt7b mutants (Parr et al., 2001). In contrast, a second mutant Wnt7b allele fails to cause a chorio-allantoic fusion defect and leads to perinatal lethality due to defects in lung vascular development (Shu et al., 2002). Despite a truncated transcript being produced from this allele, it has been suggested to be functionally null (Shu et al., 2002). As extra-embryonic Porcn mutants phenocopy the chorio-allantoic fusion defect, my data suggests that the truncated transcript produced from this second Wnt7b allele is hypomorphic, but not completely non-functional.

107

5.4 WNT redundancy in development

Numerous processes in the development and life of mammals are dependent on WNT signaling (Clevers and Nusse, 2012; Wang et al., 2012). In the case of canonical WNT signaling, these processes have mostly been investigated by genetic ablation or stabilization of the key effector and transcriptional activator Beta-Catenin (Aoki and Taketo, 2008). The identity and source of the WNT ligand causing a specific effect however, is usually not known. Further, whether this effect is caused by a single WNT ligand from one source, multiple ligands from the same source, or multiple ligands from different sources is extremely difficult to elucidate. For a better understanding of both the WNT pathway at the molecular level, and WNT activities on the level of the organism, it will be important to develop strategies and tools to investigate these questions. Ablation of Porcn by tissue-specific deletion of a floxed allele is a new tool for the investigation of WNT signaling in mice.

While numerous compound WNT ligand phenotypes are more severe than the respective single WNT mutants (Table 1-2), the phenotypes I have observed in zygotic and extra-embryonic Porcn mutant embryos phenocopy two single WNT ligand mutants (Wnt3 and Wnt7b respectively) and are not more severe, as would be expected if there were functional redundancy. This suggests that other WNT ligands co-secreted from WNT3 and WNT7B secreting cells respectively are not required and have no additional functions. These observations question the extent of functional redundancy between WNT ligands.

One approach to investigate WNT redundancy taking advantage of this tool would be to delete Porcn using Cre recombinase under the control of WNT ligand promoter (e.g. Wnt1-Cre). A phenotype more severe than the single WNT mutant (e.g. Wnt1) would reveal (A) additional functions for other WNTs secreted from these cells. An identical phenotype would confirm that (B) all other WNTs secreted from these cells are not required, whereas a less severe phenotype would indicate that (C) other WNTs secreted from these cells have opposing functions or (D) there are alternative mechanisms of secretion for the Cre-driving WNT ligand gene from these cells. While my results are consistent with in vitro studies showing that PORCN is required for WNT secretion (Najdi et al., 2012; Takada et al., 2006), it remains an open question whether this the case in all cell types.

108

5.5 PORCN functions at the cellular level

Porcn is a member of the MBOAT superfamily (Hofmann, 2000). Consistent with the early fly epistasis experiments placing PORCN in the WNT signaling pathway (Siegfried et al., 1994; van den Heuvel et al., 1993), it was later shown that PORCN is required for the acylation of a WNT ligand in mammals (Takada et al., 2006). The long standing assumption that all mammalian WNTs require PORCN-mediated acylation, was only recently confirmed (Najdi et al., 2012).

In my studies, I was able to show that PORCN is required in WNT3A producing ES cells in order to activate the canonical WNT signaling pathway. I have further shown that all 4 isoforms produced by alternative splicing from the Porcn locus are sufficient to rescue this defect. Consistent with my results, recent over-expression studies have shown that all 10 WNT ligands tested can be acylated by all 4 PORCN isoforms, but show global differences in efficiency between the PORCN isoforms (Proffitt and Virshup, 2012). Despite this significant functional overlap, individual PORCN isoforms show different expression patterns (Caricasole et al., 2002) and different effects in differentiating embryocarcinoma cells (Tanaka et al., 2003). It remains elusive how cells use the regulated expression of Porcn isoforms to modulate their WNT signaling activity. Ectopic expression of individual isoforms in Porcn mutant cells are a promising avenue to elucidate isoform-specific effects.

While the effect of PORCN on WNT ligands has been fairly well described in recent years, potential WNT-unrelated functions of PORCN are still poorly defined and limited to in vitro studies. In my studies, I have shown that PORCN ablation has no effect on NODAL signal secretion or activity. Similar studies have also excluded the Notch and Hedgehog signaling pathways (Chen et al., 2009). One recent study however, has found that PORCN knock-down in human cancer cells caused a WNT-independent proliferation defect that was further not dependent on PORCN catalytic activity (Covey et al., 2012). These studies highlight the need for unbiased screens for PORCN targets and interactors. The availability of targeted genome editing technology (e.g. zinc-finger nucleases) allows the targeted ablation of PORCN in cultured cell lines (Proffitt and Virshup, 2012). Subsequent proteome analyses by mass spectrometry could be used to identify PORCN targets. Further, in the absence of a functional PORCN-specific antibody, the generation of a mouse line carrying a tagged allele in the endogenous locus would allow for the identification of interacting proteins. These studies might not only identify WNT-

109 unrelated functions but also help elucidate additional molecular details of WNT acylation and secretion.

5.6 Implications for human disease

5.6.1 Focal Dermal Hypoplasia

In 2005, two graduate students of the Rossant laboratory (Dr. Owen Tamplin and Dr. Brian Cox) came up with the idea for a screen that would reveal X-linked mutants that caused male embryonic lethality in the mouse and potentially humans as well, based on 1-to-1 homology and X-chromosomal linkage. I got involved in the execution of the screen and one of the genes we identified was Porcn (Cox et al., 2010). While the screen was in progress, two groups reported that mutations in human PORCN were the cause for FDH (OMIM#305600) (Grzeschik et al., 2007; Wang et al., 2007), a syndrome that had first been described in 1962 and was also known as Goltz Syndrome (Goltz et al., 1962). Consistent with our observations of lethality in mutant male mouse embryos, human FDH was mainly observed in females and all males described showed postzygotic mutations. While the absence of zygotic PORCN mutant males in humans is highly suggestive of embryonic/fetal lethality, evidence in form of PORCN mutant aborted fetuses is lacking. Further, the time-point or cause of the male embryonic lethality was unknown.

Using the mouse as a model for inaccessible stages of human embryonic development, my studies of the mouse Porcn gene show that zygotic and epiblast specific male mutant embryos die at gastrulation stages (E7.5) (Chapters 2.4.4 and 3.4.2). Based on the conserved role of PORCN in WNT signaling (Najdi et al., 2012) and WNT signaling in inducing gastrulation (Davidson et al., 2012), I suggest that this is also the case in humans. In human embryonic development, hemizygous PORCN mutants would die early in the third week of pregnancy, which would result in an abortion approximately 4-5 weeks after the last menses and thus not necessarily be recognized as a miscarriage.

In contrast to the male embryonic lethality, familial cases have been described in which females have inherited a mutant PORCN allele from mother or father (Grzeschik et al., 2007; Wang et al., 2007). These heterozygous females are the basis for the identification of the human syndrome FDH. Due to XCI, the phenotypes are highly variable and range from patches of skin defects to severe morphological deformities and mental retardation (Bornholdt et al., 2009b).

110

In an effort to generate a mouse model for the human disease, I have generated zygotic Porcn heterozygous female mice. In contrast to humans (Grzeschik et al., 2007), maternal allele mutant mouse embryos die at E11.5 due to a failure in chorio-allantoic fusion (Chapter 3.4.3), which requires WNT7B secretion from the extra-embryonic chorion (Parr et al., 2001). Due to imprinted XCI in these heterozygous females, the paternal X chromosome carrying the wildtype Porcn allele is inactivated in the extra-embryonic tissue (Barakat and Gribnau, 2012), which are thus functionally mutant. Consistent with this extra-embryonic role for Porcn in mice, epiblast- specific deletion of the maternal allele in females does not cause chorio-allantoic fusion defects, but variable fetal defects and perinatal lethality (Barrott et al., 2011). The observed discrepancy between zygotic human and mouse phenotypes could indicate that PORCN-mediated WNT signaling is not required for chorio-allantoic fusion in humans and/or imprinted XCI is not as stringent in humans as in mice. Supporting the second explanation, imprinted XCI has been reported to be less stringent in humans (Zeng and Yankowitz, 2003).

Circumventing the extra-embryonic requirement for Porcn in mice, I generated zygotic paternal allele mutant heterozygous females (Chapter 4). In contrast to maternal allele mutants, these females survived to term at normal Mendelian ratios. Similar to epiblast-specific maternal allele mutants (Barrott et al., 2011), these fetuses had a highly variable phenotype due to individual patterns of XCI. The majority (95%) of these females died perinatally, most likely due to body wall closure defects, diaphragmatic hernias and kidney defects.

