! Tfap2a, Irf6 & Grhl3 : A NOVEL NETWORK THAT REGULATES BOTH NEURULATION AND CRANIOFACIAL DEVELOPMENT By Youssef Ayoub Adly Kousa A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirments for the degree of Biochemistry and Molecular Biology Ð Doctor of Philosophy 2014 !ABSTRACT Tfap2a, Irf6 & Grhl3 : A NOVEL NETWORK THAT REGULATES BOTH NEURULATION AND CRANIOFACIAL DEVELOPMENT By Youssef Ayoub Adly Kousa Interferon Regulatory Facto rs transcrip tionally regulate development and differentiation of the innat e and adaptive immune systems. Within this family, IRF6 is unique because it regulates cutaneous and orofacial development in humans and mice. Common variants in IRF6 are associated with 12% of all orofacial clefting risk. Critically , a DNA variant in the IRF6 enhancer MCS9.7, rs642961, is found in 30% of the world Õs population. Biochemically, we know that rs642961 abrogates one of four TFAP2a binding sites, suggesting regulatory func tion . Mutations in TFAP2a can lead to Branio -oculo -facial Syndrome, a dominantly inherited orofacial clefting syndrome that include s upper lip pits. However, functional studies have not shown if Tfap2a regulates MCS9.7 activity or endogenous Irf6 expressio n in the mouse. In addition, rare mutations in IRF6 , located within 1q32 -q41, lead to Van der Woude and Popliteal Pterygium Syndromes, dominantly inherited orofacial clefting disorders. Currently, 70% of VWS families have mutations in IRF6 . While the remai ning 30% have unknown etiology, p rior linkage analysis suggest locus he terogeneity . We use a mouse models to determine how common variants in IRF6 may be associated with orofacial clefting and to investigate locus heterogeneity in VWS. We find that knocki ng out Tfap2a leads to loss of MCS9.7 enhancer activity and Irf6 expression in vivo. On the other hand, Irf6 also appears to stabiliz e Tfap2a protein in epidermis . The !necessity of Tfap2a for Irf6 expression contribute s to our understanding of the associat ion between rs642961 and isolated orofacial clefting . Significantly, we also find that Irf6 transcriptionally activities Grhl3 in epithelium. Consistent with prior work showing locus heterogeneity, we find that mutations in GRHL3 can also le d to Van der Wo ude Syndrome. These results suggest that TFAP2a , IRF6 , and GRHL3 share a conserved genetic pathway that is required for proper development of the lip and palate in humans and mice . In the mouse, loss of Grhl3 and Tfap2a leads to skin, limb, craniofacial and neural tube defects. Because Irf6 is an intermediate node between Tfap2a and Grhl3 in oral epithelium, we predict and find that changes in Irf6 expression can lead to neural tube defects. Over -expressing Irf6 le ads to rostral neural tube defects , inclu ding loss of the cranial vault , i.e. acrania , and a split face . In addition, both reducing and over -expressing Irf6 leads to caudal neural tube defects, a curled and kinked tail, respectively. Consistent with orofacial genetic regulation , we find that Irf6 represses Tfap2a in rostral neural tube development. In the caudal neural tube, we find that Irf6 activates both Tfap2a and Grhl3 expression and that Tfap2a and Grhl3 interact in caudal neurulation. Consistently, human sequencing reveals a rare IRF6 mutat ion in an individual with spina bifida. Finally, we show that Irf6 expression in skin development rescues perinatal lethality but not limb, tail and palatal development. These results suggest that Tfap2a -Irf6 -Grhl3 regulate the development of multiple ecto dermal lineages. We conclude that c ross -fertilization in orofacial and neural tube deve lopment provides candidate genes and potential therapeutic strategies for two congeni tal diseases with significant morbidity and mortality. ! Copyright by YOUSSEF AYOUB ADLY KOUSA 2014 "!! I dedicate this work to my parents, my family and my teachers, younger and older; my parents for having the courage to come into an unknown land and language with nothing in hopes of a better futur e for my brother and I. I dedicate this book to Evon, David and Elijah, for not only supporting my passion but for motivating me to achieve more and for making life outside the lab truly enjoyable. I would not be here without my teachers, professors, mento rs, co -workers, and undergrads...for teaching me that I have lots to learn from everyone. "#!!ACKNOWLEDGMENTS First and foremost, I would like to thank my lab. It all starts with my mentor, Dr. Brian C. Schutte, for letting me join his lab, creating a outstanding research environment and for teaching me how to be a scientist who strives for both depth and breadth. I would also like to thank the people in the lab that made research fun and possible. Arianna Smith and Tamer Mansour not only made these years e njoyable but also constantly helped me to think through experiments, work through models and interpret results. They are not only my co -workers but are also my friends and I deeply appreciate them. I also want to thank Dr. Wal id Fakhouri for teaching me te chniques and how to remain skeptical, the essence of being a good scientist. I am also deeply indebted to Raeuf Roushangar, Nicole Patel and Ari Walter for all of the time they put toward this work, including days, nights, weekends and holidays. Mouse work is difficult and together R aeuf and Nicole performed over 5 000 genotyping reactions. Ari Walter single handedly sectioned more than 100 boxes worth of tissue, totaling nearly 5000 slides. This work would not be possible without their dedication. I also want to thank Dr. Dina Moussa for her efficiency and dedication despite numerous family and work obligations. This work would also not be possible without Eric Fuller, Keith McAuley, Matt Thomas, Kendra Siegersma, Krysta Wierzbicki, Silus DeBacker and Rac hel Han who gave freely of their time and have taught me so much about the art of teaching and learning. I also want to thank our collaborators who are technically at the University of Iowa but who I consider to be part of our lab. These especially include Dr. "## !!Martine Dunnwald, Dr. Rob Cornell, Dr. Jeff Murray, Dr. Elizabeth Leslie , Dr. Leah Biggs and Tiffany Smith . My time as a PhD student would have been dramatically different if not for their mentorship and friendship. I want to thank Dr. John Wang, for giving me a chance at Michigan State University, seeing that I could be successful and for being a person I truly admire and cherish . I am also truly indebted to the DO -PhD program at Michigan State University . Dr. Justin McCormick and Mrs. Bethany Heinle n have dedicated many years to this program and at every turn in the road they provided guidance and support. I am also very thankful for having , Dr. David Arnosti and Dr. John LaPres , brilliant scientists, as mentors and teachers . Finally, I want to sincerely thank Dr. An dy Amalfitano, Dr. Terrie Taylor, Dr. Karl Seydel and Dr. Douglas Postels for being mentors, friends and role models . They embody the spirit, intellect and clinical acumen that defines a physician -scientist. I can only hope to be as insightful, knowledge able and creative as my mentors. I want to al so thank Dr. Jon Kaguni, Mrs. Jessica Lawrence and Kaillathe ÔPappanÕ Padmanabhan for not only helping and guiding but for also going above and beyond the call of duty on multiple occasions. I would also like to thank Rick Lillich and Rob Patten at Mager S cientific for sharing their insights, expertise and equipment to make this work possible. I would also like to sincerely thank Amy Porter and Kathy Joseph at Investigative Histopathology Laboratory at Michigan State University. They literally handled every piece of tissue we sectioned and stained. Therefore, I am deeply indebted to them for their dedication and persistence in processing our precious tissues. "### !! I want to thank the people of Michigan State University who made the journey so complete , includin g Dr. John Fyfe , Dr. Sherif Ibrahim, Dr. Nermin Kady, Satyaki Sengupta, Dr. Bill Henry, Dr. Pat Venta, Dr. Jerry Dodgson, Dr. Ellen Wilch, Dr. Karen Friderici, Dr. Vilma Yuzbasiyan -Gurkan, Maciej Parys, Dr. Sandhya Payankaulam, Cho Rattanasinchai , Dr. Kath y Gallo , Sarah Roosa, Dr. Yasser Aldhamen and the ULAR Staff facilitating the human e treatment of research animals, including Mr. Don Harrington, Mr. Patrick Lee, Mr. Randy Shoemaker and Ms. Danielle Ferguson. I want to also thank my friends in the DO -PhD program, incl uding Darin Quach, Paul Beach, Steve Proper, Tyrell Simkins, Joe Prinsen, Eric Schauberger, My -Trang Dang, Nadine El -Ayache and Yirong Zhu. At the College of Osteopathic Medicine, I also want to thank Dr. Margaret Kingry , Ms. Robin Hastings and Erin Doelli ng for providing the support and instruction necessary to navigate the DO -PhD program. It would have been nearly impossible to finish this program in eight years without Dr. KingryÕs support. Ms. Robin Hastings was essential in this process as well and I a m deeply indebted to her. I am also very thankful for Dean William Strample, Dr. Reza Nassiri, Dr. H. Steven Williams, Dr. William Falls, and Ms. Beth Courey. I also want to thank Dr. Myriam Peyrard -Janvid and Dr. Juha Kere at the Karolinska Institutet, D r. Richard Finnell and Dr. Huiping Zhu at UT Austin, Dr. Laura Mitchell and Dr. AJ Agopian at UT Houston, Dr. Gary Shaw at Stanford University, Dr. Bogi Anderson and William Gordon at UC Irvine, Dr. Allison Ashley -Koch and Dr. Simon #$!!Gregory at Duke Univers ity, Dr. Akira Kinoshita at Nagasaki University, Dr. Yang Chai at UCLA and Dr. Trevor Williams at UC Denver . Most importantly, I would like to thank the National Institute of Dental and Craniofacial Research for supporting this work. I am also indebted t o Dr. Leslie Frieden for always providing the support and feedback necessary to understand the process of getting and keeping NIH funding. $!!PREFACE In my time as a graduate student, I have often found myself responding in one of three ways to research papers. The first response is ÒMan, I wish I could have written this.Ó The second is ÒGlad someone wrote this.Ó The third, more of a question, is ÒWhy would anyone write this?Ó I have written this thesis to avoid the question. ÒThe dogma s of the quiet past, are inadequate to the stormy present. The occasion is piled high with difficulty, and we must rise -- with the occasion. As our case is new, so we must think anew, and act anew .Ó Abraham Lincoln $#!!TABLE OF CONTENTS LIST OF TAB LESÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ ..ÉÉÉÉÉÉ..xiv LIST OF FIGURESÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ ..ÉÉÉÉÉÉÉ.....xv %&'!()!*++,&-.*(.)/0111111 ÉÉÉÉÉÉÉÉÉÉÉ.ÉÉÉÉÉÉÉÉÉÉÉxvii CHAPTER 1 - Irf6 regulates development and differentiation in multiple ectodermal lineages ........................................................................ÉÉÉÉÉÉÉ.É...1 Historical.............. ....................................................ÉÉÉÉÉÉÉÉÉÉÉ....2 Variable expressivity suggests that VWS is a clinical model for iCLP ....É..É.....2 Clinical context an d impact of orof acial cleftingÉÉÉÉÉÉ.....ÉÉÉÉÉ....4 Mutations in IRF6 cause VWS ...................................ÉÉ...........ÉÉÉÉÉÉ..6 Variants within IRF6 are associated with iCLP ................................É......É.......10 IRF6.................... ....................................................ÉÉÉÉÉÉ.ÉÉÉÉÉ..11 IRF6 structure -function in development and disease .......................................... ..13 Regulating IRF6 expression......................................ÉÉÉÉÉÉÉÉÉÉ.... .19 Irf6 knockout mice.. .....................................ÉÉÉÉÉÉÉÉÉÉ..................21 Genotype -phenotype correlation: From morphology to molecule.... .....................22 Rescuing the knockout phenotype... ......................................................................24 Neurulation: Neural tube d evelopment... ...............................................................27 Conclusion. ............................................................................................................29 APPENDIX.. ..............................................................ÉÉÉÉÉÉÉÉÉÉ....30 BIBLIOGRAPHY ..................................................................................................34 CHAPTER 2 - Dominant mutations in GRHL3 cause Van der Woude syn drome and disrupt oral periderm development ...............................................................................50 Abstract....................................................................ÉÉÉÉÉÉÉÉÉÉÉ. .53 Introduction........................... .....................................................É............É........54 Results..................................................................ÉÉÉÉÉÉ.....ÉÉÉÉÉ. 55 Grainy -head like 3 is the VWS2 gene ...........................................É.........55 Affect of GRHL3 alleles on zebrafish development .ÉÉÉ..ÉÉÉ.......57 Grhl3 -/- murine embryos have cleft palate at low penetrance....................58 The oral phenotypes of Irf6 and Grhl3 heterozygous murine mutants are independent.................... ............................................................................60 Discussion.. ..............................................................................ÉÉ.....ÉÉÉÉ.61 Supplemental Data Description..... ........................................................................64 Acknowledgments....... ...........................................................................................64 Web Resources............ ...........................................................................................64 Materials and Methods............ ...............................................................................65 Human DNA samples ................................................................................65 Targeted exome sequen cing .......................................................................65 Genotyping .................................................................................................66 $## !!Mutation screening by Sanger sequencing ................................................66 Phenotype Analysis ....................................................................................67 Transfection of human GRHL3 mutation variants into Zebrafish embryos ......................................................................................................67 Murine crosses ...........................................................................................68 Morphological, histological and molecular analyses of mice ....................69 Imaging ......................................................................................................69 Statistical analysis ......................................................................................70 APPENDIX....................... .........................................ÉÉÉÉÉÉÉÉÉÉ....71 BIBLIOGRAPHY ..................................................................................................99 CHAPTER 3 - Irf6 regulates Tfap2 ! and Grhl3 in neurulation ..................ÉÉ.É104 Abstract................... .................................................ÉÉÉÉÉÉÉÉÉÉ....106 Introduction................................................................................É............É...... 106 Results................................................... ...............ÉÉÉÉÉÉ.....ÉÉÉÉ...109 Tfap2 ! is necessary for MCS9.7 activity and Irf6 expression .................109 Irf6 regulates Tfap2 ! ..ÉÉÉÉÉÉ...É..........................................É112 Irf6 homeostasis is required for neurulation and Tfap2 ! expression in epid ermis......... .........................................................................................113 Endogenous and t ransgenic Irf6 expression regulate Tfap2 ! in neurulation. ..............................................................................................117 Tfap2 ! and Grhl3 interact in caudal neurulation .....................................119 Shared IRF6 mutation in Spina Bifida and VWS ....................................121 Discussion.. ..................................................................................................ÉÉ122 Supplemental Data Description..... ......................................................................126 Author Contributions ...........................................................................................126 Acknowledgments....... .........................................................................................126 Web Resources............ .........................................................................................126 Materials and Methods............ .............................................................................127 Murine crosses .........................................................................................127 Irf6 hypomorphic allele.............. .............................................................128 Murine gen otyping......................... .........................................................129 Morphological, histological and molecular analyses of murine tissue...129 Skeletal prep. ...........................................................................................131 Bioimaging upright/fluorescent microscope................ ............................131 Transcriptional profiling using quantitative -PCR ....................................132 Western blotting........................................................... ............................134 Human sequencing and genotyping.......................................... ...............134 Statistical analysis ....................................................................................134 APPENDIX....................... .........................................ÉÉÉÉÉÉÉÉÉÉ..135 BIBLIOGRAPHY ................................................................................................182 CHAPTER 4 - Epithelial Irf6 rescues lethality but not craniofacial and limb development ....................................................................................................................187 Abstract............................. .......................................ÉÉÉÉÉÉÉÉÉÉÉ188 Introduction................................................................................É............É...... 189 $### !!Results.................................................................. ÉÉÉÉÉÉ.....ÉÉÉ...É191 Titrating Irf6 dose shows that mandible -maxilla oral adhesions are not sufficient for clefting ................................................................É.....É..191 Irf6 expression using the KRT14 promoter completely rescues cutaneous defec ts..ÉÉÉÉ............................................................................ÉÉ193 Epithelial Irf6 rescues perinatal lethality............................... ..................194 Completely penetrant clefting despite basal Irf6 expression....... ............196 Palate-tongue oral adhesions prevent palatal elevat ion...........................19 7 Discussion.. ........................................................................ÉÉÉÉÉÉ..........198 Author Contributions ...........................................................................................202 Acknowledgments....... .........................................................................................203 Web Resources............ .........................................................................................203 Materials and Methods............ .............................................................................203 Murine crosses .........................................................................................204 Morphological and histological analysis..................... ............................204 Molecular analyse s of murine tissu e.. ......................................................205 Skeletal prep...................................................... .......................................205 Bioimaging upright/fluorescent microscope and stereomicroscope ........206 Transcriptional profiling using quantitative -PCR ....................................206 Statistical analysis ....................................................................................207 APPENDIX....................... .........................................ÉÉÉÉÉÉÉÉÉÉ..208 BIBLIOGRAPHY ................................................................................................229 CHAPTER 5 - Conclusion and Future Directions .....................................................232 Major Themes: Orofacial Clefting. ...........................ÉÉÉÉÉÉÉÉÉÉ...233 Major Themes: Neural Tube Defects............................ÉÉÉÉÉ........É É...234 Genetic risk for orofacial and neural tube defects................. ..............................236 Expanding the gene regulatory n etwork : Tfap2 and Grhl paralogs................. ....237 Novel implications for Irf6 knockout phenotype.............. ...................................239 Proposed gene regulatory network.......................................... .............................241 Therapeutic consideration in congenital disease................................................. .242 Preventat ive strategies in orofacial clefting and neural tube defects.................. .243 APPENDIX....................... .........................................ÉÉÉÉÉÉÉÉÉÉ..247 BIBLIOGRAPHY ................................................................................................250 $#" !!LIST OF TABLES Table 1: GRHL3 mutations in eight Van der Woude syndrome families ÉÉÉÉÉ.....83 Table 2 : Comparison of VWS phenotypes caused by mutations in IRF6 and GRHL3.....85 Table 3 : Human GRHL3 primers used in Sanger sequencing mutation screening ÉÉ....95 Table 4 : Genomic SNPs used for confirming de novo mutation s......................................96 Table 5 : Frequency of genotypes and resorbing embryos from Irf6 +/-xGrhl3 +/- cross....9 7 Table 6 : Association betwe en IRF6 (rs642961) and risk of NTDs ÉÉÉÉÉÉÉÉ.160 Table 7 : Association between IRF6 (rs75012801) and risk of NTDs ÉÉÉÉÉÉÉ.161 Table 8 : Association between IRF6 (rs17317411) and risk of NTDs ÉÉÉÉÉÉÉ.162 Table 9 : Tfap2a +/-;Tg MCS9.7 -LacZ ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ..............171 Table 1 0: Irf6 ey/- vs. Irf6 ey/+ É....ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ...........172 Table 1 1: Irf6 +/+ ; Tg KRT14 -Ir6 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ................173 Table 1 2: TgKRT14 -Ir6 ; Tg KRT14 -Ir6 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ................174 Table 1 3: Tfap2a +/-;Irf6 +/-ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ....ÉÉÉÉÉ..175 Table 1 4: Tfap2a -/-ÉÉ...ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ................176 Table 1 5: Tfap2a +/-;Tg KRT14 -Ir6 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.........177 Table 1 6: Tfap2a +/-;Irf6 +/-;Tg KRT14 -Ir6 ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.178 Table 1 7: Tfap2a +/-;Grhl3 +/-ÉÉÉÉÉÉ.................ÉÉÉÉÉÉÉÉÉÉ.ÉÉ179 Table 1 8: Murine qPCR p rimer sequences ÉÉÉÉÉÉÉÉ.......ÉÉÉÉÉÉÉ...180 Table 1 9: IRF6 sequencing primers ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ.....................181 Table 20 : Irf6 -/-; Tg KRT14:: Irf6 ÉÉÉÉÉ..........................ÉÉÉÉ...ÉÉÉÉ.......228 $"!!LIST OF FIGURES Figure 1: Analogous processes lead to palate and neural tube developmentÉÉ....... ......31 Figure 2 : Structure of IRF6 .ÉÉÉÉÉ.........ÉÉÉÉÉÉÉÉÉÉÉ........ÉÉ......32 Figure 3 : Proposed genetic network for orofacial and rostra l, caudal ne ural tube development............................ ÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉÉ...ÉÉÉ........33 Figure 4 : Mutations in GRHL3 cause Van der Woude syndrome..... ÉÉÉÉ................72 Figure 5 : VWS-associated alleles of GRHL3 disrupt the development of the periderm when expressed in zebrafish embryos ..ÉÉÉÉÉ.........ÉÉÉÉÉÉÉÉ.................75 Figure 6 : Grhl3 is required for murine periderm and palatal development............. ..........77 Figure 7 : No evidence for genetic interaction between Irf6 and Grhl3 in murine pal atal development..ÉÉÉÉÉ......... .ÉÉ.........ÉÉÉÉÉÉÉÉ......................................80 Figure 8 : Pedigrees of the eight VWS families with GRHL3 mutation................É .É....86 Figure 9 : Multiple alignment and protein domains of GRHL3 gene product s from human, mouse and zebrafish..ÉÉÉÉÉ......... .ÉÉ.........ÉÉÉÉÉÉÉÉ..........................91 Figure 1 0: Molecular changes in the oral epithelium of Irf6 -/- and Grhl3 -/- embryos .......93 Figure 1 1: Tfap2 ! is necessary for MCS9.7 activity and Irf6 expression ........................136 Figure 1 2: Irf6 regulates Tfap2 !......................................................................................141 Figure 1 3: Irf6 homeostasis is required for n eurulation and Tfap2 ! expression in epidermis ..ÉÉÉÉÉ..........ÉÉ.........ÉÉÉÉÉ.................ÉÉÉ........................145 Figure 1 4: Endogenous and Transgenic Irf6 expression regulate Tfap2 ! in neurulation ..ÉÉÉÉÉ..........ÉÉ.........ÉÉÉÉÉ.................ÉÉÉ.....................150 Figure 15 : Tfap2 ! and Grhl3 interact in caudal neurulation ...........................................155 Figure 1 6: Shared IRF6 mutation in Spina Bifida and VWS ...........................................158 Figure 17: Gen eration of Irf6 hypomorphic allele ...........................................................163 Figure 1 8: Irf6 transcripitonally regulates Trp63 , Tgm1 and Krt1 but not Krt14 and Tfap2c in skin .ÉÉ..........ÉÉ.........ÉÉÉÉ.......ÉÉ..................................ÉÉ......165 $"#!! Figure 19 : Irf6 homeostasis is required for epidermal development.. .............................167 Figure 20 : Irf6 Transcriptionally regulates Tfap2c but not Krt14 in tail development ..170 Figure 21 : Irf6 compound heterozygosity c auses completely p enetrant oral adhesions but not clefting .ÉÉÉÉ..........ÉÉ.........ÉÉÉÉÉÉÉÉ..................................ÉÉ209 Figure 22: Irf6 expression using the KRT14 promoter rescues cutaneous defects in knockout embryos. ....ÉÉ...ÉÉ.........ÉÉÉÉÉÉÉÉ..................................ÉÉ.213 Figure 23: Epidermal expression of Irf6 rescues perinatal lethality without altering skeletal defects , limb clubbing or syndactyly ...................................................É...........217 Figure 24 : Rescued pups have completely penetrant palatal clefting and oral adhesions ..........................................................................................................................220 Figure 25 : Obliteration of the esophageal lumen contribut es to postnatal lethality ........222 Figure 26 : Tongue -palate oral adhesions obstr uct palatal developm ent..........................224 Figure 27 : O ral adhesions to the tongue prevent re-orientation and apposition of palatal shelves ÉÉ.........ÉÉÉÉÉÉÉÉ............................ÉÉ.........ÉÉÉÉÉÉÉÉ226 Figure 28 : Summery of genetic network in orofacial and neural tube development É...248 Figure 29 : Proposed IRF6 gene regulatory network É....................ÉÉ......ÉÉÉ.....249 $"## !!!"#$%& !ABBREVIATI ONS PPS.......................................................................................popliteal pterygium syndromes BOFS ...................................................................................Branio -Oculo -Facial Syndrome GRHL3 ..................................................................................................Grainy -Head Like 3 LOD ..................................................................................!Logarithm (base 10) of the odds REN .....................................................................................................................Renin gene TDT................................................................................Transmission Disequilibrium Test SNP.................................................................................Single Nucleotide Polymorphisms Trp63 ..............................................................................Transformation Related Protein 63 TP63 .............................................................................................Tumor Related Protein 63 EEC.......................................Ectrodactyly, Ectodermal Dysplasia and Cleft lip and Palate R84C...................................................................................................Arginine 84 Cytosine Maspin ..........................................................................Mammary Serine Protease Inhibitor SCC............................................................................................. Squa mous Cell Carcinoma HNSCC ..............................................................Head and Neck Squamous Cell Carcinoma TSS...................................................................................................Transcription Start Site Irf6gt/+ ......................................................................................... Irf6 gt/+ Irf6 gene trap allele Irf6 R84C/+ ...................................................Arginine 84 Cytosine mutation in a mouse allele Irf6 Clft1/+ ...........................................ENU forward screen leading to mutation at Proline -39 Ikka ..........................................Inhibitor of Nuclear Factor Kappa -B Kinase subunit alpha TP53 ...........................................................................................Transformation Protein 53 G2/M ..............................................................Growth 2/Mitosis DNA Damage Checkpoint $"### !!NF-kB ..............................Nucleor Factor Keppa -light -chain -enhancer of activat ed B cells IKK ......................................................................................................................IkB Kinase Ripk4 .........................................................................Receptor -Interacting Protein Kina se 4 Kdf1 ...........................................................................Keratinocytes Differentiation Factor 1 BPS.............................................................................................Bartsocas -Papas Syndrome Krt1 .........................................................................................................................Keratin 1 Krt6 ........................................................................................................................Kera tin 6 Krt14 .....................................................................................................................Keratin 14 Celsr1 ...................................................Cadherin, EGF, LAG seven -pass G -type receptor 1 GRN...............................................................................................Gene Regulator Network !"" Chapter 1 - Irf6 regulates development and differentiation in multiple ectodermal lineages #""Historical Clefts of the lip and palate have afflicted humanity since the dawn of civilization (1). According to Cervenka et al (1966) orofacial clefting with lip pits were first described in 1845 as a Òvery rareÓ congenital malformation by Demarquay (2). In 1947 Test and Falls described lip pits, cleft lip and palate syndrome in five generations of an affected family (3). Dr. Anne van der Woude, then at the University of Michigan, summarized the literatu re in 1954 and concluded an autosomal dominant inheritance pattern (4). From this point forth, l ip pits along with orofacial clefting became k nown Van der Woude Syndrome (VWS). In 1962 Dr. Levy highlighted the presence of lip pits, symmetrically located on the medial edge of lower lips , pathognomonic feature of VWS (5). The earlies t recorded survey of VWS prevalence among all forms of cleft lip and palate took place in 1971 when Dr. Dronamraju reported that eight of 260 clefting families (3%) had VWS (6). The incidence of VWS has been directly estimated at 3.6/100,000 live births (7). Variable expressivity suggests tha t VWS is a clinical model for iCLP Variable expressivity and incomplete penetrance continue to be important aspect of VWS. Dr. Baker studied one affected family and concluded variable expressivity as some presented with lip pits but not orofacial clefting (8). Cervenka et al reported 80% penetrance in 25 VWS cases, with lip pit present in 69.6 % and clefting present in 36.0% of individ uals (2). In 1980 Janku et al. (1980) (9) reported that penetrance was closer to 96.7%, with lip pit present in 88% and cleft ing present in 21% of individuals. Burdick et al (1985) examined 864 individuals from 164 families and found that VWS penetrance ranged from 89% to 99% using different methods, with cleft lip and palate occurring $""more commonly than cleft palate (10). Hypodontia, bifid uvula, hypernasal voice, lip mounds that secret mucus instead, Hirshsprung disease , congenital he art defects, popliteal webs, limb anomalies and accessory nipples have also been associated with VWS (4, 9, 11-13). By either estimate, a portion of VWS families has clefting but not lip pits. As such, patients wit h VWS may only present with orofacial clefting, mimicking patients with isolated cleft lip and palate (iCLP). ICLP is a common, complex disease with multiple genetic and environmental contributing factors (14). Considering the phenotypic overlap, Murray et al (1990) hypothesized that discovering the genetic etiology of VWS may also elucidate the genetic architecture of iCLP (15). However, unlike iCLP , VWS does not exhibit sex -specific differences (10, 16, 17). Variable expressivity in V an der Woude Syndrome may result from locus heterogeneity, different types of mutations (mis -sense vs. truncation), location of mutation (DNA Binding Domain vs. Protein Interaction Domain vs. Activation Domain), affect on protein location (sequestration in sub -cellular organelle), affect on protein activity or stability (resistance to degradation/activation/turnover), regulation of expression (enhancer , promoter ), or could be environmental modifiers. Consistent with this, prior work showed that 17p11.2 -11.1 increased VWS clefting risk (18). However, a more recent hypothesis -driven search for common variants that modify VWS did not yield a formally significant association (19). Understanding the basis for this phenotypic variation may provide preventative strategies to reduce disease severity. For example, if VWS expressivity could be limited to lip pits, disease burden could be significantly reduced. %"" While the phenotypi c evidence suggests that VWS can be a clinical model for iCLP, is there evidence that VWS could also be a genetic model for iCLP? As a Mendelian disorder, family studies are a crit ical component in VWS research (2, 20, 21). Houdayer et al (2001) used a family design to ask if VWS and iCLP were associated in a parametric linkage analysis and Transmission Disequilibrium Tests (TDT) (22). While parametric linkage was not suppor tive, TDT provided evidence for a ge netic link between VWS and iCLP (22). TDT measures the over -transmitted allele from parents to affected offspring and as such is robust to population structure, e.g. population s tratification. Clinical context and impact of orofacial cle fting As a result of life -long utilization of medical resources and loss of productivity, each individual born with a CLP will require $200,000 for medical treatment (23). Despite these enormous resources , surgical and cli nical intervention are often inadequate and result in physical and psychological sequelae (24). In addition, individuals born with a CLP have an increased risk for cancer (25) and neurologica l (11, 17, 26-31), musculoskeletal and cardiovascular diseases (17). Importantly, individuals born with cleft lip and palate had an increased risk of mortality between birth and 55 years of age (32). In addition, recent studies have shown that patients with VWS have an increased risk of surgical c omplications after cleft repair (33). As a r esult, understanding the gene regulatory network leading to orofacial clefting and these associated phenotypes may lead to preventative strategies and therapies for other disease. &""Orofacial clefting results from defective palate and lip closure between t he 6 th and 10 th week of human gestation. First, and foremost, prevention is possible in at least a subset of cases because reduced maternal folate, alcohol consumption and ma ternal smoking can contribute risk toward iCLP (34, 35). Current standard of care for children born with a cleft lip and p late includes surgical closure of a cleft lip by 4 months and closure of the palate by 12 months. Closure of alveolar clefts with bone grafts should be complete by 11 years o f age, correction of residual abnormalities by 12 years of age and final nasal contours and breathing problems by 17 years of age. In addition to these surgeries, children born with CLP need to undergo speech therapy until the age of 11. According to the A merican Society of Plastic Surgeons, children born with CLP require a multi -disciplinary team to receive appropriate care. This team includes a pediatrician, pediatric dentist, otolaryngologist, auditory specialist, speech pathologist, genetic counselor an d a social worker. Costs include surgical procedure, hospital, anesthesia, medication, garments and devices and clinical tests. However, recent work also suggests brain anomalies in patients with orofacial clefting, suggesting