Human FDH is considered a rare syndrome, as the reported prevalence for FDH is <1/1,000,000 based on the number of observed live-births (Germain, 2006). Data on pre-natal FDH however is lacking. The data from my mouse model suggests that, similar to Porcn mutant male embryos, a large proportion (~95%) of heterozygous female fetuses die in utero or perinatally. If this extent of lethality is replicated in humans, the actual prevalence of female fetal FDH could be 20 times higher at 1/50,000 pregnancies. Further, when pregnancies with PORCN mutant male embryos are included, the overall prevalence of embryonic/fetal FDH could be up to 1/25,000 pregnancies. Supporting this estimate, the same prevalence has been observed in X-linked Charcot-Marie-Tooth disease (CMTX1, OMIM#302800, (Ma et al., 2002)), which is also a X- linked dominant disorder, but is not associated with pre-natal lethality that would mask the prevalence. Based on the mouse data presented here and by others (Barrott et al., 2011; Liu et al., 2012), approximately 98% of pregnancies with embryonic PORCN mutations would result in

111 lethality, as all mutant male embryos die during gastrulation and 95% of the female fetuses die during fetal development or perinatally. Consistent with these findings, fetal FDH-related perinatal lethality has also been reported in humans (Dias et al., 2010) and some studies suggest that only mildly affected patients survive beyond birth (Bornholdt et al., 2009a). It is thus possible that FDH is actually not a rare syndrome, but a disorder that affects 1/25.000 pregnancies with largely embryonic/fetal lethal outcome and has thus a bigger impact on human pregnancies and maternal health than previously appreciated. Targeted sequence analysis of aborted fetal tissues could be used to confirm this hypothesis.

In contrast to other studies (Barrott et al., 2011; Liu et al., 2012), several females survived to adulthood and thus represent the first adult mouse model of zygotic FDH. Using extensive phenotyping protocols, I was able to detect several features observed in human FDH, such as skeletal and skin defects. In addition, I have discovered that Porcn heterozygous females exhibit several characteristic features that have been linked to ciliopathies (Vogel et al., 2010; Ware et al., 2011), such as severe kidney defects (perinatal lethal), mild hydrocephalus and mucus transport defects resulting in rhinitis, otitis and bronchopneumonia. These observations lead me to speculate that PORCN might be involved in ciliary assembly or function. While it is possible that such an effect would be mediated through WNT signaling, it could also represent a WNT- unrelated function. Heterozygous, and thus functionally mosaic, females are not the ideal model to investigate this defect. Tissue-specific ablation in tissues with known cilia function, such as the lung epithelium (using Nkx2.1-Cre), would allow for a more detailed investigation of this defect. Combination of such in vivo approaches with in vitro rescue experiments using catalytically inactive PORCN and supplementing WNT ligands could elucidate whether this defect is WNT ligand mediated. Further, ZFN-mediated PORCN ablation in ciliated human male cell lines could be used to confirm whether this link between PORCN and cilia function is conserved and relevant in humans and human disease.

5.6.2 PORCN as a therapeutic target

The importance of PORCN on human health extends beyond fetal and adult FDH. Numerous aspects of tissue homeostasis are controlled by WNT signaling and ectopic activation of WNT signaling has been shown to be involved in cancer (Clevers and Nusse, 2012), as highlighted by the discovery of the first mammalian WNT ligand based on its activation due to the integration

112 of a mammary tumor virus (MMTV) (Nusse and Varmus, 1982). As PORCN is an enzyme required for the secretion and function of all WNT ligands, it is an attractive therapeutic target for all WNT-stimulated diseases. The first chemical inhibitor of PORCN function was published in 2009 (Chen et al., 2009), and two years later multiple chemical scaffolds with inhibitor activity but better bioavailability have been published (Dodge et al., 2012). The major drawback for PORCN as a therapeutic target were potential side effects, as genetic studies had established that WNT signaling is critical for the constant self-renewal of the intestinal epithelium (Pinto et al., 2003). However, recent studies in a mouse breast cancer model show that PORCN inhibition is effective in preventing MMTV-Wnt1 tumor growth without affecting the intestinal epithelium (Proffitt et al., 2012). Consistent with these encouraging results, the Novartis PORCN inhibitor LGK974 is currently in phase I clinical trials (NCT01351103) (Lum and Clevers, 2012). Complementing small molecule-mediated inhibition, tissue-specific genetic ablation of Porcn in mouse models will contribute to elucidate mechanisms of PORCN function and to improve therapeutic strategies to treat human disease.

5.7 The future of Porcn research

Although porc/Porcn was first identified as a WNT pathway component in the fly over two decades ago (Perrimon et al., 1989), it has largely been ignored in studies of mammalian WNT signaling, as most of these studies focused on events in the WNT signal receiving cells. Only recently has the specialized secretion of WNT ligands gained more attention (Herr et al., 2012). Hampered by its X-chromosomal localization in the genome, the Porcn gene has been particularly difficult to investigate in the mammalian system. My studies have been focused on the roles of Porcn in early mouse embryonic development, and while I have been able to define the earliest requirements for Porcn in embryonic and extra-embryonic tissues, many questions remain unanswered.

The focus of future research on PORCN should be in defining its function at the molecular level. Unbiased mass-spectrometry-based approaches could identify novel molecules interacting with PORCN. These studies might uncover and elucidate potential WNT-independent targets and functions that require PORCN, but will also further the more detailed understanding of the specialized secretion pathway of WNT ligands. Tissue-specific deletions of Porcn, both in the developing embryo and adult animals, could complement these studies and could define in vivo

113 sources of WNT ligands during development and in adult tissue homeostasis. Ultimately, these studies can lead to a more complete view of signaling events in vivo: starting from the biogenesis and secretion of the signal, through its interaction with receptors, to downstream effects in the signal receiving cells.

114

References

Agalliu, D., Takada, S., Agalliu, I., McMahon, A. P. and Jessell, T. M. (2009). Motor neurons with axial muscle projections specified by Wnt4/5 signaling. Neuron 61, 708–720.

Alkindi, S., Battin, M., Aftimos, S. and Purvis, D. (2012). Focal Dermal Hypoplasia Due to a Novel Mutation in a Boy with Klinefelter Syndrome. Pediatric Dermatology.

Andre, P., Wang, Q., Wang, N., Gao, B., Schilit, A., Halford, M. M., Stacker, S. A., Zhang, X. and Yang, Y. (2012). The Wnt coreceptor Ryk regulates Wnt/planar cell polarity by modulating the degradation of the core planar cell polarity component Vangl2. J Biol Chem 287, 44518–44525.

Ang, S. L., Conlon, R. A., Jin, O. and Rossant, J. (1994). Positive and negative signals from mesoderm regulate the expression of mouse Otx2 in ectoderm explants. Development (Cambridge, England) 120, 2979–2989.

Angers, S. and Moon, R. T. (2009). Proximal events in Wnt signal transduction. Nat Rev Mol Cell Biol 10, 468–477.

Aoki, K. and Taketo, M. M. (2008). Tissue-specific transgenic, conditional knockout and knock-in mice of genes in the canonical Wnt signaling pathway. Methods Mol Biol 468, 307–331.

Arnold, S. J., Stappert, J., Bauer, A., Kispert, A., Herrmann, B. G. and Kemler, R. (2000). Brachyury is a target gene of the Wnt/beta-catenin signaling pathway. Mech Dev 91, 249– 258.

Barakat, T. S. and Gribnau, J. (2012). X chromosome inactivation in the cycle of life. Development (Cambridge, England) 139, 2085–2089.

Barrott, J. J., Cash, G. M., Smith, A. P., Barrow, J. R. and Murtaugh, L. C. (2011). Deletion of mouse Porcn blocks Wnt ligand secretion and reveals an ectodermal etiology of human focal dermal hypoplasia/Goltz syndrome. Proceedings of the National Academy of Sciences 108, 12752–12757.

Barrow, J. R., Howell, Rule, Hayashi, S., Thomas, Capecchi, M. R. and McMahon, A. P. (2007). Wnt3 signaling in the epiblast is required for proper orientation of the anteroposterior axis. Dev Biol.

Barrow, J. R., Thomas, K. R., Boussadia-Zahui, O., Moore, R., Kemler, R., Capecchi, M. R. and McMahon, A. P. (2003). Ectodermal Wnt3/beta-catenin signaling is required for the establishment and maintenance of the apical ectodermal ridge. Genes Dev 17, 394–409.

Belenkaya, T. Y., Wu, Y., Tang, X., Zhou, B., Cheng, L., Sharma, Y. V., Yan, D., Selva, E. M. and Lin, X. (2008). The retromer complex influences wnt secretion by recycling wntless from endosomes to the trans-Golgi network. Dev Cell 14, 120–131.

115

Belteki, G., Haigh, J., Kabacs, N., Haigh, K., Sison, K., Costantini, F., Whitsett, J., Quaggin, S. E. and Nagy, A. (2005). Conditional and inducible transgene expression in mice through the combinatorial use of Cre-mediated recombination and tetracycline induction. Nucleic Acids Res 33, e51.

Ben-Haim, N., Lu, C., Guzman-Ayala, M., Pescatore, L., Mesnard, D., Bischofberger, M., Naef, F., Robertson, E. J. and Constam, D. B. (2006). The nodal precursor acting via activin receptors induces mesoderm by maintaining a source of its convertases and BMP4. Dev Cell 11, 313–323.

Berge, Ten, D., Kurek, D., Blauwkamp, T., Koole, W., Maas, A., Eroglu, E., Siu, R. K. and Nusse, R. (2011). Embryonic stem cells require Wnt proteins to prevent differentiation to epiblast stem cells. Nature 13, 1070–1075.

Bhanot, P., Brink, M., Samos, C. H., Hsieh, J. C., Wang, Y., Macke, J. P., Andrew, D., Nathans, J. and Nusse, R. (1996). A new member of the frizzled family from Drosophila functions as a Wingless receptor. Nature 382, 225–230.

Biechele, S., Cox, B. J. and Rossant, J. (2011). Porcupine homolog is required for canonical Wnt signaling and gastrulation in mouse embryos. Dev Biol 355, 275–285.

Boothroyd, A. E. and Hall, C. M. (1987). The radiological features of Goltz syndrome: focal dermal hypoplasia. A report of two cases. Skeletal Radiol 17, 505–508.

Bornholdt, D., Oeffner, F., König, A., Happle, R., Alanay, Y., Ascherman, J., Benke, P. J., del Carmen Boente, M., van der Burgt, I., Chassaing, N., et al. (2009a). PORCN mutations in focal dermal hypoplasia: coping with lethality. Hum Mutat 30, E618–28.

Bornholdt, D., Oeffner, F., König, A., Happle, R., Alanay, Y., Ascherman, J., Benke, P., Del Carmen Boente, M., van der Burgt, I., Chassaing, N., et al. (2009b). PORCN mutations in focal dermal hypoplasia: coping with lethality. Hum Mutat.

Bornholdt, D., Oeffner, F., König, A., Happle, R., Alanay, Y., Ascherman, J., Benke, P., Del Carmen Boente, M., van der Burgt, I., Chassaing, N., et al. (2009c). PORCN mutations in focal dermal hypoplasia: coping with lethality. Hum Mutat 30, 1472–1473.

Brown, A. M., Wildin, R. S., Prendergast, T. J. and Varmus, H. E. (1986). A retrovirus vector expressing the putative mammary oncogene int-1 causes partial transformation of a mammary epithelial cell line. Cell 46, 1001–1009.

Buechling, T., Chaudhary, V., Spirohn, K., Weiss, M. and Boutros, M. (2011). p24 proteins are required for secretion of Wnt ligands. EMBO Rep 12, 1265–1272.

Burrus, L. W. and McMahon, A. P. (1995). Biochemical analysis of murine Wnt proteins reveals both shared and distinct properties. Exp Cell Res 220, 363–373.

Caricasole, A., Ferraro, T., Rimland, J. M. and Terstappen, G. C. (2002). Molecular cloning and initial characterization of the MG61/PORC gene, the human homologue of the Drosophila segment polarity gene Porcupine. Gene 288, 147–157.

116

Carmon, K. S., Gong, X., Lin, Q., Thomas, A. and Liu, Q. (2011). R-spondins function as ligands of the orphan receptors LGR4 and LGR5 to regulate Wnt/beta-catenin signaling. Proceedings of the National Academy of Sciences 108, 11452–11457.

Caron, A., Xu, X. and Lin, X. (2012). Wnt/β-catenin signaling directly regulates Foxj1 expression and ciliogenesis in zebrafish Kupffer's vesicle. Development (Cambridge, England) 139, 514–524.

Carroll, T. J., Park, J.-S., Hayashi, S., Majumdar, A. and McMahon, A. P. (2005). Wnt9b plays a central role in the regulation of mesenchymal to epithelial transitions underlying organogenesis of the mammalian urogenital system. Dev Cell 9, 283–292.

Chazaud, C. and Rossant, J. (2006). Disruption of early proximodistal patterning and AVE formation in Apc mutants. Development (Cambridge, England) 133, 3379–3387.

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, D., Jarrell, A., Guo, C., Lang, R. and Atit, R. P. (2012a). Dermal β-catenin activity in response to epidermal Wnt ligands is required for fibroblast proliferation and hair follicle initiation. Development (Cambridge, England) 139, 1522–1533.

Chen, Q., Takada, R. and Takada, S. (2012b). Loss of Porcupine impairs convergent extension during gastrulation in zebrafish. J Cell Sci 125, 2224–2234.

Chen, W.-S., Antic, D., Matis, M., Logan, C. Y., Povelones, M., Anderson, G. A., Nusse, R. and Axelrod, J. D. (2008). Asymmetric homotypic interactions of the atypical cadherin flamingo mediate intercellular polarity signaling. Cell 133, 1093–1105.

Ching, W., Hang, H. C. and Nusse, R. (2008). Lipid-independent secretion of a Drosophila Wnt protein. J Biol Chem 283, 17092–17098.

Ciruna, B. and Rossant, J. (2001). FGF signaling regulates mesoderm cell fate specification and morphogenetic movement at the primitive streak. Dev Cell 1, 37–49.

Clevers, H. C. and Nusse, R. (2012). Wnt/β-Catenin Signaling and Disease. Cell 149, 1192– 1205.

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. Development (Cambridge, England) 139, 1931–1940.

Coombs, G. S., Yu, J., Canning, C. A., Veltri, C. A., Covey, T. M., Cheong, J. K., Utomo, V., Banerjee, N., Zhang, Z. H., Jadulco, R. C., et al. (2010). WLS-dependent secretion of WNT3A requires Ser209 acylation and vacuolar acidification. J Cell Sci 123, 3357–3367.

Covey, T. M., Kaur, S., Tan Ong, T., Proffitt, K. D., Wu, Y., Tan, P. and Virshup, D. M. (2012). PORCN moonlights in a Wnt-independent pathway that regulates cancer cell

117

proliferation. PLoS ONE 7, e34532.

Cox, B. J., Vollmer, M., Tamplin, O. J., Lu, M., Biechele, S., Gertsenstein, M., van Campenhout, C., Floss, T., Kühn, R., Wurst, W., et al. (2010). Phenotypic annotation of the mouse X chromosome. Genome Res 20, 1154–1164.

Daane, J. M., Enders, A. C. and Downs, K. M. (2011). Mesothelium of the murine allantois exhibits distinct regional properties. J. Morphol. 272, 536–556.

Dabdoub, A., Donohue, M. J., Brennan, A., Wolf, V., Montcouquiol, M., Sassoon, D. A., Hseih, J.-C., Rubin, J. S., Salinas, P. C. and Kelley, M. W. (2003). Wnt signaling mediates reorientation of outer hair cell stereociliary bundles in the mammalian cochlea. Development (Cambridge, England) 130, 2375–2384.

Davidson, K. C., Adams, A. M., Goodson, J. M., McDonald, C. E., Potter, J. C., Berndt, J. D., Biechele, T., Taylor, R. J. and Moon, R. T. (2012). Wnt/β-catenin signaling promotes differentiation, not self-renewal, of human embryonic stem cells and is repressed by Oct4. Proceedings of the National Academy of Sciences. de Lau, W., Barker, N., Low, T. Y., Koo, B.-K., Li, V. S. W., Teunissen, H., Kujala, P., Haegebarth, A., Peters, P. J., van de Wetering, M., et al. (2011). Lgr5 homologues associate with Wnt receptors and mediate R-spondin signalling. Nature 476, 293–297.

De Vries, W. N., Binns, L. T., Fancher, K. S., Dean, J., Moore, R., Kemler, R. and Knowles, B. B. (2000). Expression of Cre recombinase in mouse oocytes: a means to study maternal effect genes. Genesis 26, 110–112.

De Vries, W. N., Evsikov, A. V., Haac, B. E., Fancher, K. S., Holbrook, A. E., Kemler, R., Solter, D. and Knowles, B. B. (2004). Maternal beta-catenin and E-cadherin in mouse development. Development (Cambridge, England) 131, 4435–4445.

Dias, C., Basto, J., Pinho, O., Barbêdo, C., Mártins, M., Bornholdt, D., Fortuna, A., Grzeschik, K.-H. and Lima, M. (2010). A nonsense porcn mutation in severe focal dermal hypoplasia with natal teeth. Fetal Pediatr Pathol 29, 305–313.

Diez-Roux, G., Banfi, S., Sultan, M., Geffers, L., Anand, S., Rozado, D., Magen, A., Canidio, E., Pagani, M., Peluso, I., et al. (2011). A high-resolution anatomical atlas of the transcriptome in the mouse embryo. PLoS Biol 9, e1000582.

Dodge, M. E., Moon, J., Tuladhar, R., Lu, J., Jacob, L. S., Zhang, L.-S., Shi, H., Wang, X., Moro, E., Mongera, A., et al. (2012). Diverse chemical scaffolds support direct inhibition of the membrane-bound O-acyltransferase porcupine. J Biol Chem 287, 23246–23254.

Doubravska, L., Krausova, M., Gradl, D., Vojtechova, M., Tumova, L., Lukas, J., Valenta, T., Pospichalova, V., Fafilek, B., Plachy, J., et al. (2011). Fatty acid modification of Wnt1 and Wnt3a at serine is prerequisite for lipidation at cysteine and is essential for Wnt signalling. Cell Signal 23, 837–848.

Feiguin, F., Hannus, M., Mlodzik, M. and Eaton, S. (2001). The ankyrin repeat protein Diego

118

mediates Frizzled-dependent planar polarization. Dev Cell 1, 93–101.

Fotaki, V., Larralde, O., Zeng, S., McLaughlin, D., Nichols, J., Price, D. J., Theil, T. and Mason, J. O. (2010). Loss of Wnt8b has no overt effect on hippocampus development but leads to altered Wnt gene expression levels in dorsomedial telencephalon. Dev Dyn 239, 284–296.

Franch-Marro, X., Wendler, F., Griffith, J., Maurice, M. M. and Vincent, J.-P. (2008). In vivo role of lipid adducts on Wingless. J Cell Sci 121, 1587–1592.

Fu, J. and Hsu, W. (2012). Epidermal Wnt Controls Hair Follicle Induction by Orchestrating Dynamic Signaling Crosstalk between the Epidermis and Dermis. J Investig Dermatol.