that additional resources may still be required (11). Despite team -based medical treatment and the enormous cost, several risks are associated with CLP repair. These commonly include bleeding, infection, irregular healing of scars and puckering of tissues (contractures), asymmetries and remaining deformities, anesthesia risks, allergies to suture material and glue, damage to deep structures, such as blood vessels, nerves and muscles, and possibility of surgery revision. In addition, changes in nose shape and teeth alignment may result from cleft repair. Finally, teeth '""abnormalities associated with CLP may require additional repair (36). Finally, because the mouth and palate are integral tissues, CLP morbidity also includes poor feeding, growth retardation and repeated ear infections. Considering t he complications and the number of healthcare providers need for treatment, there has been a shift of CLP repair to teaching hospitals and an associated increase in c ost (37). While o ngoing clinical investigation in CLP treatment has led to a dramatic decrease in the associated morbidity (38), the challenges are even greater for developing countries (39). The chall enges highlighted above illustrate the need for prevention, rather than treatment of CLP. Folate and multi -vitamin use are playing a role in reducing disease risk and burden (35, 40). Mutations in IRF6 cause VWS The first study into the genetic etiology of VWS was by Schneider (1973) (41). Here, the author used the ÒR ed Blood Cell antigenÓ for genetic linka ge as well as several biochemical assays, including electrophoretic studies of glucose -6-phosphate -dehydrogenase, hapatoglobin, phosglucomutase, and haemoglobin (41). A subsequent study examined additional genetic markers and several biochemical assays but also report ed a LOD (logarithm of odds) as a measure of linkage (42). Spence et al. (1983) reported on VWS linkage using ten genetic markers, including three on chromosome one, but did not find a plausible genetic link (43). Wienker et al. (1987) studied 27 informative polymorphic markers and excluded several based on an analysis of a five -generation kindred. Interestingly, despite a paucity of markers examined, the authors reported positive linkage to the ÒVWS:DuffyÓ antigen , located on chromosome 1 (44). (""A critical finding that further refin ed the VWS locus came with the discovery of a cytogenetic anomaly by Bocian and Walker (45). In their rep ort, an interstitial deletion of chromosome 1 at q32 -q41 was found in a 41 -month -old girl of Polish -German descent who had lip pits. Murray et al. (1990) used a candidate -gene -and -region approach to study multiple generations in six families with VWS. The authors successful identified linkage with the renin (REN) gene and the D1S65 locus in 1q using Restriction Fragment Length Polymorphisms, providing a LOD score of 10.83 (15). Additional studies confirmed linkage between REN and VWS (46). A report by Sander et al. (1994) showed a VWS family with deletion of D1S205, a highly polymorphic microsatellite within the 1q32-41 region (47). The observed microdeletion led to refinement of the VWS locus to an approximate 4.1 Mega base pair stretch within 1q32 -41. Additional work by Houdayer et al (1999) supported the locus homogeneity of VWS and the chromosomal location reported previously (48). Another family allowed Schutte and colleagues to map the VWS l ocus to a 1.6 Mega base pair interval between D1S491 and D1S205. Cloning of the critical region allowed the production of a single YAC clone with an 850 Kb segment containing the microdeletion (49). Subsequent studies expanded the region (50) and work by Schutte et al. (2000) provided a 900 Kb gene map (51). The gene map included 11 novel and four previously described genes, along with nine putative genes (51). Additional mapping further confirmed these findings (52, 53). Several important twin studies have contributed to our understanding of VWS (54). While zygosity was not determined, t he VWS twins were discordant : one had lip pits and a cleft palate while the other had preauricular skin tags (54). Dizygotic twins discordant )""for the VWS were described by Levy et al (1962), with one having lip pits and cleft palate while the other had a likely unrelated hemangioma (5). Cervenka et al (1963) characterized the first published description of concordant twin s for VWS, who were the n identified as Òprobably monozygoticÓ based on facial features and blood typing (2). While these twins had lip pits, only one had a unilateral cleft lip. Another monozygotic twin concordant for VWS came nearly 30 years later when Hersh and Verdi (1992) showed siblings with unilateral cleft lip and palate, and lip pits (55). At this point , four monozygotic twins concordant for VWS were reported (56, 57). Of these studies, Jobling et al (2011) was significant because it showed monozygotic twins who were concordant for VWS but had highly dissimilar features, with one twin showing lip pits only while the other sibling also showed a cleft lip and a cleft palate (56). Thus, multiple examples of monochorionic, diamniotic twins with variable expressivity are seen, suggesting that somatic mutations or stochastic interactions may be a common feature of VWS expressivity. Nearly 150 years after Demarquay first described a clefting syndrome with lip pits, Kondo et al (2002) used discordant monozygotic twins to discover Interferon Re gulatory Factor 6 (IRF6 ) as the VWS gene (21). Specifically, t arg eted sequencing in the twins found mutations in IRF6 in the affected individual but not the unaffected sibling . Prevalence screening in 45 unrelated VWS families further showed both truncation and point mutations in IRF6. In addition, analysis of three families with popliteal pterygium syndromes (PPS) showed linkage to the Van der Woude Syndrome locus, at 1q32 -q41 (58, 59). PPS, like VWS, can include orofacial clefting, lip pits, hypodontia and skin *""anomalies. However, unlike VWS, PPS also includes webbing in the back of knee (popliteal fossa), genital anomalies (hypoplasia of the labia majora, cryptorchidism or bifid scrotum), webbing between toes or fingers (syndactyly), triangular folds of skin over nails and tissue connecting the upper and lower eyelids (ankyloblepharon) (60-62). Seeing linkage to the same locus and p henotypic similarity, IRF6 was sequenced and mutations found in 13 families with PPS (21). Both protein truncation s and substitutions were found throughout the IRF6 open reading frame in patients with VWS . In contr ast, the preponderance of mutations leading to PPS are single nucleotide substitution in exons three and four, the DNA binding domain (21, 63). Since identification, numerous replication studies found IRF6 mutati ons in Van der Woude Syndrome families displaying a broad phenotypic spectrum and geographic al distribution (64-86). While most genetic studies identified point mutations in IRF6 , deletions a s large as 2.98 Mb, involving 25 genes , have also been reported in VWS (47, 50, 84). Considering genomic deletions , haploinsufficient etiology is strongly supported . Furthermore, phenotypic variation resulting fr om deletions relative to point mutations suggest s that additional genes or regulatory sequences in 1q41 -q32 may be interacting with IRF6 . de Lima et al (2009) conducted a comprehensive study of IRF6 mutations leading to VWS and found that 80% of newly disco vered disease causing mutations were found in exons 3, 4, 7 and 9 (80). This information may guide targeted sequencing for IRF6 mutations in patients with VWS and PPS . Importantly, de Lima et al (2009) did not find IRF6 mutations in 30% of VWS families. However, prior work suggested locus heterogeneity, linking a large Finnish VWS family to 1p34, instead of 1q32 -q41, where !+""IRF6 is located (87). The second VWS locus at 1p34 contains nearly 700 genes, necessitating a targeted approach for elucidation of a second VWS gene. Rorick and colleagues (2011) provided evidence suggesting that WDR65 was a candidate gene and identified one disease -associated variant (88). At this point, it is not clear if this variant is etiologic. Variants within IRF6 are associated w ith iCLP Considering that mutations in IRF6 lead to VWS, and that VWS is a clinical model for iCLP, is it possible that IRF6 may also be contributing risk to isolated orofacial clefting? Consistent with this rationale, t hree recent studies show a robust li nk between IRF6 and isolated orofacial clefting (89-91). The first , by Zucchero et al (2004) , show that a non -synonymous substitution (V274I) within IRF6 is associated with 12% of all orofacial clefting (89). Considering that V274I is the ancestral allele, the association with cleft lip and palate seems counter -intuitive because clefting is lethal in all non -human primates . Furthermore, V274I does not alter the protein -codin g sequence of IRF6. Instead, the authors predicted that V274I is in Linkage Disequilibrium (LD) with the etiologic variant. Consistently, sequencing of highly conserved regions within this LD block (140 Kb in length) revealed an association to a non-coding variant 9.7 Kb upstream of the IRF6 transcription start site (91). In contrast to V274I, the single nucleotide poly morphism (SNP), rs642961 is not ancestral . Furthermore, rs642961 lies within a 608 bp sequence tha t is highly conserved and has enhancer activity that highly recapitulates endogenous IRF6 expression in vivo (92). The variant abrogates one of four binding sites for Transcription Factor Activating Protein 2 alpha (TFAP2a). However, functional !!""studies in cell culture on this associated variant did not reveal mechanistic insights. Plausibly, but untested, the variant reduces IRF6 enhancer activity because the trans -activating factor (TFAP2a) binds less robustly, alt ers endogenous IRF6 expression and therefore increases risk for a loss -of-function disease known to result from haploinsufficient IRF6 mutations. Interestingly, while the pathogenic affect of rs642961 is unaltered with prenatal vitamins , two variants wit hin the IRF6 locus do interact with prenatal multi -vitamin supplementation (93-95). These data suggest that both environmental and non -environmental pathways may be associated with IRF6 function in orofacial develo pment . If IRF6 and environment interact in iCLP, they may also interact in VWS. Environmental interaction with IRF6 in VWS and iCLP may be leveraged to alter disease penetrance and/or expressivity. IRF6 Interferon Regulatory Factor 6 is a member of the I RF family of transcription factors, which are widely known to regulate innate and adaptive immune function (96). Unlike other members of IRF family, IRF6 regulates orofacial, skin and limb development. IRF6 is composed of nine exons, with the start codon in exon th ree and a stop codon in exon nine (97). Kondo et al. (2002) detected two IRF6 transcript s (one at 4.4 Kb, the second larger) with Northern Blot analysis from whole mouse embryos from E4.5 to E18.5, with apparent di fferential regulation (21). Spatial dimension of Irf6 expression is seen in the brain, eyes, heart, liver, lung, placenta, skin, testes and tongue but not the spleen (21). !#""During palatal development, Irf6 expression is seen most robustly in the oral epithelium, which includes the periderm, a flat epithelial layer that envelops embryos, and the basa l cell layer, a cuboidal epithelium that lies beneath periderm cells but is su perficial to the dermis (98). IRF6 protein is 467 amino acids (97, 99) and western blot analys is shows two bands for IRF6; with one at 59 and the other at 63 kDa, a phosphorylated form of the protein is found in cell culture and murine mammary epithelium (97, 100). Consis tent with IRF6 function in human orofacial development, Irf 6 expression is seen in murine palatal epithelium from E12.5 to E17.5 (Fig. 1) . Two cell types that constitute the early oral epithelium and express Irf6 are the periderm and the basal cell layer. The periderm is a flat, squamous monolayer that coats the palatal shelves and may be prevent ing pathological oral adhesion to surrounding oral structures, including t ongue, maxilla and mandible. B asel epithelial cells are a cuboidal monolayer early in deve lopment (12.5 -E13.5) but proliferate to give rise to the periderm and other intermediate cell types (E12 .5-E15.5) and undergo cell death to allow fusion . The palatal shelves start as mesenchymal buds covered by epithelium at E12.5. During the next 24 hours, the palatal shelves, mesenchyme and epithelium, proliferate to expand, to inhabit the space between the tongue and mandible bilaterally . Between E13.5 and E14.0, the palatal shelves e levate and pivot toward midline, ultimately apposing above the tongue. At E14.5, periderm cells along the medial contact points are lost, allowing the basal cell s to from the medial edge to adhere and then to interdigitate (adhere) . Loss of the medial edge epithelium allows a mesenchymal bridge to form , generating the nasal a nd oral cavities (E15.5 - E17.5) (99). !$"" Unlike the fused mouse palate, Irf 6 expression is not ob served in the naturally cleft chick palate (99). To date , no study has directly tested if knocki ng down Irf6 in the mouse or over -expressing it in the chick is necessary and sufficient for palatal development. Irf6 expression is also seen at the fusion point of the lateral and medial nasal processes and the maxill ary processes, which fuse at E11.5 in the mouse to for m the upper lip (99). IRF6 structure -function in development and d isease Base d on the crystal structure of IRF1, IRF6 appears to contain a highly conserved penta -tryptophan winged -helix -loop -helix DNA binding domain in exons three and four (Fig. 2 ) (21, 101). While IRF 6 is structurally char acterized as transcription factor, it is mainly detected in the cytoplasm and rarely visualized in the nucleus with various antibodies. Therefore, d oes IRF6 transcriptionally regulate downstream targets and is this activity important for development? If so , why donÕt we find IRF6 in the nucleus? Several lines of evidence suggest that IRF6 binds DNA an d transcriptionally regulates gene expression in critical developmental pathways. First, i njection of cDNA containing the IRF6 DNA binding domain (dominant neg ative construct) leads to more s evere developmental defects than knocking down the whole transcript (morpholino) in zebrafish and xenopus embryos (102, 103). These data suggest that while embryonic development is g rossly resistant to some perturbations in IRF6 dose, cDNA construct s that directly or indirectly affect IRF6 DNA binding result in more severe developmental defects. Furthermore, in mice and humans, mutations in the DNA binding domain of IRF6 lead to more severe developmental phenotypes . For example, i n humans , a mutation in the DNA binding !%""domain, R84C , is associated with more severe developmental defects than deletion mutations (80). As a murine allele, R84C heter ozygous embryos have more severe oral adhesions than embryos heterozygous for the gene trap (null) allele (104, 105). Unlike physiological adhesions or fusions between apposing surfac es of the palatal shelves, path ological adhesions or fusions occur between the palate and mandible or palate and tongue or mandible and maxilla. Biochemically, R84C appears to reduce IRF6 DNA-binding affinity (101) and a concomitant reduction in transactivation of a luciferase reporter is observed (106). In support of transcriptional regulation , we recently showed that irf6 transcriptionally regula tes grhl3 in zebrafish embryos via a highly conserved binding element. Consistently, grhl3 mRNA partially rescues zebrafish embryos injected with a dominant negative irf6 (103). In primary human keratinocytes, a genome wide screen showed that IRF6 binds within this highly conserved element and that knocking down IRF6 leads to a reduction of GRHL3 expression (107). During palatal development, we found that Irf6 is require d for Grhl3 expression in the epithelium and oral periderm (103). Considering that IRF6 is required for palatal development and that IRF6 regulates GRHL3 in oral epithelium, we predicated that mutations in GRHL3 could also contribute to orofacial clefting. Considering that GRHL3 is at 1p36 and linkage to a second VWS loc us at 1p34, we performed exome sequencing and examined this region for mutations. Consistently, we found GRHL3 mutations only in the affected members of the pedigree. In addition, screen VWS families negative for IRF6 mutations (30%, N=45) and discover sev en families with GRHL3 mutations. In vivo assays showed that VWS associated GRHL3 !&""mutations disrupted endogenous gene function, suggesting dominant negative function . In the mouse, loss of Grhl3 leads to neural tube defects (108). In the oral cavity, loss of Grhl3 leads to bi-lateral oral adhesions and palatal clefting. However, compared to loss of Irf6 in the mouse, the oral adhesions were less severe a nd the cleft was less penetrant. These results are consistent w ith Grhl3 working downstream of Irf6 in palatal development. Embryos heterozygous for Irf6 had oral adhesions at the tooth germ and embryos heterozygous for Grhl3 had oral adhesions and fusions posterior to the tooth germ. Embryos doubly heterozygous for Irf6 and Grhl3 had a combination of both phenotypes at and more posterior to the tooth germ but neither was more severe, suggesting function in the same cell types and time point if not the same location (Chapter 2 ). Together, these data suggest that IRF 6 transcriptionally regulates critical genes and tissues during development in multiple species. While th ese results highlight Irf6 transcriptional activity , recent work has also sho wn that IRF6 directly binds fewer than 2,200 genes (107). In contrast, TP63 , a transcription factor co-expressed with IRF6 throughout epithelial developme nt, binds over 7,500 targets (109). While differences in peak threshold may account for the number of bindi ng sites reported, only 2.6% (56 /2177) of putative downstr eam targets bound by IRF6 were altered with knockdown (107). In contrast, 1,213 genes bound by TP63 are differential expressed with knockdown studies (109). Therefore, as a transcription factor, IRF6 does not seem to have robust transcriptional activity. Lack of nuclear staining , considering similar results with multiple antibodies, may therefore result from relatively minor transcriptional activity . However, more rapid IRF6 turnover via the proteasome (110) or !'""cytoplasmic sequestration and exocytosis (100) may also be contributing to the nuclear /cytoplasmic localization ratio. Examining IRF6 expression in multiple tissues and time points along with i nhibition of nuclear export with Leptomycin B may further elucidate the tendencies and targets of IRF6 . In fact, the majority of genes affect ed by IRF6 perturbation are not bound IRF6 (83%, 276/332) . A useful example to contrast is TP63. In humans, mutations in TP63 can lead to Ectrodactyly, Ectoder mal Dysplasia (EEC), which includes CLP (OMIM #604292) . While T P63 drives IRF6 expression, IRF6 seems to be post -translationally targeting TP63 for degradation via the proteasome (111), in a negative feedback loop critical f or palatal development (112). In the mouse, embryos doubly heterozygous for Trp63 and Irf6 can develop a cleft palate. Considering cytoplasmic localization and regulation of TP63 protein stability, post -translational regulation by IRF6 seems increasingly important . Protein -protein interaction by IRF6 leading to Trp63 degradation are likely to be mediated by a less well -conserved protein -binding domain in exon s seven and eight (97). Protein -protein interaction by IRF6 have also been shown with the Mammary Serine Protease Inhibitor (Maspin) (97). However, in contrast to the inhibitory affect on TP63, IRF6 cooperatively binds t o Maspin to regulate differentiation in mammary epithelium. In fact, transient re -expression of IRF6 reduced breast cancer invasiveness. In the skin, loss of IRF6 is associated with squamous cell carcinoma (107, 113). Unlike breast cancer, this may result from an increase in TP63, which is a proliferative factor in the epidermis (107). IRF6 also transcriptionally regulates OVOL1, a transcription factor regulating epithelial differentiation and a repressor of the oncogenic protein c-Myc (107, !(""114). Mutations in IRF6 are als o found in 5% of patients with head and neck squamous cell carcinoma (HNSCC) (115). While epithelial origin suggests oncogenic similarity to the epidermis, the regulatory partners and pathway of IRF6 in HNSCC are undetermined. Aside from TP63 and Maspin , little is known about protein -protein interactions mediated by IRF6 . Important targets for future work include the E3 ubiquiti n ligases that regu late and are regulated by IRF6 . Skin development, like palate developme nt, includes both epithelium (known as epidermis in skin) and an underlying mesenchyme (known as dermis in skin) . Like palatal development , the early epidermis (E9.5 Ð E12.5) include s squamous periderm cells, marked with Krt6, and cuboidal basal cells, marked with Krt14. While the periderm persists, basal cells of the epidermis proliferate between E13.0 and E16.5, leading to four cell types. Starting from the dermis and ending at the visible skin layers , basal cell s are marked by Krt14 and Trp63, supraspinous cells are marked by Krt1, granular cells are marked by Loricrin, and cornified cells are marked by Krt6. As the epidermis develops, so does the dermis, leading to embryonically ma ture skin that acts as a permeability barrier by E17.5 . Cells retaining periderm characteristics are found as late as E17.5 , but are eventually sloughed off prior to birth . Early in cutaneous development (E9.5 -E12.5), Irf6 is expressed in both the periderm and basal cells. From E13.0 to E16.5, Irf6 expression is also seen in the intermediate supra -spinous cells. In embryonically mature epithelium (E17.5), Irf6 expression is primarily seen in the spinous cell layer and, to a lesser degree, the basal and gran ular cells. Pathological histological and molecular changes observed in Irf6 knockout embryos include ectopic K rt14 and Trp63 expression, !)""proliferative supra -basal cells and loss of terminal differentiation (105). Consistently, loss of IRF6 leads to reduced basal cell differentiation and results in a hyperproliferative epidermis that may give rise to squamous cell carcinoma (107). One model to explain these results would include either asymmetric Irf6 deposition into a daughter basal cell s to dr ive differentiation or de novo Irf6 expression in an otherwise pre-programmed daughter cell . Together, these data suggest that Irf6 expression drive s differentiation and stratification of the epidermis from basal to spinous to granular cells. In the c -terminus, exon nine is serine rich and appears to harbor the regulatory domain for IRF6 in mammary epithelium (97). The generalization of an IRF6 c -terminal activation/repression domain seems plausible considering a n analogous domain in IRF3 and IRF5 (116). In addition to biochemical and structural results consistent with a regulatory domain, this c -terminal domain in IRF3 and IRF5 is highly sensitive to mutagenesis (116). Like IRF3 and IRF5, activation of IRF6 most likely results from phosphorylation at amino acid 416 (97). In zebrafish (Rob Cornell, unpublished data) and human keratinocytes, creation of a phosphomimetic IRF6 by converting ser ine and threonine to aspartic ac id leads to constitutive activation and nuclear localization (106). The proteins regulating IRF6 post -translational activation/repression are unknow n. Exons five and six appear to encode a less conse rved proline -rich region. In direct cont rast to the distribution of the DNA binding domain , of 19 etiologic mutations found in exons five and six, 16 are protein trunctions . Underrepresentation of missen se mutations along with less conservation, suggests that most coding changes in exons five !*""and six rarely led to orofacial clefting. We caution, however, against the assumption that paucity of mutations suggests a non -functional domain. Rather, this domain may be associated with other developmental processes and phenotypes. Regulating IRF6 e xpression Regulation of IRF6 expression is implicated in two disease processes. Rahimov and collaborators (2008) discovered an IRF6 enhancer in a Multi -Species Conserv ed Sequence 9.7 Kb upstream of the IRF6 Transcription Start Site ( MCS 9.7) (91). MCS9.7 highly recapitulates endogenous Irf6 expression in skin and oral epithelium (92). A DNA variant, rs642961, in the IRF6 enhancer is associated with isolated cleft lip and palate (CLP) but not cleft palate only (91). Biochemically, rs642961 abrogates one of four TFAP2 binding site s within MCS9.7. The Transcription Factor Activating Protein 2 (TFAP2) family of transcription factors is composed of five members that homo or heterodimerize to repress or activate gene expression through a common, conserved binding element . A role in palatal development is most clearly demonstrated for TFAP2a. Mutations in TFAP2a can lead to Branio -Oculo -Facial Syndrome (BOFS OMIM # 113620), a dominantly inherited orofacial clefting disorder that can also include malformation of the eyes, ears and skin (117). Like VWS, BOFS can also include lip pits. Similar to Irf6 knockout embryos (reviewed below) , loss of Tfap2 ! leads to severe craniofacial, limb and skin defects (118, 119). However, Tfap2 ! knockout embryos are unique in the biomedical literature for absence of a thoracic and abdominal body wall as well as neural tube defects . Facial clefting, which results from failed neural tu be closure, precludes analysis of palatal development in Tfap2 !-/- embryos . However, tissue -specific #+""deletion of Tfap2 ! suggests a requirement for palatal development that is independent of neural tube closure (120). In vitro and in primary human keratinocytes , TFAP2a binds to MCS9.7 (91, 109). TFAP2a and TP63 also appear to cooperativel y regulate IRF6 expression in primary keratinocytes (109). In vivo, Tfap2a regulates MCS9.7 and Irf6 expression in murine epidermis (Chapter 3) . As such, rs642961 may reduce TFAP2a trans -activation of IRF6 , contrib uting to orofacial clefting risk. Furthermore, l oss of Irf6 expression in Tfap2a knockout embryos lead s to pathological molecular changes associated with loss of Irf6 (105, 107) but not Tfap2a (109). As such, Irf6 may be more important for Tfap2 ! function than previously recognized. Considering numerous etiologic TFAP2a mutations in the DNA binding domain , altered trans -activation of IRF6 may also be contributing to skin, l ip pit and orofacial clefting in BOFS (Chapter 3) . In addition to TFAP2a, MCS9.7 also harbors bindings sites for TP63 and MAFB, the latter recently associated with CLP (111, 121). More recent work also suggests th at IRF6 is downstream of Notch signaling in keratinocytes . Notch appears to be acting through novel enhancers located 2.4 and 3.5 Kb upstream of the IRF6 TSS (122). Epigenetic regulation of IRF6 expression has also been documented . Bisulfite sequencing showed that a ~300 bp C pG island in the IRF6 promoter was methylated (107). Methylation of the IRF6 promoter represses expression and may increase risk for squamous cell carcinoma (107). IRF6 expression is also found in post -utero mammary gland development and exhibited apical localization fol lowed by luminal secretion into #!""milk (100). While regulation of this process is unknown, luminal secretion may be involved in delivering IRF6 to the neonate, e.g. IgG, or in purging it from maternal mammary tissue. Irf6 knockout mice There are three mouse lines for Irf6 . A g ene trap all ele (Irf gt/+ ), inserted 36 base pair into intron 1 , has several splice donor/ acceptor sites and stop codons , resulting in loss of Irf6 translation (105). Targeted insertion of R84C, the human mutation disrupting the D NA binding domain and most frequently leading to PPS (Irf R84C/+ ), led to the second murine allele (104). A more recent third allele (Irf6 clft1 /+), resulted from a forward genetic screen using N-ethyl -N-Nitrosourea (ENU) mutagenesis (123). Interestingly, this approach led to a missense mutation at Proline -39, which was previously reported in a VWS pedigree (21). IRF6 gene d eletions an d both mutations are found in human disease , providing robust tools to study human pathogenesis. Murine embryos that lack Irf6 has clubbed limbs, syndactyly, a bifid xiphoid, a shortened fused tail, palatal clefting and a grossly smaller head (104, 105). These embryos also have lingual (124) and mandibular defects (125). However, the most severe ly affect ed cell type is the epithelium. Epithelial abnor malities include a hyperproliferative epidermis that fails to differentiate, a permeable skin barrier , esophageal adhesions and pervasive oral adhesions. Intraoral adhesions seem to prevent palatal elevation leading to a cleft palate . While untested, the mechanical force generated by palate -tongue cohesion may be restr aining the palatal shelves vertically and preventing elevation and midline ##""reorientation . That is, the force normally generated by the palatal shelves during elevation may not be sufficient with oral adhesions. Genoty pe-phenotype correlation: From morphology to m olecule Loss of Irf6 is phenocopied by four other genetic knockouts; Stratifin (14-3-3sigma ), Ikka , Kdf1 , and, to a lesser extent, Rpik4 (126-129). 14-3-3sigma is a tumor suppressor protein that interact s with TP53 via a positive feedback loop to regulate the G2/M cell cycle checkpoint (130, 131). 14-3-3sigma also enhances Protein Kinase C activity an d contains a Pleckstrin homology domain, critical in protein -protein interaction with serine/threonine phosphorylation (132-134). Irf6 genetically interacts with 14-3-3sigma in skin, limb, craniofac ial and oral cav ity development. If the genetic interaction is direct, 14-3-3sigma may be involved in phosphorylation and post -translational activation of Irf 6. Mutations in 14-3-3sigma have not been assoc iated with syndromic human disease. However, hypermethylation of a CpG regulatory island reduced 14-3-3sigma expression in 91% of breast carcinoma cells (135), and likely constitutes an early oncogenic event (136). In humans, homozygous recessive mutations in Nuclear Factor Kappa -B Kinase subunit alpha ( IKKA) lead to Severe Fetal Encasement Malformation , also called Cocoon Syndrome (OMIM # 613630), which appears to include body wall, skin, limb and neural tube defects (137). IKKA, also known as CHUK , is a serine/threonine protein kinase that regulates the activation of NF -kB by marking its repressors (IkB Kinase ) for ubiquitin -mediated degradation . However, in the skin, Ikka is a tumor suppressor protein and #$""function s independent ly of NF -kB and IkB Kinase (138). Instead, Ikka works downstream of Tgfb signaling in a complex with Smad2/3 that allows nuclear translocation independent of Smad4 (139). In skin, Tgfb signaling regulates Irf6 (140), which in turn regulates OVOL1 (107), in a molecular cascade highly analogous to Ikka (139). In the palate , Tgfb signaling regulates Irf6 through Smad4 (141) but the mole cular context of Ikka in this tissue are less clearly delineated . Despite the phenotypic similarity in skin and palate, and the common upstream and downstream molecular targets, preliminary work appears to show that Ikka does not interact with Irf6 in the mouse (104). While analyzing Irf6 expression in Ikka knockout murine skin would further elucidate this point , testing epistasis in the mouse is highly specific but not sensitive , i.e. absence of proof is not proof of absence . As such, Ikka may be upstream of Irf6 in skin and palate development. Like 14-3-3sigma , the Receptor -Interacting serine/threonine P rotein Kinase 4 ( Ripk4 ) regulates keratinocytes differentiation and interacts with the Protein Kinase C (142). Like Ikka , Ripk4 activates NF-kB (143). However, Ripk4 knockout embryos appear to be the least severely affect ed of the group (144). In contr ast, human mutations in RIPK4 can lead to a lethal type of Po pliteal Pterygium Syndrome, called Bartsocas -Papas Syndrome (BPS) (OMIM # 263650). Like PPS , caused by mutation in IRF6 , BPS is associated with popliteal webbing, ankyloblepharon, cleft lip and p alate and syndactyly (145, 146). Like Cocoon Syndrome, caused by mutations in Ikka , BPS is associated with severe craniofacial defects, leading to superficial visualization of the nasal cavity, in what may be a for m of facial clefting. As of this writing, a test for epistasis between Ripk4 and each #%""of the other three murine models that it phenocopies, i.e. Irf6 , Ikka, and 14-3-3sigma , has not been reported. A recently discovered gene, Keratinocytes Differentiation Factor 1 ( Kdf1 ), like Irf6 , appears to interact with Trp63 and 14-3-3sigma in skin, limb and craniofacial development (128). While the molecular nature of Kdf1 is undetermined, cytoplasmic localization and a ssociation with the cellular membrane would suggest a signaling molecule. Rescuing the knockout phenotype Seeing multiple epithelial defects, several st udies have tried to rescue the knockout phenotype with epithelial specific promoters , i.e. KRT14 or KRT 5, to drive expression in basal epithelial cells . Using the KRT14 promoter to drive Ikka in Ikka knockout embryos led to rescue of skin, skeletal and limb defects (147). Skeletal and limb rescue is intriguing becau se it involves both cartilaginous and bony structures that lie beneath the epiderm al cells driving KRT14 , strongly suggesting Ikka non-cell autonomous function. However, unlike the skin and limbs, a curled tail persisted, suggesting additional cell autonom ous function for Ikka in neural tube development. Furthermore, pups did not feed, as suggested by absence of a milk mark in the abdomen. L ikely , esophageal adhesions occlude the gastrointestinal tract and result from minimal KRT14 promoter activity in basa l cells of the esophagus. In addition , using two different KRT5 transgenic lines to drive Ikka in Ikka knockout embryos leads to rescue of skin, limb and skeletal developmen t. However, rescue of tail morphology, classically a consequence of neural #&""tube clo sure, diverged more prominently between the two transgenic lines. While physiological expression of Ikka in the epidermis did not rescue tail development in one line, a super -physiological dose of Ikka completely rescue s the tail in the other . First, a s KRT14 and KRT5 are co -expressed intermediate filaments, these results suggest that at a certain dose, the KRT14 promoter may also rescue the curled tail noted above. More importantly, how is Ikka expression in the epidermis rescuing a curled tail? Is it thro ugh rescue of the epidermis or is it through non -cell autonomous signaling of Ikka in neural tube ? Considering highly similar epidermal rescues with both doses of Ikka yet divergent tail rescue, we favor a non -cell autonomous process. An analogous experi ment using the KRT14 promoter to drive Ripk4 in Ripk4 knockout pups rescue s skin defect s. As seen with Ikka , KRT14 spatio -temporal regulation of Ripk4 was not sufficient to rescue esophageal adhesions (129). In a c lever test for epistasis, epithelial expression of Ripk4 using the KRT14 promoter did not rescue Ikka and 14-3-3sigma knockout embryos. Considering less severely affect ed knockout embryos and failure to rescue loss of Ikka and 14-3-3sigma , Ripk4 may be in a parallel, but converging pathway or require both Ikka and 14-3-3sigma for function. In direct contrast to non-cell autonomous Ikka function in limb and skeletal development , using the KRT14 promoter to drive Irf6 only rescues epidermal defects. Importa ntly, while the skin grossly appeared taut, both histological and molecular analysis revealed complete rescue. Despite epidermal rescue, limb, skeletal , tail and craniofacial defects persisted. The limb are free of adhesion to the body wall but clubbing an d syndactyly #'""persisted. In the axial skeleton, a bifid xiphoid remained in rescue pups as seen in Irf6 knockout pups . Despite epithelial re -expression of Irf6 , palatal clefting is completely penetrant at P0. At E15.5, o ral adhesions persisted between the t ongue and palate and the mandible and maxi lla but these are less seve re. Dramatically, oral adhesions gripped the midline oriented palatal shelves to the tongue, physical ly restraining horizontal movement . Together, th ese data suggest that Irf6 function s in limb and skeletal development throu gh a cell autonomous mechanism (Chapter 4). In support of this model, we recently showed Irf6 enhancer activity in limb bone and cartilage development. In addition to genetic rescue, e xperimental embryonic gene therap y protocols to prevent disease in animal models have been developed for cystic fibrosis (148), Duchenne muscular dystrophy (149), Herlitz junctional epidermolysis bullosa (150, 151), Thrombotic thrombocytopenic purpura (152) and congenital blindness (153). Likewise, gene delivery to the oral epitheliu m and developing epidermis is highly feasible during development (154). In mature skin, epithelial stratification (cornified layer) and keratin secretion forms a physical barrier, leaving viral and bacterial pathog ens refractory to host penetration. However, during early embryonic development, a cornified layer is not present, leaving the tissue highly susceptible to transduction. As such, intra -aminotic injection of a viral vector with tropism