Fu, J., Ivy Yu, H.-M., Maruyama, T., Mirando, A. J. and Hsu, W. (2011). Gpr177/mouse Wntless is essential for Wnt-mediated craniofacial and brain development. Dev Dyn 240, 365–371.

Fu, J., Jiang, M., Mirando, A. J., Ivy Yu, H.-M. and Hsu, W. (2009a). Reciprocal regulation of Wnt and Gpr177/mouse Wntless is required for embryonic axis formation. Proc Natl Acad Sci USA 106, 18598–18603.

Fu, J., Jiang, M., Mirando, A. J., Yu, H.-M. I. and Hsu, W. (2009b). Reciprocal regulation of Wnt and Gpr177/mouse Wntless is required for embryonic axis formation. Proceedings of the National Academy of Sciences 106, 18598–18603.

Gadue, P., Huber, T. L., Nostro, M. C., Kattman, S. J. and Keller, G. M. (2005). Germ layer induction from embryonic stem cells. Exp Hematol 33, 955–964.

Gadue, P., Huber, T. L., Paddison, P. J. and Keller, G. M. (2006). Wnt and TGF-beta signaling are required for the induction of an in vitro model of primitive streak formation using embryonic stem cells. Proc Natl Acad Sci USA 103, 16806–16811.

Galli, L. M. and Burrus, L. W. (2011). Differential Palmit(e)oylation of Wnt1 on C93 and S224 Residues Has Overlapping and Distinct Consequences. PLoS ONE 6, e26636.

Galli, L. M., Barnes, T. L., Secrest, S. S., Kadowaki, T. and Burrus, L. W. (2007). Porcupine-mediated lipid-modification regulates the activity and distribution of Wnt proteins in the chick neural tube. Development (Cambridge, England) 134, 3339–3348.

Gao, B., Song, H., Bishop, K., Elliot, G., Garrett, L., English, M. A., Andre, P., Robinson, J., Sood, R., Minami, Y., et al. (2011). Wnt Signaling Gradients Establish Planar Cell Polarity by Inducing Vangl2 Phosphorylation through Ror2. Dev Cell 20, 163–176.

George, S. H. L., Gertsenstein, M., Vintersten, K., Korets-Smith, E., Murphy, J., Stevens, M. E., Haigh, J. J. and Nagy, A. (2007). Developmental and adult phenotyping directly from mutant embryonic stem cells. Proc Natl Acad Sci USA 104, 4455–4460.

Germain, D. (2006). www.orpha.net: Focal Dermal Hypoplasia. http://www.orpha.net/consor/cgi-

119

bin/Disease_Search.php?lng=EN&data_id=2004&Disease_Disease_Search_diseaseGroup =Porcn&Disease_Disease_Search_diseaseType=Gen&disease%2528s%2529/group%20of %20diseases=Focal-dermal-hypoplasia&title=Focal-dermal- hypoplasia&search=Disease_Search_Simple.

Glinka, A., Dolde, C., Kirsch, N., Huang, Y.-L., Kazanskaya, O., Ingelfinger, D., Boutros, M., Cruciat, C.-M. and Niehrs, C. (2011). LGR4 and LGR5 are R-spondin receptors mediating Wnt/β-catenin and Wnt/PCP signalling. EMBO Rep 12, 1055–1061.

Goltz, R. W. (1992). Focal dermal hypoplasia syndrome. An update. Arch Dermatol 128, 1108– 1111.

Goltz, R. W., PETERSON, W. C., GORLIN, R. J. and RAVITS, H. G. (1962). Focal dermal hypoplasia. Arch Dermatol 86, 708–717.

Gonçalves, L., Filipe, M., Marques, S., Salgueiro, A. M., Becker, J. D. and Belo, J. A. (2011). Identification and functional analysis of novel genes expressed in the Anterior Visceral Endoderm. Int J Dev Biol 55, 281–295.

Goss, A. M., Tian, Y., Cheng, L., Yang, J., Zhou, D., Cohen, E. D. and Morrisey, E. E. (2011). Wnt2 signaling is necessary and sufficient to activate the airway smooth muscle program in the lung by regulating myocardin/Mrtf-B and Fgf10 expression. Dev Biol 356, 541–552.

Goss, A. M., Tian, Y., Tsukiyama, T., Cohen, E. D., Zhou, D., Lu, M. M., Yamaguchi, T. P. and Morrisey, E. E. (2009). Wnt2/2b and beta-catenin signaling are necessary and sufficient to specify lung progenitors in the foregut. Dev Cell 17, 290–298.

Gross, J. C., Chaudhary, V., Bartscherer, K. and Boutros, M. (2012). Active Wnt proteins are secreted on exosomes. Nature 14, 1036–1045.

Grzeschik, K.-H., Bornholdt, D., Oeffner, F., König, A., del Carmen Boente, M., Enders, H., Fritz, B., Hertl, M., Grasshoff, U., Höfling, K., et al. (2007). Deficiency of PORCN, a regulator of Wnt signaling, is associated with focal dermal hypoplasia. Nat Genet 39, 833– 835.

Guo, N., Hawkins, C. and Nathans, J. (2004). Frizzled6 controls hair patterning in mice. Proc Natl Acad Sci USA 101, 9277–9281.

Hadjantonakis, A.-K. K., Cox, L. L., Tam, P. P. L. and Nagy, A. (2001). An X-linked GFP transgene reveals unexpected paternal X-chromosome activity in trophoblastic giant cells of the mouse placenta. Genesis 29, 133–140.

Hadjantonakis, A.-K. K., Gertsenstein, M., Ikawa, M., Okabe, M. and Nagy, A. (1998a). Non-invasive sexing of preimplantation stage mammalian embryos. Nat Genet 19, 220–222.

Hadjantonakis, A.-K. K., Gertsenstein, M., Ikawa, M., Okabe, M. and Nagy, A. (1998b). Generating green fluorescent mice by germline transmission of green fluorescent ES cells. Mech Dev 76, 79–90.

120

Hadjantonakis, A.-K. K., Macmaster, S. and Nagy, A. (2002). Embryonic stem cells and mice expressing different GFP variants for multiple non-invasive reporter usage within a single animal. BMC Biotechnol 2, 11.

Hao, H.-X., Xie, Y., Zhang, Y., Charlat, O., Oster, E., Avello, M., Lei, H., Mickanin, C., Liu, D., Ruffner, H., et al. (2012). ZNRF3 promotes Wnt receptor turnover in an R- spondin-sensitive manner. Nature 485, 195–200.

Harada, N., Tamai, Y., Ishikawa, T., Sauer, B., Takaku, K., Oshima, M. and Taketo, M. M. (1999). Intestinal polyposis in mice with a dominant stable mutation of the beta-catenin gene. EMBO J 18, 5931–5942.

Heisenberg, C. P., Tada, M., Rauch, G. J., Saúde, L., Concha, M. L., Geisler, R., Stemple, D. L., Smith, J. C. and Wilson, S. W. (2000). Silberblick/Wnt11 mediates convergent extension movements during zebrafish gastrulation. Nature 405, 76–81.

Herr, P. and Basler, K. (2012). Porcupine-mediated lipidation is required for Wnt recognition by Wls. Dev Biol 361, 392–402.

Herr, P., Hausmann, G. and Basler, K. (2012). WNT secretion and signalling in human disease. Trends in molecular medicine.

Herrmann, B. G. (1991). Expression pattern of the Brachyury gene in whole-mount TWis/TWis mutant embryos. Development (Cambridge, England) 113, 913–917.

Hierholzer, A. and Kemler, R. (2010). Beta-catenin-mediated signaling and cell adhesion in postgastrulation mouse embryos. Dev Dyn 239, 191–199.

Ho, H.-Y. H., Susman, M. W., Bikoff, J. B., Ryu, Y. K., Jonas, A. M., Hu, L., Kuruvilla, R. and Greenberg, M. E. (2012). Wnt5a-Ror-Dishevelled signaling constitutes a core developmental pathway that controls tissue morphogenesis. Proceedings of the National Academy of Sciences 109, 4044–4051.

Hofmann, K. (2000). A superfamily of membrane-bound O-acyltransferases with implications for wnt signaling. Trends Biochem Sci 25, 111–112.

Hsieh, J.-C., Lee, L., Zhang, L., Wefer, S., Brown, K., DeRossi, C., Wines, M. E., Rosenquist, T. and Holdener, B. C. (2003). Mesd encodes an LRP5/6 chaperone essential for specification of mouse embryonic polarity. Cell 112, 355–367.

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.

Hunter, S. M. (1999). Non‐surgical method for the induction of delayed implantation and recovery of viable blastocysts in rats and mice by the use of tamoxifen and Depo‐Provera. Molecular Reproduction and ….

Ikeya, M. and Takada, S. (1998). Wnt signaling from the dorsal neural tube is required for the

121

formation of the medial dermomyotome. Development (Cambridge, England) 125, 4969– 4976.

Ikeya, M., Lee, S. M., Johnson, J. E., McMahon, A. P. and Takada, S. (1997). Wnt signalling required for expansion of neural crest and CNS progenitors. Nature 389, 966–970.

Imajo, M., Miyatake, K., Iimura, A., Miyamoto, A. and Nishida, E. (2012). A molecular mechanism that links Hippo signalling to the inhibition of Wnt/β-catenin signalling. EMBO J 31, 1109–1122.