to epithelial tissue may provide robust targeting. Like the epithelium covering the skin, epithelium covering the oral cavity is also highly amenable to transduction. Considering that embryonic development of the lip and palate occurs between the 6 th and 10 th week of human ges tation, such efforts could be highly #(""targeted using ultrasound to visualize the structures. Circulation of amniotic fluid in and through the embryos also ensures transduction of viral vectors into the oral cavity. Like numerous orofacial clefting genes, Irf6 pathogenesis results from abnormal epithelial development. As such , gene delivery approaches may provide a feasible therapeutic modality for multiple, single gene clefting disorders. Furthermore, transduction of the periderm layer, a cell type lost befo re birth, limits side effects on post -embryonic development. Finally, immune -privileged status of amniotic fluid limits innate and adaptive blunting of the therapy. " Neurulation: Neural tube d evelopment Like palate development, neural tube development is a highly orchestrated process that begins as flat epithelial layers followed by a period of proliferation to establish neural plates (E7.5 Ð E8.5 ) (Fig. 1) . Unlike the palate, convergent extension and a median hinge point allows the bilateral neural plate s to orient that growth toward a midline pivot (E8.5 Ð E9.5) . Also like palatal development, a dditional growth allows midline oriented neural plates to appose (E10.0). Adhesion of the neural plates is mediated by bilateral lamelipodial cell protrusions. Mi dline cell death and epithelial remodeling ultimately leads to fusion of the neural plates an d formation of the neural tube (reviewed fully in (155). As such, palate and neural tube development occur in a highly analogous manner. However, unlike palate development, neural tube developme nt is a highly complex process that involves multiple independent closure points. Defects in the rostral closure point can lead to anencephaly, as seen with Tfap2a knockout embryos . Defects in the caudal closure point can lead to an open lumbo -sacral defec t or a curled tail , as seen with #)""Grhl3 knockout embryos (108). Defects in the intermediate closure points can lead to a craniorachischisis, as seen with the Cadherin, EGF, LAG seven -pass G -type receptor 1 (Celsr1 ) knockout embryos (155). Unex pectedly, we found that over -expressing Irf6 lead s to rostral neural tube defect with variable penetrance and expressivity (Chapter 3) . While 6% of embryos over -expressing Irf6 had exencephaly, 5% of embryos had anencephaly, phenocopying Tfap2a knockout embryos. We further show that modulating Irf6 expression in vivo completely and negatively correlates with Tfap2a mRNA. Despite an increase in Tfap2a transcript, we also found that reducing Irf6 le d to a reduction of Tfap2a protein. Irf6 is expressed in both the rostral and caudal neural plates, the neural tube and the non -neural ectoderm. Consistent with Tfap2a dose regulating neural tube development, we also found that 10% of Tfap2a heterozygous embryos have exencephaly. Finally, we show that reducing endog enous Irf6 in Tfap2a +/-;Irf6 +/- double heteroz ygous embryos completely rescues rostral neural tube defects seen with Tfap2a haploinsufficiency. In addition, we found that reducing Irf6 expression le d to a completely penetrant caudal neu ral tube defect, a curled tail. Like skin development, we show that Tfap2a regulates Irf6 in the caudal neural tube. Transcriptional profiling shows that Irf6 positively regulates both Tfap2a and Grhl3 in the caudal neural tube. These data suggest that Tfap2a interacted with Grhl3 via Irf6 . Consistent with this model, we found that 13% of Tfap2a +/-;Grhl3 +/- double heterozygous embryos have a curled. #*""Conclusion In summery, we show that Tfap2a regulates Irf6 , which in turn regulates Grhl3 . The link between Tfap2a and Irf6 may explain the pathophysiological process involved in 12% of all oral facial clefting risk. Similarly, the link between Irf6 and Grhl3 in zebrafish and mouse led to the discovery of an additional orofacial clefting gene and to the etiology underpinning previ ously documented locus heterogeneity in VWS . Similar to the pathway in orofacial development, we further show that Irf6 regulates neural tube development through genetic interactions with Tfap2a and Grhl3 . In the caudal neural tube, we show that Tfap2a int eracts with Grhl3 (Fig. 3) . Remarkably, we also show that Irf6 expression in epithelium is not sufficient to rescue palatal development and that oral adhesions can physically restrain mid -line oriented palatal shelve s from adhesion . Finally, despite well -documented roles for Irf6 in epithelium, we show that Irf6 rescue of skin is not sufficient to rescue limb , skeletal and tail development. While the role of Irf6 in tail development is orthogonally discovered in this work, these data strongly suggest multip le additional , as yet un -documented cell -autonomous roles for Irf6 in embryonic development. Broadly, this work is significant because it shows that orfacial cleft ing genes also play a role neural tube development, underscoring the commonality of mol ecular pathways stemming from common ectodermal lineages. Considering these results, future work should seek to identify the role of this pathway in skin caner. In addition, considering that multiple mouse models phenocopy the Irf6 knockout, the role of Ikka , 14-3-3sigma , Ripk4 and Kdf1 in neural tube development should be analyzed. $+"" APPENDIX $!""APPENDIX Figure 1: Analogous processes le ad to palate and neural tube development. Top row: palate development begins at E12.5 as a flat epith elium, in blue, with an underlying mesenchyme, in yellow. A period of rapid proliferation leads to formation of palatal shelves alongside the tongue (t) and mandible (m). Reorientation of the palatal shelves leads to a midline pivot and a horizontal suspe nsion above the tongue. Apposition of the palatal shelves leads to adhesion, or interdigitation of the epithelial cells. Breakdown of that epithelium leads to fusion of the shelves, forming a mesenchymal bridge that separates the nasal from the oral cavit ies. Bottom row: Like palate development, neural tube development happens through a highly choreographed progress. Flat epithelial layers, including the non -neural superficial ectoderm (blue) and neural plate (white, NP) undergo a period of rapid prolifera tion to expand. Neural tube specific processes (including a median hinge point and convergent extension, not shown) provide direction. Like palate development, a pivot toward midline is followed by adhesion. Breakdown of the epithelium leads to fusion. $#"" Figur e 2: Structure of IRF6 . Exon one and two are not translated (orange). Exons three, beginning of translation, and four are the DNA binding domain of IRF6 (blue). Exons five and six are less highly conserved (yellow). The majority of exon seven and eight mark the Interferon Association Domain, or the protein binding domain of IRF6 (green). Exon nine is less highly conserved and includes a c -terminal helix thought to regulate IRF6 activation and repression (yellow). The 3ÕUTR of IRF6 is shown in orang e. $$"" Figur e 3: Proposed genetic network for orofacial and rostral , caudal neural tube development. Top : Orofacial development is likely to proceed through a negative feedback loop between Irf6 and Tfap2 a that lies upstream of Grhl3 . Middle: Rostral neural tube development is likely mediated through a negative feedback loop between Tfap2a and Irf6 . A role for Grhl3 in this rostral neural tube pathway has not been investigated. Bottom: Caudal neural tube development is mediated through a positi ve $%"" BIBLIOGRAPHY $&""BIBLIOGRAPHY 1 Hoffman, H. and Hudgins, P.A. (2002) Head and skull base features of nine Egyptian mummies: evaluation with high -resolution CT and reformation t echniques. AJR Am J Roentgenol , 178, 1367-1376. 2 Cervenka J, G.R., Anderson VE (1967) The syndrome of pits of the lower lip and cleft lip and/or palate. Genetic considerations. Am J Hum Genet. , 19, 416-432. 3 Test, A.R. and Falls, H.F. (1947) Dominant inheritance of cleft lip and palate in five generations. J Oral Surg (Chic) , 5, 292-297. 4 Van Der Woude, A. (1954) Fistula labii inferioris congenita and its association with cleft lip and palate. Am J Hum Genet , 6, 244-256. 5 Levy, J. (1962) [Twins in a family with lower lip malformation]. Acta Genet Stat Med , 12, 33-40. 6 Dronamraju, K.R. (1971) Genetic studies of a cleft palate clinic population. Birth Defects Orig Artic Ser , 7, 54-57. 7 Burdick, A.B. (1986) Genetic epidemiology and control of genet ic expression in van der Woude syndrome. J Craniofac Genet Dev Biol Suppl , 2, 99-105. 8 Baker, B.R. (1964) A Family with Bilateral Congenital Pits of the Inferior Lip. Oral Surg Oral Med Oral Pathol , 18, 494-497. 9 Janku, P., Robinow, M., Kelly, T., Bral ley, R., Baynes, A. and Edgerton, M.T. (1980) The van der Woude syndrome in a large kindred: variability, penetrance, genetic risks. Am J Med Genet , 5, 117-123. 10 Burdick, A.B., Bixler, D. and Puckett, C.L. (1985) Genetic analysis in families with van de r Woude syndrome. J Craniofac Genet Dev Biol , 5, 181-208. 11 Nopoulos, P., Richman, L., Andreasen, N.C., Murray, J.C. and Schutte, B. (2007) Abnormal brain structure in adults with Van der Woude syndrome. Clin Genet , 71, 511 -517. 12 Nopoulos, P., Richman , L., Andreasen, N., Murray, J.C. and Schutte, B. (2007) Cognitive dysfunction in adults with Van der Woude syndrome. Genet Med , 9, 213-218. 13 Shprintzen, R.J., Goldberg, R.B. and Sidoti, E.J. (1980) The penetrance and variable expression of the Van der Woude syndrome: implications for genetic counseling. Cleft Palate J , 17, 52-57. $'"" 14 Cobourne, M.T. (2004) The complex genetics of cleft lip and palate. Eur J Orthod , 26, 7-16. 15 Murray, J.C., Nishimura, D.Y., Buetow, K.H., Ardinger, H.H., Spence, M.A., Sparkes, R.S., Falk, R.E., Falk, P.M., Gardner, R.J., Harkness, E.M. et al. (1990) Linkage of an autosomal dominant clefting syndrome (Van der Woude) to loci on chromosome Iq. Am J Hum Genet , 46, 486-491. 16 Huang, J.J., Hou, J.W., Tan, Y.C., Chen, K.T., L o, L.J. and Chen, Y.R. (2007) Van der Woude syndrome: clinical presentation in 64 patients. Cleft Palate Craniofac J , 44, 649-652. 17 Calzolari, E., Pierini, A., Astolfi, G., Bianchi, F., Neville, A.J. and Rivieri, F. (2007) Associated anomalies in multi -malformed infants with cleft lip and palate: An epidemiologic study of nearly 6 million births in 23 EUROCAT registries. Am J Med Genet A , 143, 528-537. 18 Sertie, A.L., Sousa, A.V., Steman, S., Pavanello, R.C. and Passos -Bueno, M.R. (1999) Linkage analys is in a large Brazilian family with van der Woude syndrome suggests the existence of a susceptibility locus for cleft palate at 17p11.2 -11.1. Am J Hum Genet , 65, 433-440. 19 Leslie, E.J., Mancuso, J.L., Schutte, B.C., Cooper, M.E., Durda, K.M., L'Heureux, J., Zucchero, T.M., Marazita, M.L. and Murray, J.C. (2013) Search for genetic modifiers of IRF6 and genotype -phenotype correlations in Van der Woude and popliteal pterygium syndromes. Am J Med Genet A , 161, 2535-2544. 20 Vignale, R., Araujo, J., Pascal, G., Reissenweber, N., Abulafia, J., Quadrelli, R., Vaglio, A., Larrandaburo, M. and Reyno, S. (1998) Van der Woude syndrome. A case report. Pediatr Dermatol , 15, 459-463. 21 Kondo, S., Schutte, B.C., Richardson, R.J., Bjork, B.C., Knight, A.S., Watanabe, Y., Howard, E., de Lima, R.L., Daack -Hirsch, S., Sander, A. et al. (2002) Mutations in IRF6 cause Van der Woude and popliteal pterygium syndromes. Nat Genet , 32, 285-289. 22 Houdayer, C., Bonaiti -Pellie, C., Erguy, C., Soupre, V., Dondon, M.G., Burglen, L., Cougoureux, E., Couderc, R., Vazquez, M.P. and Bahuau, M. (2001) Possible relationship between the van der Woude syndrome (vWS) locus and nonsyndromic cleft lip with or without cleft palate (NSCL/P). Am J Med Genet , 104, 86-92. 23 Wehby, G.L. and Casse ll, C.H. (2010) The impact of orofacial clefts on quality of life and healthcare use and costs. Oral Dis , 16, 3-10. $(""24 Manna, F., Pensiero, S., Clarich, G., Guarneri, G.F. and Parodi, P.C. (2009) Cleft lip and palate: current status from the literature an d our experience. J Craniofac Surg , 20, 1383-1387. 25 Taioli, E., Ragin, C., Robertson, L., Linkov, F., Thurman, N.E. and Vieira, A.R. (2010) Cleft lip and palate in family members of cancer survivors. Cancer Invest , 28, 958-962. 26 Conrad, A.L., Dailey, S., Richman, L., Canady, J., Karnell, M.P., Axelson, E. and Nopoulos, P. (2010) Cerebellum Structure Differences and Relationship to Speech in Boys and Girls With Nonsyndromic Cleft of the Lip and/or Palate. Cleft Palate Craniofac J, 47, 469-475. 27 Boes , A.D., Murko, V., Wood, J.L., Langbehn, D.R., Canady, J., Richman, L. and Nopoulos, P. (2007) Social function in boys with cleft lip and palate: relationship to ventral frontal cortex morphology. Behav Brain Res , 181, 224-231. 28 Rosen, H., Chiou, G.J., Stoler, J.M., Mulliken, J.B., Tarui, T., Meara, J.G. and Estroff, J.A. (2011) Magnetic Resonance Imaging for Detection of Brain Abnormalities in Fetuses With Cleft Lip and/or Cleft Palate. Cleft Palate Craniofac J , 48, 619-622. 29 Shriver, A.S., Canady, J ., Richman, L., Andreasen, N.C. and Nopoulos, P. (2006) Structure and function of the superior temporal plane in adult males with cleft lip and palate: pathologic enlargement with no relationship to childhood hearing deficits. J Child Psychol Psychiatry , 47, 994-1002. 30 Conrad, A.L., Canady, J., Richman, L. and Nopoulos, P. (2008) Incidence of neurological soft signs in children with isolated cleft of the lip or palate. Percept Mot Skills , 106, 197-206. 31 Conrad, A.L., Dailey, S., Richman, L., Canady, J ., Karnell, M.P., Axelson, E. and Nopoulos, P. (2010) Cerebellum structure differences and relationship to speech in boys and girls with nonsyndromic cleft of the lip and/or palate. Cleft Palate Craniofac J , 47, 469-475. 32 Christensen, K., Juel, K., Hers kind, A.M. and Murray, J.C. (2004) Long term follow up study of survival associated with cleft lip and palate at birth. BMJ , 328, 1405. 33 Jones, J.L., Canady, J.W., Brookes, J.T., Wehby, G.L., L'Heureux, J., Schutte, B.C., Murray, J.C. and Dunnwald, M. ( 2010) Wound complications after cleft repair in children with Van der Woude syndrome. J Craniofac Surg , 21, 1350-1353. 34 Blanton, S.H., Henry, R.R., Yuan, Q., Mulliken, J.B., Stal, S., Finnell, R.H. and Hecht, J.T. (2011) Folate pathway and nonsyndromic cleft lip and palate. Birth Defects Res A Clin Mol Teratol , 91, 50-60. $)""35 Wu, T., Liang, K.Y., Hetmanski, J.B., Ruczinski, I., Fallin, M.D., Ingersoll, R.G., Wang, H., Huang, S., Ye, X., Wu -Chou, Y.H. et al. (2010) Evidence of gene -environment interaction for the IRF6 gene and maternal multivitamin supplementation in controlling the risk of cleft lip with/without cleft palate. Hum Genet , 128, 401-410. 36 Aizenbud, D., Coval, M., Hazan -Molina, H. and Harari, D. (2011) Isolated soft tissue cleft lip: epidem iology and associated dental anomalies. Oral Dis , 17, 221-231. 37 Basseri, B., Kianmahd, B.D., Roostaeian, J., Kohan, E., Wasson, K.L., Basseri, R.J. and Bradley, J.P. (2011) Current national incidence, trends, and health care resource utilization of clef t lip -cleft palate. Plast Reconstr Surg , 127, 1255-1262. 38 Mulliken, J.B. (2004) The changing faces of children with cleft lip and palate. N Engl J Med , 351, 745-747. 39 Furr, M.C., Larkin, E., Blakeley, R., Albert, T.W., Tsugawa, L. and Weber, S.M. (2010) Extending multidisciplinary management of cleft palate to the developing world. J Oral Maxillofac Surg , 69, 237-241. 40 van Rooij, I.A., Ocke, M.C., Straatman, H., Zielhuis, G.A., Merkus, H.M. and Steegers -Theunissen, R.P. (2004) Periconceptional fola te intake by supplement and food reduces the risk of nonsyndromic cleft lip with or without cleft palate. Prev Med , 39, 689 -694. 41 Schneider, E.L. (1973) Lip pits and congenital absence of second premolars: varied expression of the Lip Pits syndrome. J M ed Genet , 10, 346-349. 42 Eastman, J.R., Bixler, D. and Escobar, V. (1978) Linkage studies in Van der Woude syndrome. J Med Genet , 15, 217-218. 43 Spence, M.A., Glass, L., Crandall, B.F., Stewart, R.E., Miles, J., Falk, R.E., Field, L.L. and Sparkes, R.S . (1983) Genetic linkage studies with cleft lip and palate: report of two family studies. J Craniofac Genet Dev Biol , 3, 207-212. 44 Wienker, T.F., Hudek, G., Bissbort, S., Mayerova, A., Mauff, G. and Bender, K. (1987) Linkage studies in a pedigree with V an der Woude syndrome. J Med Genet , 24, 160-162. 45 Bocian, M. and Walker, A.P. (1987) Lip pits and deletion 1q32 ----41. Am J Med Genet , 26, 437-443. 46 Sander, A., Moser, H., Liechti -Gallati, S., Grimm, T., Zingg, M. and Raveh, J. (1993) Linkage of Van der Woude syndrome (VWS) to REN and exclusion of the candidate gene TGFB2 from the disease locus in a large pedigree. Hum Genet , 91, 55-62. $*""47 Sander, A., Schmelzle, R. and Murray, J. (1994) Evidence for a microdeletion in 1q32-41 involving the gene respo nsible for Van der Woude syndrome. Hum Mol Genet , 3, 575-578. 48 Houdayer, C., Soupre, V., Rosenberg -Bourgin, M., Martinez, H., Tredano, M., Feldmann, D., Feingold, J., Aymard, P., Munnich, A., Le Bouc, Y. et al. (1999) Linkage analysis of 5 novel van der Woude syndrome kindreds to 1q32 -q41 markers further supports locus homogeneity of the disease trait. Ann Genet , 42, 69-74. 49 Schutte, B.C., Sander, A., Malik, M. and Murray, J.C. (1996) Refinement of the Van der Woude gene location and construction of a 3.5 -Mb YAC contig and STS map spanning the critical region in 1q32 -q41. Genomics , 36, 507-514. 50 Schutte, B.C., Basart, A.M., Watanabe, Y., Laffin, J.J., Coppage, K., Bjork, B.C., Daack-Hirsch, S., Patil, S., Dixon, M.J. and Murray, J.C. (1999) Microdel etions at chromosome bands 1q32 -q41 as a cause of Van der Woude syndrome. Am J Med Genet , 84, 145-150. 51 Schutte, B.C., Bjork, B.C., Coppage, K.B., Malik, M.I., Gregory, S.G., Scott, D.J., Brentzell, L.M., Watanabe, Y., Dixon, M.J. and Murray, J.C. (2000 ) A preliminary gene map for the Van der Woude syndrome critical region derived from 900 kb of genomic sequence at 1q32 -q41. Genome Res , 10, 81-94. 52 Wong, F.K., Karsten, A., Larson, O., Huggare, J., Hagberg, C., Larsson, C., Teh, B.T. and Linder -Aronso n, S. (1999) Clinical and genetic studies of Van der Woude syndrome in Sweden. Acta Odontol Scand , 57, 72-76. 53 Beiraghi, S., Miller -Chisholm, A., Kimberling, W.J., Sun, C.E., Wang, Y.F., Russell, L.J., Khoshnevisan, M., Storm, A.L., Long, R.E., Jr., Wit t, P.D. et al. (1999) Confirmation of linkage of Van der Woude syndrome to chromosome 1q32: evidence of association with STR alleles suggests possible unique origin of the disease mutation. J Craniofac Genet Dev Biol , 19, 128-134. 54 Neuman, Z. and Shulma n, J. (1961) Congenital sinuses of the lower lip. Oral Surg Oral Med Oral Pathol , 14, 1415-1420. 55 Hersh, J.H. and Verdi, G.D. (1992) Natal teeth in monozygotic twins with Van der Woude syndrome. Cleft Palate Craniofac J , 29, 279-281. 56 Jobling, R., Fe rrier, R.A., McLeod, R., Petrin, A.L., Murray, J.C. and Thomas, M.A. (2011) Monozygotic twins with variable expression of Van der Woude syndrome. Am J Med Genet A , 155A, 2008-2010. 57 Tokat, C., Bilkay, U., Songur, E. and Akin, Y. (2005) Van der Woude syn drome in twins. J Craniofac Surg , 16, 936-939. %+"" 58 Lees, M.M., Winter, R.M., Malcolm, S., Saal, H.M. and Chitty, L. (1999) Popliteal pterygium syndrome: a clinical study of three families and report of linkage to the Van der Woude syndrome locus on 1q32. J Med Genet , 36, 888-892. 59 Wong, F.K. and Gustafsson, B. (2000) Popliteal pterygium syndrome in a Swedish family --clinical findings and genetic analysis with the van der Woude syndrome locus at 1q32 -q41. Acta Odontol Scand , 58, 85-88. 60 Bixler, D., Pol and, C. and Nance, W.E. (1973) Phenotypic variation in the popliteal pterygium syndrome. Clin Genet , 4, 220-228. 61 Escobar, V. and Weaver, D. (1978) Popliteal pterygium syndrome: a phenotypic and genetic analysis. J Med Genet , 15, 35-42. 62 Hammer, J., Klausler, M. and Schinzel, A. (1989) The popliteal pterygium syndrome: distinct phenotypic variation in two families. Helv Paediatr Acta , 43, 507-514. 63 Leslie, E.J., Standley, J., Compton, J., Bale, S., Schutte, B.C. and Murray, J.C. (2013) Comparative analysis of IRF6 variants in families with Van der Woude syndrome and popliteal pterygium syndrome using public whole -exome databases. Genet Med , 15, 338-344. 64 Matsuzawa, N., Yoshiura, K., Machida, J., Nakamura, T., Niimi, T., Furukawa, H., Toyoda, T., Natsume, N., Shimozato, K. and Niikawa, N. (2004) Two missense mutations in the IRF6 gene in two Japanese families with Van der Woude syndrome. Oral Surg Oral Med Oral Pathol Oral Radiol Endod , 98, 414-417. 65 Shotelersuk, V., Srichomthong, C., Yoshiura, K. and Niikawa, N. (2003) A novel mutation, 1234del(C), of the IRF6 in a Thai family with Van der Woude syndrome. Int J Mol Med , 11, 505-507. 66 Wang, X., Liu, J., Zhang, H., Xiao, M., Li, J., Yang, C., Lin, X., Wu, Z., Hu, L. and Kong, X. (2003) Novel mu tations in the IRF6 gene for Van der Woude syndrome. Hum Genet , 113, 382-386. 67 Kim, Y., Park, J.Y., Lee, T.J. and Yoo, H.W. (2003) Identification of two novel mutations of IRF6 in Korean families affected with Van der Woude syndrome. Int J Mol Med , 12, 465-468. 68 Kayano, S., Kure, S., Suzuki, Y., Kanno, K., Aoki, Y., Kondo, S., Schutte, B.C., Murray, J.C., Yamada, A. and Matsubara, Y. (2003) Novel IRF6 mutations in Japanese patients with Van der Woude syndrome: two missense mutations (R45Q and P396S) a nd a 17 -kb deletion. J Hum Genet , 48, 622-628. %!""69 Gatta, V., Scarciolla, O., Cupaioli, M., Palka, C., Chiesa, P.L. and Stuppia, L. (2004) A novel mutation of the IRF6 gene in an Italian family with Van der Woude syndrome. Mutat Res , 547, 49-53. 70 Item, C.B., Turhani, D., Thurnher, D., Yerit, K., Sinko, K., Wittwer, G., Adeyemo, W.L., Frei, K., Erginel -Unaltuna, N., Watzinger, F. et al. (2005) Van Der Woude syndrome: variable penetrance of a novel mutation (p.Arg 84Gly) of the IRF6 gene in a Turkish famil y. Int J Mol Med , 15, 247-251. 71 Wang, X.F., Xiao, M.Z., Shi, J.N., Zhang, H.B., Hu, L.D. and Kong, X.Y. (2005) [IRF6 gene mutation analysis in a van Der Woude syndrome family in Henan province]. Shanghai Kou Qiang Yi Xue , 14, 234-237. 72 Ghassibe, M., Revencu, N., Bayet, B., Gillerot, Y., Vanwijck, R., Verellen -Dumoulin, C. and Vikkula, M. (2004) Six families with van der Woude and/or popliteal pterygium syndrome: all with a mutation in the IRF6 gene. J Med Genet , 41, e15. 73 Ghassibe, M., Bayet, B., R evencu, N., Verellen -Dumoulin, C., Gillerot, Y., Vanwijck, R. and Vikkula, M. (2005) Interferon regulatory factor -6: a gene predisposing to isolated cleft lip with or without cleft palate in the Belgian population. Eur J Hum Genet , 13, 1239-1242. 74 Ye, X .Q., Jin, H.X., Shi, L.S., Fan, M.W., Song, G.T., Fan, H.L. and Bian, Z. (2005) Identification of novel mutations of IRF6 gene in Chinese families with Van der Woude syndrome. Int J Mol Med , 16, 851-856. 75 Du, X., Tang, W., Tian, W., Li, S., Li, X., Liu, L., Zheng, X., Chen, X., Lin, Y. and Tang, Y. (2006) Novel IRF6 mutations in Chinese patients with Van der Woude syndrome. J Dent Res , 85, 937-940. 76 Du, X.Y., Tang, W., Tian, W.D., Li, X.Y., Liu, L. and Zheng, X.H. (2006) [Identification of three novel mutations of IRF6 in Chinese families with Van der Woude syndrome]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi , 23, 82-83. 77 Brosch, S., Baur, M., Blin, N., Reinert, S. and Pfister, M. (2007) A novel IRF6 nonsense mutation (Y67X) in a German family with Van de r Woude syndrome. Int J Mol Med , 20, 85-89. 78 Tan, E.C., Lim, E.C., Yap, S.H., Lee, S.T., Cheng, J., Por, Y.C. and Yeow, V. (2008) Identification of IRF6 gene variants in three families with Van der Woude syndrome. Int J Mol Med , 21, 747-751. 79 Paranai ba, L.M., Martelli -Junior, H., Oliveira Swerts, M.S., Line, S.R. and Coletta, R.D. (2008) Novel mutations in the IRF6 gene in Brazilian families with Van der Woude syndrome. Int J Mol Med , 22, 507-511. %#""80 de Lima, R.L., Hoper, S.A., Ghassibe, M., Cooper, M.E., Rorick, N.K., Kondo, S., Katz, L., Marazita, M.L., Compton, J., Bale, S. et al. (2009) Prevalence and nonrandom distribution of exonic mutations in interferon regulatory factor 6 in 307 families with Van der Woude syndrome and 37 families with poplit eal pterygium syndrome. Genet Med , 11, 241-247. 81 Yeetong, P., Mahatumarat, C., Siriwan, P., Rojvachiranonda, N., Suphapeetiporn, K. and Shotelersuk, V. (2009) Three novel mutations of the IRF6 gene with one associated with an unusual feature in Van der Woude syndrome. Am J Med Genet A , 149A, 2489-2492. 82 Malik, S., Kakar, N., Hasnain, S., Ahmad, J., Wilcox, E.R. and Naz, S. (2010) Epidemiology of Van der Woude syndrome from mutational analyses in affected patients from Pakistan. Clin Genet , 78, 247-256. 83 Scioletti, A.P., Brancati, F., Gatta, V., Antonucci, I., Peissel, B., Pizzuti, A., Mortellaro, C., Tete, S., Gherlone, E., Palka, G. et al. (2010) Two novel mutations affecting splicing in the IRF6 gene associated with van der Woude syndrome. J Crani ofac Surg , 21, 1654-1656. 84 Salahshourifar, I., Halim, A.S., Sulaiman, W.A., Ariffin, R., Naili Muhamad Nor, N. and Zilfalil, B.A. (2011) De novo interstitial deletion of 1q32.2 -q32.3 including the entire IRF6 gene in a patient with oral cleft and other dysmorphic features. Cytogenet Genome Res , 134, 83-87. 85 Birkeland, A.C., Larrabee, Y., Kent, D.T., Flores, C., Su, G.H., Lee, J.H. and Haddad, J., Jr. (2011) Novel IRF6 mutations in Honduran Van der Woude syndrome patients. Mol Med Report , 4, 237-241. 86 Minones -Suarez, L., Mas -Vidal, A., Fernandez -Toral, J., Llano -Rivas, I. and Gonzalez -Garcia, M. (2011) A novel mutation in the IRF6 gene associated with facial asymmetry in a family affected with Van der Woude Syndrome. Pediatr Dermatol . 87 Koillinen, H., Wong, F.K., Rautio, J., Ollikainen, V., Karsten, A., Larson, O., Teh, B.T., Huggare, J., Lahermo, P., Larsson, C. et al. (2001) Mapping of the second locus for the Van der Woude syndrome to chromosome 1p34. Eur J Hum Genet , 9, 747 -752. 88 Rorick, N.K. , Kinoshita, A., Weirather, J.L., Peyrard -Janvid, M., de Lima, R.L., Dunnwald, M., Shanske, A.L., Moretti -Ferreira, D., Koillinen, H., Kere, J. et al. (2011) Genomic strategy identifies a missense mutation in WD -repeat domain 65 (WDR65) in an individual wi th Van der Woude syndrome. Am J Med Genet A , 155A, 1314-1321. 89 Zucchero, T.M., Cooper, M.E., Maher, B.S., Daack -Hirsch, S., Nepomuceno, B., Ribeiro, L., Caprau, D., Christensen, K., Suzuki, Y., Machida, J. et al. (2004) Interferon %$""regulatory factor 6 (I RF6) gene variants and the risk of isolated cleft lip or palate. N Engl J Med , 351, 769-780. 90 Beaty, T.H., Murray, J.C., Marazita, M.L., Munger, R.G., Ruczinski, I., Hetmanski, J.B., Liang, K.Y., Wu, T., Murray, T., Fallin, M.D. et al. (2010) A genome -wide association study of cleft lip with and without cleft palate identifies risk variants near MAFB and ABCA4. Nat Genet , 42, 525-529. 91 Rahimov, F., Marazita, M.L., Visel, A., Cooper, M.E., Hitchler, M.J., Rubini, M., Domann, F.E., Govil, M., Christense n, K., Bille, C. et al. (2008) Disruption of an AP -2alpha binding site in an IRF6 enhancer is associated with cleft lip. Nat Genet , 40, 1341 -1347. 92 Fakhouri, W.D., Rhea, L., Du, T., Sweezer, E., Morrison, H., Fitzpatrick, D., Yang, B., Dunnwald, M. and Schutte, B.C. (2012) MCS9.7 enhancer activity is highly, but not completely, associated with expression of Irf6 and p63. Dev Dyn , 241, 340-349. 93 Bliek, J.B., Rothenberg, S.P. and Steegers -Theunissen, R.P. (2006) Maternal folate receptor autoantibodies a nd cleft lip and/or palate. Int J Gynaecol Obstet , 93, 142 -143. 94 Mills, J.L., Molloy, A.M., Parle -McDermott, A., Troendle, J.F., Brody, L.C., Conley, M.R., Cox, C., Pangilinan, F., Orr, D.J., Earley, M. et al. (2008) Folate -related gene polymorphisms as risk factors for cleft lip and cleft palate. Birth Defects Res A Clin Mol Teratol , 82, 636-643. 95 Blanton, S.H., Henry, R.R., Yuan, Q., Mulliken, J.B., Stal, S., Finnell, R.H. and Hecht, J.T. (2010) Folate pathway and nonsyndromic cleft lip and palate. Birth Defects Res A Clin Mol Teratol . 96 Honda, K. and Taniguchi, T. (2006) Toll -like receptor signaling and IRF transcription factors. IUBMB Life , 58, 290-295. 97 Bailey, C.M., Khalkhali -Ellis, Z., Kondo, S., Margaryan, N.V., Seftor, R.E., Wheaton, W.W. , Amir, S., Pins, M.R., Schutte, B.C. and Hendrix, M.J. (2005) Mammary serine protease inhibitor (Maspin) binds directly to interferon regulatory factor 6: identification of a novel serpin partnership. J Biol Chem , 280, 34210-34217. 98 Richardson, R.J., D ixon, J., Jiang, R. and Dixon, M.J. (2009) Integration of IRF6 and Jagged2 signalling is essential for controlling palatal adhesion and fusion competence. Hum Mol Genet , 18, 2632-2642. 99 Knight, A.S., Schutte, B.C., Jiang, R. and Dixon, M.J. (2006) Devel opmental expression analysis of the mouse and chick orthologues of IRF6: the gene mutated in Van der Woude syndrome. Dev Dyn , 235, 1441-1447. %%""100 Bailey, C.M., Margaryan, N.V., Abbott, D.E., Schutte, B.C., Yang, B., Khalkhali -Ellis, Z. and Hendrix, M.J. ( 2009) Temporal and spatial expression patterns for the tumor suppressor Maspin and its binding partner interferon regulatory factor 6 during breast development. Dev Growth Differ , 51, 473-481. 101 Little, H.J., Rorick, N.K., Su, L.I., Baldock, C., Malhotr a, S., Jowitt, T., Gakhar, L., Subramanian, R., Schutte, B.C., Dixon, M.J. et al. (2009) Missense mutations that cause Van der Woude syndrome and popliteal pterygium syndrome affect the DNA -binding and transcriptional activation functions of IRF6. Hum Mol Genet , 18, 535-545. 102 Sabel, J.L., d'Alencon, C., O'Brien, E.K., Van Otterloo, E., Lutz, K., Cuykendall, T.N., Schutte, B.C., Houston, D.W. and Cornell, R.A. (2009) Maternal Interferon Regulatory Factor 6 is required for the differentiation of primary s uperficial epithelia in Danio and Xenopus embryos. Dev Biol , 325, 249-262. 103 de la Garza, G., Schleiffarth, J.R., Dunnwald, M., Mankad, A., Weirather, J.L., Bonde, G., Butcher, S., Mansour, T.A., Kousa, Y.A., Fukazawa, C.F. et al. (2013) Interferon regu latory factor 6 promotes differentiation of the periderm by activating expression of grainyhead -like 3. J Invest Dermatol , 133, 68-77. 104 Richardson, R.J., Dixon, J., Malhotra, S., Hardman, M.J., Knowles, L., Boot -Handford, R.P., Shore, P., Whitmarsh, A. and Dixon, M.J. (2006) Irf6 is a key determinant of the keratinocyte proliferation -differentiation switch. Nat Genet , 38, 1329 -1334. 105 Ingraham, C.R., Kinoshita, A., Kondo, S., Yang, B., Sajan, S., Trout, K.J., Malik, M.I., Dunnwald, M., Goudy, S.L., L ovett, M. et al. (2006) Abnormal skin, limb and craniofacial morphogenesis in mice deficient for interferon regulatory factor 6 (Irf6). Nat Genet , 38, 1335-1340. 106 Su, L. -I. (2011), In Faculty of Life Science . University of Manchester, Manchester, Vol. Doctor of Philosophy, pp. 231. 107 Botti, E., Spallone, G., Moretti, F., Marinari, B., Pinetti, V., Galanti, S., De Meo, P.D., De Nicola, F., Ganci, F., Castrignano, T. et al. (2011) Developmental factor IRF6 exhibits tumor suppressor activity in squamous cell carcinomas. Proc Natl Acad Sci U S A, 108, 13710-13715. 108 Ting, S.B., Wilanowski, T., Auden, A., Hall, M., Voss, A.K., Thomas, T., Parekh, V., Cunningham, J.M. and Jane, S.M. (2003) Inositol - and folate -resistant neural tube defects in mice lackin g the epithelial -specific factor Grhl -3. Nat Med , 9, 1513-1519. 109 McDade, S.S., Henry, A.E., Pivato, G.P., Kozarewa, I., Mitsopoulos, C., Fenwick, K., Assiotis, I., Hakas, J., Zvelebil, M., Orr, N. et al. (2012) Genome -wide analysis of p63 binding sites identifies AP -2 factors as co -regulators of epidermal differentiation. Nucleic Acids Res , 40, 7190-7206. %&"" 110 Bailey, C.M., Abbott, D.E., Margaryan, N.V., Khalkhali -Ellis, Z. and Hendrix, M.J. (2008) Interferon regulatory factor 6 promotes cell cycle arre st and is regulated by the proteasome in a cell cycle -dependent manner. Mol Cell Biol , 28, 2235-2243. 111 Moretti, F., Marinari, B., Lo Iacono, N., Botti, E., Giunta, A., Spallone, G., Garaffo, G., Vernersson -Lindahl, E., Merlo, G., Mills, A.A. et al. (20 10) A regulatory feedback loop involving p63 and IRF6 links the pathogenesis of 2 genetically different human ectodermal dysplasias. J Clin Invest , 120, 1570-1577. 112 Thomason, H.A., Zhou, H., Kouwenhoven, E.N., Dotto, G.P., Restivo, G., Nguyen, B.C., Li ttle, H., Dixon, M.J., van Bokhoven, H. and Dixon, J. (2010) Cooperation between the transcription factors p63 and IRF6 is essential to prevent cleft palate in mice. J Clin Invest , 120, 1561-1569. 113 Stransky, N., Egloff, A.M., Tward, A.D., Kostic, A.D., Cibulskis, K., Sivachenko, A., Kryukov, G.V., Lawrence, M.S., Sougnez, C., McKenna, A. et al. (2011) The mutational landscape of head and neck squamous cell carcinoma. Science , 333, 1157 -1160. 114 Nair, M., Teng, A., Bilanchone, V., Agrawal, A., Li, B. a nd Dai, X. (2006) Ovol1 regulates the growth arrest of embryonic epidermal progenitor cells and represses c -myc transcription. J Cell Biol , 173, 253-264. 115 Stransky, N., Egloff, A.M., Tward, A.D., Kostic, A.D., Cibulskis, K., Sivachenko, A., Kryukov, G. V., Lawrence, M.S., Sougnez, C., McKenna, A. et al. (2011) The mutational landscape of head and neck squamous cell carcinoma. Science , 333, 1157 -1160. 116 Chen, W., Lam, S.S., Srinath, H., Jiang, Z., Correia, J.J., Schiffer, C.A., Fitzgerald, K.A., Lin, K . and Royer, W.E., Jr. (2008) Insights into interferon regulatory factor activation from the crystal structure of dimeric IRF5. Nat Struct Mol Biol , 15, 1213-1220. 117 Li, H., Sheridan, R. and Williams, T. (2013) Analysis of TFAP2A mutations in Branchio -Oculo -Facial Syndrome indicates functional complexity within the AP -2alpha DNA-binding domain. Hum Mol Genet , 22, 3195-3206. 118 Zhang, J., Hagopian -Donaldson, S., Serbedzija, G., Elsemore, J., Plehn -Dujowich, D., McMahon, A.P., Flavell, R.A. and Williams, T. (1996) Neural tube, skeletal and body wall defects in mice lacking transcription factor AP -2. Nature , 381, 238-241. 119 Schorle, H., Meier, P., Buchert, M., Jaenisch, R. and Mitchell, P.J. (1996) Transcription factor AP -2 essential for cranial closure and craniofacial development. Nature , 381, 235-238. %'""120 Brewer, S., Feng, W., Huang, J., Sullivan, S. and Williams, T. (2004) Wnt1 -Cre-mediated deletion of AP -2alpha causes multiple neural crest -related defects. Dev Biol , 267, 135-152. 121 Thomason, H.A ., Zhou, H., Kouwenhoven, E.N., Dotto, G.P., Restivo, G., Nguyen, B.C., Little, H., Dixon, M.J., van Bokhoven, H. and Dixon, J. (2010) Cooperation between the transcription factors p63 and IRF6 is essential to prevent cleft palate in mice. J Clin Invest , 120, 1561-1569. 122 Restivo, G., Nguyen, B.C., Dziunycz, P., Ristorcelli, E., Ryan, R.J., Ozuysal, O.Y., Di Piazza, M., Radtke, F., Dixon, M.J., Hofbauer, G.F. et al. (2011) IRF6 is a mediator of Notch pro -differentiation and tumour suppressive function in keratinocytes. EMBO J . 123 Stottmann, R.W., Bjork, B.C., Doyle, J.B. and Beier, D.R. (2010) Identification of a Van der Woude syndrome mutation in the cleft palate 1 mutant mouse. Genesis , 48, 303-308. 124 Goudy, S., Angel, P., Jacobs, B., Hill, C., Mai nini, V., Smith, A.L., Kousa, Y.A., Caprioli, R., Prince, L.S., Baldwin, S. et al. (2013) Cell -autonomous and non -cell-autonomous roles for IRF6 during development of the tongue. PLoS One , 8, e56270. 125 Boell, L., Pallares, L.F., Brodski, C., Chen, Y., C hristian, J.L., Kousa, Y.A., Kuss, P., Nelsen, S., Novikov, O., Schutte, B.C. et al. (2013) Exploring the effects of gene dosage on mandible shape in mice as a model for studying the genetic basis of natural variation. Dev Genes Evol , 223, 279-287. 126 Herron, B.J., Liddell, R.A., Parker, A., Grant, S., Kinne, J., Fisher, J.K. and Siracusa, L.D. (2005) A mutation in stratifin is responsible for the repeated epilation (Er) phenotype in mice. Nat Genet , 37, 1210-1212. 127 Song, L., Dong, W., Gao, M., Li, J. , Hu, M., Guo, N. and Huang, C. (2010) A novel role of IKKalpha in the mediation of UVB -induced G0/G1 cell cycle arrest response by suppressing Cyclin D1 expression. Biochim Biophys Acta , 1803, 323-332. 128 Lee, S., Kong, Y. and Weatherbee, S.D. (2013) Forward genetics identifies Kdf1/1810019J16Rik as an essential regulator of the proliferation -differentiation decision in epidermal progenitor cells. Dev Biol , 383, 201-213. 129 Rountree, R.B., Willis, C.R., Dinh, H., Blumberg, H., Bailey, K., Dean, C., Jr. , Peschon, J.J. and Holland, P.M. (2010) RIP4 regulates epidermal differentiation and cutaneous inflammation. J Invest Dermatol , 130, 102-112. 130 Yang, H.Y., Wen, Y.Y., Chen, C.H., Lozano, G. and Lee, M.H. (2003) 14 -3-3 sigma positively regulates p53 and suppresses tumor growth. Mol Cell Biol , 23, 7096 -7107. %("" 131 Hermeking, H., Lengauer, C., Polyak, K., He, T.C., Zhang, L., Thiagalingam, S., Kinzler, K.W. and Vogelstein, B. (1997) 14 -3-3 sigma is a p53 -regulated inhibitor of G2/M progression. Mol Cell , 1, 3-11. 132 Dellambra, E., Patrone, M., Sparatore, B., Negri, A., Ceciliani, F., Bondanza, S., Molina, F., Cancedda, F.D. and De Luca, M. (1995) Stratifin, a keratinocyte specific 14 -3-3 protein, harbors a pleckstrin homology (PH) domain and enhances prote in kinase C activity. J Cell Sci , 108 ( Pt 11) , 3569-3579. 