Janda, C. Y., Waghray, D., Levin, A. M., Thomas, C. and Garcia, K. C. (2012). Structural basis of Wnt recognition by Frizzled. Science 337, 59–64.

Jenny, A., Reynolds-Kenneally, J., Das, G., Burnett, M. and Mlodzik, M. (2005). Diego and Prickle regulate Frizzled planar cell polarity signalling by competing for Dishevelled binding. Nat Cell Biol 7, 691–697.

Jho, E.-H., Zhang, T., Domon, C., Joo, C.-K., Freund, J.-N. and Costantini, F. (2002). Wnt/beta-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway. Mol Cell Biol 22, 1172–1183.

Kadowaki, T., Wilder, E. L., Klingensmith, J., Zachary, K. and Perrimon, N. (1996). The segment polarity gene porcupine encodes a putative multitransmembrane protein involved in Wingless processing. Genes Dev 10, 3116–3128.

Kang, M., Piliszek, A., Artus, J. and Hadjantonakis, A.-K. K. (2012). FGF4 is required for lineage restriction and salt-and-pepper distribution of primitive endoderm factors but not their initial expression in the mouse. Development (Cambridge, England) 1–13.

Kelly, O. G., Pinson, K. I. and Skarnes, W. C. (2004). The Wnt co-receptors Lrp5 and Lrp6 are essential for gastrulation in mice. Development (Cambridge, England) 131, 2803–2815.

Kemler, R., Hierholzer, A., Kanzler, B., Kuppig, S., Hansen, K., Taketo, M. M., De Vries, W. N., Knowles, B. B. and Solter, D. (2004). Stabilization of beta-catenin in the mouse zygote leads to premature epithelial-mesenchymal transition in the epiblast. Development (Cambridge, England) 131, 5817–5824.

Kemp, C., Willems, E., Abdo, S., Lambiv, L. and Leyns, L. (2005). Expression of all Wnt genes and their secreted antagonists during mouse blastocyst and postimplantation development. Dev Dyn 233, 1064–1075.

Kibar, Z., Vogan, K. J., Groulx, N., Justice, M. J., Underhill, D. A. and Gros, P. (2001). Ltap, a mammalian homolog of Drosophila Strabismus/Van Gogh, is altered in the mouse neural tube mutant Loop-tail. Nat Genet 28, 251–255.

Kilian, B., Mansukoski, H., Barbosa, F. C., Ulrich, F., Tada, M. and Heisenberg, C.-P. (2003). The role of Ppt/Wnt5 in regulating cell shape and movement during zebrafish gastrulation. Mech Dev 120, 467–476.

122

Kim, K.-A., Wagle, M., Tran, K., Zhan, X., Dixon, M. A., Liu, S., Gros, D., Korver, W., Yonkovich, S., Tomasevic, N., et al. (2008). R-Spondin family members regulate the Wnt pathway by a common mechanism. Mol Biol Cell 19, 2588–2596.

Kinzler, K. W., Nilbert, M. C., Su, L. K., Vogelstein, B., Bryan, T. M., Levy, D. B., Smith, K. J., Preisinger, A. C., Hedge, P. and McKechnie, D. (1991). Identification of FAP locus genes from chromosome 5q21. Science 253, 661–665.

Komekado, H., Yamamoto, H., Chiba, T. and Kikuchi, A. (2007). Glycosylation and palmitoylation of Wnt-3a are coupled to produce an active form of Wnt-3a. Genes Cells 12, 521–534.

Korkut, C., Ataman, B., Ramachandran, P., Ashley, J., Barria, R., Gherbesi, N. and Budnik, V. (2009). Trans-synaptic transmission of vesicular Wnt signals through Evi/Wntless. Cell 139, 393–404.

Kurayoshi, M., Yamamoto, H., Izumi, S. and Kikuchi, A. (2007). Post-translational palmitoylation and glycosylation of Wnt-5a are necessary for its signalling. Biochem J 402, 515–523.

Kwon, G. S. and Hadjantonakis, A.-K. K. (2009). Transthyretin mouse transgenes direct RFP expression or Cre-mediated recombination throughout the visceral endoderm. Genesis 47, 447–455.

Kwon, G. S., Viotti, M. and Hadjantonakis, A.-K. K. (2008). The endoderm of the mouse embryo arises by dynamic widespread intercalation of embryonic and extraembryonic lineages. Dev Cell 15, 509–520.

Lawrence, P. A., Casal, J. and Struhl, G. (2002). Towards a model of the organisation of planar polarity and pattern in the Drosophila abdomen. Development (Cambridge, England) 129, 2749–2760.

Leoyklang, P., Suphapeetiporn, K., Wananukul, S. and Shotelersuk, V. (2008). Three novel mutations in the PORCN gene underlying focal dermal hypoplasia. Clin Genet.

Lewandoski, M., Wassarman, K. M. and Martin, G. R. (1997). Zp3-cre, a transgenic mouse line for the activation or inactivation of loxP-flanked target genes specifically in the female germ line. Curr Biol 7, 148–151.

Li, V. S. W., Ng, S. S., Boersema, P. J., Low, T. Y., Karthaus, W. R., Gerlach, J. P., Mohammed, S., Heck, A. J. R., Maurice, M. M., Mahmoudi, T., et al. (2012). Wnt Signaling through Inhibition of β-Catenin Degradation in an Intact Axin1 Complex. Cell 149, 1245–1256.

Lighthouse, J. K., Zhang, L., Hsieh, J.-C., Rosenquist, T. and Holdener, B. C. (2011). MESD is essential for apical localization of megalin/LRP2 in the visceral endoderm. Dev Dyn 240, 577–588.

Liu, P., Wakamiya, M., Shea, M. J., Albrecht, U., Behringer, R. R. and Bradley, A. (1999).

123

Requirement for Wnt3 in vertebrate axis formation. Nat Genet 22, 361–365.

Liu, W., Shaver, T. M., Balasa, A., Ljungberg, M. C., Wang, X., Wen, S., Nguyen, H. and Van den Veyver, I. B. (2012). Deletion of Porcn in mice leads to multiple developmental defects and models human focal dermal hypoplasia (Goltz syndrome). PLoS ONE 7, e32331.

Loguercio Leite, J. C., Faermann, R., Rodrigues Stein, N., Zandoná, D. and Maegawa, G. (2005). Focal dermal hypoplasia associated with split sternum--Goltz syndrome. Clin. Dysmorphol. 14, 37–39.

Lu, W., Yamamoto, V., Ortega, B. and Baltimore, D. (2004). Mammalian Ryk is a Wnt coreceptor required for stimulation of neurite outgrowth. Cell 119, 97–108.

Lum, L. and Clevers, H. C. (2012). The Unusual Case of Porcupine. Science 337, 922–923.

Lyu, J., Yamamoto, V. and Lu, W. (2008). Cleavage of the Wnt receptor Ryk regulates neuronal differentiation during cortical neurogenesis. Dev Cell 15, 773–780.

Ma, W., Farrukh Nizam, M. and Grewal, R. P. (2002). X-linked Charcot-Marie-Tooth disease caused by a novel point mutation in the connexin-32 gene. Neurol Sci 23, 195–197.

Macfarlan, T. S., Gifford, W. D., Driscoll, S., Lettieri, K., Rowe, H. M., Bonanomi, D., Firth, A., Singer, O., Trono, D. and Pfaff, S. L. (2012). Embryonic stem cell potency fluctuates with endogenous retrovirus activity. Nature 487, 57–63.

Majumdar, A., Vainio, S., Kispert, A., McMahon, J. A. and McMahon, A. P. (2003). Wnt11 and Ret/Gdnf pathways cooperate in regulating ureteric branching during metanephric kidney development. Development (Cambridge, England) 130, 3175–3185.

Mamo, S., Gal, A. B., Bodo, S. and Dinnyes, A. (2007). Quantitative evaluation and selection of reference genes in mouse oocytes and embryos cultured in vivo and in vitro. BMC Dev Biol 7, 14.

Marikawa, Y., Tamashiro, D. A. A., Fujita, T. C. and Alarcon, V. B. (2009). Aggregated P19 mouse embryonal carcinoma cells as a simple in vitro model to study the molecular regulations of mesoderm formation and axial elongation morphogenesis. Genesis 47, 93– 106.

Mason, J. O., Kitajewski, J. and Varmus, H. E. (1992). Mutational analysis of mouse Wnt-1 identifies two temperature-sensitive alleles and attributes of Wnt-1 protein essential for transformation of a mammary cell line. Mol Biol Cell 3, 521–533.

McMahon, A. P. and Bradley, A. (1990). The Wnt-1 (int-1) proto-oncogene is required for development of a large region of the mouse brain. Cell 62, 1073–1085.

McMahon, A. P. and Moon, R. T. (1989). Ectopic Expression of the Proto-Oncogene int-1 in Xenopus Embryos Leads to Duplication of the Embryonic Axis. Cell 58, 1075–1064.

Mikels, A. J. and Nusse, R. (2006). Purified Wnt5a protein activates or inhibits beta-catenin-

124

TCF signaling depending on receptor context. PLoS Biol 4, e115.

Miller, C. and Sassoon, D. A. (1998). Wnt-7a maintains appropriate uterine patterning during the development of the mouse female reproductive tract. Development (Cambridge, England) 125, 3201–3211.

Miller, M. F., Cohen, E. D., Baggs, J. E., Lu, M. M., Hogenesch, J. B. and Morrisey, E. E. (2012). Wnt ligands signal in a cooperative manner to promote foregut organogenesis. Proceedings of the National Academy of Sciences 109, 15348–15353.

Miura, S., Singh, A. P. and Mishina, Y. (2010). Bmpr1a is required for proper migration of the AVE through regulation of Dkk1 expression in the pre-streak mouse embryo. Dev Biol 341, 246–254.

Mohamed, O. A., Clarke, H. J. and Dufort, D. (2004). Beta-catenin signaling marks the prospective site of primitive streak formation in the mouse embryo. Dev Dyn 231, 416–424.

Mohamed, O. A., Jonnaert, M., Labelle-Dumais, C., Kuroda, K., Clarke, H. J. and Dufort, D. (2005). Uterine Wnt/beta-catenin signaling is required for implantation. Proc Natl Acad Sci USA 102, 8579–8584.

Monkley, S. J., Delaney, S. J., Pennisi, D. J., Christiansen, J. H. and Wainwright, B. J. (1996). Targeted disruption of the Wnt2 gene results in placentation defects. Development (Cambridge, England) 122, 3343–3353.

Morkel, M., Huelsken, J., Wakamiya, M., Ding, J., van de Wetering, M., Clevers, H. C., Taketo, M. M., Behringer, R. R., Shen, M. M. and Birchmeier, W. (2003). Beta-catenin regulates Cripto- and Wnt3-dependent gene expression programs in mouse axis and mesoderm formation. Development (Cambridge, England) 130, 6283–6294.

Morrison, G. M., Oikonomopoulou, I., Migueles, R. P., Soneji, S., Livigni, A., Enver, T. and Brickman, J. M. (2008). Anterior definitive endoderm from ESCs reveals a role for FGF signaling. Cell Stem Cell 3, 402–415.

Mulroy, T., McMahon, J. A., Burakoff, S. J., McMahon, A. P. and Sen, J. (2002). Wnt-1 and Wnt-4 regulate thymic cellularity. Eur. J. Immunol. 32, 967–971.

Nagy, I. I., Railo, A., Rapila, R., Hast, T., Sormunen, R., Tavi, P., Räsänen, J. and Vainio, S. J. (2010). Wnt-11 signalling controls ventricular myocardium development by patterning N-cadherin and beta-catenin expression. Cardiovasc Res 85, 100–109.

Najdi, R., Proffitt, K. D., Sprowl, S., Kaur, S., Yu, J., Covey, T. M., Virshup, D. M. and Waterman, M. L. (2012). A uniform human Wnt expression library reveals a shared secretory pathway and unique signaling activities. Differentiation 84, 203–213.

Neumann, S., Coudreuse, D. Y. M., Van Der Westhuyzen, D. R., Eckhardt, E. R. M., Korswagen, H. C., Schmitz, G. and Sprong, H. (2009). Mammalian Wnt3a is released on lipoprotein particles. Traffic 10, 334–343.

125

Nichols, J. and Ying, Q.-L. (2006). Derivation and propagation of embryonic stem cells in serum- and feeder-free culture. Methods Mol Biol 329, 91–98.

Nichols, J., Chambers, I., Taga, T. and Smith, A. (2001). Physiological rationale for responsiveness of mouse embryonic stem cells to gp130 cytokines. Development (Cambridge, England) 128, 2333–2339.

Niehrs, C. (2012). The complex world of WNT receptor signalling. Nat Rev Mol Cell Biol 13, 767–779.

Nielsen, H., Engelbrecht, J., Brunak, S. and Heijne, von, G. (1997). Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 10, 1–6.

Nishioka, N., Inoue, K.-I., Adachi, K., Kiyonari, H., Ota, M., Ralston, A., Yabuta, N., Hirahara, S., Stephenson, R. O., Ogonuki, N., et al. (2009). The hippo signaling pathway components lats and Yap pattern Tead4 activity to distinguish mouse trophectoderm from inner cell mass. Dev Cell 16, 398–410.

Nishisho, I., Nakamura, Y., Miyoshi, Y., Miki, Y., Ando, H., Horii, A., Koyama, K., Utsunomiya, J., Baba, S. and Hedge, P. (1991). Mutations of chromosome 5q21 genes in FAP and colorectal cancer patients. Science 253, 665–669.

Niwa, H., Yamamura, K. and Miyazaki, J. (1991). Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108, 193–199.

Noordermeer, J., Klingensmith, J., Perrimon, N. and Nusse, R. (1994). dishevelled and armadillo act in the wingless signalling pathway in Drosophila. Nature 367, 80–83.

Nostro, M. C., Cheng, X., Keller, G. M. and Gadue, P. (2008). Wnt, Activin, and BMP Signaling Regulate Distinct Stages in the Developmental Pathway from Embryonic Stem Cells to Blood. Cell Stem Cell 2, 60–71.

Nusse, R. and Varmus, H. E. (1982). Many tumors induced by the mouse mammary tumor virus contain a provirus integrated in the same region of the host genome. Cell 31, 99–109.

Nusse, R., Brown, A., Papkoff, J., Scambler, P., Shackleford, G., McMahon, A. P., Moon, R. T. and Varmus, H. (1991). A new nomenclature for int-1 and related genes: the Wnt gene family. Cell 64, 231.

Nüsslein-Volhard, C. and Wieschaus, E. (1980). Mutations affecting segment number and polarity in Drosophila. Nature 287, 795–801.

Ohkawara, B., Glinka, A. and Niehrs, C. (2011). Rspo3 binds syndecan 4 and induces Wnt/PCP signaling via clathrin-mediated endocytosis to promote morphogenesis. Dev Cell 20, 303–314.

Paller, A. S. (2007). Wnt signaling in focal dermal hypoplasia. Nat Genet 39, 820–821.

126

Panáková, D., Sprong, H., Marois, E., Thiele, C. and Eaton, S. (2005). Lipoprotein particles are required for Hedgehog and Wingless signalling. Nature 435, 58–65.

Parr, B. A. and McMahon, A. P. (1995). Dorsalizing signal Wnt-7a required for normal polarity of D-V and A-P axes of mouse limb. Nature 374, 350–353.

Parr, B. A. and McMahon, A. P. (1998). Sexually dimorphic development of the mammalian reproductive tract requires Wnt-7a. Nature 395, 707–710.

Parr, B. A., Cornish, V. A., Cybulsky, M. I. and McMahon, A. P. (2001). Wnt7b regulates placental development in mice. Developmental Biology 237, 324–332.

Patthy, L. (2000). The WIF module. Trends Biochem Sci 25, 12–13.

Peifer, M., McCrea, P. D., Green, K. J., Wieschaus, E. and Gumbiner, B. M. (1992). The vertebrate adhesive junction proteins beta-catenin and plakoglobin and the Drosophila segment polarity gene armadillo form a multigene family with similar properties. J Cell Biol 118, 681–691.

Perrimon, N., Engstrom, L. and Mahowald, A. P. (1989). Zygotic lethals with specific maternal effect phenotypes in Drosophila melanogaster. I. Loci on the X chromosome. Genetics 121, 333–352.

Pinto, D., Gregorieff, A., Begthel, H. and Clevers, H. C. (2003). Canonical Wnt signals are essential for homeostasis of the intestinal epithelium. Genes Dev 17, 1709–1713.

Port, F., Hausmann, G. and Basler, K. (2011). A genome-wide RNA interference screen uncovers two p24 proteins as regulators of Wingless secretion. EMBO Rep 12, 1144–1152.

Port, F., Kuster, M., Herr, P., Furger, E., Bänziger, C., Hausmann, G. and Basler, K. (2008). Wingless secretion promotes and requires retromer-dependent cycling of Wntless. Nature 10, 178–185.

Posfai, E., Kunzmann, R., Brochard, V., Salvaing, J., Cabuy, E., Roloff, T. C., Liu, Z., Tardat, M., van Lohuizen, M., Vidal, M., et al. (2012). Polycomb function during oogenesis is required for mouse embryonic development. Genes Dev 26, 920–932.

Potok, M. A., Cha, K. B., Hunt, A., Brinkmeier, M. L., Leitges, M., Kispert, A. and Camper, S. A. (2008). WNT signaling affects gene expression in the ventral diencephalon and pituitary gland growth. Dev Dyn 237, 1006–1020.

Pöpperl, H., Schmidt, C., Wilson, V., Hume, C. R., Dodd, J., Krumlauf, R. and Beddington, R. S. P. (1997). Misexpression of Cwnt8C in the mouse induces an ectopic embryonic axis and causes a truncation of the anterior neuroectoderm. Development (Cambridge, England) 124, 2997–3005.

Proffitt, K. D. and Virshup, D. M. (2012). Precise regulation of porcupine activity is required for physiological wnt signaling. J Biol Chem 287, 34167–34178.

127

Proffitt, K. D., Madan, B., Ke, Z., Pendharkar, V., Ding, L., Lee, M. A., Hannoush, R. N. and Virshup, D. M. (2012). Pharmacological inhibition of the Wnt acyltransferase PORCN prevents growth of WNT-driven mammary cancer. Cancer Res.

Rashbass, P., Wilson, V., Rosen, B. and Beddington, R. S. (1994). Alterations in gene expression during mesoderm formation and axial patterning in Brachyury (T) embryos. Int J Dev Biol 38, 35–44.