133 Yaffe, M.B., Rittinger, K., Volinia, S., Caron, P.R., Aitken, A., Leffers, H., Gamblin, S.J., Smerdon, S.J. and Cantley, L.C. (1997) The structural basis for 14 -3-3:phosphopeptide binding spec ificity. Cell , 91, 961-971. 134 Rittinger, K., Budman, J., Xu, J., Volinia, S., Cantley, L.C., Smerdon, S.J., Gamblin, S.J. and Yaffe, M.B. (1999) Structural analysis of 14 -3-3 phosphopeptide complexes identifies a dual role for the nuclear export signal of 14 -3-3 in ligand binding. Mol Cell , 4, 153-166. 135 Ferguson, A.T., Evron, E., Umbricht, C.B., Pandita, T.K., Chan, T.A., Hermeking, H., Marks, J.R., Lambers, A.R., Futreal, P.A., Stampfer, M.R. et al. (2000) High frequency of hypermethylation at the 1 4-3-3 sigma locus leads to gene silencing in breast cancer. Proc Natl Acad Sci U S A , 97, 6049-6054. 136 Umbricht, C.B., Evron, E., Gabrielson, E., Ferguson, A., Marks, J. and Sukumar, S. (2001) Hypermethylation of 14 -3-3 sigma (stratifin) is an early eve nt in breast cancer. Oncogene , 20, 3348-3353. 137 Lahtela, J., Nousiainen, H.O., Stefanovic, V., Tallila, J., Viskari, H., Karikoski, R., Gentile, M., Saloranta, C., Varilo, T., Salonen, R. et al. (2010) Mutant CHUK and severe fetal encasement malformatio n. N Engl J Med , 363, 1631-1637. 138 Hu, Y., Baud, V., Oga, T., Kim, K.I., Yoshida, K. and Karin, M. (2001) IKKalpha controls formation of the epidermis independently of NF -kappaB. Nature , 410, 710-714. 139 Descargues, P., Sil, A.K., Sano, Y., Korchynsky i, O., Han, G., Owens, P., Wang, X.J. and Karin, M. (2008) IKKalpha is a critical coregulator of a Smad4 -independent TGFbeta -Smad2/3 signaling pathway that controls keratinocyte differentiation. Proc Natl Acad Sci U S A , 105, 2487-2492. 140 Le, M., Naridz e, R., Morrison, J., Biggs, L.C., Rhea, L., Schutte, B.C., Kaartinen, V. and Dunnwald, M. (2012) Transforming growth factor Beta 3 is required for excisional wound repair in vivo. PLoS One , 7, e48040. 141 Iwata, J., Suzuki, A., Pelikan, R.C., Ho, T.V., Sa nchez -Lara, P.A., Urata, M., Dixon, M.J. and Chai, Y. (2013) Smad4 -Irf6 genetic interaction and TGFbeta -mediated %)""IRF6 signaling cascade are crucial for palatal fusion in mice. Development , 140, 1220 -1230. 142 Chen, L., Haider, K., Ponda, M., Cariappa, A., Rowitch, D. and Pillai, S. (2001) Protein kinase C -associated kinase (PKK), a novel membrane -associated, ankyrin repeat -containing protein kinase. J Biol Chem , 276, 21737-21744. 143 Meylan, E., Martinon, F., Thome, M., Gschwendt, M. and Tschopp, J. (2002 ) RIP4 (DIK/PKK), a novel member of the RIP kinase family, activates NF -kappa B and is processed during apoptosis. EMBO Rep , 3, 1201-1208. 144 Holland, P., Willis, C., Kanaly, S., Glaccum, M., Warren, A., Charrier, K., Murison, J., Derry, J., Virca, G., B ird, T. et al. (2002) RIP4 is an ankyrin repeat -containing kinase essential for keratinocyte differentiation. Curr Biol , 12, 1424-1428. 145 Mitchell, K., O'Sullivan, J., Missero, C., Blair, E., Richardson, R., Anderson, B., Antonini, D., Murray, J.C., Sha nske, A.L., Schutte, B.C. et al. (2012) Exome sequence identifies RIPK4 as the Bartsocas -Papas syndrome locus. Am J Hum Genet , 90, 69-75. 146 Kalay, E., Sezgin, O., Chellappa, V., Mutlu, M., Morsy, H., Kayserili, H., Kreiger, E., Cansu, A., Toraman, B., A bdalla, E.M. et al. (2012) Mutations in RIPK4 cause the autosomal -recessive form of popliteal pterygium syndrome. Am J Hum Genet , 90, 76-85. 147 Sil, A.K., Maeda, S., Sano, Y., Roop, D.R. and Karin, M. (2004) IkappaB kinase -alpha acts in the epidermis to control skeletal and craniofacial morphogenesis. Nature , 428, 660-664. 148 Keswani, S.G., Le, L.D., Morris, L.M., Lim, F.Y., Katz, A.B., Ghobril, N., Habli, M., Frischer, J.S. and Crombleholme, T.M. (2011) Submucosal gland development in the human fetal t rachea xenograft model: implications for fetal gene therapy. J Pediatr Surg , 46, 33-38. 149 Koppanati, B.M., Li, J., Reay, D.P., Wang, B., Daood, M., Zheng, H., Xiao, X., Watchko, J.F. and Clemens, P.R. (2010) Improvement of the mdx mouse dystrophic pheno type by systemic in utero AAV8 delivery of a minidystrophin gene. Gene Ther , 17, 1355-1362. 150 Muhle, C., Neuner, A., Park, J., Pacho, F., Jiang, Q., Waddington, S.N. and Schneider, H. (2006) Evaluation of prenatal intra -amniotic LAMB3 gene delivery in a mouse model of Herlitz disease. Gene Ther , 13, 1665-1676. 151 Endo, M., Zoltick, P.W., Radu, A., Qiujie, J., Matsui, C., Marinkovich, P.M., McGrath, J., Tamai, K., Uitto, J. and Flake, A.W. (2011) Early intra -amniotic gene transfer using lentiviral vecto r improves skin blistering phenotype in a murine model of Herlitz junctional epidermolysis bullosa. Gene Ther . %*"" 152 Niiya, M., Endo, M., Shang, D., Zoltick, P.W., Muvarak, N.E., Cao, W., Jin, S.Y., Skipwith, C.G., Motto, D.G., Flake, A.W. et al. (2009) Cor rection of ADAMTS13 deficiency by in utero gene transfer of lentiviral vector encoding ADAMTS13 genes. Mol Ther , 17, 34-41. 153 Dejneka, N.S., Surace, E.M., Aleman, T.S., Cideciyan, A.V., Lyubarsky, A., Savchenko, A., Redmond, T.M., Tang, W., Wei, Z., Rex , T.S. et al. (2004) In utero gene therapy rescues vision in a murine model of congenital blindness. Mol Ther , 9, 182-188. 154 Wu, C., Endo, M., Yang, B.H., Radecki, M.A., Davis, P.F., Zoltick, P.W., Spivak, R.M., Flake, A.W., Kirschner, R.E. and Nah, H.D . (2013) Intra -amniotic transient transduction of the periderm with a viral vector encoding TGFbeta3 prevents cleft palate in Tgfbeta3( -/-) mouse embryos. Mol Ther , 21, 8-17. 155 Copp, A.J., Greene, N.D. and Murdoch, J.N. (2003) The genetic basis of mammalian neurulation. Nat Rev Genet , 4, 784-793. "!!"#! Chapter 2 - Dominant mutations in GRHL3 cause Van der Woude syndrome and disrupt oral periderm development !!"$!Myriam Peyrard -Janvid ,1,16* Elizabeth J. Leslie ,2,16 Youssef A. Kousa ,3,16 Tiffany L . Smith ,4,16 Martine Dunnwald ,2,16 M„ns Magnusson ,5 Brian A. Lentz ,2 Per Unneberg ,6 Ingegerd Fransson ,1 Hannele K. Koillinen ,7 Jorma Rautio ,8 Marie Pegelow ,9 Agneta Karsten ,9 Lina Basel -Vanagaite ,10,11 ,12 William Gordon ,13 Bogi Andersen ,13 Thomas Svensson ,5 Jeff rey C. Murray ,2 Robert A. Cornell ,4 Juha Kere ,1,5,1 4** and Brian C. Schutte ,15 Affiliations: 1Department of Bi osciences and Nutrition, Karolinska Institutet, and Center for Biotechnology, 14183 Huddinge , Sweden 2Department of Pediatrics and Interdisciplinary Program in Genetics , University of Iowa, Iowa City 52242, Iowa, USA 3Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing 48824, Michigan, USA 4Department of Anatomy and Cell Biology, University of Iowa, Iowa City 52242, Iowa, USA 5Department of Biosciences and Nutrition, Science for Life Laboratory, Karolinska Institutet, 17121 Solna, Sweden 6Department of Biochemistry and Biophysics Science for Life Laboratory , Stockholm University, 17121 Solna, Sweden 7Departmen t of Clinical Genetics, Helsinki University Hospital, 00029 Helsinki, Finland 8Cleft Palate and Craniofacial Center, Department of Plastic Surgery, Helsinki University Hospital, 00029 Helsinki, Finland 9Department of Orthodontics, Stockholm Craniofacial Team, Institute of Odontology, Karolinska Institu tet, 17177 Stockholm, Sweden !!"%!10Pediatric Genetics Unit , Schneider Children's Medical Center of Israel and Raphael Recanati Genetic Institute, Rabin Medical Center, Petah Tikva 49100, Israel 11Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv 69978, Israel 12Felsenstein Medical Research Center, Petah Tikva, 49100, Israel 13Department of Biological Chemistry, University of California Irvine, Irvine 92697, California, USA 14Research Programs Unit, University of Helsinki, and Folkh−lsan Institute of Genetics, 000 14 Helsinki, Finland 15Department of Microbiology and Molecular Genetics, Michigan State University, East Lansing 48824, Michigan, USA 16These authors contributed equally to this work Correspondence to: *Myriam.Peyrard@ki.se **Juha.Kere@ki.se !!"&!ABSTRACT Mutations in the inter feron regulatory factor 6 ( IRF6 ) gene account for ~70% of cases of Van der Woude syndrome (VWS), the most common syndromic form of cleft lip and palate. In eight of 45 VWS families lacking a mutation in IRF6 , we found coding mutations in the grainy head -like 3 (GRHL3) gene . Using a zebrafish -based assay, the disease -associated GRHL3 mutations abrogate d periderm development and were consistent with a dominant -negative effect , in contrast to haploinsufficiency seen in most VWS cases caused by IRF6 mutations . In mouse, all embryos lacking Grhl3 exh ibit ed abnormal oral periderm and 17% develop ed a cleft palate. Analysis of the oral phen otype of double heterozygote (Irf6 +/-;Grhl3 +/-) murin e embryos failed to detect epistasis between the two genes, suggestin g that they function in separa te but convergent pathways during palat ogen esis. Taken t ogether, our data demonstrate d that mutations in two genes, IRF6 and GRHL3, can lead to nearly identical phenotypes of orofacial cleft . They support ed the hypothese s that both genes are essential for the presence of a functional oral periderm and that failure of this process contributes to VWS. !!"'!INTRODUCTION Grainy head -like 3 (GRHL3, MIM 608317 ) belongs to a family of three human genes that encode transcription factor orth ologs of the Drosophila gene grainy head (grh ). Among multiple conserved roles, t his gene family is required for the development and repair of the epide rmal barrier layer 1-3. In zebrafish, grhl1 and grhl3 were shown to be required for the development of the periderm 4, the transient layer of squamous epithelial cells located on the surfa ce of developing embryos. Interferon regulatory factor 6 ( irf6) is also required for periderm development in zebrafish 5, and directly regulates the expression of grhl3 4; 6 . In addition, over -expression of Grhl3 partially rescued periderm development in zebrafish embryos that expressed a dominant -negative mutant form of irf6 4. These data suggest that Grhl3 is an important player in the Irf6 -dependent pathway of periderm development. IRF6 belongs to the IRF family of transcription factors that are known best for their roles in immune function 7. However, IRF6 (MIM 607199) is required for skin, limb and craniofacial development 8-10. In mice, embryos that lack Irf6 expression fail to develop the epidermal barrier 9; 10 . While reminiscent of embryos that lack Grhl3 2, the cutaneous phenotype of Irf6 mutant embryos appears to be more severe macroscopically . In addition, Irf6 mutant embryos have extensive oral epithelial adhesions 9; 10 , a phenotype not reported in the Grhl3 mutant . The oral epithelial adhesions in Irf6 knockout embryos lead to cleft palate 9; 10 , and appear to stem from periderm dysfunction 4; 11 . In humans, mutations in IRF6 cause Van der Woude syndrome (VWS ü MIM 119300), the most common synd romic form of orofacial clefting , or popliteal pterygium syndrome (PPS , MIM 119500). Individuals with VWS can have cleft lip (CL), cleft palate (CP), or cleft lip and palate (CLP) . In addition, 85% of affected individuals have pits in their lower lip 12. To date, mutations in IRF6 have been identified in 70% of families with VWS 8; 13; 14 . The possibility that locus !!""!heterogeneity accou nts for some of the remaining 30% of VWS mutations is underscored by linkage in one large pedigree from Finland to a locus on 1p33-p36 rather than to IRF6 at 1q32 -q41 15. In this family, most affected individuals have an orofacial cleft and the proband has lip pit s, the hallmark of VWS . Because of the autosomal dominant inheritance pattern and the presence of the lip pit s, this family was diagnosed with VWS and the linked region was named the VWS2 locus 15. Here we report disease -causing mutations in the GRHL3 gene in the above mentioned original Finnish family as well as in seven additional families with VWS , therefore demonstrating that GRHL3 is the second gene for which mutations lead to VWS . While w e observe d no consistently unique phenotypes in these families , individuals with a GRHL3 mutation are more likely to have CP and less likely to have CL or lip pits than individuals with an IRF6 mutation . In addition , we use d zebrafish and murine models to show that Grhl3 , like Irf6, has a conserved role in the development of the periderm . Our observations from all three species suppor t the conclusion that a functional oral periderm is essential for the proper palatogenesis. RESULTS Grainy -head like 3 is the VWS2 gene A single large VWS family of Finnish origin (Fig. 8 ) showed linkage to a ~40cM region on 1p33 -36, pointing to a second VWS locus 15, i.e . VWS2 (MIM 606713). From this family, w e selected eight affected individuals , including the proband who is the only one with lip pits, and three healthy individ uals , for whole -exome sequencing . We searched the ~700 genes contained in th e entire linkage region (~46 Mb ) for variants common to all eight affected family members but not seen in any of the three health y members. This resulted in t hree segregating exonic variants : in GRHL3 (1:24666175 ; NM_198174.2 : c.969 -970insTG) , PHACTR4 (1:28806971 ; rs200581707; !!"(!NM_001048183.1: c.1615 G>A) and KTI12 (1:52499097-52499071; NM_138417.2: c.337_363delCCGATCGCGGGACCTCAGGTGGCGGGC ). The GRHL3 and PHACTR4 variants were confirmed by TaqM an genotyping and the KTI12 variant by allelic discrimination based on differential melting temperature . The PHACTR4 variant was found in two out of 8252 European American chromosomes in the NHLBI/ESP database , and is therefore unlikely to be the causative variant for VWS. In a set of 561 Finnish controls, the KTI12 variant w as found at a frequency of 12.4% , and is therefore a common, non-causative variant . The GRHL3 varia nt was not found in any of the Finnish controls nor in NHLBI/ESP, making GRHL3 a strong candidate gene in the VWS2 locus . To test whether mutations in GRHL3 accounted for VWS in other families, we screened 44 families of variable ethnicity where no causative IRF6 mutations had been previously detected . We identified GRHL3 variants in seven fami lies, including four protein -truncati ng mutations and four missense mutations (Fig. 4 ). All mutations except c.1661A>G (coding for the p. Asn554Ser missense mutation) were predicted by Polyphen2 and SIFT to be damaging/deleterious and two were confirmed de novo events (Table 1 ). In one of th e seven families (VWS -III), we found two variants located in trans . Variant c.268_278delTACTACCATGG was inherited from the probandÕs affected father and from the healthy paternal grandfather, while variant c.1661A>G was inherited from the probandÕs healthy mother (Fig S1). In addition, one family (VWS -IV) was previously determined to have a novel IRF6 missense variant ( c.239A>G ) that was not conclusively determined to be causative for VWS 23, raising the possibility that variants in both IRF6 and GRHL3 could contribute to VWS in one family (Fig. 8 ). We tested for phenotyp ic variation between the VWS and VWS2 loci . The phenotype s observed in the individuals with mutations in GRHL3 overlap with the classic VWS phenotype !!")!(Fig. 8 ). However , individuals positive for a GRHL3 mutation were significantly more likely to have CP ( 70% (GRHL3) vs. 27% ( IRF6 ), p-value = 2.0 !10-6) and less likely to have CL/P (CL or CLP) (11% vs. 46%, p-value = 0.001) than individuals with IRF6 mutations (Table 2) . Lip pits were less frequent among individuals with GRHL3 mutations (52% vs. 76%), however this difference was not statistically significant (p -value = 0.05) . The presence of dental and limb ano malies did not differ significantly between the two groups. Affect of GRHL3 alleles on zebrafish development To distinguish whether the human GRHL3 alleles that cause VWS are nulls or dominant -negative, we developed an in vivo assay to measure the function of the gene on the development of the periderm in zebrafish 4. The assay is based on the observation that over -expression of wild type grhl3 in zebrafi sh or frog embryos (Xenopus laevis ) is sufficient to induce, in deep cells , ectopic expression of genes whose expression is normally restricted to the periderm , e.g. keratin 4 (krt4) 4; 24 . Also, simultaneous reduction of grhl1 and grhl3 , or over -expression of a n engineered dominant -negative variant of frog grhl1 , prevent s the expression of krt4 in epithelial cells of the zebrafish periderm, and cause s embryonic death during epiboly 4. Thus, we injected wild type and mutant alleles of human GRHL3 mRNA into zebrafish embryos and scored for embryonic viability and krt4 expression . At shield stage (6 h post fertilization, hpf), most embryos injected with a control mRNA ( lacZ ) developed normally , and krt4 expression was confined to the periderm (Fig. 5 A,E). In most embryos injected with wild type GRHL3 epiboly was slightly delayed in comparison to lacZ -injected control embryos (Fig. 5B), and krt4 was ectopic ally express ed in deep cells (Fig. 5 F). In contrast , the majority of embryos injected with GRHL3 mRNA carrying the c.1171C>T variant from VWS -II stall ed before (4 hpf) or during epiboly stage, and then ruptured through the animal hemisphere (Fig. !!"*!5C). This phenotype resembles th at of embryos injected with the dominant -negative alleles of Xenopus grhl1 or zebrafish irf6 4; 5 . We tested four other VWS-associated alleles of GRHL3 with this in vivo assay, including both alleles found in VWS -III. For all four alleles, embryonic development stalled and the embryo ruptured at a timepoint and frequency similar to embryos inje cted with the c.1171C>T variant from VWS -II (Fig 2 D). To test whether the effect of these mutations was cell -autonomous, we generated mosaic embryos by co-injecting GRHL3 mRNA and biotin into one cell at the 16 -cell stage of zebrafish development . In this assay, cells that inherit ed the GRHL3 mRNA were marked by biotin staining. In embryos injected with control mRNA ( LacZ ), we observed normal krt4 expression in all periderm cells , regardless of the biotin staining (Fig. 5 G). In embryos injected with GRHL3 mRNA containing the c.893G>A variant (from VWS -IV), the cells from the periderm inherit ing the mutated mRNA (biotin -positive) lack ed krt4 expression , but biotin -negative cells express ed krt4 (Fig. 5 H). We conclude that mutant GRHL3 variant interfere d with the development of the periderm in a cell -autonomous fashion. In summary, each of the five GRHL3 mutations appear ed to encode a protein with dominant -inhibitory effect that disrupt ed the development of the periderm through a cell -autonomous mechanism . Grhl3-/- murine embryos have cleft palate at low penetrance To identify a potential common mechanism for orofacial clefts in individuals with VWS, we compared the oral phenotype of murine embryos that lack Irf6 (Irf6 -/-) to embryos that lack Grhl3 (Grhl3 -/-). Wild type embryos at E15.5 had normal oral epithelium and a fully fused palate ( Fig. 6A), whereas Irf6 -/- embryos (n = 4) had extensive epithelial adhesions between the palatal shelves and the lingual, mandibular and maxillary surfaces ( Fig. 6 B) 9; 10 . These adhesions prevented the palatal shelves from elevating and led to a cleft palate in all embryos. Similarly, all !!"+!Grhl3 -/- embryos at E15.5 had bilateral oral epithelial adhesions (n = 6) and one of these embryos had a cleft palate ( Fig. 6 C). Thus, Grhl3 , like Irf6 , is required for palatal development. To compare the histological changes in these two mutant strains, we immunostained with keratin 6 (Krt 6), a marker for the periderm 25 and tumor protein p63 ( p63), a marker f or the basal epithelial layer 26. We detected Krt6 in the oral periderm o f wild type embryos ( Fig. 6 D), but Krt6 expression was strongly reduced in the epithelium superficial to the tooth germs in both Irf6 -/- and Grhl3 -/- mutant embryos ( Fig. 6 E,F ). Similar results were observed for activated Notch1 (Act N1) (Fig. 10 ), another protein expressed in the periderm 11. Thus, we concluded that both Irf6 and Grhl3 were required for proper development of the oral periderm in the mouse. In addition to its potential role in the periderm, Irf6 regulates the differentiation of the keratinocytes in the epidermis 9; 10 and the oral cavity 11. In the oral cavity, wild type embryos had a uniform, single layer of basal epithelium ( Fig. 6 D), whereas the basal layer in Irf6 -/- embryos was disorganized and thicker, and p63 was ectopically expressed in the cells of the suprabasal layer ( Fig. 6 E). In Grhl3 -/- embryos, the basal epithelial layer appeared grossly normal with normal expression of p63 ( Fig. 6 F). We also looked at the medial edge epithelium (MEE), the epithelium located at the medial edge of the palatal shelves that must dissolve for proper palatal fusion. In wild type embryos ( Fig. 6 G) and Grhl3 -/- (Fig. 6 I), the MEE dissolved to form a confluent bridge of mesenchymal cells across the palate as shown by the loss of expression of p63. In contrast, while we do not know the exact location of the MEE i n Irf6 -/- embryos, expression of p63 persisted throughout the epithelium of the palatal shelves ( Fig. 6 H) 11. Thus, Irf6 -/- embryos have at least two problems during palatal development: the presence of oral epithelial adhesions and the failure of the MEE to dissolve. In contrast, Grhl3 -/- embryos only have oral epithelial adhesio ns due to the loss of periderm. Since mutations in both these genes !!(#!cause VWS, these results are consistent with the hypothesis that abnormal periderm function con tributes to CL/P in humans. The oral phenotypes of Irf6 and Grhl3 heterozygous murine mutants are independent Based on ChIP -seq experiments on a human keratinocyte cell line and epistasis experiments in zebrafish embryos, we hypothesized that Irf6 and Grhl3 function in a common pathway 4; 6 . To test for epistasis during murine palat ogenesis , we generated embryos that were heterozygous for both Irf6 and Grhl3 (Irf6 +/-;Grhl3 +/-). As expected, we did not observe any oral epithelial adhesions in wild -type embryos ( Fig. 7 A,D ). In Irf6 +/- embryos we detected bilateral oral adhesions at the tooth germ sites ( Fig. 7 B). We also observed bilateral epithelial abnormalities in Grhl3 +/- embryos ( Fig. 7 E), but they differed from those seen in the Irf6 +/- embryos in three respects. First, whereas o ral adhesions in Irf6 +/- embryos were more prominent at the tooth germ sites ( Fig. 7 B), epithelial abnormalities in Grhl3 +/- embryos were located throughout the oral cavity and most frequently posterior to the tooth germs ( Fig. 7 E). Second, epithelial abno rmalities included oral fusions ( Fig. 7 E), which do not occur in Irf6 +/- embryos. Here, we distinguish oral epithelial adhesions fro m oral fusions histologically. Whereas adhesions have a loss of periderm that allows cell interactions between two adjacent epithelial layers, fusions have a loss of both the periderm and the basal epithelial layers that allows cell interactions between the underlying mesenchymal cells from adjacent tissues. Finally, whereas oral adhesions in Irf6 +/- occurred most frequently be tween the mandible and maxilla, oral fusions in Grhl3 +/- embryos occurred between the mandible and either the palate or the maxilla. In the Irf6 +/-;Grhl3 +/- double heterozygous embryos, we found oral adhesions at areas superficial to the tooth germ ( Fig. 7 C), similar to Irf6 +/- embryos, as well as oral adhesions and fusions posterior to the tooth germ ( Fig. 7F), similar to Grhl3 +/- embryos. Thus, the oral histopathology of the Irf6 +/-;Grhl3 +/- double !!($!heterozygote embryos provides no evidence for epistasis a nd suggests that Irf6 and Grhl3 function in independent but converging pathways during oral periderm development. . As previously observed in the single knockout Irf6 -/- and Grhl3 -/- embryos, we detected a reduction in expression of Krt6 in both heterozygous embryos ( Fig. 7 G vs 4H,I) and a more apparent reduction of Krt6 in the double heterozygous embryos ( Fig. 7 J). At higher magnification, the loss in Krt6 staining coincided with the loss of oral periderm cells ( Fig. 7 K vs 4L-N). We did not detect any change in p63 expression in the Irf6 +/- embryos ( Fig. 7 O vs 4P). However, in the Grhl3 +/- (Fig. 7 Q) and the Irf6 +/-;Grhl3 +/- (Fig. 7 R) embryos , we observed a loss of expression of p63, indicating a loss of the basal epithelial cells at the sites of the oral fusions. Again, these molecular data suggest that Irf6 and Grhl3 function independently during palatal development. Although we did not dete ct epistasis between Irf6 and Grhl3 during palatal development, we observed a 12% (6/51) rate of resorbing embryos ( Table S3). This frequency was significantly higher than expected (3%, p -value = 0.0008) for the C57Bl/6 murine strain 27. In addition, while we observed a Mendelian distribution of pups at birth (postnatal day 0, P0), Irf6 +/-;Grhl3 +/- pups were significantly under -represented at P21 (p -value = 0.01). Thus, pre -natal and post -natal lethality from crosses that generated the double heterozygous pups suggest positive epistasis between Irf6 and Grhl3 at other timepoints and/or tissues during development. DISCUSSION Using a combination of whole -exome and Sanger sequencing methods, we identified mutations in GRHL3 in eight families with VWS that had no causative mutations in IRF6 , thus demonstrating that , when mutated, GRHL3 is the gene responsible for VWS at the VWS2 locus . Although previous studies had found IRF6 mutations in 70% of families with VWS, there had been very !!(%!little evidence for locus heterogeneity. Despite 15 published linkage studies on 49 families from throughout the world 28, only one pedigree demonstrated linkage outside of t he IRF6 locus 15. Since this family originated from Finland, a relatively isolated population, and since, at that time, only one member of the family had lip pits, the cardinal fe ature of VWS, the broader impact of this family on VWS genetics was uncertain. However, the finding of causative mutations in seven additional families from broad geographic and phenotypic spectra supports the clinical and biological significance of this l ocus for VWS, and demonstrates that locus heterogeneity contributes to the genetic architecture of VWS. The results from our mutation screen also suggest a complex allelic architecture for GRHL3 in VWS. Based on the precedent of IRF6 , we hypothesized that VWS due to mutation at the second locus ( VWS2 ), would be caused by haploinsufficiency of GRHL3. Consistent with this hypothesis, we observed both missense and protein truncation mutations. In addition , the DECIPHER database (Database of Chromosomal Imbalan ce and Phenotype in Humans using Ensembl Resources) 29 includes a 1.9Mb de novo deletion encompassing GRHL3 in a n individual with CP, club foot, developmental delay, prominent forehead, and a thin upper lip . In our small number of cases, we also observed a case of compound heterozygous alleles for GRHL3 (proband in VWS -III) and another case with a rare variant in both IRF6 and GRHL3 (proband in VWS -IV). However, all five GRHL3 variants used in the zebrafish as say, including both alleles of the compound heterozygote individual , uniformally tested as dominant -negative. If VWS -associated GRHL3 alleles also have a dominant -negative effect in human tissues, it is not clear why they would be found in a coupled state, how the protein truncation alleles remain stable and whether GRHL3 part icipates in protein complexes. Further genetic and biochemical studies will be required to understand the effects of these alleles in human tissues. !!(&!The analysis of human phenotypes su ggests two clinical hypotheses. First, since individuals with GRHL3 mutations were more likely to have CP and less likely to have CL/P than individuals with IRF6 mutations, this association may be used to prioritize these two genes for mutation screens in VWS cases. We note that this association was made from a small number of individuals with GRHL3 mutations (n = 27) and that 9 individuals originated from one fam ily (VWS -I). However, when we restricted the analysis to a family -based phenotype (n = 8), we observed the same trends, although not achieving statistical significance due to low power. Second, like IRF6 , common DNA variants in GRHL3 may also be associated with isolated forms of orofacial clefting 30, especially for CP, given the increased likelihood of CP in individuals with a mutation in GRHL3. However, multiple genome -wide association studies for CL/P 31 and one for CP 32 have not provided strong evidence for common variants at the GRHL3 locus. While these studies suggest that common DNA variants in GRHL3 do no account for significant risk for CL/P or CP, GRHL3 remains an excellent candidate gene for isolated orofacia l clefts. Finally, our analysis of phenotypes in Irf6 and Grhl3 mutant mice identified common and distinct oral abnormalities. Previous studies revealed that Irf6 deficiency in mice could lead to an orofacial cleft by at least two pathophysiological mechan isms: abnormal periderm differentiation and failure of the medial edge epithelium (MEE) to dissolve 11; 33 . Since the MEE was able to dissolve normally in embryos that lack Grhl3 , the common feature of Irf6 and Grhl3 mutants is failed periderm differentiation, strengthening the previously hypothesized role of periderm in development of the lip and pal ate. In conclusion, these studies identif y GRHL3 as the second gene which when mutated leads to Van der Woude syndrome , thus confirming locus heterogeneity for this syndrome . Further , they strengthen the connection between cleft palate and abnormal periderm development. We !!('!anticipate that these findings will imp rove the molecular diagnostic for VWS and other forms of orofacial clefting SUPPLEMENTAL DATA DESCRIPTION Supplemental data include three figures and three tables . ACKNOWLEDGEMENTS We greatly appreciate the many individuals affected with VWS , their family members and clinicians for participating in this study. We would like to thank Arianna L. Smith and Mager Scientific for technical assistance ; Nicole Patel for the artistic renderi ngs of murine embryos at E13.5 and E15.5 ; P−ivi Lahermo for providing the Finnish controls and Dr. Pat Venta for critiques . Financial support for this research was provided by the Swedish Research Council 521-2007-3133 (MP-J) and 2009-5091 (JK) , NIH grants DE021071 (RAC), DE13513 (BCS), F31DE022696 (YAK), DE08559 (JCM), GM008629 (EJL), AR061586 (MD), AR44882 (BA) and the Sigrid Jus”lius Foundation (JK). The authors do not have any conflicts of interest. WEB RESOURCES Online Mendelian Inheritance in Man: www.omim.org Picard toolkit: picard.sourceforge.net GATK: www.broadinstitute.org/gatk/ Primer3: biotools.umassmed.edu/bioapps/primer3_www.cgi NHLBI/ESP database: evs.gs.washington.edu/EVS/ UCSC genome browser: genome -euro.ucsc.edu/cgi -bin/hgGateway !!("! MATERIAL S AND METHODS Human DNA samples DNA samples from 45 families of multiple ethnicities and who were completely sequenced for IRF6 without identifying a causative mutation were used in this study. All subjects were examined by clinical geneticists or genetic counselors who made diagnoses as described previously 15-17. Written informed consent w as obtained for all subjects and all protocols were approved by the local ethical boards in Helsinki (Finland) or in Stockholm (Sweden), or by the Institutional Review Boards at the University of Iowa (U.S.A.). Three hundred and sixty unrelated individuals without a history of oral cleft from the Philippines were used as controls for the GRHL3/c.1171C>T Filipino variant, while 561 unrelated Finnish individuals (blood donors) were used as controls for the GRHL3/ c.969 -970insTG , the PHACTR4 /c.1615G> A/rs200581707 and the KTI12 /c.337_363delCCGATCGCGGGACCTCAGGTGGCGGGC Finnish variants. Targeted exome sequencing Genomic DNA from eight affected and three healthy individuals from the VWS2 Finnish family underwent SureSelect Target Enrichment (Agilent Tech nologies) in order to perform sequence capture of the exome. Enriched samples were sequenced on an Illumina HiSeq instrument. Reads were aligned to reference sequence with the bwa read mapper 18. A high -quality variant call set was generated based on a best -practice workflow 19, in which we utilized the Picard and Genome analysis toolkit (GATK) for data process ing and analysis. !!((!Genotyping Genotyping of the GRHL3 c.969 -970insTG (Finnish) and c.1171C>T (Filippino) variants, the PHACTR4 /c.1615G>A/rs200581707 (Finnish) variant, was performed using TaqMan SNP Genotyping Assays (Life Technologies, Grand Island, NY) on the ABI Prism 7900HT or ABI 7500 and analyzed with the SDS 2.3 or SDS 1.4 software (Applied Biosystems, Foster City, CA), respectively. Family relationships for appa rently de novo variants ( c.1171C>T and c.1559_1562delGGAG ) were confirmed by genotyping 16 markers distributed across the genome (Table S2). The KTI12 /c.337_363delCCGATCGCGGGACCTCAGGTGGCGGGC variant was genotyped using PCR amplification using SYBR green labelling of the wild type (100 bp) and the deleted (73 bp) alleles, and checked for their respective melting temperatures/curves. Mutation screening by Sanger sequencing Primers for GRHL3 we re designed to amplify the exons of all isoforms of GRHL3 using Primer3. The exons of all four GRHL3 transcript variants were screened in a total of 13 PCR amplicons (Table S1). PCR reactions were incubated at 94¡C for 5 min followed by 35 amplification cy cles (45 s at 94¡C, 45 s at 60¡C, 45 s at 72¡C) and a final extension at 72¡C for 7 min. PCR products were sent for sequencing using an ABI 3730XL (Functional Biosciences, Inc., Madison, WI). Chromatograms were transferred to a UNIX workstation, base -called with PHRED (v.0.961028), assembled with PHRAP (v. 0.960731), scanned by POLYPHRED (v. 0.970312), and viewed with the CONSED program (v. 4.0). The effects of missense variants were predicted using the Variant Effect Predictor program 20 which generates scores from Polyphen2 and SIFT. !!()!Phenotype Analysis Affected individuals with GRHL3 mutations (n = 27) were assigned a phenotype classification of cleft lip with or without cleft palate (CL/P which includes CL and CLP cases), cleft palate (CP), lip pits only, CL/P with lip pits, or CP with lip pits based on the clinical diagnoses. Additional phenotypic classifications described the presence of dental anomalies (hypodontia, dental aplasia, or malocclusion), limb anomalies (syndactyly, polydactyly, club foot or contractures), or popliteal pterygia. Fr om the set of families positive for IRF6 mutations 8; 13; 14; 17; 21 , affected individuals were also assigned to the same phenotype classifications (n = 632). Exclusion criteria for this analysis were individuals wit h a cleft but without identified familial mutation (i.e. potential phenocopies), and individuals diagnosed with VWS without a known IRF6 or GRHL3 mutation. Transfection of human GRHL3 mutation variants into Zebrafish embryos Full -length, wild -type human GRHL3 cDNA variant 4 (v4) was obtained as a cDNA clone from Open Biosystems (MHS1010 -9204655) and shuttled by Gateway cloning into the CS2+ destination vector (kindly provided by Dave Turner, University of Michigan). This construct was used for in vitro sy nthesis of wild type GRHL3 mRNA. Specific mutations from VWS -affected individuals were generated in the GRHL3 mRNA (v4) using PCR -mediated mutagenesis and the resulting cDNAs engineered into CS2+, resulting in the truncation of the first 6 bp of 5ÕUTR and the last 70 bp of 3ÕUTR from mutant variants. These constructs were further used for in vitro synthesis of mutant variants of GRHL3. These truncations (the first 6 bp of 5ÕUTR and the last 70 bp of 3ÕUTR from mutant variants) had no functional consequence, as we tested a similarly truncated and cloned wild type GRHL3, and GRHL3 mRNA synthesized from this construct behaved equivalently to full -length GRHL3 in the zebrafish -based functional assay. !!(*!Capped mRNA was synthesized in vitro (mMESSAGE mMACHINE SP6 ki t, Ambion Inc., Austin, TX), purified using the MEGAclear kit (Ambion Inc., Austin, TX) and approximately 1 ng of mRNA was injected into wild type zebrafish embryos (Scientific Hatcheries outbred strain, Huntington Beach, USA), at the one cell or, for mosa ic injections, at the 16 -cell stage. Embryos were fixed at 50% epiboly or corresponding time -point (5 -6 hpf), and whole mount in situ hybridization for krt4 was performed as previously described 22. Plasmids used for probe synthesis are available upon request. Embryos were injected with biotinylated -dextran (Invitrogen, D -1956) and processed for visualization as previously described 4. Animal use protocols were approved by the Public Health Service Assurance. Murine crosses We crossed mice heterozygous for the Irf6 genetrap allele ( Irf6 +/gt ; here referred to as Irf6 +/-) 9 with mice heterozygous for the Grhl3 knockout allele ( Grhl3 +/-) 3 to generate wild type, Irf6 +/-, Grhl3 +/- and Irf6 +/-;Grhl3 +/- double heterozygous embryos. Grhl3 knockout embryos were obtained by crossing Grhl3 +/- mice. Presence of a copulation plug was denoted as E0.5. Pregnant dams were injected intraperitoneally with BrdU (Sigma) two hours before euthanization at a dose of 100 mg per gram pregnant dam body weight. Embryos were collected at indicated timepoint s and genotyped for Irf6 and Grhl3 null alleles as described previously 3; 9 . Both alleles were maintained on a C57BL/6 background. Animal use protocols were approved by the Institutional Animal Care and Use Committe es at Michigan State University and the University of California, Irvine, U.S.A. !!