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

Rodriguez, T. A., Casey, E. S., Harland, R. M., Smith, J. C. and Beddington, R. S. (2001). Distinct enhancer elements control Hex expression during gastrulation and early organogenesis. Developmental Biology 234, 304–316.

Roelink, H., Wagenaar, E., Lopes da Silva, S. and Nusse, R. (1990). Wnt-3, a gene activated by proviral insertion in mouse mammary tumors, is homologous to int-1/Wnt-1 and is normally expressed in mouse embryos and adult brain. Proc Natl Acad Sci USA 87, 4519– 4523.

Rogers, D. C., Fisher, E. M., Brown, S. D., Peters, J., Hunter, A. J. and Martin, J. E. (1997). Behavioral and functional analysis of mouse phenotype: SHIRPA, a proposed protocol for comprehensive phenotype assessment. Mamm Genome 8, 711–713.

Rubinfeld, B., Souza, B., Albert, I., Müller, O., Chamberlain, S. H., Masiarz, F. R., Munemitsu, S. and Polakis, P. (1993). Association of the APC gene product with beta- catenin. Science 262, 1731–1734.

Rudloff, S. and Kemler, R. (2012). Differential requirements for -catenin during mouse development. Development (Cambridge, England) 139, 3711–3721.

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.

Saijoh, Y., Adachi, H., Sakuma, R., Yeo, C. Y., Yashiro, K., Watanabe, M., Hashiguchi, H., Mochida, K., Ohishi, S., Kawabata, M., et al. (2000). Left-right asymmetric expression of lefty2 and nodal is induced by a signaling pathway that includes the transcription factor FAST2. Mol Cell 5, 35–47.

Saldanha, J., Singh, J. and Mahadevan, D. (1998). Identification of a Frizzled-like cysteine rich domain in the extracellular region of developmental receptor tyrosine kinases. Protein Sci. 7, 1632–1635.

Schambony, A. and Wedlich, D. (2007). Wnt-5A/Ror2 regulate expression of XPAPC through an alternative noncanonical signaling pathway. Dev Cell 12, 779–792.

Schuijers, J. and Clevers, H. C. (2012). Adult mammalian stem cells: the role of Wnt, Lgr5 and

128

R-spondins. EMBO J 31, 2685–2696.

Schwabe, G. C., Trepczik, B., Süring, K., Brieske, N., Tucker, A. S., Sharpe, P. T., Minami, Y. and Mundlos, S. (2004). Ror2 knockout mouse as a model for the developmental pathology of autosomal recessive Robinow syndrome. Dev Dyn 229, 400–410.

Shawlot, W. and Behringer, R. R. (1995). Requirement for Lim1 in head-organizer function. Nature 374, 425–430.

Shawlot, W., Deng, J. M. and Behringer, R. R. (1998). Expression of the mouse cerberus- related gene, Cerr1, suggests a role in anterior neural induction and somitogenesis. Proc Natl Acad Sci USA 95, 6198–6203.

Shawlot, W., Wakamiya, M., Kwan, K. M., Kania, A., Jessell, T. M. and Behringer, R. R. (1999). Lim1 is required in both primitive streak-derived tissues and visceral endoderm for head formation in the mouse. Development (Cambridge, England) 126, 4925–4932.

Sheldahl, L. C., Slusarski, D. C., Pandur, P., Miller, J. R., Kühl, M. and Moon, R. T. (2003). Dishevelled activates Ca2+ flux, PKC, and CamKII in vertebrate embryos. J Cell Biol 161, 769–777.

Shimizu, H., Julius, M. A., Giarré, M., Zheng, Z., Brown, A. M. and Kitajewski, J. (1997). Transformation by Wnt family proteins correlates with regulation of beta-catenin. Cell Growth Differ 8, 1349–1358.

Shu, W., Jiang, Y. Q., Lu, M. M. and Morrisey, E. E. (2002). Wnt7b regulates mesenchymal proliferation and vascular development in the lung. Development (Cambridge, England) 129, 4831–4842.

Siegfried, E., Wilder, E. L. and Perrimon, N. (1994). Components of wingless signalling in Drosophila. Nature 367, 76–80.

Slusarski, D. C., Corces, V. G. and Moon, R. T. (1997a). Interaction of Wnt and a Frizzled homologue triggers G-protein-linked phosphatidylinositol signalling. Nature 390, 410–413.

Slusarski, D. C., Yang-Snyder, J., Busa, W. B. and Moon, R. T. (1997b). Modulation of embryonic intracellular Ca2+ signaling by Wnt-5A. Developmental Biology 182, 114–120.

Später, D., Hill, T. P., O'sullivan, R. J., Gruber, M., Conner, D. A. and Hartmann, C. (2006). Wnt9a signaling is required for joint integrity and regulation of Ihh during chondrogenesis. Development (Cambridge, England) 133, 3039–3049.

Stark, K., Vainio, S., Vassileva, G. and McMahon, A. P. (1994). Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4. Nature 372, 679– 683.

Stenman, J. M., Rajagopal, J., Carroll, T. J., Ishibashi, M., McMahon, J. A. and McMahon, A. P. (2008). Canonical Wnt signaling regulates organ-specific assembly and differentiation of CNS vasculature. Science 322, 1247–1250.

129

Stephenson, R. O., Rossant, J. and Tam, P. P. L. (2012). Intercellular interactions, position, and polarity in establishing blastocyst cell lineages and embryonic axes. Cold Spring Harbor Perspectives in Biology 4.

Stephenson, R. O., Yamanaka, Y. and Rossant, J. (2010). Disorganized epithelial polarity and excess trophectoderm cell fate in preimplantation embryos lacking E-cadherin. Development (Cambridge, England).

Stevens, J. R., Miranda-Carboni, G. A., Singer, M. A., Brugger, S. M., Lyons, K. M. and Lane, T. F. (2010). Wnt10b deficiency results in age-dependent loss of bone mass and progressive reduction of mesenchymal progenitor cells. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 25, 2138–2147.

Su, L. K., Vogelstein, B. and Kinzler, K. W. (1993). Association of the APC tumor suppressor protein with catenins. Science 262, 1734–1737.

Tai, C. C., Sala, F. G., Ford, H. R., Wang, K. S., Li, C., Minoo, P., Grikscheit, T. C. and Bellusci, S. (2009). Wnt5a knock-out mouse as a new model of anorectal malformation. J. Surg. Res. 156, 278–282.

Takada, R., Satomi, Y., Kurata, T., Ueno, N., Norioka, S., Kondoh, H., Takao, T. and Takada, S. (2006). Monounsaturated fatty acid modification of Wnt protein: its role in Wnt secretion. Dev Cell 11, 791–801.

Takada, S., Stark, K. L., Shea, M. J., Vassileva, G. and McMahon, A. P. (1994). Wnt-3a regulates somite and tailbud formation in the mouse embryo. Genes Dev 8, 174–189.

Takaoka, K., Yamamoto, M. and Hamada, H. (2011). Origin and role of distal visceral endoderm, a group of cells that determines anterior-posterior polarity of the mouse embryo. Nature 13, 743–752.

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.

Tamai, K., Semenov, M., Kato, Y., Spokony, R., Liu, C., Katsuyama, Y., Hess, F., Saint- Jeannet, J. P. and He, X. (2000). LDL-receptor-related proteins in Wnt signal transduction. Nature 407, 530–535.

Tanaka, K., Kitagawa, Y. and Kadowaki, T. (2002). Drosophila segment polarity gene product porcupine stimulates the posttranslational N-glycosylation of wingless in the endoplasmic reticulum. J Biol Chem 277, 12816–12823.

Tanaka, K., Kitagawa, Y. and Kadowaki, T. (2003). Misexpression of mouse porcupine isoforms modulates the differentiation of P19 embryonic carcinoma cells. Cell Biol Int 27, 549–557.

Tanaka, K., Okabayashi, K., Asashima, M., Perrimon, N. and Kadowaki, T. (2000). The evolutionarily conserved porcupine gene family is involved in the processing of the Wnt

130

family. Eur J Biochem 267, 4300–4311.

Taylor, J., Abramova, N., Charlton, J. and Adler, P. N. (1998). Van Gogh: a new Drosophila tissue polarity gene. Genetics 150, 199–210.

Thomas, K. R. and Capecchi, M. R. (1990). Targeted disruption of the murine int-1 proto- oncogene resulting in severe abnormalities in midbrain and cerebellar development. Nature 346, 847–850.

Thomas, P. and Beddington, R. S. P. (1996). Anterior primitive endoderm may be responsible for patterning the anterior neural plate in the mouse embryo. Curr Biol 6, 1487–1496.

Tilghman, S. M. (1996). Lessons learned, promises kept: a biologist's eye view of the Genome Project. Genome Res 6, 773–780.

Torban, E., Wang, H.-J., Groulx, N. and Gros, P. (2004). Independent mutations in mouse Vangl2 that cause neural tube defects in looptail mice impair interaction with members of the Dishevelled family. J Biol Chem 279, 52703–52713.

Tortelote, G. G., Hernández-Hernández, J. M., Quaresma, A. J. C., Nickerson, J. A., Imbalzano, A. N. and Rivera-Pérez, J. A. (2012). Wnt3 function in the epiblast is required for the maintenance but not the initiation of gastrulation in mice. Developmental Biology 1– 37.