(+!Morphological, histological and molecular analyses of mice Gross morphological analysis of the Irf6 +/- by Grhl3 +/- cross was done at E13.5, E17.5, P0 and P21. Embryos were then fixed in 4% paraformaldehyde, embedded in paraffin and sectioned at 7 mm intervals. Haematoxylin and Eosin staining was performed as described 9. For immunostaining, antigen retrieval was performed in sodium citrate, followed by blocking steps in BSA and a Goat anti -mouse Fab fragment (Jackson ImmunoResearch Laboratories, 115 -007-003). Primary antibody was incubated overnight at 4¡C and secondary antibody was incubated for 1.5 h at room temperature. We used primary antibodies against Keratin 6 (Covance, PRB -190 169P), tumor protein p63 (Santa Cruz, 4A4, SC -8431), Irf6 (Sigma -Aldrich, SAB2102995) and Activated Notch1 (Act N1, Cell Signaling, Val1744, D3B8, 4 147S). We used the following secondary antibodies: goat anti -rabbit (Molecular Probes, A21429), goat anti -mouse (Molecular Probes, A11029) and goat anti -rat (Molecular Probes, A11006). Nuclei were stained with DAPI (Invitrogen, D3571) followed by slide mou nting in ProLong Gold Antifade Reagent (Invitrogen, P36930). Imaging Histological and immunostained sections were imaged with a Nikon Eclipse 90i upright microscope using a Plan APO 10x/0.45 DIC, a CFI Plan Apo Lambda 20x/0.75 and a Plan APO 40x/0.95 DIX M/N2 objectives. A Nikon DS -Fi1 high -definition camera head and a DigitalSight PC-use control unit were used for Haematoxylin and Eosin imaging. A X -Cite Series 120Q laser and a CoolSnap HQ2 photometric camera were used to obta in immunofluorescent images. NIS Elements Advanced Research v3.10 was used for RAW image deconvolution and Adobe Photoshop Elements v9.0 was used for figure formation. !!)#!Statistical analysis FisherÕs exact test in STATA (v12.1) was used to compare the frequ encies of VWS -associated phenotypes between individuals with GRHL3 mutations and those with IRF6 mutations. The threshold p -value for this analysis was calculated using a Bonferroni correction (p = 0.05; 8 phenotypes = 0.006). We used Chi -Squared Analysis to compare the observed genotype distributions of mice with the predicted Mendelian frequencies. Previous reports show that resorption rates in C57Bl/6 mice range between 1 -3%. We used a two -tailed FisherÕs exact test to compare the upper limit of this ran ge with the observed resorption rates. !!!!!!!!)$!!!! APPENDIX !!"#!APPENDIX Figure 4: Mutations in GRHL3 ca use Van der Woude syndrome . !!"$!Figure 4. (contÕd) !!"%!Figure 4. (contÕd) (A,B) Clinical images of the proband from families VWS -II (A) and VWS -VII (B) display the cardinal feature of VWS, i.e. lip pits (arrowhead). Sequence tracks from each individual are shown to the right with an arrow pointing to the base affected by the mut ation. Note that the sequence for c.1559_1562delGGAG is to be read from the reverse strand. (C) GRHL3 has four alternative transcripts variants, v1 to v4 (UCSC genome browser), with three alternative first exon (1, 1Õ and 1Ó) and two alternative last exons (16 and 16Õ). Translation starts in the first exon of each variant except for v4 where translation starts in exon 2, and stops in the last exon of each variant. The genomic location and cDNA change of each of the nine mutations observed are indicated (acc ording to v3, NM_198174.2). The mutation found in the original Finnish family (VWS -I) is indicated by a filled -circle. Colors for the exons are corresponding to their coding for the GRHL3 protein domains. (D) Schematic representation of the GRHL3 protein product v2, (NP_ 937816) with at scale, the three known protein domains : the trans activation (orange), the DNA binding (green) and the dimerization (pink) domains. The position of each change in the protein sequence is also indicated. Please note that as no mutation was found in exon 16, the denomination for each amino acid changes is valid both in v2 and v3. More d etails of the v2 full protein sequence can also be found in Fig. 9 . !!"&! Figure 5: VWS-associated alleles of GRHL3 disrupt the development of the periderm when expressed in zebrafish embryos . !!"'!Figure 5. (contÕd) (A-C) Lateral views of live sibling embryos injected with indicated mRNA. Embryo shown in C, injected with the GRHL3 mRNA carrying the c.1171C>T mutation , ruptured through the animal hemisphere shor tly after th e image was taken (67% [n = 48] of wild type GRHL3-injected embryos reached at least 50% epiboly stage, while 76% [n = 115] of mutant -injected embryos burst without initiating epiboly) . (D) Histogram showing fraction of embryos that rupture d when injected with indicated mRNA. Percentage is the average from 3-4 separate experiments of 20 -40 embryos each . (E,F ) Animal pole views of embryos injected with indicated mRNA and processed to detect krt4 expression . Insets, cross sections of the same embryos showing (E) krt4 expression confined to the periderm and (F) ectopically in deep cells. (G,H ) Animal pole views of mosaic embryos injected with mRNA and biotinylated -dextran at 16 -cell stage, fixe d at shield stag e, and processed for krt4 expression (blue) and biotin distribution (brown). Periderm cells possess ed (black arrowhead) or lack ed (white arrowhead) biotin stain , demonstrating that they were, or were not, derived from an RNA injected cell, respectively. Daughter c ells derived from the cell injected with the c.893G>A mutant variant of GRHL3 lack krt4 expression. Scale bars represent 500 µm (A-C,E,F ), 100 µm (E,F inset) , and 20 µm (G,H ). !!""! Figure 6: Grhl3 is required for murine periderm and palatal development. !!"(!Figure 6. (contÕd) !!")!Figure 6. (contÕd) (A-C) Haematoxylin and Eosin staining of coronal sections of posterior palate at E15.5 (AÕ). Wild type embryos showed complete fusion of palatal shelves (*) (A). In contrast, Irf6 -/- embryos have bilateral oral adhesions (arrows) and a fully penetrant clef t palate (*) (B). Similarly, Grhl3 -/- embryos have bilateral oral adhe sions (arrows) (C ). However, in Grhl3 -/- embryos, adhesions were restricted to areas superficial to the tooth germ and palatal surfaces, and a cleft palate was o bserved in 1 of 6 embryos (*) (C). (D-F) Immunostaining for Krt6 (red) and p63 (green). Krt6 was expressed uniformly in the periderm superficial to the tooth germ (arrow) of wild type embryos (D ) (from boxed structure in A ), but was very weak ly expressed in Irf6 -/- (E) and Grhl3 -/- (F) embryos. P63 was expressed uniformly in the basal epithelium of wild type (D) and Grhl3 -/- (F) embryos, but was expressed ectopically in suprabasal cells in Irf6 -/- embryos (E). (G -I) Loss of p63 expression marks normal dissolution of the medial edge epithelium ( MEE) (arrowhead) in wild type (G) and Grhl3 -/- (I) embryos. In contrast, p63 expression persisted around the palatal epithelium in Irf6 -/- embryos (H ). Nuclei are coun terstained with DAPI (blue) (D -I). Sc ale bars are 2 mm for images A -C; 20 mm for D -F; 50 mm for G -I. Labeled oral structures are mandible (mn), maxilla (mx), palatal shelf (p), tongue (t) and tooth germ (tg). !!(*! Figure 7: No evidence for genetic interaction between Irf6 and Grhl3 in murine palatal development. !!(+!Figure 7. (contÕd) !!(#!Figure 7. (contÕd) Haematoxylin and Eosin staining of coronal sections of E13.5 palate at (AÕ) and posterior (DÕ) to the tooth germ. Compared to wild type embryos (A,D), Irf6 +/- embryos had bilateral oral adhesions (arrowheads) at the tooth germ site (B). In contrast, Grhl3 +/- littermates had oral adhesions (arrowheads) and fusions (arrow) located predominantly posterior to the tooth germ (E). Irf6 +/-;Grhl3 +/- embryos (C,F) have oral adhesions (arrowheads) at the tooth germ (C) as well as adhesions (arrowheads) and fusions (arrow) posterior to the tooth germ (F). Krt6 immunostaining (red) of the oral periderm (G -N). Compared to wild type embryos (G and enlarged in K ), Krt6 express ion in Irf6 +/- (H, enlarged in L ), Grhl3 +/- (I, enlarged in M ), and Irf6 +/-;Grhl3 +/- (J, enlarged in N ) embryos was markedly reduced along the oral surface of the palatal shelves and the mandible. Loss of Krt6 expression coincides with oral adhesions (arrow heads) and fusions (arrow ) (G-N). P63 immunostaining (green) of the basal epitheliu m was continuous in wild type (O) and Irf6 +/- (P) embryos. In contrast, p 63 staining of Grhl3 +/- (Q) and Irf6 +/-;Grhl3 +/- (R) embryos was discontinuous. Oral fusions are seen between surfaces of the palate and mandible with mesenchymal communication (arrows) punctuating islands of p63 positive epithelial cells (arrow heads). Scale bars are 2 mm (A -F, G -J and O -R) and 20 mm (K -N). Labeled oral structures are mandible (mn ), maxilla (mx), palatal shelf (p), tongue (t) and tooth germ (tg). !!($!Table 1: GRHL3 mutations in eight Van der Woude syndrome families VWS pedigree Origin DNA change b Protein change d Genomic position f Exon De novo / Familial Ia Finland c.970_971insTG p.Phe324Leufs*22 1:24666175 8 Familial II Philippines c.1171C>T p.Arg391Cys e 1:24668728 9 De novo III Israel c.[268_278delTACTA CCATGG];[1661A>G] c p.[Tyr90Hisfs*4];[Asn554Ser] c 1:24662973 -24662983; 24676579 4 15 Familial IV Pakistan c.893G>A p.Arg298His e 1:24664534 7 N/A V U.K. c.1419+1G>T Splice donor site IVS11+1 1:24669516 IVS11 Familial VI U.S.A. c.1559G>A p.Arg520Gln e 1:24673973 14 N/A VII Swedish c.1559_1562delGGAG p.Glu522Leufs*10 1:24673973 -24673976 14 De novo VIII U.S.A./ African American c.1575delG p.Val526Cysfs*7 1:24673989 14 Familial !!(%!Table 1. (contÕd) a Family originally studied by linkage analysis in 15 and presently, by exome sequencing b Position on GRHL3 cDNA vari ant 3 (v3) NM_198174.2 c Mutations occurring in the same family but on separate chromosomes as indicated d Position on GRHL3 protein product NP_937817.3 e Missense mutation predicted to be damaging by Polyphen2 and SIFT using the Variant Effect Preditor program f Position according to the human genome reference hg19 N/A Not applicable as parent DNA unavailable !!(&!Table 2 : Comparison of VWS phenotypes caused by mutations in IRF6 and GRHL3 Has Phenotype? CL/P a CP Cleft only b Lip Pits Lip Pits Only Dental anomalies c Limb defects d Pterygia e Yes 3 19 12 14 5 2 2 0 No 24 8 15 13 22 25 25 27 GRHL3 (n = 27) % 11 70 44 52 19 7 7 0 Yes 267 159 141 445 158 70 45 10 No 365 473 491 187 474 562 587 622 IRF6 (n = 632) % 46 27 24 76 27 12 8 2 p-value 0.001 2.0!10-6 0.02 0.05 0.65 0.76 1 1 a Includes cleft lip (CL) and cleft lip and palate (CLP) b Includes cleft palate (CP), CL or CLP but without lip pits. c Dental anomalies include hypodontia, dental aplasia, and malocclusion d Includes syndactyly, polydactyly, club foot, contractures and pterygium e Only pterygia coun !!('! Figure 8: Pedigrees of the eight VWS families with GRHL3 mutation .!!("!Figure 8. (contÕd) !!((!Figure 8. (contÕd) !!()! Figure 8. (contÕd) !!)*!Figure 8. (contÕd) In each family, the corresponding GRHL3 mutation is named under each individual where it is detected. Mutations were detected by whole -exome sequencing (* in VWS -I only), TaqMan genotyping (^) or Sanger sequencing (#). Mutation carriers without any detecte d phenotypic characteristics of VWS are indicated with a black dot in their symbol. Phenotypical characteristics of affected individuals are cleft palate (CP), cleft lip and palate (CLP) or unknown (?). The proband in VWS -IV has been shown to be carrier of a rare variant in IRF6 (K80R) 1. VWS -I was previously described in 2, VWS -IV in 1 under the denomination VWS -SM13 and VWS -VII in 3 under the denomination VWS -12. !!"#! Figure 9: Multiple alignment and protein domains of GRHL3 gene products from human, mouse and zebrafish . !!"$!Figure 9. (contÕd) The alignment was done with the input p rotein sequences from human (NP_937816 for H.s.GRHL3v2, Homo sapiens variant 2) , Mouse (NP_001013778.1 for M.m.Grhl3, Mus musculus) and zebrafish (XP_001332938.3 for D.r.Grhl3, Danio rerio) and using Clustal Omega 1.1.0 software (www.ebi.ac.uk/Tools/msa/clustalo ). Amino acids conserved in the 3 species, are denoted by a star below the alignment. In the human protein sequence, the corresponding coding exons are numbered 1 to 16 and are identifiable by a blue or a black protein sequence. Amino acids in red are encod ed by two neighboring exons. In the human GRHL3 protein sequence, the three known protein domains are underlined in orange for the transactivation domain (exons 2 and 3, amino acids 25 -74), in green for the DNA -binding domain (exons 6 to 10, amino acids 230-423) and in pink for the dimerization domain (exons 13 to 16, amino acids 493 -602). For each of the exonic mutations/variants detected in our set of families, the first amino acid affected by the mutation is labeled in red (protein truncation) or in gree n (missense), and a red arrow indicates the location of the splice site mutation IVS11+1 (from VWS -V). The known repressive and activating protein domains of the murine Grhl3 are underlined in the mouse sequence by a continuous line (repressive, amino acid s 1 -102 and 296 -603) or dotted line (activating, amino acids 102 -296) 4. A blue arrow above human exon 2 indicates the start of the GRHL3 protein produced by variant 4 (v4), and used in the zebrafish experiments . !"#!!Figure 10 : Molecular changes in the oral epithelium of Irf6 -/- and Grhl3-/- embryos . !"$!Figure 10. (contÕd) Comparison of expression in E15.5 wild type (A,E ), Irf6 -/- (B,F) and Grhl3 -/- (C,D,G,H ) embryos. Images in columns 3 and 4 are taken at the tooth germs sites in two different Grhl3 -/- embryos to illustrate the dynamic changes in gene expression in areas without (C,G) and with oral adhesions ( D,H ), respectively. Irf6 expression is seen in ep ithelial cells of the tooth germ and oral ep ithelium in wild type embryos (A ) but absen t in Irf6 -/- embryos (B ). Irf6 expression is detected in the oral epithelium (arrowhead) of Grhl3 -/- embryos, but reduced at sites of oral adhesion (arrow) (C,D). Activa ted Notch1 (Act N1) is seen in the oral periderm of wild type embryos (E) while it is undetectable in Irf6 -/- littermate embryos (F ). Grhl3 -/- embryos (G,H ) show loss of Act N1 in areas of oral adhesion (arrow) but not in adjacent healthy epithelium (arrowhead). Scale bar is 20 µm (A -H). Labeled oral structures are mandible (mn) and tooth germ (tg). !"#!Table 3 : Human GRHL3 primers used in Sanger sequencing mutation screening GRHL3 exons Primer Sequence a 1 and 1' F CTCACCAAGGAAGGAATTGG R TAGCTTGAGACTGGGGCTTG 1'' F GTCTTAGCCGAGCAGCCATAG R GTAGTGGATTTGGGAACCTCCT 2 F GTGGCAGGAAGAGGCAGTTTC R CAAAGGCCCAGAGATGAGG 3 F AAAGCTGCAGGAGGGGATT R TCAGCACTGTGCCTCCTGT 4 and 5 F GCATGCTGGATGGACCTAAA R TTCATCCCCCACTTCTCATT 6 and 7 F TTTTCCAAGGTCAAACAGCA R GACAGAGGTCAGAGCCAGGT 8 F GAGTGAGGCCCAGTTTTTAATG R CGTCGGAGCAAATGACACTA 9, 10 and 11 F CTTGGCAGTCTAGCGGAAAC R GAAGCCTCCTCTTTGTGTGC 12 F CTGAGCAGAATGGGCTAGAA R AGGCGTGTGGTTGTTTCTCT 13 and 14 F TGATGGGCTAAGGGACTCAC R GATAACATCGCAGAGGCACA 15 F GCACACCCAGATGTTAATGG R AGAGGTGACCAGTGGCTTTG 16 F ACCACATCCCCTCTTCCATT R TAGCCATCTCTTTCCAGCAGAC 16' F TTGCTTCTGATACTCCCCACTT R CAGCCCTCTGCTTTTCTCTG a All primers had a melting temperature of 61 oC !!"$! Table 4 : Genomic SNPs used for confirming de novo mutation s SNP Chromosomal band rs1051614 1q21.3 rs10204426 2p11.2 rs237887 3p26.1 rs1063499 5p13.1 rs654351 6p24.3 rs1366883 7q21.13 rs4458901 8p23.2 rs2515617 9q31.1 rs2136892 10q21.1 rs1729410 11q23.3 rs1053900 14q32.2 rs140685 15q12 rs3744262 17p13.1 rs2296241 20q13.2 rs1789953 21q22.3 rs2051616 22q13.31 !"#!Table 5 : Frequency of genotypes and resorbing embryos from Irf6 +/-xGrhl3+/- cross !"#$%&!"#'%&!"()*+,- !./!.0#!.12- !3,456 !78449*- !3 2 5 6 5 11 32 !"#$ :;: <%"&'( :;: !5 3 8 15 5 20 28 !"#$ :;=<%"&'( :;: !4 3 7 10 9 19 26 !"#$ :;: <%"&'( :;=!11 4 15 12 12 24 39 !"#$ :;=<%"&'( :;=!5 4 9 9 1 10 19 >9-,*)8?@ !3 3 6 N/A N/A NA 6 (p=0.0008) 3,456 !25 14 45 46 27 73 118 2=A5619 !0.18 0.96 0.34 0.60 0.01 0.12 0.06 !"$!Figure 5. (contÕd) We intercrossed mice heterozygous for the Irf6 genetrap allele ( Irf6 +/gt here called Irf6 +/-) with mice heterozygous for the Grhl3 knockout allele ( Grhl3 +/-) to generate wild -type ( Irf6 +/+ ;Grhl3 +/+ ) embryos; Irf6 +/- or Grhl3 +/- single mutant embryos; and Irf6 +/-;Grhl3 +/- double heterozygous embryos. We detected significant embryonic resorptions at combined E13.5 and E17.5. Furthermore, we found a significant reduction in Irf6 +/-;Grhl3 +/- double heterozygous mice at weaning. !!!""! !!!!!! BIBLIOGRAPHY !#$$!BIBLIOGRAPHY 1. Mace KA, Pearson JC, McGinnis W (2005) An epidermal barrier wound repair pathway in Drosophila is mediated by grainy head. Science 308:381 -385 2. Ting SB, Caddy J, Hislop N, Wilanowski T, Auden A, Zhao LL, Ellis S, Kaur P, Uchida Y, Holleran WM, Elias PM, Cunningham JM, Jane SM (2005) A homolog of Drosophila grainy head is essential for epidermal integrity in mic e. Science 308:411 -413 3. Yu Z, Lin KK, Bhandari A, Spencer JA, Xu X, Wang N, Lu Z, Gill GN, Roop DR, Wertz P, Andersen B (2006) The Grainyhead -like epithelial transactivator Get -1/Grhl3 regulates epidermal terminal differentiation and interacts functiona lly with LMO4. Dev Biol 299:122 -136 4. de la Garza G, Schleiffarth JR, Dunnwald M, Mankad A, Weirather JL, Bonde G, Butcher S, Mansour TA, Kousa YA, Fukazawa CF, Houston DW, Manak JR, Schutte BC, Wagner DS, Cornell RA (2013) Interferon regulatory factor 6 promotes differentiation of the periderm by activating expression of grainyhead -like 3. J Invest Dermatol 133:68 -77 5. Sabel JL, d'Alencon C, O'Brien EK, Van Otterloo E, Lutz K, Cuykendall TN, Schutte BC, Houston DW, Cornell RA (2009) Maternal Interferon Regulatory Factor 6 is required for the differentiation of primary superficial epithelia in Danio and Xenopus embryos. Dev Biol 325:249 -262 6. Botti E, Spallone G, Moretti F, Marinari B, Pinetti V, Galanti S, De Meo PD, De Nicola F, Ganci F, Castrignano T, Pesole G, Chimenti S, Guerrini L, Fanciulli M, Blandino G, Karin M, Costanzo A (2011) Developmental factor IRF6 exhibits tumor suppressor activity in squamous cell carcinomas. Proc Natl Acad Sci U S A 108:13710 -13715 7. Tamura T, Yanai H, Savitsky D, T aniguchi T (2008) The IRF family transcription factors in immunity and oncogenesis. Annu Rev Immunol 26:535 -584 8. Kondo S, Schutte BC, Richardson RJ, Bjork BC, Knight AS, Watanabe Y, Howard E, et al. (2002) Mutations in IRF6 cause Van der Woude and popliteal pterygium syndromes. Nat Genet 32:285 -289 9. Ingraham CR, Kinoshita A, Kondo S, Yang B, Sajan S, Trout KJ, Malik MI, Dunnwald M, Goudy SL, Lovett M, Murray JC, Schutte BC (2006) Abnormal skin, limb and craniofacial morphogenesis in mice deficien t for interferon regulatory factor 6 (Irf6). Nat Genet 38:1335 -1340 10. Richardson RJ, Dixon J, Malhotra S, Hardman MJ, Knowles L, Boot -Handford RP, Shore P, Whitmarsh A, Dixon MJ (2006) Irf6 is a key determinant of the keratinocyte proliferation -differentiation switch. Nat Genet 38:1329 -1334 !#$#!11. Richardson RJ, Dixon J, Jiang R, Dixon MJ (2009) Integration of IRF6 and Jagged2 signalling is essential for controlling palatal adhesion and fusion competence. Hum Mol Genet 18:2632 -2642 12. Burdick AB, Bixler D, Puckett CL (1985) Genetic analysis in families with van der Woude syndrome. J Craniofac Genet Dev Biol 5:181 -208 13. de Lima RL, Hoper SA, Ghassibe M, Cooper ME, Rorick NK, Kondo S, Katz L, Marazita ML, Compton J, Bale S, Hehr U, Dixon MJ, Daack -Hirsch S, Boute O, Bayet B, Revencu N, Verellen -Dumoulin C, Vikkula M, Richieri -Costa A, Moretti -Ferreira D, Murray JC, Schutte BC (2009) Prevalence and nonrandom distribution of exonic mutations in interferon regulatory factor 6 in 307 fam ilies with Van der Woude syndrome and 37 families with popliteal pterygium syndrome. Genet Med 11:241 -247 14. Leslie EJ, Standley J, Compton J, Bale S, Schutte BC, Murray JC (2012) Comparative analysis of IRF6 variants in families with Van der Woude syndr ome and popliteal pterygium syndrome using public whole -exome databases. Genet Med 15:338 -344 15. Koillinen H, Wong FK, Rautio J, Ollikainen V, Karsten A, Larson O, Teh BT, Huggare J, Lahermo P, Larsson C, Kere J (2001) Mapping of the second locus for the Van der Woude syndrome to chromosome 1p34. Eur J Hum Genet 9:747 -752 16. Schutte BC, Bjork BC, Coppage KB, Malik MI, Gregory SG, Scott DJ, Brentzell LM, Watanabe Y, Dixon MJ, Murray JC (2000) A preliminary gene map for the Van der Woude syndrome critical region derived from 900 kb of genomic sequence at 1q32 -q41. Genome Res 10:81 -94 17. Peyrard -Janvid M, Pegelow M, Koillinen H, Larsson C, Fransson I, Rautio J, Hukki J, Larson O, Karsten AL, Kere J (2005) Novel and de novo mutations of the IRF6 gene detec ted in patients with Van der Woude or popliteal pterygium syndrome. Eur J Hum Genet 13:1261 -1267 18. Li H, Durbin R (2010) Fast and accurate long -read alignment with Burrows -Wheeler transform. Bioinformatics 26:589 -595 19. DePristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, Hartl C, Philippakis AA, del Angel G, Rivas MA, Hanna M, McKenna A, Fennell TJ, Kernytsky AM, Sivachenko AY, Cibulskis K, Gabriel SB, Altshuler D, Daly MJ (2011) A framework for variation discovery and genotyping using next -gener ation DNA sequencing data. Nat Genet 43:491 -498 20. McLaren W, Pritchard B, Rios D, Chen Y, Flicek P, Cunningham F (2010) Deriving the consequences of genomic variants with the Ensembl API and SNP Effect Predictor. Bioinformatics 26:2069 -2070 21. Pegelow M, Koillinen H, Magnusson M, Fransson I, Unneberg P, Kere J, Karsten A, Peyrard -Janvid M (2013) Association and mutation analyses of the IRF6 gene in families with !#$%!non-syndromic and syndromic cleft lip and/or cleft palate. Cleft Palate -Craniof J 2013 Feb 8. [Epub ahead of print] 22. Thisse C, Thisse B (2008) High -resolution in situ hybridization to whole -mount zebrafish embryos. Nat Protoc 3:59 -69 23. Malik S, Kakar N, Hasnain S, Ahmad J, Wilcox ER, Naz S (2010) Epidemiology of Van der Woude syndrome from mutational analyses in affected patients from Pakistan. Clin Genet 78:247 -256 24. Chalmers AD, Lachani K, Shin Y, Sherwood V, Cho KW, Papalopulu N (2006) Grainyhead -like 3, a transcription factor identified in a microarray screen, promotes the specif ication of the superficial layer of the embryonic epidermis. Mech Dev 123:702 -718 25. Mazzalupo S, Coulombe PA (2001) A reporter transgene based on a human keratin 6 gene promoter is specifically expressed in the periderm of mouse embryos. Mech Dev 100:65 -69 26. Koster MI, Kim S, Mills AA, DeMayo FJ, Roop DR (2004) p63 is the molecular switch for initiation of an epithelial stratification program. Genes Dev 18:126 -131 27. Krishnan L, Guilbert LJ, Wegmann TG, Belosevic M, Mosmann TR (1996) T helper 1 resp onse against Leishmania major in pregnant C57BL/6 mice increases implantation failure and fetal resorptions. Correlation with increased IFN -gamma and TNF and reduced IL-10 production by placental cells. J Immunol 156:653 -662 28. Schutte BC, Dixon MJ, Murr ay JC (2008) Interferon Regulatory Factor 6 (IRF6) is causal in Van der Woude and popliteal pterygium syndromes and contributes to the risk for non -syndromic cleft lip and palate. In: Epstein C (ed) Inborn Errors of Metabolism Oxford University Press 29. Firth HV, Richards SM, Bevan AP, Clayton S, Corpas M, Rajan D, Van Vooren S, Moreau Y, Pettett RM, Carter NP (2009) DECIPHER: Database of Chromosomal Imbalance and Phenotype in Humans Using Ensembl Resources. Am J Hum Genet 84:524 -533 30. Zucchero TM, Coo per ME, Maher BS, Daack -Hirsch S, Nepomuceno B, Ribeiro L, Caprau D, et al. (2004) Interferon regulatory factor 6 (IRF6) gene variants and the risk of isolated cleft lip or palate. N Engl J Med 351:769 -780 31. Ludwig KU, Mangold E, Herms S, Nowak S, Reutt er H, Paul A, Becker J, et al. (2012) Genome -wide meta -analyses of nonsyndromic cleft lip with or without cleft palate identify six new risk loci. Nat Genet 44:968 -971 32. Beaty TH, Ruczinski I, Murray JC, Marazita ML, Munger RG, Hetmanski JB, Murray T, e t al. (2011) Evidence for gene -environment interaction in a genome wide study of nonsyndromic cleft palate. Genet Epidemiol 35:469 -478 !#$&!33. Knight AS, Schutte BC, Jiang R, Dixon MJ (2006) Developmental expression analysis of the mouse and chick orthologues of IRF6: the gene mutated in Van der Woude syndrome. Dev Dyn 235:1441 -1447 34. Malik S, Kakar N, Hasnain S, Ahmad J, Wilcox ER, Naz S (2010) Epidemiology of Van der Woude syndrome from mutational analyses in affected patients from Pakistan. Clin Genet 78:247 -256 35. Koillinen H, Wong FK, Rautio J, Ollikainen V, Karsten A, Larson O, Teh BT, Huggare J, Lahermo P, Larsson C, Kere J (2001) Mapping of the second locus for the Van der Woude syndrome to chromosome 1p34. Eur J Hum Genet 9: 747-752 36. Peyrard -Janvid M, Pegelow M, Koillinen H, Larsson C, Fransson I, Rautio J, Hukki J, Larson O, Karsten AL, Kere J (2005) Novel and de novo mutations of the IRF6 gene detected in patients with Van der Woude or popliteal pterygium syndrome. Eur J Hum Genet 13:1261 -1267 37. Kudryavtseva EI, Sugihara TM, Wang N, Lasso RJ, Gudnason JF, Lipkin SM, Andersen B (2003) Identification and characterization of Grainyhead -like epithelial transactivator (GET -1), a novel mammalian Grainyhead -like factor. Dev D yn 226:604 -617 ! 104! Chapter 3 - Irf6 regulates Tfap2 ! and Grhl3 in neurulation 105!Youssef A. Kousa 1, Huiping Zhu 2, Allison Ashley -Koch 3, Tamara D. Busch 4, Akira Kinoshita 5, Walid D. Fakhouri 6, Raeuf R. Roushangar 1, Trevor J. Williams 7, Yang Chai 8, Brad A. Amendt 9, Jeffrey C. Murray 4, Simon Gregory 3, Gary M. Shaw 10, Alexander G. Bassuk 4, Richard H. Finnell 2, Brian C. Schutte 11 1Biochemistry and Molecular Biology Department, Michigan State University, 48824 East Lansing, Michigan, USA 2Dell Pediatric Research Institute, Department of Nutritional Sciences, University of Texas at Austin, 78723 Austin, Texas, USA 3Duke Molecular Physiology Institute, Department of Medicine and Molecular Genetics and Microb iology, Duke University, 27701 Durham, NC, USA 4Department of Pediatrics, University of Iowa, 52242 Iowa City, Iowa, USA 5Department of Human Genetics, Nagasaki University, Nagasaki, Japan 6Department of Diagnostic & Biomedical Sciences , School of Dentistr y, University of Texas at Houston, 77054 Houston, Texas, USA 7Department of Craniofacial Biology, University of Colorado Denver at Anschutz Medical Campus, 80045 Aurora, Colorado, USA 8Center for Craniofacial Molecular Biology, Ostrow School of Dentistry, University of Southern California, 90033 Los Angeles, California, USA 9Department of Anatomy and Cell Biology, University of Iowa, 52242 Iowa City, Iowa, USA 10Department of Pediatrics, Stanford University School of Medicine, 94305 Stanford, California, USA 106!11Department of Microbiology and Molecular Genetics, Michigan State University, 48824 East Lansing, Michigan, USA Abstract Mutations in IRF6 cause orofacial clefting. A DNA variant in the IRF6 enhancer MCS9.7 is present in 30% of the world Õs population , disrupts a TFAP2! binding site and contributes 12% of all orofacial clefting risk. Here, we show that Tfap 2! positively regulates Irf6 through MCS9.7 and that Irf6 regulates Tfap 2!. Moreover, we show that a null allele of Irf6 completely rescues haploinsufficiency of Tfap 2! and that both over and under -expressing Irf6 causes defects in neurulation. We also show that Tfap2 ! interacts with Grhl3 through Irf6 in caudal neurulation. Finally, w e sequenc e IRF6 in pa tients with spina bifida and find a rare coding mutation previously identified in orofacial clefting . This discovery illuminates a novel pathway that regulates orofacial and neural tube defects and may contribute to our understanding of a common, derived IRF6 regulatory variant that increases risk for orofacial clefting. Introduction IRF6 encodes a member of the Interferon Regulatory Factor of transcription factors. While the IRF family widely regulates immun ity1, IRF6 is required for craniofacial, skin and limb development 2. Mutations in IRF6 cause two Mendelian orofacial clefting disorders, Van der Woude syndrome (VWS; OMIM#119300) and Popliteal Pterygium Syndrome (PPS ; O MIM#119500) 2. Furthermore, a DNA variant (rs642961) within the IRF6 locus is found in 30% of the worlds population and contributes 12% of all non - 107!syndromic orofacial clefting risk 3. The variant is located 9.7 Kb upstream of IRF6 within a multi -species conserved enhancer sequence ( MCS9.7). MCS9.7 recapitulates endogenous Irf6 expression 3, 4 . Interestingly, the variant abrogates one of four binding sites for Transcription Factor Activating Protein 2 ! (TFAP2 ! )3. Mutations in TFAP2! can lead to Branchio -oculo -facial syndrome (BOFS; OMIM#113620) which, li ke VWS and PPS, can include orofacial clefting, lip pits and cutaneous abnormalities 5-7. Prior work in primary kertinocytes show s that both TFAP2 ! and TP63 co -regulate IRF6 expression via MCS9.7 8. However, in vivo regulation of MCS9.7 by TFAP 2! has not been demonstrated. Considering developmental def ects in VWS, PPS and BOFS, both Irf6 and Tfap 2! murine models illuminate pathophysiological mechanisms . In the mouse, loss of Irf6 leads to a hyperproliferative epidermis, craniofacial defects that include a cleft p alate and appendage deformities considered secondary to skin adhesions 9, 10 . While the phenotype is dissimilar to Irf6 , loss of Tfap 2! also leads to skin, limb and craniofacial defects 11, 12 . However, loss of Tfap2 ! also leads to multiple neurulation defects. Indeed, the Tfap 2! knockout mouse is unique in the biomedical literature for what can be considered whole -body clefting. Neurulation defects span the entire neural tube and include kin ked tail (closure point 1/posterior neuropore, caudal), bilateral outgrowth of unfused mandibular bones and a bifurcated tongue (split -face malformation) and anencephaly (closure point 2 /3, rostral). Strikingly, loss of Tfap 2! leads to schi sis, or extrusio n, of both thoracic and abdominal cavity contents (thoraco -abdomino -schisis). A 108!role for Irf6 in neurulation or abdomino -thoracic wall closure has not been shown in mouse or man, even in cases where disrupted gene function leads to VWS and PPS. Also, a rol e for TFAP2 ! in human neural tube development has not been in BOFS or isolated spina bifida. Many of the biological functions attributed to Tfap 2! and Irf6 are derived from their function in skin, an abundant and accessible tissue. Among epidermal transcr iption factors, Irf6 is unique because it has been reported to have both transcriptional and post -translational regulation of down -stream targets. One prominent transcriptional target of Irf6 in zebrafish is Grhl3 13. Like Tfap 2!, loss of Grhl3 leads to skin, limb, craniofacial and both rostral and caudal neural tube defects 14. Significantly, we recently showed that mutations in GRHL3 can also lead to Van der Woude Syndrome (VWSII; OMIM# 606713) (Peyrard, 2013) . On the other hand, TP63 drive s Irf6 expression via MCS9.7 15 and IRF6 down -regulate TP63 via the proteasome 16. Importantly, mutation in TP63 also lead to dominantly inherited orofacial clefting in Ectrodactyly, Ectodermal Dysplasia (EEC; OMIM#604292) 17. In this work, we dissect the nature of Tfap2 !-Irf6 -Grhl3 genetic interaction in vivo . We show that Tfap2 ! is necessary for MCS9.7 enh ancer activity in the mouse, a genetic relationship that may explain orofacial clefting risk within the IRF6 locus . We titer Irf6 dose in vivo and find that reducing Irf6 expression leads to a curly tail , a caudal neural tube defect, through regulation of Tfap2a and Grhl3 . Furthermore, we find that 11% of embryos over -expressing Irf6 have craniofacial and neural tube defects (NTD) that 109!phenocopy Tfap2 ! heterozygous and knockout embryos. Mo reover, while loss of Irf6 has multiple pathophysiological consequences, we report here that a null allele of Irf6 completely rescues craniofacial and neural tube defects in Tfap2 ! haploinsufficiency. Finally, to assess the role of IRF6 in h uman neurulation, we sequenced 92 individuals with spina bifida. We found a rare missense mutation at a highly conserved amino acid in exon 9 of IRF6 (D427Y) , previously reported in orofacial clefting but not in any control database . However, sequencing of the IRF6 3ÕUTR shows no association with spina bifida risk. O ur discovery illuminates Tfap2 !-Irf6 -Grhl3 as a novel gene regulatory network in both orofacial and neural tube development. Results Tfap2 ! is necessary for MCS9.7 activity and Irf6 expression We previously showed that Tfap 2! binds MCS9.7 in vitro . Additional work in primary keratinocytes shows that Tfap 2! and Trp63 co-regulate Irf6 expression via MCS9.7. Here , we predict that Tfap 2! is necessary for MCS9.7 expression in vivo. We analyzed changes in MCS9.7 activity with a LacZ reporter construct crossed into the Tfap 2! knockout mouse (N=66; genotype distribution based on Mandelian genetics, not statistically significant, p -value = 0.71 ; Supp Table 9). At E17 .5, we foun d gross changes in "-Gal staining, most prominently differing from wildtype littermate s in limb, skin and craniofacial structures (Fig. 11 a,b). At E13 .5, we detected more subtle differences in "-Gal activity (Fig. 11 c,d) . Specifically, a s opposed to the linear, evenly stained tail of Wt embryos, Tfap 2!-/- embryos either lacked or had punctuate staining near the distal segments of the closed posterior neuropore . Unlike E17.5 embryos, the in tensity of the 110!LacZ stain at E13 .5 did not differ in cutaneous and limb structures. Combined, these results suggest that Tfap 2! is necessar y for MCS9.7 activity in vivo . We predict that loss of Tfap2 ! trans -activation at MCS9.7 would result in altered Irf6 expression. As shown previously, MCS9.7 is acti ve and Tfap2 ! , Irf6 are co -expressed in the spinous layer. Histologically, loss of Tfap2a leads to a disorganized basal cell layer with a hypertrophic epidermis at E17.5 ( Fig. 11 e-g). We marked the basal cell layer with Tfap2 ! and Trp63 ( Fig. 11 i,j), the spinous layer with Krt1 (Fig. 11 m,n), the cornified and granular layer with Loricrin (Fig. 11 o,p) in wild type and Tfap2 ! knockout embryos . Despite morphological changes, we detected the epidermal cell layer Tfap2a -/- embryos. In wildtype skin, Trp63 is restricted to the basal cell layer and Tfap 2! is expressed in both the basal and spinous cells . Conversely, is Irf6 expressed in the basal layer but emerges more robustly in the spinous and granular cells . Consistent with our prediction, we did not detect Irf6 expression in the spinous layer of Tfap 2! knockout embryos. This pattern is in direct contrast with the adjacent granular and cornified layer s, where a different cohort of Irf6 trans -factors are active 18, 19 . Loss of Tfap2 ! also lead to persistence of Krt6 ( Fig. 11 u,v), a marker of stress, and altered desmocolin expression, perhaps contributing to epidermal folding( Fig. 11 w,x). We previously found that Irf6 restricts Krt14, Trp63, and prolifer ation to the basa l cell layer 10. We predict here that loss of Irf6 in the spinous layer would lead to ectopic Krt14, Trp63 and, as a measure of cellular proliferation , Ki-67. Importantly, previous work shows that loss of Tfap 2! was not sufficient to alter Trp63 expression 8. Consistent with 111!our prediction, loss of Irf6 expression in the spinous layer coincided with ectopic cellular proliferation (Fig. 11 h,i) and spinous -cell expression of both Trp63 and Krt14 (Fig. 11s,t) . This data suggests that loss of Irf6 can uniquely contribute to the pathophysiology of Tfap 2! knockout embryos. Considering TFAP2 ! mutations in the DNA binding domain, reduced IRF6 expression may contribute to BOFS pathology. In addition to qualitative , localized changes in Irf6 gene expression , we also wanted to assess if Irf6 transcript level might be reduced in Tfap 2! knockout skin . However, in comparing Tfap 2!+/+ , Tfap 2!+/- and knockout embryos, we did not find quantitative changes in Irf6 mRNA level despite a significant reduction in Krt1 transcript , a marker for the cell layer where Irf6 is expressed (Fig. 11 y). Consistent ectopic expression of Trp63 , we found a significant increase in Trp63 transcript level. An analogous change in Krt14 expression was not observed . Furthermore, p revious reports have shown that Tfap 2! and Tfap2c 20 play redundant roles in skin, limb and craniofacial development and share cis -acting regulator elements. Therefore, l oss of MCS9.7 expression in Tfap 2! knockout mice may also result from changes in Tfap2c expression that is co -incident with loss of Tfap 2!. However, in Tfap 2! knockout mice, Tfap2c is significantly increased (Fig. 11 y). Taken together, these data suggest that Tfap 2! is necessary for MCS9.7 activation and Irf6 expression. Moreover, as rs642961 abrogates one of four Tfap2 binding sites within MCS9.7, this data suggests that rs642961 is a functional variant. 112!Irf6 regulates Tfap2 ! Seeing that Tfap 2! is a critical trans -acting factor for Irf6 expression , we wanted to assess if Irf6 also regulates Tfap 2!, as previously suggested for Trp63 . At E14.5 we find that Tfap 2! is expressed in both the periderm, as marked by Krt6, and basal cell layer, as marked by Krt1 4 and Lef1 (Fig. 12 a,b) . In Irf6 -/- embryos, we found an expanded epidermis. While Lef1 continues to mark the basal cell layer in Irf6 -/- embryos , we found ectopic expression of Krt14 and Tfap 2! (Fig. 12 c,d) . Similarly, at E17.5, Irf6 restricts Tfap 2! expression to the basal and spinous layer whereas loss of Irf6 leads to expression of Tfap 2! throughout the epidermis, as previously shown for Trp63 (Fig. 12 e,f) . Transcriptionally, we find that loss of Irf6 in the epidermis leads to over -expression of Trp63 , Krt14 , Tfap 2! and Tfap2c (Fig. 12 g). We hypothesized that an expended epidermis with more Trp63 and Tfap 2! positive cells would lead to a quantitative increase in their relative protein content. Consistent with the transcriptional changes and prior reports 16, we find a significant increase of Trp63 in Irf6 knockout embryos (Fig. 12h), Unexpectedly , despite the 4-fold increase of Tfap 2! mRNA, we found a reduction in Tfap 2! protein . Therefore, loss of Ir6 leads to diverging quantitative changes in Tfap 2! mRNA and protein. These results suggest that multiple regulatory mechanism s link Ir6 and Tfap2 ! expression. These mechanisms may have arisen as a result of important roles in pleiotropic developmental disease. Irf6 homeostasis is required for neurulatio n and Tfap2 ! expression in epidermis To determine the gene regulatory network between Irf6 and Tfap 2! in vivo, we modulated Irf6 expression in vivo . To reduce Irf6 expression, we created a hypomorphic 113!allele (Irf6 ey, Supp Fig. 17 ) and examined embryos at E13.5 and E17.5 (N=47, MandelÕs p-value = 0.5; Supp Table 10). To create an allelic series , we combined the Irf6 hypomormophic allele (Irf6 ey) with a genetrap (functionally null) allele ( Irf6 -). At E17.5 , Irf6 ey/+ (n=17) heterozygotes appeared grossly normal (Fig. 13 a). Unexpectedly, compound heterozygous embryos (Irf6 ey/-, n=18) had a completely penetrant curled tail . At E13.5 , tail abnormalities were not grossly visible in either Irf6 ey/+ (n=3) or Irf6 ey/- (n=6) embryos (data not shown) . These results are striking because while Irf6 knockout embryos have a shortened, dysmorphic tail, we presumed that severe skin adhesions led to tail anomalies. To over -express Irf6 , we used the KRT14 promoter to drive ectopic expressio n (TgKRT14 -Irf6 )21. As a basal keratin , we predict over -expression would begin at E9.5 , as previously described 22. The genotypic distribution from this mating did not differ from the predict (N=246; MendelÕs p -value = 0.7 ; Supp Table 11). Surprisingly, while 89% of transgenic embryos appeared grossly normal ( Irf6 tg-wt), we found that 11% (N=14) of transgenic embryos ( Irf6 tg-mut ) showed severe rostral and caudal neural tube defects. Of those, eight embryos (6%, Irf6 tg-ex) presented with excencephaly (Fig. 13 a). The rest, six embryos (5%), presented with anencephaly, thoracoabdominoschisis and a kinked tail, a phenotype highly analogous but not identical to Tfap2 ! knock out embryos (Irf6 tg-an). This data suggests that over -expressing Irf6 could repress Tfap2 !. However, c onsidering that ectodermal loss of Tfap2 ! leads to pathology as early as E9.5, this data also suggests that the KRT14 promoter is driving Irf6 earlier than previously recognized for either gene. 114!Developmentally, palate and neural tube closure occur in highly analogous process . Essentially, a flat epithelium and underlying mesenchyme rapidly expand, pivot toward midline, appose, adhere and fuse . However, n eural tube development is a more complex process, involving multiple independent closure or fusion points. Defects in the rostral closure point s can lead to exencephaly and anencephaly , which are highly analogous in mouse and humans . Defects in the caudal closure point can result in curled or kinke d tail in the mouse and spina bifida in humans. The incidence of both rostral and c audal closure defects is 1/1000, perhaps constituting the most severe and least treatable cohort of developmental anomalies. While more than 200 mouse models of neural tube defects have been reported, few of these have been conclusively tested for a conserved role in human disease and fewer still have shown an association. We wanted to understand how altering endogenous Irf6 expression can lead to five grossly different developmental phenotypes . We hypothesized that differing phenotypes resulted from diffe rent levels of Irf6 expression. Transcriptionally in skin, we found five different leve ls of Irf6 expression across five different phenotypes tested (Fig. 13 b). Remarka bly, variable transgene expression accounts for two of these levels. Transcriptionally, we found a dose -dependent negative correlation bet ween Irf6 and Tfap 2! across five different murine models . Strikingly, the most severely affect ed transgenic embryo had the greatest amount of Irf6 and the least amount of Tfap2 !. Furthermore, while Irf6 expression is necessary to restrict Trp63 mRNA level, over -expression of Irf6 was not sufficient to further reduce Trp63 transcript. A similar rela tionship is also observed for Krt14 and Tfap2c . Paradoxically, while Irf6 expression 115!was not necessary to maintain Tgm1 and Krt1 levels, over -expression was sufficient for repression (Supp Fig 18). Having observed that more severely affect ed transgenic e mbryos expressed more Irf6 , we asked if both the penetrance and variable expressivity of the transgene could be modified in vivo . To test this hypothesis, we intercrossed Irf6 tg-wt mice. We found that 36% of transgene positive embryos are affect ed ( Irf6 tg-mut , N=33; MandelÕs p -value = 0.35; S upp Table 12 ). While 10% of transgenic embryos had exencephaly, 26% of the embryos had anencephaly. Therefore, by intercrossing Irf6 tg-wt mice, we increased the penetrance of murine neural tube defects (11% vs. 36% affect ed) and the severity (5% vs. 26% anencephaly). These findings suggest that murine development is highly sensitive to Irf6 dose. Our data suggests that Irf6 transcriptionally represses Tfap2 ! . However, the protein level, loss of Irf6 lead to a reduct ion of Tfap 2!. Therefore, we hypothesized that both over - and under -expressing Irf6 woul d reduce Tfap2 !. Consistent with transcriptional changes , we found that over -expressing Irf6 lead to a reduction in Tfap 2! (Fig. 13 c). However, consistent with post -translational stabilization , we found that lowering Irf6 expression also lead to reduction in Tfap 2! in a dose -dependent manner despite a relative increase in Tfap 2! mRNA . Thus, both over and under -expressing Irf6 lead s to reductions in Tfap 2! protei n. 116!To further characterize how changes in Irf6 expression affect skin development and Tfap 2! localization , we examined five different murine phenotypes and Irf6 doses histologically and molecularly at E17.5. We found that over -expression of Irf6 lead s to epidermal hyperplasia , with a disorganized basal cell layer and vacuolated super basal cells (Supp Fig 19 ). Loss of Irf6 expression resulted in a hypotrophic epidermis that also included vacuolated cells. Epidermal thickness positively correlated with cha nges in the spinous cell layer, as marked by Krt1 and Krt14 ( Fig. 13 d,e) , suggesting that Irf6 expression is both necessary and sufficient for spinous layer development and epidermal thickness . Reducing Irf6 expression lead to the stratification of Tfap 2! into the granular layer and Trp63 in to the spinous cell layer. Conversely, over -expression of Irf6 leads to a loss of Tfap 2! but not Trp63 (Fig. 13 f,g) . In skin, like the neural tube , both over and under expression of Irf6 lead to pathology, including ectopic expression of Krt6, Krt14, and Ki -67 (Supp Fig 19 ). No appreciable change in Activated Caspase 3 expression was noted despite Irf6 function in tumor suppression . Instead, Irf6 expression correlated with Tgm1 and Loricin immunostaining, suggesting n ecessity. In viewing seria l sections, we found well -circumscribed foci of highly proliferative, Krt14 positive foci in the epidermis resulting from dermal projections . These included hair follicles and basal cells , but were engulfed by more superficial epidermal cells. These epidermal plumes seem to result from binding of non -contiguous adjacent basal cell layers and altered desmosome expression , as seen in Tfap2a knockout embryos ( Fig. 11 w,x) . 117!Endogenous and t ransgenic Irf6 expression regulate Tfap2 ! in n eurulation Seeing cross -regulation of Irf6 and Tfap2 !, we hypothesized that Irf6 and Tfap 2! interact genetically. We tested the genetic interaction in vivo by interbreeding mice that are heterozygous for the Irf6 genetrap ( Irf6 +/-) and Tfap 2! LacZ knockin ( Tfap 2!LacZ/+ ) alleles. The distribution of embryonic genoty pes was not different from the M endelian prediction (N = 300 embryos; MandelÕs p -value = 0.30 ; Supp Table 13 ). We found that 10.3% of embryos are resorbing , a sign ificant number relative to the expected rate of 1 -3%23(p-value = 0.0007, FisherÕs Exact, two -tail, t -test). Whereas Tfap 2!LacZ /LacZ knockin mice have completely penetrant neural tube and abdominal wall defects, we found that 10.3% of Tfap 2!LacZ/+ embryos have excencephaly, frontonasal hypoplasia, low -set unat tached pinna, and disproportionately short upper and lower limbs (Fig. 14 a,b) (N=68, 7 affect ed). W e replicated this finding in embryos heterozygous for the Tfap 2! knockout allele (Tfap2 !+/-) (N=69, 3 affect ed, 7% ; Mande lÕs p -value = 0.02; Supp Table 14 ). Strikingly, and consisten t with our allelic series, a null allele of Irf6 completely rescues haploinsufficiency of Tfap 2! in Irf6 +/-;Tfap 2!LacZ/+ embryos (Fig. 14 c,d) (N=69, 0 affect ed, p -value = 0.0063, FisherÕs Exact Test, two -tail t -test). We conclude that intergenic suppression of Irf6 rescues Tfap 2! haploinsufficiency. These results are consistent with our findings, which suggest that embryonic development is exquisitely sensitive to Irf6 and Tfap 2! dose and that endogenous Irf6 expression regulates Tfap 2!. As noted above, the transgenic phenotype suggested Irf6 and KRT14 expression prior to E9.5. As such, we asked if Irf6 and Krt14 are expressed in at E8.75, prior to neural tube development. Consistent with intergenic suppression, w e detected Irf6 in the neural plate, 118!neural plate border and non -neural ectoderm with two different antibodies (Fig. 14 eÕ,fÕ) . Further, w e found co -localization of Irf6 and Tfap2 ! in the neural folds and non -neural ectoderm. At E9.0, we dual -stained for Irf6 and RhoB, a marker of early migrating neural crest cells. We found Irf6 expression in the neural tube and co -localization with RhoB 24 in delamina ting neural crest cells ( Fig. 14 g). Consistently, dual staining for Irf6 and Krt14 shows co -expression in the neural tube (Fig. 14 h). We did not detect Irf6 expression in migratory neural crest cells ( Fig. 14 k). Molecular staining for Irf6 and RhoB in transgenic embryos shows ectopic Irf6 expression in the neural tube, non -neural ectoderm and cephalic mesenchyme ( Fig. 14 l). Irf6 and Tfap2 ! co -localized in the neural tube, non-neural ectoderm and delaminating neural crest cells (Fig. 14 m). Signif icantly, in embryos over -expressing Irf6 we did not detect Tfap2 ! in the neural tube and delaminating neural crest cells (Fig. 14 n). Consistently, embryos over -expressing Irf6 had neural tube closure defects rostrally and caudally, including defects in the optic cup and disordered cephalic mesenchymal tissue (Fig. 14 i,j ). Given our immuno -staining , which suggested co -localization of Irf6, Krt14 and Tfap 2! in the neural tube and delaminating neural crest cells , we wanted to test the hypothesis that endogenous and transgenic Irf6 expression cooperatively antagonize Tfap 2! expression in vivo . Considering c o-localiza tion and intergenic suppression of Tfap 2! by reducing Irf6 , we reasoned that over -expressing Irf6 via the KRT 14 promoter in Tfap 2! heterozy gous embryos (TgKRT14 -Irf6 ;Tfap 2!LacZ/+ ) would increase the penetrance and the 119!severity of neural tube defects . Importantly, genotypic distribution did not differ from MendelÕs prediction (N=63 , MendelÕs p -value = 0.18; Supp Table 15 ). In total, 16 embryos genotyped as TgKRT14 -Irf6 ;Tfap 2!LacZ/+ . While half of these embryos appeared grossly normal, the other half is more severely affect ed than Tfap 2! knockout embryos (Fig. 14 o). We further hypothesiz ed that if transgene and endogenous Irf6 expression co -antagonize Tfap 2!, than TgKRT14 -Irf6 ;Irf6 +/-;Tfap 2!LacZ/+ embryos would be less severely affect ed than TgKRT14 -Irf6 ;Tfap 2!LacZ/+ littermate s because it would reduce total endogenous Irf6 expression. The Irf6 null allele did not affect the genotypic distribution (N=66 , MendelÕs p -value = 0.07; Supp Table 16 ). Significantly, as predict , reducing endogenous Irf6 expression (TgKRT14 -Irf6 ;Irf6 +/-;Tfap 2!LacZ/+ , n=11) partially rescued all affect ed tissue s relative to TgKRT14 -Irf6 ;Tfap 2!LacZ/+ littermates (Fig. 14 o). Strikingly, TgKRT14 -Irf6 ;Irf6 +/-;Tfap 2!LacZ/+ phenocopied Tfap 2!-/-, further supporting the role of Irf6 in regulating endogenous Tfap 2! levels . Tfap2 ! and Grhl3 interact in caudal neurulation We wanted to determine how lowering Irf6 expression in Irf6 ey/- embryos c ould lead to a curly tail. Significantly , we showed that Irf6 binds to and transcriptionally activates Grhl3 and that irf6 is epistatic to grhl3 in zebrafish 13. In primary human kertainocyt es, IRF6 binds to and positively regulates GRHL3 25. This association is significant because in mouse, a spontaneous mutatio n in Grhl3 leads to a hypomoprhic allele and the curly tail mouse (ct/ct) , one the oldest and best -described murine neural tube defect phenotypes. Furthermore, knocking out Grhl3 leads to exence phaly, a defective epidermis, an open lumbosacral neural tube defect and a curly tail 14. A genetic cross 120!between Irf6 and Grhl3 heterozygous mice suggested epistatis via early embryonic resorptions and post -natal lethality of double heterozygous pups, but gros s phenotypic anomlies were not observed (Peyrard, xx) . Considering epidermal expression data and analogous tail phenotypes, we predict that Irf6 regulates Grhl3 in neural tube development. To test our hypothesis, we compared Grhl3 in grossly normal ( Irf6 ey/+ ) and abnormal (Irf6 ey/-) tail tissue . We predict that redu cing Irf6 would lead to a reduction in Grhl3 . Consistently, reduc ing Irf6 leads to a reduction in Grhl3 at E17.5 (Fig. 15 a). Considering kinked tail in the knockout and interaction with Irf6 , we also examined Tfap 2! levels. Remarkably, and in direct contrast to the skin, reducing Irf6 expression in the tail also lead to a reduction in Tfap 2!. Prior genomic screen also showed tha t Irf6 binds within Tfap2c 25 and that tissue -specific deletion of Tfap2c leads to a curled tail 26. As predict , reducing Irf6 expression lead to a reduction in Tfap2c at E17.5 (Supp Fig. 20) . Consistent with onset of tail dysmorphology, molecular changes in Irf6 , Tfap2 ! , Grhl3 and Tfap2c occurred at E17.5 . Considering changes in MCS9.7 enhancer activity in the tail of Tfap2 ! knockout embryos (Fig. 11 c,d) , this work shows that Irf6 , Tfap2 !, Tfap2c and Grhl3 constitute a gene regulatory network in caudal neurulation. Our model (Fig. 15 c) suggests that Irf6 is an intermediate node between Tfap2! more proximally and Grhl3 more distally. Based on our model, we further predict that Tfap2 ! and Grhl3 interact in caudal neural tube development . Consistent with our prediction, we find that of 15 % Tfap 2!+/-;Grhl3 +/- double heterozygous embryos have a curly tail (Fig. 121!15b) (N=20, 3 affect ed; MandelÕs p -value = 0.11, Supp Table 17). Consistent with the curly tail mouse, we find incomplete penetrance. However, our prior work shows that Grhl3 and Irf6 double heterozygous embryos do not have a curly tail . Absence of a curly tail in Irf6 +/-;Grhl3 +/- double heterozygous embryos suggests that additional targets , including Tfap2c , may be playing a redundant role with Grhl3 downstream of Irf6 in caudal neurulation . Shared IRF6 mutation in Spina Bifida and VWS Considering the role of Irf6 in murine neural tube development, we hypothesized that common and rare variants in IRF6 could contribute risk to neural tube defects in humans. To test our hypothesis in the cauda l neural tube, we sequenced IRF6 protein coding sequence in 96 patients with Spina Bif ida. Consistent with our hypothesis, we found a non-synonymous substitution in exon 9 that alters a highly conserved amino acid residue, D427Y (Fig. 16) . The variant , previously reported in patient with Van der Woude Syndrome 27, is predict to be damaging and probably deleterious by PolyPhen2.0 and SIFT, and is not found in nearly 7000 control samples (EVS and 1000 genomes). Most strikingly, structural analysis reveals that D427Y occurs at the junction of the c -terminal alpha h elix of IRF6. Prior work on IRF6 paralogs has shown that mutations in this region may affect the c-terminal domain and prev ent activation and dimerization 28. Considering detection of rare variants, w e further predict that common and rare non -coding variants within IRF6 may be associated with spina b ifida risk. Our analysis suggests no association with three IRF6 variants, rs642961 (MAF 17%), rs17371411 122!(MAF 8%), rs75012801 (MAF 0.8%) (Tables 6 -8). Additional analysis has also sho wn that rs17371411 is not associated with anencephaly in humans. While we did not detect an association, these three variants only account for 28% of the genetic variation at IRF6 . As such, the role of IRF6 in human neurulation remains largely unanswered. DISCUSSION Our data show that Irf6 , Tfap 2! and Grhl3 interact in the development of multiple ectodermal lineages, including skin, craniofa cial and the neural tube . Mechanistically, while Tfap 2! regulates Irf6 enhancer activity and expression in skin and tail, we found that Irf6 regulates Tfap2! through both transcriptional repression and post -translational stabilization. These data may provide a molecular rationale that underpins common co -occurring orofacial clefting, lip pits and skin defects in VWS , PPS and BOF S. More broadly, this pathway provides a functional li nk that may be perturbed in 30% of the worldÕs population who have rs642961, a DNA variant that alters Tfa p2! binding at MCS9.7 and is associated with 12% of all orofacial clefting. Evoluti onarily, rs642961 is perplexing because it is a derived, common variant and yet increases risk for orofacial clefting in multiple, ethnically diverse populations . Considering that increasing Irf6 expression increased risk for rostral neural tube defects, uniformly lethal events in mouse and humans, a variant that dampens Irf6 expression may provide a selective survival advantage during development. In support of this hypothesis , we also show here that a null Irf6 allele rescues perinatal le thal neural tube defects in Tfap 2! haploinsufficient embryos, a seldom -described biological relationship in the 123!mouse 29, 30 . Similar to the heterozygous advantage provided by sickle cell trait in the context of endemic malaria risk, we provide evidence here that dampening Irf6 expression may confer a developmental advantage. In addition, recent evidence suggests that IRF6 and TFAP 2! function is critical in preventing post -embryonic disease. For example, previous work shows that IRF6 functions as a tumor suppressor in breast cancer and more recent work shows that mutations in IRF6 can lead to squamous cell carcinoma 31, 32 . Furthermore, a variant within TFAP 2! modifies BRCA2 breast cancer risk 33. Therefore, if this pathway is conserved in other tissues, additional biological and or novel therapeutic applications may be explored. Interestingly, D427Y was previously identified in a patient with Van der Woude Syndrome. While neural tube defects have not been reported in VWS, these results suggest that common IRF6 function is perturbed in orofacial clefting and neural tube defects. Fu rthermore, considering a common Tfap 2!-Irf6 -Grhl3 pathway in neural tube and orofacial development in mouse and man, we suggest a novel category of genes that are mutually involved in both biological processes. While underpinned by strikingly different cel lular processes, both neural tube and palate formation essentially occurs through analogous tissue processes, including proliferation, growth , pivot toward midline, adhesion and fusion of apposing surfaces. Considering analogous genetic and morphological p athways, we suggest use of Ôorofacial and neural t ube cleftingÕ genes. 124!Strikingly, we found that over -expression of Irf6 lead s to anencephaly and a kinked tail and that reductions in Irf6 expression lead s to a curly tail. While posterior n europore dysfun ction has been blended together, differences in the molecular etiology suggested here support their phenotypic distinction s. Biologically, Irf6 joins PTEN as a tumor suppressor genes that when under -expressed leads to loss -of-function and when over -express ed leads to gain -of-function in vivo 34. Surprisingly, few studies have examined in vivo over -expression of tumor suppressors. However, as we have shown here, over -expressing this important class of genes may provide insight into novel pathways, targets or tissue -specific functions. To date, the only known murine alleles that display both anencephly and a thoracoabdominoschesis are Tfap 2! knockout embryos and those that over -express Irf6. However, Palladin and Grhl2 knockout embryos also display anencephaly and abdominal -schesis but have an intact thoracic wall. Considering the morphological similarity, future work should address if these genes are part of t he same pathway as Irf6 and Tfap 2!, although prior work suggested that Tfap 2! did not interact with Grhl2 35. Similarly, knocking out Ikka 36, 14-3-3 sigm a9, Ripk 437 and Kdf1 38 produced murine embryos that phenocopy the Irf6 knockout. Considering the sh ortened abnormal tail in Irf6 knockout embryos was consistent with a role in neural tube development, future work may uncover analogous roles for Ikka , 14-3-3sigma , Ripk 4 and Kdf1 . Consistent with this hypothesis, homozygous recessive mutations in IKKa lead to Cacoon Syndrome, which includes an abdomino -schesis and neural tube defects, as evidenced by brain anomlies and exencephaly in both reported featuses. Taken together, neural tube 125!biology may gain a new pathway that originated in skin and palate de velopment and human genetics may consider a new cohort of candidate genes. Like the complex genetics of human neural tube defects, many inbred murine neural tube defects models dis play v ariable penetrance and expressivity. To that end, we discovered that variable expressivity and penetrance in our mouse models stems from exquisite sensitivity to Irf6 dose. As such, genes upstream and downstream of Irf6 in neural tube development may provide novel candidate genes. Epidemiological data has sug gested a common pathway for oro facial and neural tube development as evidenced by reduced risk for either defect with folate supplementation 39. One line of reasoning suggests that folate supplementation affect s methlyation of target genes and as such buffers gene transcription to fine -tune d evelop ment (methylation hypothesis). Consistently, p revious work has shown that the IRF6 promoter is methylated to suppress transcription 25. If folate supplementation affect s this promoter, IRF6, and this core -network, could provide a molecular rationale that integrates folate epidemiology but not the folate pathway specifically for both oral and neural tube development . Considering multiple mechanisms that dynamically dampen Irf6 expression, including a negative feedback loop reported here in the rostral neural tube , IRF6 promoter methy lation as an end target of folate supplementation provides an attractive target for future study. 126!SUPPLEMENTAL DATA DESCRIPTION Supplemental data include four figures and eleven tables. AUTHOR CONTRIBUTIONS YAK and BCS conceiv ed of the work. YAK designed in vitro, cell culture and murine crosses , performed all murine crosses and experiments , completed statistical analyses, analyzed all data , prepared the figures and tables and wrote the manuscript. BCS and YAK designed human sequenc ing with RHF and HZ. HZ and TDB performed human sequencing and statistical analysis. YAK performed comparative sequence analysis. BCS designed and AK made the Irf6 ey hypomorphic allele. WDF performed and analyzed in vitro experiment s and RRR performed mouse genotyping and statistical analyse s. TJW and YC contributed critical reagents. RHF, GMS, AGB and JCM carried out patient recruitment. YAK, BAA, JCM, AGB and RHF supervised exp eriments. BCS established and maintained key collaborations and supervised the project. All co -authors reviewed the manuscript and edits were reviewed/incorporated by YAK and BCS. ACKNOWLEDGEMENTS Financial support to YAK (#F31DE022696 -01) came from the NIH National Institute of Dental and Craniofacial Research. WEB RESOURCES Online Mendelian Inheritance in Man: www.omim.org PolyPhen -2: http://genetics.bwh.harvard.edu/pph2/ 127!Primer3: biotools.umassmed.edu/bioapps/primer3_www.cgi NHLBI/ESP database: evs.gs .washington.edu/EVS/ UCSC genome browser: genome -euro.ucsc.edu/cgi -bin/hgGateway MATERIALS AND METHODS Murine crosses All animal use protocols and procedures are approved by the Institutional Animal Care and Use Committees, AUF Number 05/12 -093-00, at Michigan State University. In all matings, presence of a copulation plug is denoted as E0.5. Tfap2 !-/-;TgMCS9.7 -LacZ : We used a recently characterized TgMCS9.7 -LacZ allele, a transgene with the Irf6 enhancer MCS9.7 fused to the LacZ gene, to assess the necessity of Tfap2 ! in the function of this Irf6 enhancer. We crossed TgMCS9.7 -LacZ with Tfap2 !+/- mice, a knockout allele via homologous recombination, to produce Tfap2 !+/-; Tg MCS9.7 -LacZ mice. We than crossed Tfap2 !+/-; Tg MCS9.7 -LacZ males with Tfap2 !+/- female littermates to obtain Tfap2 !-/-; TgMCS9.7 -LacZ embryos at E13.5 and E17.5. Irf6 +/-;Tfap2 !LacZ/+ : To test the genetic interaction between Irf6 and Tfap2 !, we intercrossed Irf6 +/- mice, heterozygous for a gentrap allele knocked into the first intron of Irf6 , with Tfap2 !LacZ/+ mice, LacZ knocked into exon 7 of Tfap2 ! . We generated single heterozygous mice for each genotype by intermating Irf6 +/- and Tfap2 !LacZ/+ and the cross is don e with both males and females carrying each of the respective alleles to generate Irf6 +/-;Tfap2 !LacZ/+ embryos at timepoints spanning E13.5 -E18.5. TgKRT14 -Irf6 : We characterized tgKRT14 -Irf6 allele by first intercrossing transgene positive mice with litter mates that are transgene negative. As with the Irf6 +/- and Tfap2 !LacZ/+ cross, we controlled for differential parent -of-origin 128!transmission by conducting the crosses with either males or females carrying the transgene. To test the affect of Irf6 dose, we intercrossed TgKRT14 -Irf6 mice. To test the relevance of Irf6 expression from the transgene relative to endagenous Irf6 expression, we first generated TgKRT14 -Irf6 ;Irf6 +/- mice by intercrossing transgene positive embryos with those heterozygous for the Irf 6 genetrap allele. In the second generation, we intercossed TgKRT14 -Irf6 ;Irf6 +/- with Tfap2 !LacZ/+ mice. Irf6 hypomorphic allele We generated the Irf6 hypomorphic allele by creating a targeting vector from a mouse BAC clone (RPCI22 -516G1) digested with restriction enzymes. The targeting vectors spanned Irf6 genomic sequence from Intron 2 through Intron 6. We placed a Pgk -Neo cassette within a BamHI site within Intron 4. Specifically, a 1.8 kb of KpnI/Bam HI fragment for 5Õ -arm and 3.9 kb of Bam HI/ Hind III fragment for 3Õ -arm were cloned into pBluescript II SK( -) (Agilent Technologies). 3 kb of Bam HI fragment containing exons 3 and 4, coding DNA binding (IR F) domain, was cloned into ploxP3 -Neo-pA vector (kind gift from Professor Takeshi Yagi, Osaka University). 5.8 kb of Xho I fragment which contains floxed exons and Pgk -Neo cassette was subcloned into Bam HI site between 5Õ - and 3Õ - arms. NotI digested target ing construct was electroporated into mouse R1 ES cells. After G418 selection, ES cells were screened by PCR. Primer set of 5Õ - GAGAAATAGGGCCTTCACGGTG -3Õ (sense) and 5Õ - TGTGCCCTCTGATGCTGGAACAG -3Õ (antisense) for 5Õ -side, 5Õ - TCGCCTTCTTGACGAGTTCTTCTG -3Õ (sense, in Pgk -Neo cassette) and 5Õ - GCTCAACTCCCTTTGTGACTGTCC -3Õ (antisense) for 3Õ side were used. 129!Positive ES clones were used for establishment of Irf6 hypomprphic mouse ( neo ). Murine genotyping Both embryos and pups were euthanized before tissue collection. We extracted DNA from each individual embryos/yolk sac or pup. We obtained genomic DNA by first incubating the tissue in lysis buffer with proteinase K (03115887001, Roche) and than isolated t he DNA using ethanol precipitation. We performed PCR to identify genotype. PCR protocols for the Irf6 genetrap allele, Tfap2 ! neomycin and LacZ knockin alleles and MCS9.7 were completed as described previously. Genotyping for the transgenic TgKRT14 -Irf6 was done using newly designed KRT14 forward primer 5ÕTTACAAAACCCTTTCACATACATTGTCGCATTGG3Õ and KRT14 reverse primer 5ÕTTGGGGTGGGAACCACGATACACCT3Õ to yield an expected amplicon of 328 bp. Genotyping for the Irf6 hypomorphic allele was complete with primers that flank the LoxP recombined sites with forward primer 5Õ-GCAGAGTGGAGCACACTTCA -3Õ and reverse primer 5Õ-AAGCATGTCTATTTGGGGGTTA -3Õ. Expected amplicon sizes were 283 bp for the wildtype allele and 592 bp for the hypomorphic allele. PCR protocols included a 4 min denaturation step at 94 ¡C. Following denaturation , 30-35 amplification cycles (94¡C for 30 sec, 60 ¡C for 30 sec, 72 ¡C for 40 sec) were completed before a final 5 min extension step. Morphological, histological and molecular analyses of murine tissue Gross morphological assessment of all embryos was complete upon dissection. Embryos are placed in freshly -prepared 4% paraformaldehyde, paraffin embedded and sectioned at 130!7 µm intervals. Hematoxylin (GHS332, Sigma) and Eosin (E511 -25, Fisher Chemica l) staining was performed using a series of hydration and dehydration steps. Immunostaining is performed with an initial set of hydration steps and followed by antigen retrieval in sodium citrate (pH6.0) on a hot plate and than Triton X -100 (VWR) permeabil ization for 30 minutes. To reduce background, we undertake two blocking steps, each lasting 1 hour. The first is in BSA (A7906, Sigma) diluted in PBS at 10mg/ml and the second is in a Goat anti -mouse Fab fragment (Jackson ImmunoResearch Laboratories, 115 -007-003) also diluted in PBS as a function of the embryonic timepoint tested (20ug/ml for E13.5, 30ug/ml for E15.5, and 40ug/ml for E17.5). Primary antibody is incubated for 16 to 19 hours at 4¡C. Primary Antibodies tested are as follows: Irf 6 (Sigma -Aldric h, SAB2102995), Tfa p2! (3B5, sc -12726, Santa Cruz), T Trp63 (Santa Cruz, 4A4, sc -8431), Keratin 6 (Covance, PRB -169P), Keratin 14 (Novocastra, NCL -L-LL002), Keratin 1 (NCL -CK1, Novocastra), Loricrin ( PRB -145P, Covance), RhoB (56.4H7, University of Iowa Hybridoma Facility )24, Activated Caspase 3 (Abcam, Ab13847), Ki-67 (ab15580, Abcam), BrdU (Abcam, Ab6326). Secondary antibodies are incubated for 1.5 hr at room temperature. Secondary Antibodies tested are as follows : goat anti -mouse (Molecular Probes, A11029), goat anti -rat (Molecular Probes, A11006), goat anti -rabbit (Molecular Probes, A21429 and O -6381). We labeled nuclei with DAPI (Invitrogen, D3571). Finally, to prevent loss of signal over time and protect the se ctions, we mount slides in ProLong Gold Antifade Reagent (Invitrogen, P36930). 131!Skeletal p rep Embryos were fixed in formalin at 4¡C until time of processing. Once genotyped, embryos were placed in 70% Ethanol for 24 hours at 4 ¡C, partially dissected to remove skin and subcutaneous fat, and then transferred into 95% Ethanol for at least 48 hours also at 4 ¡C, at which point dehydration was grossly apparent. We than removed the 95% Ethanol and placed in the embryos in 2% KOH for approximately 72 hours at ro om temperature, at which point skeletal structures become clearly visible. We replaced the 2% KOH with Alcian blue solution (Sigma, A5268 -10G), which stains cartilage. The embryos began to once again dehydrate and in the process absorbed the stain. This process took about 72 hours and was followed by a de -staining step in 95% Ethanol for 6 Ð 24 hours at room temperature. At this point, we removed the Ethanol and added Alizarin Red Solution (Sigma -Aldrich, A5533 -25G) for 24 -36 hours at room temperature. To p erform a final clearing step, we placed the skeleton in a 1%KOH/20% Glycerol solution until it become totally clear. Final specimens were kept in a 1:1 glycerol:95% Ethanol solution until imaging. All incubations performed at room temperature were done on an elliptical rocker. Bioimagin g upright/f luorescent microscope Imaging was completed with a Nikon Eclipse 90i upright microscope. The two objectives used for this work are the Plan APO 10x/0.45 DIC and Plan APO 40x/0.95 DIC M/N2. For light microscope i maging of Hematoxylin and Eosin sections we used the Nikon DS -Fi1-U2 5 mega -pixel color digital camera. Fluorescent imaging is complete using a HQ2 photometric CoolSnap camera and an X -Cite 120Q illuminator. We used NIS Elements 132!Advanced Research v3.10 ima ging software to capture the images in RAW format. Image processing is limited to sharpening and deconvolution with Gauss -Laplace Sharpening on the NIS -Elements AR software. Adobe Photoshop Elements v9.0 is used for plate configuration and enhancement of c olor and contrast. Stereomicroscope . We used a Nikon SMZ800 with a motorized stage, a Nikon DS -Fi2 high -definition camera and a Nikon Digital Sight DS -U3 control unit for high -magnification imaging of whole -mount embryos. For whole -mount, 1x imaging of embryos we used a SMZ1000 Nikon microscope. A Fiber Optic Light Ring is used for both microscopes. We use NIS -Elements Basic Research software version 4.11 to process the embryonic images and Adobe Photoshop Elements v9.0 for plate configuration Transcrip tional profilin g using quantitative -PCR Skin is collected from embryos directly after harvesting at the timepoint specified. Skin sample is snap frozen in liquid nitrogen and stored in -80¡C until time of RNA extraction. TRIzol RNA extraction (15596 -026, Ambion) is used with minor modifications to the manufacturers protocol. Briefly, we homoginze the tissue manually, add 200 ul of chloroform and shake vigorously for 15 seconds. After a brief incubation at room temperature, we centrifuge for 15 minutes at 4 ¡C and transfer the aqueous phase to a new vial. To prevent DNA contamination we add 30 units of RNase -Free DNase (79254, Qiagen) for 45 minutes at 37 ¡C and we inactivate the enzyme at 65 ¡C for 5 minutes. To extract the RNA, we than add 50 µl NaAcetate (3M, pH4.0), 500 µl acidic phenol and 200µl chloroform and finally centrifuge for 15 minutes at 12,000 RPM at 4 ¡C. The aquous phase is transferred to a new vial and 20 µl NaAcetate, 500 µl of propanol and 133!yeast -tRNA (15401 -029, Invitrogen) are added for RNA preci pitation. To pellet the RNA, we centrifuge the sample at 14,000 RPM at 4 ¡C for 20 minutes. The supernatant is removed and the RNA pellet is washed with 75% ethanol, centrifuged and resuspended in RNase -free H 2O and incubated at 55¡C for 10 minutes. RNA is quantified and stored at -70¡C until preparation of cDNA. cDNA. We use a total of 440ng of RNA in addition to Oligo dT primers (18418 -012, Invitrogen) and dNTP mix (18427 -013, Invitrogen) plus water for a total volume of 13 µl for an initial heat mixture step at 65 ¡C for 5 minutes. Furthermore, we added SuperScriptIII Reverse Transcriptase (18080 -093, Invitrogen) and reaction components as per the manufacturers protocol. Including Recombinant RNasin Ribonuclease Inhibitor (N2511, Prom ega) and RNA -Free H 2O, total cDNA reaction volume was 20 µl. We preformed quantification using SYBR Green (4309155, Applied Biosystems) as recommended in a total reaction volume of 10 µl, including 5.5 ng of cDNA and 0.31 picomoles for each of the primer pai r. A complete list of primers is included in Supp table xx. We preformed all reactions with three technical replicates per biological replicate. Negative controls for the cDNA did not include Reverse transcriptase and RNaseOut. qPCr data was collected duri ng the extension step of each cycle and melting curves were generated for each reaction. We used the cycle threshold (Ct), set within the linear range of amplification, to analyze the data. We obtained the delta -Ct relative to Beta -Actin and the delta -delt a Ct and the fold change, relative to wildtype embryo levels for the gene of interest. A complete list of qPCR primer sequences is in Supp Table 18 . 134!Western b lotting We performed protein extraction on murine skin at E17.5. First, we used a mortar and pestle to ground the tissue in liquid nitrogen. The samples were than placed in RIPA buffer with a cocktail of protease inhibitors (11836153011, Roche Diagnostics). After a brief incubation, samples were sonicated on ice to prevent overheating. Finally, we centrifuged the samples for 15 min using 14,000 RPM at 4 ¡C , collected the supernatant and determined protein concentration. We loaded 50 ug of protein per sample. Membranes were blocked in 5% milk and washed in TBST. Irf6 (SPEA, Schutte Lab) and Tfap2 ! (3B5, sc -12726, Santa Cruz) antibodies were diluted in 5% milk at a concentration of 1:750. Human sequencing and g enotyping A complete list of qPCR prim er sequences is in Supp Table 19 . Statistic al analysis We used GraphPad Prism software, version 5, for data analysis. A StudentÕs t -test is used to determine significance based on the variance between the two sample populations with an f -test. For this analysis, a p -value of 0.05 is considered to be significan t. 135! APPENDIX 136!APPENDIX Figure 11: Tfap2 ! is necessary for MCS9.7 activity and Irf6 expression. 