Tsukiyama, T. and Yamaguchi, T. P. (2012). Mice lacking Wnt2b are viable and display a postnatal olfactory bulb phenotype. Neurosci. Lett. 512, 48–52.

Unterseher, F., Hefele, J. A., Giehl, K., De Robertis, E. M., Wedlich, D. and Schambony, A. (2004). Paraxial protocadherin coordinates cell polarity during convergent extension via Rho A and JNK. EMBO J 23, 3259–3269.

Vainio, S., Heikkilä, M., Kispert, A., Chin, N. and McMahon, A. P. (1999). Female development in mammals is regulated by Wnt-4 signalling. Nature 397, 405–409.

Valenta, T., Gay, M., Steiner, S., Draganova, K., Zemke, M., Hoffmans, R., Cinelli, P., Aguet, M., Sommer, L. and Basler, K. (2011). Probing transcription-specific outputs of β- catenin in vivo. Genes Dev 25, 2631–2643. van Amerongen, R. (2012). Alternative wnt pathways and receptors. Cold Spring Harbor Perspectives in Biology 4. van Amerongen, R. and Berns, A. (2006). Knockout mouse models to study Wnt signal transduction. Trends Genet 22, 678–689. van Amerongen, R. and Nusse, R. (2009). Towards an integrated view of Wnt signaling in development. Development (Cambridge, England) 136, 3205–3214. van Amerongen, R., Fuerer, C., Mizutani, M. and Nusse, R. (2012). Wnt5a can both activate and repress Wnt/β-catenin signaling during mouse embryonic development. Dev Biol 369,

131

101–114. van den Heuvel, M., Harryman-Samos, C., Klingensmith, J., Perrimon, N. and Nusse, R. (1993). Mutations in the segment polarity genes wingless and porcupine impair secretion of the wingless protein. EMBO J 12, 5293–5302.

Veeman, M. T., Slusarski, D. C., Kaykas, A., Louie, S. H. and Moon, R. T. (2003). Zebrafish prickle, a modulator of noncanonical Wnt/Fz signaling, regulates gastrulation movements. Curr Biol 13, 680–685.

Vendrell, V., Vázquez-Echeverría, C., López-Hernández, I., Alonso, B. D., Martínez, S., Pujades, C. and Schimmang, T. (2012). Roles of Wnt8a during formation and patterning of the mouse inner ear. Mech Dev.

Verras, M., Papandreou, I., Lim, A. L. and Denko, N. C. (2008). Tumor hypoxia blocks Wnt processing and secretion through the induction of endoplasmic reticulum stress. Mol Cell Biol 28, 7212–7224.

Vogel, P., Read, R., Hansen, G. M., Freay, L. C., Zambrowicz, B. P. and Sands, A. T. (2010). Situs Inversus in Dpcd/Poll-/-, Nme7-/- , and Pkd1l1-/- Mice. Veterinary Pathology 47, 120–131.

Wang, J., Sinha, T. and Wynshaw-Boris, A. (2012). Wnt signaling in Mammalian development: lessons from mouse genetics. Cold Spring Harbor Perspectives in Biology 4.

Wang, X., Reid Sutton, V., Omar Peraza-Llanes, J., Yu, Z., Rosetta, R., Kou, Y.-C., Eble, T. N., Patel, A., Thaller, C., Fang, P., et al. (2007). Mutations in X-linked PORCN, a putative regulator of Wnt signaling, cause focal dermal hypoplasia. Nat Genet 39, 836–838.

Wang, Y., Guo, N. and Nathans, J. (2006). The role of Frizzled3 and Frizzled6 in neural tube closure and in the planar polarity of inner-ear sensory hair cells. J Neurosci 26, 2147–2156.

Ware, S. M., Aygun, M. G. and Hildebrandt, F. (2011). Spectrum of Clinical Diseases Caused By Disorders of Primary Cilia. Proceedings of the American Thoracic Society 8, 444–450.

Wilkinson, D. G., Bhatt, S., Cook, M., Boncinelli, E. and Krumlauf, R. (1989). Segmental expression of Hox-2 homoeobox-containing genes in the developing mouse hindbrain. Nature 341, 405–409.

Willert, K., Brown, J. D., Danenberg, E., Duncan, A. W., Weissman, I. L., Reya, T., Yates, J. R. and Nusse, R. (2003). Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature 423, 448–452.

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.

Witte, F., Dokas, J., Neuendorf, F., Mundlos, S. and Stricker, S. (2009). Comprehensive expression analysis of all Wnt genes and their major secreted antagonists during mouse limb

132

development and cartilage differentiation. Gene Expr Patterns 9, 215–223.

Wray, J., Kalkan, T., Gomez-Lopez, S., Eckardt, D., Cook, A., Kemler, R. and Smith, A. (2011). Inhibition of glycogen synthase kinase-3 alleviates Tcf3 repression of the pluripotency network and increases embryonic stem cell resistance to differentiation. Nature 13, 838–845.

Xie, H., Tranguch, S., Jia, X., Zhang, H., Das, S. K., Dey, S. K., Kuo, C. J. and Wang, H. (2008). Inactivation of nuclear Wnt-{beta}-catenin signaling limits blastocyst competency for implantation. Development (Cambridge, England) 135, 717–727.

Yamaguchi, T. P., Bradley, A., McMahon, A. P. and Jones, S. (1999a). A Wnt5a pathway underlies outgrowth of multiple structures in the vertebrate embryo. Development (Cambridge, England) 126, 1211–1223.

Yamaguchi, T. P., Takada, S., Yoshikawa, Y., Wu, N. and McMahon, A. P. (1999b). T (Brachyury) is a direct target of Wnt3a during paraxial mesoderm specification. Genes Dev 13, 3185–3190.

Yamamoto, M., Mine, N., Mochida, K., Sakai, Y., Saijoh, Y., Meno, C. and Hamada, H. (2003). Nodal signaling induces the midline barrier by activating Nodal expression in the lateral plate. Development (Cambridge, England) 130, 1795–1804.

Yamanaka, Y., Lanner, F. and Rossant, J. (2010). FGF signal-dependent segregation of primitive endoderm and epiblast in the mouse blastocyst. Development (Cambridge, England) 137, 715–724.

Yamanaka, Y., Tamplin, O. J., Beckers, A., Gossler, A. and Rossant, J. (2007). Live imaging and genetic analysis of mouse notochord formation reveals regional morphogenetic mechanisms. Dev Cell 13, 884–896.

Yang, J. T., Rayburn, H. and Hynes, R. O. (1995). Cell adhesion events mediated by alpha 4 integrins are essential in placental and cardiac development. Development (Cambridge, England) 121, 549–560.

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 6, 1302–1306.

Yasunaga, M., (null), (null), Okada, M., Jakt, L. M., Torikai-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.

Ying, Q.-L., Wray, J., Nichols, J., Batlle-Morera, L., Doble, B. W., Woodgett, J. R., Cohen, P. and Smith, A. G. (2008). The ground state of embryonic stem cell self-renewal. Nature 453, 519–523.

Zeng, L., Fagotto, F., Zhang, T., Hsu, W., Vasicek, T. J., Perry, W. L., Lee, J. J., Tilghman, S. M., Gumbiner, B. M. and Costantini, F. (1997). The mouse Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway that regulates embryonic axis formation. Cell 90,

133

181–192.

Zeng, S.-M. and Yankowitz, J. (2003). X-inactivation patterns in human embryonic and extra- embryonic tissues. Placenta 24, 270–275.

Zhai, L., Chaturvedi, D. and Cumberledge, S. (2004). Drosophila wnt-1 undergoes a hydrophobic modification and is targeted to lipid rafts, a process that requires porcupine. J Biol Chem 279, 33220–33227.

Zheng, H.-F., Tobias, J. H., Duncan, E., Evans, D. M., Eriksson, J., Paternoster, L., Yerges- Armstrong, L. M., Lehtimäki, T., Bergström, U., Kähönen, M., et al. (2012). WNT16 Influences Bone Mineral Density, Cortical Bone Thickness, Bone Strength, and Osteoporotic Fracture Risk. PLoS Genet 8, e1002745.

Zhu, X., Zhu, H., Zhang, L., Huang, S., Cao, J., Ma, G., Feng, G., He, L., Yang, Y. and Guo, X. (2012a). Wls-mediated Wnts differentially regulate distal limb patterning and tissue morphogenesis. Dev Biol 365, 328–338.

Zhu, X., Zhu, H., Zhang, L., Huang, S., Cao, J., Ma, G., Feng, G., He, L., Yang, Y. and Guo, X. (2012b). Wls-mediated Wnts differentially regulate distal limb patterning and tissue morphogenesis. Dev Biol 365, 328–338.

Zoltewicz, J. S., Ashique, A. M., Choe, Y., Lee, G., Taylor, S., Phamluong, K., Solloway, M. and Peterson, A. S. (2009). Wnt signaling is regulated by endoplasmic reticulum retention. PLoS ONE 4, e6191.

134

Chapter 2: A modified version of this chapter has been published (PMID: 21554866): Biechele, S., Cox, B. J. and Rossant, J. (2011). Porcupine homolog is required for canonical Wnt signaling and gastrulation in mouse embryos. Dev Biol 355, 275–285.

Chapter 3: A modified version of this chapter is currently under review at Development.