137!Figure 11 . (contÕd) 138!Figure 11 . (contÕd) 139! Figure 11 . (contÕd) 140!Figure 11 . (contÕd) (a-d) Loss of Tfap2 ! alters MCS9.7 enhancer activity during two different developmental time points and in different tissue. (a,b) Representative images of transgene positive ( TgMCS9.7 -LacZ ) Tfap2 !+/- (n=20) (a) and Tfap2 !-/- (n=4) embryos at E17.5 (b). (c,d) Representative images of Tfap2 !+/-;TgMCS9.7 -LacZ (n=4) and Tfap2 !-/-;TgMCS9.7 -LacZ (n=7) embryos at E14.0. We did not detected differences in MCS9.7 activity between Tfap2 !+/- (n=16) and Tfap2 !+/+ (n=3) at either time poi nt (data not shown). (e -x) Loss of Tfap2 ! leads to loss of Irf6 expression in spinous cells of the epidermis. (e -g) Hematoxylin and Eosin staining of Tfap2 !+/+ and Tfap2 !-/- skin at E17.5. Scale bars - e,f: 200 µm; g -h: 20 µm; (i -x) Immunofluorescence compa ring Tfap2 !+/+ and Tfap2 !-/- skin at E17.5, with molecular markers as indicated. In all sections, DAPI staining of nuclei (blue). Tfap2 ! (red) and Trp63 (green) (i -j); Irf6 (red) (f -g); Krt1 (red) (m -n); Loricrin (red) (o -p); Ki -67 (red) and Trp63 (green) (h-i); Krt14 (red) and Trp63 (s-t); Krt6 (red) (u -v); Desmocollin (red) (w -x). Scale bars - i-x: 20 µm. (y) Quantitative PCR comparing skin transcriptional profiles in Tfap2 !+/+ (n=7, white), Tfap2 !+/- (n=6, checker) and Tfap2 !-/- (n=3, black). Bar graphs presented as a mean ± SEM. A StudentÕs t -test is used to analyze significance and a p -value is shown above the groups being compared. 141! Figure 12: Irf6 regulates Tfap2 ! . 142!Figure 1 2. (contÕd) 143!Figure 1 2. (contÕd) 144! Figure 1 2. (contÕd) (a-f) Loss of Irf6 leads to ectopic Tfap2 ! expression at two different developmental time points. (a -d) Representative images of wildtype (a,c) ( Irf6 +/+ , n=4, white ) and Irf6 -/- (b,d) embryos (n=4, black) at E14.5, immunostained for (a -b) Krt14 (red) and Lef1 (green) and (c -d) Tfap2 ! (red). In all sections, DAPI marks the nuclei (blue). In Irf6 -/- (b, d), while Lef1 expression remains restricted to the basal cell layer, Tfap2 ! is ectopically expressed. (e -f) Immunostaining of Tfap2 ! (green) and Irf6 (red) in wildtype (n=3) (e) and Irf6 -/- (b,d) embryos (n=5) at E17.5, when murine skin has embryonic reached maturation. In wildtype embryos Tfap2 ! is restricted to basal and spinous c ells (e) while Irf6 expression is more robustly seen superiorly. In wildtype embryos Tfap2 ! is restricted to basal and spinous cells (e) while Irf6 expression is more robustly seen superiorly. In contrast, loss of Irf6 leads to ectopic Tfap2 ! expression (b ). Scale bars a -f: 20 µm. (g) Loss of Irf6 in skin leads to ectopic Tfap2 ! expression. Quantitative PCR comparing skin transcriptional profiles in Irf6 +/+ (n=3) and Irf6 -/- embryos (n=3) at E18.5. Consistent with previous microarray data at E17.5, loss of Irf6 leads to transcriptional increase in TrTrp63 , Krt14 , Tfap2 ! and Tfap2c . (h) Loss of Irf6 in skin leads to reduction in Tfap2 ! expression, in contrast to Trp63 at E18.5. Quantitative analysis of western blot data from Irf6 +/+ (n=4) and Irf6 -/- (n=5) emb ryonically mature skin given as a ratio relative to Gapdh loading control. Consistent with an increase in Trp63 mRNA level, Trp63 protein levels increases in Irf6 -/- skin. In contrast to Tfap2 ! mRNA level, Tfap2 ! protein level is reduced. Bar graphs presen ted as a mean ± SEM. A StudentÕs t -test is used to analyze significance and a p -value is shown above the groups being compared. 145! Figure 13: Irf6 homeostasis is required for neurulation and Tfap2 ! expression in epidermis. 146!Figure 1 3. (contÕd) 147!Figure 1 3. (contÕd) 148!Figure 1 3. (contÕd) 149!Figure 1 3. (contÕd) (a) Neural tube development is exquisitely sensitive to Irf6 dose. Representative whole -mount of an Irf6 allelic series at E17.5. (b,c) qPCR and Western blot data shows direct correlation between Irf6 expression in skin and severity of neural tube defect a t E17.5. Transcriptional profiling (b) for Irf6 and Tfap2 ! shows a direct negative correlation embryos with 5 grossly different phenotype, including Irf6 ey/- (n=5, blue), Irf6 ey/+ (n=4, orange), Irf6 +/+ (n=3, white), Irf6 tg-wt (n=5, gray), Irf6 tg-an (n=5, black). Quantification of western blots from skin at E17.5 (c) for proteins of interests was calculated relative to an internal Gapdh loading control. Values on the y -axis represent absolute ratio. Pattern of transcriptional changes in Irf6 expression hig hly correlate with Irf6 protein quantity. However, both over -expression and under -expression of Irf6 leads to a reduction in Tfap2 ! among 5 grossly different phenotypes, including Irf6 ey/- (n=6), Irf6 ey/+ (n=3), Irf6 +/+ (n=4), Irf6 tg-wt (n=5), Irf6 tg-ex (n=3). (d-g) Skin immunofluorescence at E17.5. In all sections, DAPI marks the nuclei (blue). In a manner analogous to neural tube development, both over and under -expression of Irf6 leads to skin pathology in Irf6 ey/-, Irf6 ey/+ , Irf6 +/+ , Irf6 tg-wt and Irf6 tg-an. (d) Krt1 (red); (e) Krt14 (red); (f) Irf6 (red) and Tfap2 ! (green); (g) Ki -67 (red) and Trp63 (green). 150! Figure 14 : Endogenous and Transgenic Irf6 expression regulate s Tfap2 ! in neurulation. 151!Figure 1 4. (contÕd) 152!Figure 1 4. (contÕd) 153!Figure 1 4. (contÕd) Intergenic suppression of Irf6 rescues Tfap2 ! haploinsufficiency. (a -b) Representative whole -mount images Tfap2 !LacZ/+ haploinsufficient embryos E17.5 (n=7 affected from 68 Tfap2 !LacZ/+ heterozygous embryos examined). These results were completely reproducible with the Tfap2 ! knockout allele. (a) 10% of Tfap2 !+/- embryos have exencephaly, limb defects, frontonasal hypoloplasia with a protruding tongue, abnormally folded pinna and low -set ears and abnormal limbs and digits. (b) Skeletal prep of Tfap2 !LacZ/+ haploinsufficient embryos show absence of the skull bones, including the frontal, parietal and intra Ðparietal bones. (c -d) Representative whole -mount images Tfap2 !LacZ/+ ;Irf6 +/- embryos E17.5 (n=69, 0 affected, p=0.0063). (c) 100% of Tfap2 !LacZ/+ ;Irf6 +/- embryos have grossly normal craniofacial and limb. (d) Skeletal prep of Tfap2 !LacZ/+ ;Irf6 +/- embryos shows complete rescue of neural tube development, including craniofacial and limb str uctures. (e -n) Representative images are shown for all histological and immunostained data. (e -f) Immunostaining at E8.5 shows Irf6 and Tfap2 ! co -localization in neural tube, superficial ectoderm and neural crest cells. (eÕ) Plane of section at E8.5, for i mages e -f. In all sections, DAPI marks the nuclei (blue). (e) Irf6 (red) and Tfap2 ! (green) are co -expressed in the neural plates, non -neural superficial ectoderm, and neural folds. (fÕ) magnified view of white box (f) in (e) showing Irf6 (red) and Tfap2 ! (green) co -localization. (g -n) Histological analysis and immuno -staining reveals neural tube defects and loss of Tfap2 ! cells in embryos over -expressing Irf6 as early as E9.5. (gÕ) marks the plane of the section for (g -n). (g-h) histological analysis of w ildtype (g) and mutant embryos over -expressing Irf6 (h). Wild type embryos have both rostral and caudal neural tube closure, with intact facial mesenchyme (g). 154!Figure 1 4. (contÕd) In contrast, littermate mutant embryos show severe neural tube closure defects rostrally and closure delay caudally, abnormal optic vesicle and disorganized facial mesenchyme (h). (i -n) Immunostaining of wildtype (i -j,k,m) and mutant littermates (l,n). (i -n), Irf6 (red) and DAPI (blue). (i) RhoB (green), marks neural crest cells, co -localizes with Irf6. (j) Krt14 (green), co -expressed with Irf6, stains both the neural tube and neural crest cells. (k -l) RhoB (green). Compared to wildtype embryos (k), mutant littermates have ectopic Irf6 expression and abnormal rostral neural tube closure and neural crest cell migration (l). (m -n) Tfap2 ! (green). Wildtype embryos (m) have Irf6 (red) and Tfap2 ! co -localization in neural tube and pre -migratory neural crest cells whereas mutant littermates have ectopic Irf6 expression and loss of Tfap2 ! staining in the neural tube and neural crest but not non -neural epithelium (n). (o) Endogenous and transgenic Irf6 expression cooperate to functionally antagonize Tfap2 ! in skin, l imb craniofacial and body -wall development. While Irf6 +/-;Tfap2 !+/+ ;Tg KRT14 -Irf6 (n=12) do not develop neural tube defects, 10% of Irf6 +/-;Tfap2 !+/-;Tg KRT14 -Irf6 (n=10, 1 affected) develop anencephaly and an abdominal wall defect. In contrast, 55% of Irf6 +/+ ;Tfap2 !+/-;Tg KRT14 -Irf6 (n=18, 10 affected, 8 grossly normal) developed more severe neural tube, limb and abdominal wall defects (p -value = 0.02). 155!a Figure 1 5: Tfap2 ! and Grhl3 interact in caudal neurulation. 156!Figure 1 5. (contÕd) 157! Figure 15. (contÕd) (a) In contrast to skin, reducing Irf6 expression in tail leads to reduction in Tfap2 ! and Grhl3 . (a) qPCR data at E13.5 and E17.5 from whole -tail RNA extraction. Transcriptional profiling shows that reducing Irf6 in Irf6 ey/ - (n=5, blue) as compared to littermates with the wildteype allele, Irf6 ey/+ (n=3, orange), leads to a reduction of Irf6 and a corresponding reduction in Tfap2 ! and Grhl3 at E17.5 but not E13.5, when tail abnormalities are not yet obvious (data not shown). (b) While wildtype, and singly heterozygous pups and embryos for Grhl3 and Tfap2 ! have a grossly normal tail, 15% of double heterozygous, Tfap2 !+/-;Grhl3 +/-, embryos and pups (n=20, 3 affected) have a grossly curled tail, defined as >90 ¡ change in tail angle from base to tip. 158! a Figure 16 : Shared IRF6 mutation in Spina Bifida and VWS. 159!Figure 1 6. (contÕd) Sequencing of 96 patients with Spina Bifida reveals shared mutation in orofacial clefting and neural tube defects. The mutation, D427Y, occurs at a highly conserved amino acid within exon 9. Structural analysis shows that the mutation is found at the junction of a non -ordered liker region and a the c -terminal alpha -helix of IRF. 160! Table 6: Association between IRF6 (rs642961) and risk of NTDs rs642961 Genotype Case Control OR (%95 CI) Adjusted OR* (95% CI) Among All AA 53 42 1.3 (0.8 -2.0) 1.3 (0.8 -2.0) GA 89 113 0.8 (0.6 -1.1) 0.8 (0.6 -1.2) GG 283 292 Reference Among white AA 6 8 1.0 (0.3 -2.9) GA 33 49 0.9 (0.5 -1.4) GG 85 108 Reference Among HISP -NB AA 13 10 1.3 (0.5 -3.3) GA 14 20 0.7 (0.3 -1.6) GG 44 44 Reference Among HISP -FB AA 30 18 1.3 (0.7 -2.6) GA 36 35 0.8 (0.5 -1.4) GG 124 100 Reference Among Black AA 0 0 --- GA 3 4 0.8 (0.2 -4.2) GG 15 16 Reference Among Asian AA 3 6 1.0 (0.2 -5.0) GA 3 5 1.3 (0.3 -6.2) GG 11 23 Reference *Adjusted for race/ethnicity. 161!Table 7: Association between IRF6 (rs75012801) and risk of NTDs rs75012801 Genotype Case Control OR (%95 CI) Adjusted OR* (95% CI) Among ALL TG 5 7 0.7 (0.2 -2.4) 0.7 (0.2 -2.1) TT 420 438 Reference Among White TG 1 2 0.7 (0.1 -7.4) TT 122 161 Reference Among HISP -NB TG 1 1 1.0 (0.1 -16.8) TT 70 72 Reference Among HISP -FB TG 3 4 0.6 (0.1 -2.7) TT 188 149 Reference Among Black TG 0 0 --- TT 18 20 Reference Among Asian TG 0 0 --- TT 17 35 Reference 162! Table 8: Association between IRF6 (rs17317411) and risk of NTDs rs17317411 Genotype Case Control OR (%95 CI) Adjusted OR* (95% CI) Among ALL CC 1 3 0.4 (0.04 -3.5) 0.5 (0.05 -4.6) TC 73 66 1.2 (0.8 -1.7) 1.2 (0.9 -1.8) TT 351 382 Reference Among White CC 1 3 0.5 (0.1 -4.8) TC 34 32 1.6 (0.9 -2.7) TT 89 132 Reference Among HISP -NB CC 0 0 --- TC 6 13 0.4 (0.2 -1.2) TT 64 61 Reference Among HISP -FB CC 0 0 --- TC 27 16 1.4 (0.7 -2.7) TT 164 138 Reference Among Black CC 0 0 --- TC 2 3 0.7 (0.1 -4.8) TT 16 17 Reference Among Asian CC 0 0 --- TC 2 1 4.5 (0.4 -53.9) TT 15 34 Reference *Adjusted for race/ethnicity. Tables 6 Ð 8 show sequencing results from our examination of an association between IRF6 variant s and Spina Bifida. While sequencing is currently on-going, we have examined three variants , including rs642961 (MAF 17%), rs17371411 (MAF 8%), rs75012801 (MAF 0.8%). At this point, these three IRF6 variants do not appear to be associated with spina bifida. 163! Figure 17 : Generation of Irf6 hypomorphic allele . 164!Figure 1 7. (contÕd) (a) We inserted a neomycin cassette, bracketed with LoxP sites (black triangles within intron), into intron 4 of a Irf6 BAC clone . Spanning introns two -six, we inserted a Pgk -Neo cassette into intron 4. (b) 5 and 3 prime PCR was used to confirm wildtype (#29) and recombinant clones (#28). (c) For genotyping, two primer sets were designed. In intron two, one set is designed to bracket a LoxP site in the recombined clone (black arrows). In intron four, an additional primer set brackets the Pgk -Neo cassette (red). Primer combinations and fragment sizes are also shown. (d) PCR confirms genotypes of Irf6 wildtype and hypomorphic ( Irf6 neo or Irf6 ey) embryos. 165! Figure 18 : Irf6 transcripitonally regulates Trp63 , Tgm1 and Krt1 but not Krt14 and Tfap2c in skin . 166!Figure 1 8. (contÕd) Genotypes and biological replites include Irf6 ey/ - (n=5, blue), Irf6 ey/+ (n=4, orange), Irf6 +/+ (n=3, white), Irf6 tg-wt (n=5, gray), Irf6 tg-an (n=5, black). In addition to post -translational degradation, we found that reducing Irf6 lead to an increase in TrTrp63 mRNA. While over -expressing Irf6 was associated with reduction in TrTrp63 mRNA, this was not statistically significant. While loss of Irf6 leads to ectopic Krt14 and Tfap2c , modulation of Irf6 dose did not alter Krt14 or Tfap2c transcript. In contrast, over -expressing Irf6 lead to reduction in Krt1 and Tgm1 , consistent with Irf6 driving differentiation in epidermis. 167! Figure 19 : Irf6 homeostasi s is required for epidermal development. 168!Figure 1 9. (contÕd) 169!Figure 1 9. (contÕd) (a-f) Skin development is exquisitely sensitive to Irf6 dose. (a) Representative skin histology of an Irf6 allelic series at E17.5, stained with Hemtoxylin and Eosin. Irf6 regulates epidermal thickness, with reductions leading to hypotrophic and over -expression leading to hypertrophic skin. (b -f) Skin immunofluorescence of an Irf6 allelic series at E17.5. In manner highly analogous to neural tube development, both over and under -expression of Irf6 leads to skin pathology in Irf6 ey/ -, Irf6 ey/+ , Irf6 +/+ , Irf6 tg-wt and Irf6 tg-an. Counter -staining of nuclei with DAPI (b lue) is seen in all sections. (b) Krt6 (red); (c) Loricrin (red); (d) desmosome; (e ) Activated Caspase 3 (red) ; (f) Tgm1 (red). 170! Figure 20 : Irf6 Transcriptionally regulates Tfap2c but not Krt14 in tail development . Reducing Irf6 in tail tissue leads to an increase of Tfap2c at E13.5 but a decrease at E17.5, in direct contrast to skin tissue . Irf6 does not regulate Krt14 in the tail at either timepoint. 171!Table 9 : Tfap2!+/-;Tg MCS9.7 -LacZ We intercrossed mice hemizygous for the TgMCS9.7 -LacZ transgene with mice heterozygous for the Tfap2 ! knockout allele. To produce the F2 progeny that were Tfap2 !-/-;Tg MCS9.7 -LacZ , we intercrossed Tfap2 !+/-;Tg MCS9.7 -LacZ with Tfap2 !+/- mice. We examined embryos at E13.5 and E17.5 and found no differences from the predict distribution. We did find a significant number of resorptions. We also found that one Tfap2 !+/- had exencephaly. E13.5 E17.5 Total Litters 3 8 11 Tfap2 !+/+ 4 4 8 Tfap2 !+/- 3 13 16 Tfap2 !-/- 3 5 8 Tfap2 !+/+ ;Tg MCS9.7 -LacZ 1 6 7 Tfap2 !+/-;Tg MCS9.7 -LacZ 4 20 24 Tfap2 !-/-;Tg MCS9.7 -LacZ 7 4 11 Total 19 47 66 p-value 0.36 0.29 0.71 Resorbing 3 8 11 (p=0.0086) 172!Table 10 : Irf6 ey/- vs. Irf6 ey/+ We intercrossed mice homozgygous for the Irf6 hypomorphic allele ( Irf6 ey/ey ) with mice heterozygous for the Irf6 genetrap allele ( Irf6 +/-). We examined embryos at E13.5 and E17.5 and found no differences from predict genotype distribution. E13.5 E17.5 Total Litters 1 4 5 Irf6 ey/+ 3 17 20 Irf6 ey/- 6 18 24 Total 9 35 47 p-value 0.32 .87 .51 Resorbing 0 2 2 (p=0.51) 173!Table 11: Irf6 +/+ ; Tg KRT14 -Ir6 We intercrossed mice hemizygous for the TgKRT14::Irf6 transgene with either wild type littermates or TgMCS9.7 -LacZ mice . We examined embryos between E13.5 -E15.5 and E16.5 -E18.5 and the embryonic distribution did differ from the expected ratios with or without the MCS9.7-LacZ transgene. However, we found a significant number of resorbing embryos in either case. E13.5 -E15.5 E16.5 Ð 18.5 Total Litters 4 30 34 Irf6 +/+ 16 104 120 Irf6 +/+ ; Tg KRT14 -Ir6 12 114 126 Total 28 218 246 p-value .44 .49 0.70 Resorbing 4 19 23 (p=0.0086) 174!Table 12: TgKRT14 -Ir6; Tg KRT14 -Ir6 We intercrossed mice hemizygous for the TgKRT14::Irf6 . Genotyping protocol does not differentiate between hemizygous and homozygous embryos for the TgKRT14::Irf6 transgene. We calculated the expected ratio as 3:1, TgKRT14::Irf6 , to wildtype embryos. We examined embryos at E17.5 and not detect differences fro m predict distribution. We did not have a sufficiently large sample size to assay resorptions differences. E17.5 Ð P0 Total Litters 5 5 Irf6 +/+ 8 8 Irf6 +/+ ; Tg KRT14 -Ir6 32 32 Total 40 40 p-value .35 0.35 Resorbing 4 4 (p=0.18) 175! Table 1 3: Tfap2!+/-;Irf6 +/- We intercrossed mice heterozygous for the Irf6 genetrap allele ( Irf6 +/-) and the Tfap2 !+/- to generate Tfap2 !+/-;Irf6 +/- embryos and pups. We examined embryos at all time points and found no significant difference in embryonic distributions. However, we found a significant number of embryonic resorptions that did not affect any one genotype. E13.5 E15.0 E17.5 Ð P0 Total Litters 8 7 23 38 Tfap2 !+/+ ;Irf6 +/+ 18 13 34 65 Tfap2 !+/-;Irf6 +/+ 16 10 41 66 Tfap2 !+/+ ;Irf6 +/- 16 12 38 67 Tfap2 !+/-;Irf6 +/- 17 9 43 69 Total 69 45 157 299 p-value 0.98 .25 .43 0.34 Resorbing 1 14 17 32 (p=0.0006) 176!Table 1 4: Tfap2!-/- We intercrossed mice heterozygous for Tfap2 !+/- allele. In this cross, we found that three Tfap2 !+/- had exencephaly. E17.5 Ð P0 Total Litters 9 9 Tfap2 !+/+ 16 16 Tfap2 !+/- 29 29 Tfap2 !-/- 4 4 Total 49 49 p-value 0.02 0.02 Resorbing 8 8 (p=0.047) 177!Table 15 : Tfap2!+/-;Tg KRT14 -Ir6 We intercrossed mice heterozygous the Tfap2 !+/- with mice hemizygous for the TgKRT14::Irf6 to generate Tfap2 !+/-; Tg KRT14::Irf6 embryoe. We found no significant difference in the distribution of embryonic genotypes but found a significant increase in the number of resorbtions. E16.5 -17.5 Total Litters 8 8 Tfap2 !+/+ 19 19 Tfap2 !+/- 19 19 Tfap2 !+/+ ; Tg KRT14 -Ir6 9 9 Tfap2 !+/-; Tg KRT14 -Ir6 16 16 Total 63 63 p-value 0.24 0.24 Resorbing 9 9 (p=0.03) 178!Table 16 : Tfap2!+/-;Irf6 +/-;Tg KRT14 -Ir6 We intercrossed mice hemizygous for the TgKRT14::Irf6 (over -expressing Irf6 under the control of the KRT14 promoter) with mice heterozygous for the Irf6 genetrap allele (Irf6 +/-) and the Tfap2 !+/- generate Tfap2 !+/-;Irf6 +/-;Tg KRT14 -Irf6 and Tfap2 !+/-;Irf6 +/+ ;Tg KRT14 -Irf6 . We examined embryos at E17.5 and P0, shown are the combined genotypic distribution. While Tfap2 !+/-;Irf6 +/+ ;Tg KRT14 -Irf6 were under -represented, the genotype distribution did not differ from the predict value. E17.5 Ð P0 Litters 9 Tfap2 !+/+ ;Irf6 +/+ 12 Tfap2 !+/-;Irf6 +/+ 5 Tfap2 !+/+ ;Irf6 +/- 7 Tfap2 !+/-;Irf6 +/- 5 Tfap2 !+/+ ;Irf6 +/+ ;Tg KRT14 -Irf6 7 Tfap2 !+/-;Irf6 +/+ ;Tg KRT14 -Irf6 2 Tfap2 !+/+ ;Irf6 +/-;Tg KRT14 -Irf6 12 Tfap2 !+/-;Irf6 +/-;Tg KRT14 -Irf6 11 Total 66 p-value 0.07 Resorbing 5 (p -value =0.44) 179!Table 17: Tfap2!+/-;Grhl3 +/- We intercrossed mice heterozygous the Tfap2 !+/- and Grhl3 +/- alleles to generate Tfap2 !+/-; Grhl3 +/- embryos and pups. Both the number of resorptions and the genotype distribution did not differ from the predict values. E15.5 - P0 Litters 7 Tfap2 !+/+ ;Grhl3 +/+ 8 Tfap2 !+/-;Grhl3 +/+ 14 Tfap2 !+/+ ;Grhl3 +/- 19 Tfap2 !+/-; Grhl3 +/- 20 Total 61 p-value 0.1 1 Resorbing 3 (p=1.0) 180!Table 1 8: Murine qPCR primers sequences Gene Primer name Primer Sequence (5Õ to 3Õ) 4938 mIrf6 F AGTGTGGCCCAAAACAGAAC Irf6 4939 mIrf6 R GGGTTGCTCACCGTCATAGT 4588 Actb F TCTGGCTCCTAGCACCAT Beta Actin 4589 Actb R GGGCCGGACTCATCGTAC 4994 Grhl3 -197F GAACCTCGGAGAAGGAAGAT Grhl3 set 1 4995 Grhl3 -240R TTCTTCAGCAACCGCACAGA 4996 mGrhl3 -1779F TTGACGCGCTCATGTTGAAG Grhl3 set 2 4997 mGrhl3 -1848R AGGCCGTACTTCTCAGAGAT 4781 mTrTrp63 -707F GAAGGCAGAGCGTGCTGGTC Trp63 4782 mTrTrp63 -811R TCATTCCTCCGACGCAGCTG 5121 CK14 F AGCGGCAAGAGTGAGATTTCT Krt14 5122 CK14 R CCTCCAGGTTATTCTCCAGGG 5109 Tgm1F GCGGAGGGCTGTGGAGAAGG Tgm1 5110 Tgm1R GGGTGCGCAAACGGAAGGTG 5103 Krt1 F GACACCACAACCCGGACCCAAAACTTAGAC Krt1 5104 Krt1 R ATACTGGGCCTTGACTTCCGAGATGATG 5125 Tfap2c F ATCCCTCACCTCTCCTCTCC Tfap2c 5126 Tfap2c R CCAGATGCGAGTAATGGTCGG 5127 Tfapa F GAAGACTGCGAGGACCGTC Tfap 2! set 1 5128 Tfapa R GAAGTCGGCATTAGGGGTGTG 5191 - Tfap 2! F CGCCCTACCAGCCTATCTAC Tfap 2! set 2 5192 - Tfap 2! R GGGAGTAAGGATCTTGCGAC T Primers for Krt14 , Tfap2c and Tfap 2! set 1 were previously publishe d in Qiao et al, 2012, Cell Research; doi:10.1038/cr.2012.122. Primers for Tgm1 and Krt1 were previously published in Wang et al, 2008, JCB; doi/10.1083/jcb.200804030 181!Table 1 9: IRF6 sequencing primers Forward Primer Forward Primer IRF6_1F ttagaagcggaggagtaggg IRF6_1R accccaaacacacagatgc IRF6_2F caaagcttgtctcatgactgct IRF6_2R gggctttggaagagaaggaa IRF6_3F tggcacagcttattcccata IRF6_3R ttcaaccattgcagacatgc IRF6_4F tgtgtgtttgtgtctatgagaaagg IRF6_4R tcaggctgttttcaagttgactat IRF6_5F ggaggtccttccatgagaga IRF6_5R cagggagttcctcacctctg IRF6_6F caggagcaggggaaccttat IRF6_6R gaggatgcctctgagacagg IRF6_7F tgaatgctggttgaaaggtg IRF6_7R gcaggaaggtgaaagacagg IRF6_8F tgactaatgtgacccaggaact IRF6_8R aagatctccactaaatcaatcacc IRF6_9F gtcttcctcagggcctcttt IRF6_9R AAACTCCCAGGCCAAATCTC IRF6_10F TGGAAAAATCACCCTTCAGA IRF6_10R TCCCTAGGCTTTCTGTGTCAA IRF6_11F GCTGGCTGGTTGCTTAGAA IRF6_11R TGAAAGGGTTAGAGACTCAGCA IRF6_12F GCTGGGCAGTACTCTTCTGG IRF6_12R GTTGGAGATGGCCTGGTTTA IRF6_13F AAGCCCCAGTCCTCTTGAAT IRF6_13R TTGGCACTTTTCCAATACCC IRF6_14F CTCTTGAATCTGGGCCAGTC IRF6_14R TTTTATGGGAAAGGGACCAG IRF6_15F TCACTGTGTACCCCACCAAA IRF6_15R tgggaggaggaccagcttat 182! BIBLIOGRAPHY 183!BIBLIOGRAPHY 1. Taniguchi T, Ogasawara K, Takaoka A, Tanaka N. (2001) IRF family of transcription factors as regulators of hos t defense. Annu Rev Immunol ;19:623 -55. 2. Kondo S, Schutte BC, Richardson RJ, et al. (2002) Mutations in IRF6 cause Van der Woude and popliteal pterygium syn dromes. Nat Genet ;32:285 -9. 3. Rahimov F, Marazita ML, Visel A, et al. (2008) Disruption of an AP -2alpha binding site in an IRF6 enhancer is associate d with cleft lip. Nat Genet ;40:1341 -7. 4. Fakhouri W D, Rhea L, Du T, et al. (2012) MCS9.7 enhancer activity is highly, but not completely, associated with expression of Irf6 and p63. Dev Dyn;241:340 -9. 5. Milunsky JM, Maher TA, Zhao G, et al. (2008) TFAP2A mutations result in branchio -oculo -faci al syndrome . Am J Hum Genet ;82:1171 -7. 6. Stoetzel C, Riehm S, Bennouna Greene V, et al. (2009) Confirmation of TFAP2A gene involvement in branchio -oculo -facial syndrome (BOFS) and report of temporal bone anomalies. Am J Med Genet A ;149A:2141 -6. 7. Li H, Sheridan R , Williams T. (2013) Analysis of TFAP2A mutations in Branchio -Oculo -Facial Syndrome indicates functional complexity within the AP -2alpha DNA-binding domain. Hum Mol Genet;22:3195 -206. 8. McDade SS, Henry AE, Pivato GP, et al. (2012) Genome -wide analysis o f p63 binding sites identifies AP -2 factors as co -regulators of epidermal differentiation. Nucleic Acids Res;40:7190 -206. 9. Richardson RJ, Dixon J, Malhotra S, et al. (2006) Irf6 is a key determinant of the keratinocyte proliferation -differentiation swit ch. N at Genet ;38:1329 -34. 10. Ingraham CR, Kinoshita A, Kondo S, et al. (2006) Abnormal skin, limb and craniofacial morphogenesis in mice deficient for interferon regulatory factor 6 (Irf6). Nat Genet ;38:1335 -40. 11. Schorle H, Meier P, Buchert M, Jaenisch R, Mitchell PJ. (1996) Transcription factor AP -2 essential for cranial closure and crani ofacial development. Nature ;381:235 -8. 12. Zhang J, Hagopian -Donaldson S, Serbedzija G, et al. (1996) Neural tube, skeletal and body wall defects in mice lacking transcription factor AP -2. Nature;381:238 -41. 184!13. de la Garza G, Schleiffarth JR, Dunnwald M, et al. (2013) Interferon regulatory factor 6 promotes differentiation of the periderm by activating expression of grainyhead -like 3. J Invest Dermatol;133:68 -77. 14. Ting SB, Wilanowski T, Auden A, et al. (2003) Inositol - and folate -resistant neural tube defects in mice lacking the epithelial -spec ific factor Grhl -3. Nat Med ;9:1513 -9. 15. Thomason HA, Zhou H, Kouwenhoven EN, et al. (2010) Cooperation between the transcription factors p63 and IRF6 is essential to prevent cleft pa late in mice. J Clin Invest ;120:1561 -9. 16. Moretti F, Marinari B, Lo Iacono N, et al. (2010) A regulatory feedback loop involving p63 and IRF6 links the pathogenesis of 2 genetically different human ectodermal dysplasias. J Clin Invest ;120:1570 -7. 17. Celli J, Duijf P, Hamel BC, et al. (1999) Heterozygous germline mutations in the p53 homolog p63 are the cause of EEC syndrome. Cell 1999;99:143 -53. 18. Le M, Naridze R, Morrison J, et al. (2012) Transforming growth factor Beta 3 is required for excisional wound repair in vivo. PLoS One;7:e48040. 19. Restivo G, Nguyen BC, Dziunycz P, et al. (2011) IRF6 is a mediator of Notch pro-differentiation and tumour suppressive functi on in keratinocytes. EMBO J . 20. Li W, Cornell RA. (2007) Redundant activities of Tfap2a and Tfap2c are required for neural crest induction and development of other non -neural ectoderm derivatives in zebrafish embryos. Dev Biol ;304:338 -54. 21. Iwata J, Suzuki A, Pelikan RC, et al. (2013) Smad4 -Irf6 genetic interaction and TGFbeta -mediated IRF6 signaling cascade are crucial for palatal fusion in mice. Development;140:1220 -30. 22. Byrne C, Tainsky M, Fuchs E. (1994) Programming gene expre ssion in develo ping epidermis. Development ;120:2369 -83. 23. Krishnan L, Guilbert LJ, Wegmann TG, Belosevic M, Mosmann TR. (1996) T helper 1 response against Leishmania major in pregnant C57BL/6 mice increases implantation failure and fetal resorptions. Co rrelation with increased IFN -gamma and TNF and reduced IL -10 production by placental ce lls. J Immunol ;156:653 -62. 24. Liu JP, Jessell TM. (1998) A role for rhoB in the delamination of neural crest cells from the dors al neural tube. Development ;125:5055 -67. 185!25. Botti E, Spallone G, Moretti F, et al. (2011) Developmental factor IRF6 exhibits tumor suppressor activity in squamous cell carcinoma s. Proc Natl Acad Sci U S A;108:13710 -5. 26. Guttormsen J, Koster MI, Stevens JR, Roop DR, Williams T, Winger QA. (2008) Disruption of epidermal specific gene expression and delayed skin development in AP -2 gamma mutant mice. Dev Biol ;317:187 -95. 27. Leslie EJ, Standley J, Compton J, Bale S, Schutte BC, Murray JC. (2013) Comparative analysis of IRF6 varian ts in families with Van der Woude syndrome and popliteal pterygium syndrome using public whole -exome databases. Genet Med;15:338 -44. 28. Chen W, Lam SS, Srinath H, et al. (2008) I nsights into interferon regulatory factor activation from the crystal struct ure of dimeri c IRF5. Nat Struct Mol Biol ;15:1213 -20. 29. Jones NC, Lynn ML, Gaudenz K, et al. (2008) Prevention of the neurocristopathy Treacher Collins syndrome through inhibiti on of p53 function. Nat Med ;14:125 -33. 30. Laue K, Janicke M, Plaster N, Son ntag C, Hammerschmidt M. (2008) Restriction of retinoic acid activity by Cyp26b1 is required for proper timing and patterning of osteogenesis during zebrafi sh development. Development ;135:3775 -87. 31. Bailey CM, Khalkhali -Ellis Z, Kondo S, et al. (2005) Mammary serine protease inhibitor (Maspin) binds directly to interferon regulatory factor 6: identification of a novel serp in partnership. J Biol Chem ;280:34210 -7. 32. Stransky N, Egloff AM, Tward AD, et al. (2011) The mutational landscape of head and neck squam ous cell carcinoma. Science ;333:1157 -60. 33. Couch FJ, Wang X, McGuffog L, et al. (2013) Genome -wide association study in BRCA1 mutation carriers identifies novel loci associated with breast and ovarian cancer risk. PLoS Genet;9:e1003212. 34. Garci a-Cao I, Song MS, Hobbs RM, et al. (2012) Systemic elevation of PTEN induces a tumor -suppressive metabolic state. Cell;149:49 -62. 35. Pyrgaki C, Liu A, Niswander L. (2011) Grainyhead -like 2 regulates neural tube closure and adhesion molecule expression during neural fold fusion. Dev Biol;353:38 -49. 36. Takeda K, Takeuchi O, Tsujimura T, et al. (1999) Limb and skin abnormalities in mice lacking IKKalpha. Science ;284:313 -6. 37. Holland P, Willis C, Kanaly S, et al. (2002) RIP4 is an ankyrin repeat -contai ning kinase essential for keratinocyte differentiation. Curr Biol ;12:1424 -8. 186! 38. Lee S, Kong Y, Weatherbee SD. (2013) Forward genetics identifies Kdf1/1810019J16Rik as an essential regulator of the proliferation -differentiation decision in epidermal progenitor cells. Dev Biol;383:201 -13. 39. Shaw GM, Lammer EJ, Wasserman CR, O'Malley CD, Tolarova MM. (1995) Risks of orofacial clefts in children born to women using multivitamins containing folic acid periconceptionally. Lancet ;346:393 -6. !"#$! Chapter 4 - Epithelial Irf6 rescues lethality but not craniofacial and limb development !"##!Youssef A. Kousa 1, Dina Moussa 2, Brian C. Schutte 2 1Biochemistry and Molecular Biology Department, Michigan State University, 48824 East Lansing, Michigan, USA 2Department of Microbiology and Molecular Genetics, Michigan State University, 48824 East Lansing, Michigan, USA Abstract IRF6 regulates epithelial developmen t and differentiation . Inherited IRF6 alleles cause and contribute risk for orofacial clefting and somatic mutations are associated with skin tumorigenesis . In human and mouse, perturbing IRF6 function also leads to limb, digit and craniofacial defects. However, the critical dose , cell autonomy and tissue -type by which IRF 6 mutations co nfer pleiotropic disease are unknown . To delineate the critical dose at which disease occurs, we use an alleli c series to titrate Irf6 expression in the mouse . We find that compound heterozygous embryos for the Irf6 null and hypomorphic alleles display completely penetrant bilateral oral adhesions and abnormal periderm . To test the role of epithelium in the human -mouse disease spectrum, we drive an Irf6 transgene in the basal epithelium of embryos lacking endogenous expression . Rescue embryos appear to have a normal epidermis and, remarkably, survive parturition and early post -natal development. Despite epidermal rescue, limb clubbing and a curled tail persist . Strikingly , palatal elevation is obstructed by oral adhesions between the palatal shelves and the tongue, leading to a cleft palate in 100% of rescue embryos (N = 23). Despite Irf6 expressio n in the basal cell layer, we find an abnormal periderm at oral adhesion sites. Therefore, while sufficient for a functional epidermal barrier and post -!"#%!natal survival , basal Irf6 expression is not sufficient for palate, limb and tail development. Together, t his work suggests that inherited IRF6 alleles can contribute to pleiotropic disease through dosage sensitivity (heterozygous vs. compound hypomoprh) , involvement in multiple de velopmental pathways ( skin vs. limb clubbing and tail ) and cell-autonomous expression (basal vs. periderm expression in palate) . Introduction Mutations in Interferon regulators factor 6 ( IRF6 ) lead to Van der Woude Syndrome (VWS, # 119300) and Popliteal Pytergium Sydnrome (PPS, # 119500), two dominantly inherited orofacial clefting disorders (1). Moreover, DNA variants within the IRF6 locus also increase risk for isolated or nonsyndromic orofacial clefting (iCLP), accounting for 12% of world -wide risk (2). Loss of IRF6 expression in epithelial tissue has also been linked to skin and head and neck squamous cell carcinoma (3, 4). Furthermore, we recently showed that IRF6 is required for neurulation ( Kousa, C hapter 3). Therefore, understanding the dose, cell -type and cell autonomy by which IRF6 exerts its function may provide important insight into human development and disease. In the mouse, loss of Irf6 leads to skin, limb, craniofacial and neural tube defects (5) (Kousa, C hapter 3 ). While the range of affected tissues in Irf6 knockout embryos has provided numerous insight s, a murine phenotype that mode ls iCLP, VWS and PPS pathophysiology is lacking. For instance, while haplo insufficiency of IRF6 leads to VWS, mice heterozygous for either a null Irf6 allele or a dominant negative Irf6 allele do not develop a cleft palate (5, 6). By contrast, Irf6 knockout embryos have severe, bilateral !"%&!oral adhesions that involve all epithelial surfaces in the oral cavity . While this pathology thought to lead to cleft palate in the mouse, oral epithelial adhesions are not common in patients with VWS. Therefore, a dynamic murine model that illuminates human orofacial clefting has not been developed . In the oral cavity, Irf6 function appears to be essential during early palatogenesis . In Irf6 knockout embryos, pervasive oral adhesions result in vertically oriented palatal shelves from E13.5 Ð P0 (5, 6). Oral adhesions are thought to result from loss or dysfunction of the periderm (7), a squam ous epithelium that covers the oral cavity. More broadly, epithelial adhesions in Irf6 knockout embryos obliterate the esophageal lumen and fasten appendages to the body wall (5). However, the periderm remains an eni gmatic cell type with a poorly understood role in palatal development. In addition, it is not known what role, if any, IRF6 plays in the development of non -epithelial tissues . As such, the pathophysiological mechanisms leading to limb, digit and craniofaci al defects in VWS and PPS remain opaque. In the mouse, loss of Irf6 is highly similar to loss of Ikka (8, 9), 14-3-3 sigma (Stratifin) (6), Kdf1 (10) and, to a lesser extent, Ripk4 (11). Invariably, a prominent phenotypic featur e in these murine models is a taut, shiny epidermis that encases the organism . As a result, prior work has sought to rescue the knockout embryo by rescuing the epidermis . For example, driving Ikka with an epithelial specific promoter (KRT14 ) leads to complete rescue of the skin, limb and digits , suggesting both cell -autonomous and non-cell autonomous function (12). An analogous experiment using the KRT5 promoter, co -!"%"!expressed with KRT14 in the basal cell layer , leads to complete rescue with over -express ion of Ikka (13). While Ripk4 knockout embryos are less severely affected, Ripk4 expression under the KRT14 promoter m ore fully rescue s skin and limb defects (14). Interestingly, Ikka and 14-3-3sigma knockout embryos could not be rescued by epithelial Ripk4 expression. In this study, we set out to understand the dose and tissue by which Irf6 exerts in vivo function. We titrate Irf6 dose and find that compo und heterozygotes for the Irf6 null and hypomorphic alleles have completely penetrant, bilateral mandible -maxilla oral adhesions but not clefting . To test how epithelial Irf6 expression contributes to craniofacial, limb and digit anomalies in knockout embryos, we attempt to rescue the epidermis. We f ind Irf6 transgene expression in basal epithelium using the KRT14 promoter rescues epidermal development and perinatal lethality but not limb, tail and craniofacial defects . Unexpectedly, we find that basal Irf6 expression is not sufficient to rescue oral adhesions and palatal clefting. Furthermore, we find that oral adhesions between the palatal shelves and the tongue are completely associated with CLP, suggesting a key role in pathogenesis. Results Titrati ng Irf6 dose shows that m andible -maxilla oral adhesions are not sufficient for clefting To investigate how endogenous Irf6 expression alters palatal development, we examined craniofacial tissue from embryos with a recently characterized hypomorphic allele !"%'!(Irf6 neo ) (Kousa, Chapter 3) . To reduce endogenous expression further, we combined the hypomorphic allele with the Irf6 genetrap allele ( (5); herein referred to as Irf6 -). Based on previous work, we predict that reducing endogenous Irf6 expression would lead to oral adhesions and palatal clefting. In examining Irf6 neo /+ (N=3) , we found no evidence for oral adhesions or palatal clefting at E17.5. In contrast, Irf6 neo /- (N=6) had completely penetrant , bilater al oral adhesions between the mandible -maxilla most frequently and severely involving the mandibular tooth germ (Fig. 21A) . However, despite the severity of these oral adhesions, we did not detect palatal clefting in Irf6 neo/ - embryos . To understand this process, we marked the oral epithelium using Krt6 for the periderm (Fig. 21B,C) and Krt14 and Trp63 (Fig. 21B -E) for the basal cell layer. We previously showed that loss of Irf6 expression between the mandible and maxilla at the tooth ge rm is associated with loss of Krt6 and oral adhesions (15). Here, we predicted that reducing Irf6 expression would lead to a loss of Krt6 expression and oral adhesions. However, oral adhesions were pres ent between t he mandible and maxilla despite Krt6 expression (Fig. 21B). Together, t hese results suggest that oral adhesions between the mandible and maxilla are not sufficient for palatal clefting and that loss of Krt6 expression is not required for oral adhesions at the tooth germ . Assuming that periderm is required in preventing oral adhesions, these results suggest that Krt6 is an unreliable periderm marker in pathological states. An alternative explanation is that we have perturbed the periderm but that perid erm function is not associated with oral adhesions . To delineate between these two models, we sought to identify a novel molecular signature for the periderm. We previously showed that Grhl3 !"%(!is downstream of Irf6 in periderm (16). Here, we predict that attenuating Irf6 dose would affect Grhl3 expression in periderm. Consistently, Grhl3 expression is lost almost exclusively in adherent oral epithelium, including periderm (Fig. 21F,G ). As Grhl3 marks non-adherent periderm, we conclude that Krt6, an intermediate filament, may also be a marki ng stress in mutant oral epithelium. Analysis of proliferation using Ki -67 did not reveal qualitative changes in Irf6 neo/ - embryos (Fig. 21H,I ). Irf6 expression using the KRT14 promoter completely rescues cutaneous defects To test how epithelial development contributes to pleiotropic dysmorphology in Irf6 -/- embryos , we used a basal epithelial -specific promoter (KRT14 ) to drive Irf6 expression . In a manner analogous to Ikka rescue using the KRT 14 promoter (12), we predict ed that epithelial Irf6 expression would cell-autonomously rescue epidermal defects. We considered limb, skeletal, craniofacial and tail defects to be secondary to epidermal pathology. As such, we predicted that epidermal rescue would lead to limb, craniofacial and tail rescue . Genotyping of embryos from the experimental cross (Irf6 +/-;Tg Krt14::Irf6 x Irf6 +/-) revealed a significant difference between the expected and predicted Mendelian distribution ( Table 20; N=168 , p-value = 9.67 x 10 -6). Distribution differences were primarily driven by under -representation of Irf6 -/- (predicted 22; actual 9 ) and Irf6 +/- (predicted 44; actual 22 ). At E15.5, u nder -representation of the Irf6 -/- genot ype by approximately 13 co-occurred with 11 embryonic resorptions (p-value = 0.048, FisherÕs exact, two -tailed T -Test, ba sed on 1 -3% spontaneous resorptions in C57BL /6 mice , tested against the maximum 3% rate ) (17). However, Irf6 -/-;Tg Krt14 ::Irf6 embryos did not differ from the expected Mendelian prediction ( predicted 22; actual 23 ). Therefore, embryonic !"%)!lethality did not con tribute to a skewed production of experimental embryos (Irf6 -/-;Tg Krt14::Irf6 ). In the epidermis, loss of Irf6 leads to an expanded, hyper -proliferative super -basal layer that lacks stratification and differentiation. First, w e anal yzed skin mRNA to quantify Irf6 expression by the KRT14 promoter . We find that Irf6 -/-;Tg Krt14::Irf6 skin expresses significantly more Irf6 relative to knockout embryos (p-value = 0.00 1) (Fig. 22 A). Importantly, Irf6 expression in Irf6 +/+ and Irf6 -/-;Tg Krt14::Irf6 was not statistically different (p-value = 0.32). Histological ly, we find that Irf6 expression using the KRT14 promoter is suffici ent to rescue epidermal morphology (Fig. 22B ). Molecularly, Irf6 expression restricts Krt1 and Krt14 expression in Irf6 -/-;Tg Krt14::Irf6 , leading to a molecular profile highly analogous to Irf6 +/+ pups ( Fig. 22C -D). Irf6 expression also lead to epidermal stratification and differentiation, as seen with expression of the differentiation marker Loricrin in Irf6 -/-;Tg Krt14::Irf6 but not Irf6 -/- embryos ( Fig. 22E ). Furthermore, we did not detect Krt6 , a cutaneous marker of stress, in Irf6 -/-;Tg Krt14::Irf6 at P0 (Fig. 22 F). Epithelial Irf6 rescues perinatal lethality Examination of Irf6 -/-;Tg Krt14::Irf6 pups (N=12) revealed dramatic differences in morphology as compared to both wild -type and Irf6 -/- littermates. Most strikingly, while Irf6 -/- pups died shortly after birth, Irf6 -/-;Tg Krt14::Irf6 did not (Supp movie). Instead, Irf6 -/-;Tg Krt14::Irf6 pups appeared to be active and responsive to environmental stimuli. However, after periods of mild activity, highly comparable to wild type littermates, Irf6 -/-;Tg Krt14::Irf6 demonstrated labored breathing wit h marked abdominal retractions. Unexpectedly, this !"%*!data suggests a novel role for Irf6 in respiration. Despite perinatal survival , Irf6 -/-;Tg Krt14 ::Irf6 did not live until weaning . We did not observe gross pheno typic variation in Irf6 -/-;Tg Krt14::Irf6 experimental pups. Grossly, Irf6 -/-;Tg Krt14::Irf6 pups had normal skin that appeared more taut than wildtype littermates (Fig. 23A -B). Importantly, cutaneous rescue led to loss of adhesions between the body wall and appendages, including the tail and limbs . Unlike littermates, Irf6 -/-;Tg Krt14::Irf6 pups did not have a milk spot, suggesting dysfunction in the digestive tract. Furthermore , despite rescue of appendage -body wall adhesions, clubbing of both upper and lower limbs persisted. Furthermore, l oss of adhesions around the tail reveals a completely penetrant curl in Irf6 -/-;Tg Krt14::Irf6 pups. We analy zed the skeletons of these pups to d etermine cartilage and bony dysfunction. Importantly, the appendicular skeleton appeared different in two ways (Fig. 23C) . First, consistent with above data , loss of skin adhesions in Irf6 -/-;Tg Krt14::Irf6 permitted limb movement away from the axial skeleton. Secondly , like knockout littermates , Irf6 -/-;Tg Krt14::Irf6 pups exhibited fully penetrate syndactyly of both upper and lower limbs (Fig. 23C, top inset). Ossification of tail vertebra did not app ear to differ . In the axial skeleton, we found that Irf6 -/-;Tg Krt14::Irf6 , like Irf6 -/- pups, had a bifid xiphoid (Fig. 23C, bottom inset). This data suggests that appendage -body wall adhesions result from Irf6 expression in the epidermis . Interestingly, the se data also s uggest that limb defects, including clubbing and syndactyly, do not result from loss of epithelial Irf6 expression. !"%+!Completely penetrant clefting and esophageal adhesions despite basal Irf6 expression Prior work shows that Irf6 -/- pups have uniform ly penetrant oral clefting with ubiquitious oral and esophageal adhesions. As we did not detect a milk spot, we further predict that oral clefting and esophageal adhesions contribute to u niform postnatal lethality of Irf6 -/-;Tg Krt14::Irf6 pups. Despite Irf6 expression in basal epithelium, Irf6 -/-;Tg Krt14::Irf6 embryos (N=12 ) display completely penetrant palatal clefting at P0 (Fig. 24) . However, i n contrast to Irf6 -/- pups, Irf6 -/-;Tg Krt14::Irf6 have less extensive or al adhesions . In the anterior (Fig. 24A) and middle palate (Fig. 24B) , oral adhesions are primarily found between the mandible and maxilla . More posterior ly, oral adhesions are seen betwe en the ton gue and palate but not between the mandible and maxilla (Fig. 24C) . Unexpectedly, we also observe tongue -palate oral fusions in the posterior palate . Histological examination of t he thoracic cavity also reveal s fully penetrant esophageal adhesions in Irf6 -/-;Tg Krt14::Irf6 pups (Fig. 25) . Surprisingly, these adhesions are highly similar to but morphologically distinct from Irf6 knockout embryos (ÒMÓ vs. ÒSÓ shape) (Fig. 25A) . However, p revious reports show that skin KRT14 promoter activity is 1000-fold greater than the esophageal epithelium (14). Consistently , we find that Krt14 immunostaining in the esophagus requires 10 -fold more exposure time relative to skin (Fig. 25B) . Ther efore, in direct contrast to epidermal rescue, Irf6 expression via the KRT14 promoter is not sufficient to rescue esophageal adhesions. Together, these data suggest that morphological change in the esophagus of Irf6 -/-;Tg Krt14::Irf6 pups represent an intermediate phenotype. !"%$! Palate -tongue oral adhesions prevent palatal elevation Considering analysis of Irf6 -/-;Tg Krt14::Irf6 pups, we concluded that Irf6 expression in basal epithelium i s not sufficient to rescue palatal adhesions and clefting. To examine the pathophysiological mechanism and cell types leading to oral clefts in Irf6 -/-;Tg Krt14::Irf6 pups, we examined embryos at E15.5 (N=11 ). We examined three types of oral adhesions based on the surfaces they approximate: 1) mandible and maxilla , 2) palate and mandible and 3) palate and tongue. Between the mandible -maxilla, we found completely penetr ant, bilateral oral adhesions throughout the palate that did not differ from Irf6 -/- embryos (Fig. 26A-C). At the palate -mandible interface , oral adhesions are completely rescued in the anterior and middle palate (Fig. 26A,B). However, oral adhesions persisted between the palate and mandible in the posterior palate (Fig. 26C) . Palate-tongue oral adhesions are found throughout the pala te but are limited in severity in contrast to Irf6 -/- (Fig. 26A -C) Palate-tongue oral adhesions in the an terior and middle palate obstructed elevation and stymied horizontal growth. Critically, these palate -tongue adhesions interfered with contact between the palatal shelves (Fig. 27A) . As a result, fusion between the palatal shelves did not take place, leadi ng to a frank cleft as the head enlarges progressively until birth. To investigate the molecular mechanism underlying this highly complex process, we analyze markers of oral epithelium in mid -palate at E15.5 . First, we mark the periderm with Krt6 ( Fig. 27B) and basal cells with Trp63 and Krt14 ( Fig. 27B -C). As expected, immunostaining for Irf6 shows expression in the basal cell layer. Considering that basal !"%#!cells differentiate into periderm cells, we also expect and find Irf6 expression in the periderm ( Fig. 27D ). Importantly, and consistent with results in Irf6 -/- and Irf6 neo/ - embryos, Krt6 expression in the periderm of Irf6 -/-;Tg Krt14::Irf6 embryos is not sufficient to alter palate -tongue oral adhesions (Fig. 27, middle panel). Furthermore, we find tha t re -expression of Irf6 rescues Krt6 expression but does not attenuate adhesions at the tooth germ (data not shown). Based on results in Irf6 neo/ - embryos, we predict that re -expression of Irf6 would rescue Grhl3 expression in periderm. Consistently, Grhl3 expression is observed non-adherent oral epithelium bordering palate -tongue oral adhesions (Fig. 27E ). Unlike Krt6, these results suggest that Grhl3 expression in periderm is both necessary and sufficient in marking oral adhesions. Furt hermore, consistent with previous findings, re -introduction of Irf6 led to a wildtype pattern of Activated Caspase 3 expression in the nasal epithelium (Fig. 27F) and Ki -67 expression in palatal mesenchyme (Fig. 27G) . Importantly, Ki-67 positive cells in palatal mesenchyme of Irf6 -/-;Tg Krt14::Irf6 embryos suggests non -cell autonomous Irf6 regulation of mesenchymal proliferation. DISCUSSION We report a n Irf6 dose -dependent model of orofacial clefting . We further provide an animal model that decouples skin development from limb, tail, skeletal and craniofacial defects in Irf6 knockout embryos . Together, these results suggest that pleiotropic IRF6 disease can result from the cell -type, tissue and dose. Our previous report s show s that compound heterozygo us embryos for the Irf6 null and hypmorphic alleles have completely penetrant caudal neural tube defects. Here , we report completely penetrant !"%%!mandible -maxilla oral adhesions without clefting. Taken together, these data suggests that the neural tube is mor e sensitive to Irf6 dose than palatal clefting in the mouse. Presence of oral adhesions and a normal tail in Irf6 +/- heterozygous embryos sugges ts that oral epithelium is most sensitive to Irf6 dose. The preponderance of evidence , in this and previous reports, partially illuminates the mechanistic gap between oral adhesions and clefting. First, data from compound heterozygous embryos , shown here, suggests that oral adhesions between the maxilla and mandible are not sufficient for palatal clefting. In co ntrast , Irf6 -/-;Tg Krt14::Irf6 also have palate -tongue oral adhesions and completely penet rant palatal clefting . As seen at E15.5 in Irf6 -/-;Tg Krt14::Irf6 , palate -tongue oral adhesions may be directly interfering with the coordinated maneuvers required for palatal closure, including palatal re-orientation and adhesions. In support of this model, we recently showed that Grhl3 -/- embryos have completely penetrant bilateral mandible -maxilla oral adhesions but that clefting only results when tongue -palate adhesions are al so present. Taken together, these data support a more prominent role for functional oral adhesions between the tongue and the palatal shelves. These data are critical because it suggests that common and rare DNA variants in Irf6 can lead to orofacial clefting secondary to oral epithelial adhesions. As such, c lefting in patients with VWS, PPS and isolated orofacial clefting may result from a process that is analogous to a Pierre Robin Sequence , whereby tongue -mediated clefting results from physical attachment to the palatal shelves (adhesion) as opposed to obstruction !'&&!(micrognathia ). Thus, m odulating the molecular and physical properties of adhesive Pierre Robin Sequence may be a clinical target for future preventative strategies. While we donÕt believe that mandible -maxilla oral adhesions are playing a role in palatal development, they may be interfering with odontogenesis and contributing to hypodontia , a common finding in patients with VWS and PPS patients. . First, if oral adhesions persi st, breakdown of epithelial cells may allow mesenchymal confluence, i.e. fusions. Oral fusions are found in VWS and PPS as syngnathia and may physically obstruct dental eruption. Second, considering presence at P0, suckling or mastication may obliterate or al adhesions and epithelial integrity . Obliteration of adhered dental epithelium during odontogenes is may predispose to hypodontia . These models are not exclusive of a cell -autonomous affect of Irf6 on odontogenesis. Oral adhesions involving the oral and nasal surfaces of the palatal shelves provide a striking contrast between Irf6 -/- and Irf6 -/-;Tg Krt14::Irf6 embryos. Importantly, basal Irf6 expression completely rescues palate -mandible but not palate -tongue oral adhesions. These results suggest that bas al Irf6 expression plays an important role in palatal development . While palate -tongue oral adhesions are also less severe, localizing to distal aspects of the palate, important developmental implications are proposed . Strikingly, palate -tongue oral adhesi ons physically prevent horizontal reorientation of the palatal shelves. Despite the physical constraints, medial surfaces of the palatal shelves approximate toward midline. These data suggest that palatal reorientation, rather than elevation, plays a prom inent role in palatogenesis (18, 20). Furthermore, while palatal nasal epithelium adheres to the tongue, horizontal outgrowth supports a more fluid !'&"!determination of the Medial Edge Epithelium rather than a single cohort of cells along the palate. Importantly we observed this pattern in the anterior and middle sections of the palatal shelves. In the posterior palate, pa latal elevation is not observed, perhaps as function of palate -mandible oral adhesions. As compared to Irf6 neo /- (Kousa, Chapter 3), Irf6 -/-;Tg Krt14::Irf6 embryos have a compa rable level of epidermal Irf6 expression but more severe limb clubbing, syndactyly and tail anomalies . These data suggest that limb, tail and digit defects in Irf6 -/-;Tg Krt14::Irf6 are due to non -epithelial Irf6 expression and function. . Consistent with th is findin g, we previously found that Irf6 is expressed in the hindlimbs, cartilage primordium of the humerus (forelimb) and metacarpals (digits) (19). Interestingly, MCS9.7, an Irf6 enhancer classically associated with epithelial expression , is also active in these tissues. As such, these data suggest an important conceptual shift in our understanding of Irf6 and disease mechanism. Unlike the cohort of alleles previously reported to phenocopy the Irf6 knockout, use of KRT14 promoter to drive epithelial specific Irf6 expression did not rescue limb defects, despite skin rescue. These results suggest non -epithelial Irf6 expression and function in limb development. Furthermore, this work suggests that the common knockout phenotype for these alleles is not through a single molecular pathway but rather an endpoint achieved through multiple cellular and molecular means. In su pport of this model, Ikka limb defects were fully rescued with the KRT5 and KRT14 driver while Irf6 defects persisted . In support of this model, a test for a genetic interaction between Ikka and Irf6 !'&'!did not reveal epistasis (6). In contrast, considering epistasis between 14-3-3 sigma and both Irf6 and Kdf1 in skin, limb and craniofacial tissues (6), we would predict that a similar rescue with the KRT14 promoter for 14-3-3sigma would lead to incomplete limb rescue, as shown here for Irf6 . As opposed to incomplete rescue, an alternative model is that phenotypes described in the rescue embryos are gain -of-function from ectopic or over -expression of Irf6 in the basal cell layer. However, we do not prefer this model because the phenotypes are similar to Irf6 knockout embryos than they are to embryos over -expressing Irf6 . Furthermore, this constellation of anomalies is not seen Irf6 +/+ ;Tg Krt14::Irf6 , which h ave higher levels of Irf6 expression than Irf6 -/-;Tg Krt14::Irf6 . In that context, the results for Ripk4 offer a striking contrast. While epithelial expression of Ripk4 rescued limb defects, in manner analogous to Ikka , ectopic expression of the gene did n ot rescue either Ikka or 14-3-3sigma . Together, these results suggest that Ripk4 is 1) involved in an independent parallel pathway 2) requires Ikka and 14-3-3sigma for activation or 3) is a peripheral player in this pathway and could be either upstream or downstream . However, in comparing the phenotypes, Ripk4 knockouts are the least severely affected of the five alleles. Analogous rescue experiments for 14-3-3sigma knockout and e pithelial 14-3-3sigma expression or over -expression of Irf6 , Ikka and Ripk4 has not been reported. Author Contributions YAK and BCS conceived of the work. YAK designed and performed all murine crosses, completed statistical analyses, analyzed all data, prepared the figures and tables and !'&(!wrote the manuscript. YAK performed and analyzed morphological phenotyping and histolog ical analysis, immunostaining, qPCR, skeletal preps and captured images . DM performed histological stains and analysis, conducted and analyzed immunostaining, captured and collected images. YAK supervised all mouse genotyping and statistical analyses. BCS supervised the project. All co -authors reviewed the manuscript and edits were reviewed/incorporated by YAK and BCS. ACKNOWLEDGEMENTS Financial support to YAK (#F31DE022696 -01) came from the NIH National Institute of Dental and Craniofacial Research. WEB RESOURCES Online Mendelian Inheritance in Man: www.omim.org PolyPhen -2: http://genetics.bwh.harvard.edu/pph2/ Primer3: biotools.umassmed.edu/bioapps/primer3_www.cgi NHLBI/ESP database: evs.gs.washington.edu/EVS/ UCSC genome browser: genome -euro.ucsc.edu/cgi -bin/hgGateway MATERIALS AND METHODS Murine crosses Use, husbandry and procedures involving research animals was approved by the Michigan State University Institutional Animal Care and Use Committee (AUF # 05/ 12-093-00). Harem matings (4 females with a single breeder male) were used to enhance !'&)!pregnancy rates and presence of a copulation plug was denoted at E0.5. We used a recently characterized transgene that drives Irf6 expression under the control of the KRT 14 promoter to rescue Irf6 knockout embryos . We f irst inter -crossed Irf6 +/- and TgKrt14::Irf6 (Irf6 tg) to produce Irf6 +/-;Tg Krt14::Irf6 . We than inter -crossed Irf6 +/-;Tg Krt14::Irf6 with Irf6 +/-. Rescue embryos (Irf6 -/-;Tg Krt14::Irf6 ) had an expected yield of 12.5% . We examined embryos at two developmental time points, E15.5 and just upon birth (P0). To test the effect of Irf6 dose in the development of oral epithelium, we used a recently characterized hypomorphic allele. We than combined the hypomorphic all ele with a null Irf6 allele in the compound heterozygous embryos to reduce endogenous expression further. Genotyping was complete as described previously. Morphological and histological analysis All embryos and pups were grossly examined upon dam euthanasia or parturition. After initial inspection, embryos and pups were placed into freshly prepared 4% paraformaldehyde (245 -684, Protocol). Upon fixation at 4 ¡C for 16 -24 hours, embryos and pups were dehydrated in 50 -80% ethanol until time of embeddin g. Paraffin embedded material was sectioned at 7 µm intervals for both craniofacial tissue and thoracic cavities. Hematoxylin (GHS332, Sigma) and Eosin (E511 -25, Fisher Chemical) staining was complete essentially as described previously ( Chapter 3 ). Briefl y, we removed the paraffin with a series of short Xylene incubations. We than hydrated the tissues using a series of increasingly diluted ethanol solutions. Following short incubations in Eosin (90 seconds) and Hematoxylin (90 seconds), we dehydrated the tissue using a series of decreasingly diluted ethanol solutions. Following Xylene !'&*!incubations, the tissue was mounted (Permount, SP15 -100, Fisher Scientific) and visualized. At both E15.5 and P0 we used the eyes to determine anterior (anterior to eyes) , mid dle (at eyes) and p osterior palate (posterior to eyes). Molecular analyses of murine tissue Immunostaining was complete with the protocol an d regents described previously (Chapter 3 ). Briefly, we performed a similar series of tissue incubations in Xylene and ethanol to remove paraffin and hydrate the tissue, in a manner highly analogous to Hematoxylin and Eosin staining. Following this step, we performed antigen retrieval in sodium citrate (pH6.0) and permeabilization in Triton X -100 (VWR). We than perfor med a series of washing steps to remove the detergent. After this step, we incubated the slides in blocking reagents, including 10% BSA in PBS for one hour and 40 µg/ml of Goat anti -mouse Fab fragment in PBS (Jackson ImmunoResearch Laboratories, 115-007-003) also for one hour. Primary antibodies were incubated for 18 -24 hours at 4 ¡C. Primary antibodies include Tp63 (Santa Cruz, 4A4, sc -8431), Keratin 6 (Covance, PRB -169P), Keratin 14 (Novocastra, NCL -L-LL002), Keratin 1 (NCL -CK1, Novocastra) , Lori crin ( PRB -145P, Covance ), Activated Caspase 3 (Abcam, Ab13847), Ki -67 (ab15580, Abcam ). Skeletal prep Skeletal pre ps were processed as described previously (Chapter 3 ). Briefly, after fixing as described above, we removed skin and subcutaneous fat from embryos and than incubated in 70% and 95% ethanol for 24 hours in per solution. After 72 hours incubation !'&+!in 2% KOH, we stained the cartilage in Alcian blue (Sigma, A5268 -10G). De -staining of Alcian blue in 95% ethanol was followed by Alizarin Red stainin g (Sigma -Aldrich, A5533 -25G) for 24 -36 hours. Skeletal tissue was than placed in 1%KOH/20% Glycerol solution before images were taken. Bioimaging upright/f luorescent microscope and s tereomicroscope We image tissue on an upright microscope (Nikon Eclipse 90i upright) with a 4x, 10x and 40x objectives, as described previously (Chapter 3 ). NIS Elements Advanced Research v3.10 imaging software was used to obtain and to analyze images. Enhancement was limited to program algor ithms , applied evenly to all samples, and only included deconvolution and sharpening with Gauss -Laplace. To capture whole mount embryo images we used a SMZ1000 Nikon microscope with both Fiber Optic Gooseneck and Ring Light sources , NIS -Elements Software 4 .11. We used Adobe Photoshop Elements v9.0 to construct and produce the figures. Transcriptional profiling using quan titative -PCR We analyzed mRNA levels using methods and protocol essentially as described previously, (Chapter 3 ). Briefly, dorsal skin is collected from embryos at the time point indicated. We snap freeze the tissue in liquid nitrogen and use TRIzol RNA extraction kit (15596-026, Ambion). To prevent DNA contamination we treated the samples with RNase -Free DNase (792 54, Qiagen) for 30 minutes, which is followed by heat inactivation at 65 ¡C. To purify the RNA , we used acidic phenol and chloroform. After purification, we resuspended the RNA in RNase -free H2O and incubated at 55 ¡C for 10 !'&$!minutes. To make cDNA, we used Ol igo dT primers (18418 -012, Invitrogen), dNTP mix (18427-013, Invitrogen), SuperScriptIII Reverse Transcriptase (18080 -093, Invitrogen) and Recombinant RNasin Ribonuclease Inhibitor (N2511, Promega). The negative control for this reaction did not include ei ther the SuperScriptIII Reverse Transcriptase or the Recombinant RNasin Ribonuclease Inhibitor. We used SYBER Green (4309155, Applied Biosystems) to quantify transcript levels from total starting material of 5.5 ng of cDNA. We quantified fold c hange using the delta -delta Ct -method relative to Beta -Actin. All reactions were performed with three technical replicates per biological sample. Murine primers as shown previously ( Chapter 3, Table 18) . Statistical analysis We used both Excel, v. 2010, and GraphPad Prism Software, version 5, to analyze data. All tables and histograms were constructed within GraphPad. We used a StudentÕs t -test to determine significance and rejected the null hypothesis with a p -value equal to or below 0.05. !'&#! APPENDIX !"#$!APPENDIX A Figure 21 : Irf6 compound heterozygosity causes completely penetrant oral adhesions but not clefting . !"%#!Figure 21 . (contÕd) !"%%!Figure 21 . (contÕd) !"%"!Figure 21 . (contÕd) Head coronal section stained with Hematoxylin and Eosin of E17.5 embryos examin ing A) anterior (top) and mid palates (bottom). Compared to Irf6 neo /+, Irf6 neo /- embryos have completely penetrant mandible -maxilla oral adhesions but not clefting . Immunostaini ng for Krt6 (red)/ p63 (green) (B, tooth germ magnified in C), Krt14 (red) (D, E), Grhl3 (red) (F, G) and Ki -67 (green) (H, I). Krt6 expression is not sufficient to rescue o ral adhesions Irf6 neo /-. DAPI (blue) marks nuclei (B-I). In contrast to Krt6, Grhl3 expression is reduced in areas of oral adhesions. Scale bar (A) 500 um ; (B , D, F, H ) 100 um; (C, E, G, I) 20 um. !"%&!A Figure 22 : Irf6 expression using the KRT14 promoter rescues cutaneous defects in knockout embryos. !"%'!Figure 22 . (contÕd) !"%(!Figure 22 . (contÕd) !"%)!Figure 22 . (contÕd) qPCR analysis of RNA levels in perinatal murine skin (A). Transcriptional analysis reveals a significant increase in Irf6 expression in TgKRT14::Irf6 (n=3) compared to Irf6 -/- (n=4). No statistical significant differences are detected between TgKRT14::Irf6 and Irf6 +/+ (n=4). Skin histological analysis with Hematoxylin and Eosin reveals epidermal hypertrophy in Irf6 -/- but not TgKRT14::Irf6 and Irf6 +/+ pups (B). Immunostaining o f Krt1, Krt14, Loricrin and Krt6 (C -F). TgKRT14::Irf6 have epithelial stratification and loss of ectopic Krt1 and Krt14 expression (C -D). TgKRT14::Irf6 demonstrate epithelial differentiation (Loricrin) and loss cell stress markers in mature skin (Krt6) (E-F). Scale bars (A) 100 um, (B) 50 um. !"%*! Figure 23 : Epidermal expression of Irf6 rescues perinatal lethality without altering skeletal defects, limb clubbing or syndactyly. !"%+! Figure 23 . (contÕd) !"%$!Figure 23 . (contÕd) Profile (A) and Frontal (B) views of representative P0 pups with the following genotypes; Left: Irf6 +/+ ; Center: Irf6 -/-;TgKRT14::Irf6 ; Right: Irf6 -/-. Gross analysis of Irf6 -/-;TgKRT14::Irf6 reveals perinatal survival, limb clubbing and syndactyly, a curled tail and a somewhat taut, shiny skin compared to wildtype littermates. Unlike Irf6 -/- littermates, Irf6 -/-;TgKRT14::Irf6 appendages are not attached to the body wall and an open oral cavity is visible . Fig. 1.2. Profile views (C) of s keletal prepara tions of P0 . Analysis of Irf6 -/-;TgKRT14::Irf6 reveals limb clubbing and syndactyly, in a manner highly analogous to Irf6 -/- (arrow, top inset). In contrast, the sternum (arrow head, bottom inset) and in particular the xiphoid process appear to be modulated in Irf6 -/-;TgKRT14::Irf6 pups. !""#! Figure 24: Rescued pups have completely penetrant palatal clefting and oral adhesions !""%!Figure 24 . (contÕd) Head coronal section stained with Hematoxylin and Eosin of perinatal pups examining A) anterior, B) middle C) posterior soft palates. While the oral cavity is uniformly less severely affected , oral adhesions persist bilaterally. In anterior (A) and middle (B) palates, oral adhesions are prominently found b etween the mandible and maxilla . In the posterior palate (C), adhesions are not found between the mandible and maxilla. Scale bar (A) 500 um. !"""! Figur e 25: Obliteration of the esophageal lumen contributes to postnatal lethality . !""&!Figure 23 . (contÕd) A) Histological analysis ( Hematoxylin and Eosin) staining in transverse sections of P0 thoracic cavities. In stark contrast to the open lumen in Irf6 +/+ pups, both Tg KRT14::Irf6 , like Irf6 -/- pups have completely penetrant esophageal adhesions. However, ther e was a distinct difference in the shape of the tissue, with TgKRT14::Irf6 having an ÒMÓ while Irf6 -/- had an ÒSÓ shape. B) Immunostaining of Krt14. Staining for Krt14, whose highly conserved promoter is used to drive expression, showed a signal but requir ed a 10 -fold increase in exposure for detection. Scale bars (A) 100 um, (B) 50 um. !""'! Figure 26 : Tongue -palate oral adhesions obstruct palatal development !""(!Figure 26 . (contÕd) A-C Histological analysis of Irf6 +/+ , Irf6 -/-;TgKRT14::Irf6 and Irf6 -/- hea d coronal sections using Hematoxylin and Eosin; A) anterior, B) middle, C) posterior soft palate. Irf6 -/-;TgKRT14::Irf6 embryos show completely penetrant clefting at E15.5. A -B) In the anterior and middle palate, we did not find differences mandible -maxilla oral adhesions. In contrast, we found that palate (nasal epithlium) -tongue oral adhesions in Irf6 -/-;TgKRT14::Irf6 were markedly less severe, i.e. partially rescued, compared t o Irf6 -/- embryos . In addition, we found complete rescue of palatal adhesions between the oral surface of the palatal shelves and the mandible. C) In posterior palate, we found that palate -tongue oral adhesions were less severe but the palatal shelves were not re -orienting toward midline. Scale bar (A) 500 um. !""#! Figure 27 : Oral adhesions to the tongue prevent re -orientation and apposition of palatal shelves !""$!Figure 27 . (contÕd) Coronal section of E15.5 mid -palate histology, examined with Hematoxylin and Eosin staining (A) . Irf6 -/-;TgKRT14::Irf6 embryos have marked reduction in oral adhesions and palatal shelves elevate to reach midline. Immunostaining of Krt6 (red) /p63 (green) (B) , Krt14 (red) (C), Irf6 (red) (D), Grhl3 (red) (E), Activated Caspase 3 (red) (F), Ki -67 (red) (G). We marked the basal cell layer with Tr p63/Krt14 and the periderm with Krt6. Remarkably, oral adhesions contained Krt6 expression and prevented palatal shelve reorientation in rescue embryos . Furthermore, Irf6 expression rescued Grhl3 expression in oral epithelium. Palatal shelves in rescue embryos exhibited Act Casp 3 and Ki -67 expression highly analogous to wild type littermates. Scale bar (A) 200 um, (B) 100 um. !""%!Table 20: Irf6 -/-; Tg KRT14::I rf6 E15.5 P0 Total Litters 13 13 26 Irf6 +/+ 12 10 22 Irf6 +/- 8 17 25 Irf6 -/- 1 8 9 Irf6 +/+ ; Tg KRT14::Irf6 26 13 39 Irf6 +/-; Tg KRT14::Irf6 31 19 50 Irf6 -/-; Tg KRT14::Irf6 11 12 23 Total 89 79 168 p-value 6.6x10 -8 0.82 9.67 x10 -6 Resorbing 11 N/A 11 (p -value 0.049) We intercrossed mice hemizygous for the TgKRT14::Irf6 (over -expressing Irf6 under the control of the KRT14 promoter) with mice heterozygous for the Irf6 genetrap allele (Irf6 +/gt from here on referred to as Irf6 +/-) to generate Irf6 +/-; Tg KRT14::Irf6 mice, allowed to reach sexual maturity. We than intercrossed Irf6 +/-; Tg KRT14::Irf6 with littermates that are heterozygous for the genetrap allele ( Irf6 +/-). We examined two timepoints, one during embryonic development (E15.5) and another a round parturition (P0). We detected significant embryonic resorptions at E15.5. Furthermore, we found a significant under -representation of Irf6 -/- genotypes and corresponding phenotypes. !""&! BIBLIOGRAPHY !"'(!BIBLIOGRAPHY 1 Kondo, S., Schutte, B.C., Richardson, R.J., Bjork, B.C., Knight, A.S., Watanabe, Y., Howard, E., de Lima, R.L., Daack -Hirsch, S., Sander, A. et al. (2002) Mutations in IRF6 cause Van der Woude and popliteal pterygium syndromes. Nat Ge net , 32, 285-289. 2 Zucchero, T.M., Cooper, M.E., Maher, B.S., Daack -Hirsch, S., Nepomuceno, B., Ribeiro, L., Caprau, D., Christensen, K., Suzuki, Y., Machida, J. et al. (2004) Interferon regulatory factor 6 (IRF6) gene variants and the risk of isolated c left lip or palate. N Engl J Med , 351, 769-780. 3 Botti, E., Spallone, G., Moretti, F., Marinari, B., Pinetti, V., Galanti, S., De Meo, P.D., De Nicola, F., Ganci, F., Castrignano, T. et al. (2011) Developmental factor IRF6 exhibits tumor suppressor activ ity in squamous cell carcinomas. Proc Natl Acad Sci U S A, 108, 13710-13715. 4 Stransky, N., Egloff, A.M., Tward, A.D., Kostic, A.D., Cibulskis, K., Sivachenko, A., Kryukov, G.V., Lawrence, M.S., Sougnez, C., McKenna, A. et al. (2011) The mutational landscape of head and neck squamous cell carcinoma. Science , 333, 1157-1160. 5 Ingraham, C.R., Kinoshita, A., Kondo, S., Yang, B., Sajan, S., Trout, K.J., Malik, M.I., Dunnwald, M., Goudy, S.L., Lovett, M. et al. (2006) Abnormal ski n, limb and craniofacial morphogenesis in mice deficient for interferon regulatory factor 6 (Irf6). Nat Genet , 38, 1335-1340. 6 Richardson, R.J., Dixon, J., Malhotra, S., Hardman, M.J., Knowles, L., Boot -Handford, R.P., Shore, P., Whitmarsh, A. and Dixon, M.J. (2006) Irf6 is a key determinant of the keratinocyte proliferation -differentiation switch. Nat Genet , 38, 1329-1334. 7 Ric hardson, R.J., Dixon, J., Jiang, R. and Dixon, M.J. (2009) Integration of IRF6 and Jagged2 signalling is essential for controlling palatal adhesion and fusion competence. Hum Mol Genet , 18, 2632-2642. 8 Takeda, K., Takeuchi, O., Tsujimura, T., Itami, S., Adachi, O., Kawai, T., Sanjo, H., Yoshikawa, K., Terada, N. and Akira, S. (1999) Limb and skin abnormalities in mice lacking IKKalpha. Science , 284, 313-316. 9 Hu, Y., Baud, V., Delhase, M., Zhang, P., Deerinck, T., Ellisman, M., Johnson, R. and Karin, M. (1999) Abnormal morphogenesis but intact IKK activation in mice lacking the IKKalpha subunit of IkappaB kinase. Science , 284, 316-320. !"')!10 Lee, S., Kong, Y. and Weatherbee, S.D. (2013) Forward genetics identifies Kdf1/1810019J16Rik as an essential regulat or of the proliferation -differentiation decision in epidermal progenitor cells. Dev Biol , 383, 201-213. 11 Holland, P., Willis, C., Kanaly, S., Glaccum, M., Warren, A., Charrier, K., Murison, J., Derry, J., Virca, G., Bird, T. et al. (2002) RIP4 is an ank yrin repeat -containing kinase essential for keratinocyte differentiation. Curr Biol , 12, 1424-1428. 12 Sil, A.K., Maeda, S., Sano, Y., Roop, D.R. and Karin, M. (2004) IkappaB kinase -alpha acts in the epidermis to control skeletal and craniofacial morphoge nesis. Nature , 428, 660-664. 13 Liu, B., Willette -Brown, J., Liu, S., Chen, X., Fischer, S.M. and Hu, Y. (2011) IKKalpha represses a network of inflammation and proliferation pathways and elevates c -Myc antagonists and differentiation in a dose -dependent manner in the skin. Cell Death Differ, 18, 1854-1864. 14 Rountree, R.B., Willis, C.R., Dinh, H., Blumberg, H., B ailey, K., Dean, C., Jr., Peschon, J.J. and Holland, P.M. (2010) RIP4 regulates epidermal differentiation and cutaneous inflammation. J Invest Dermatol , 130, 102-112. 15 Peyrard -Janvid, M., Leslie, E.J., Kousa, Y.A., Smith, T.L., Dunnwald, M., Magnusson, M., Lentz, B.A., Unneberg, P., Fransson, I., Koillinen, H.K. et al. (2014) Dominant mutations in GRHL3 cause Van der Woude Syndrome and disrupt oral periderm development. Am J Hum Genet , 94, 23-32. 16 de la Garza, G., Schleiffarth, J.R., Dunnwald, M., Man kad, A., Weirather, J.L., Bonde, G., Butcher, S., Mansour, T.A., Kousa, Y.A., Fukazawa, C.F. et al. (2013) Interferon regulatory factor 6 promotes differentiation of the periderm by activating expression of grainyhead -like 3. J Invest Dermatol , 133, 68-77. 17 Krishnan, L., Guilbert, L.J., Russell, A.S., Wegmann, T.G., Mosmann, T.R. and Belosevic, M. (1996) Pregnancy impairs resistance of C57BL/6 mice to Leishmania major infection and causes decreased antigen -specific IFN -gamma response and increased produc tion of T helper 2 cytokines. J Immunol , 156, 644-652. 18 Bush, J.O. and Jiang, R. (2012) Palatogenesis: morphogenetic and molecular mechanisms of secondary palate development. Development , 139, 231-243. 19 Fakhouri, W.D., Rhea, L., Du, T., Sweezer, E., Morrison, H., Fitzpatrick, D., Yang, B., Dunnwald, M. and Schutte, B.C. (2012) MCS9.7 enhancer activity is highly, but not completely, associated with expression of Irf6 and p63. Dev Dyn , 241, 340-349. 20 Walker, B.E., Fraser, F. C. (1956) Closure of the Secondary Palate in Three Strains of Mice. J. Embryol. Exp. Morphol. , 4, 176-189.! !"#"! Chapter 5 - Conclusions and Future Directions !"##!Major Themes : Orofacial Clefting We previously identified IRF6 protein -coding mutations in 70% of families with VWS (1). While etiology in the remaining 30% of VWS cases is unknown, linkage studies suggest locus heterogeneity (VWSII, # 606713) (2). Our work, using exome sequencing, targeted prevalence detection and functional studi es shows that mutations in GRHL3 can also lead to Van der Woude Syndrome (VWS) . Considering interaction in zebrafish (3), we ask if Irf6 interacts with Grhl3 in the mouse. We find that e mbryos heterozygous for the Irf6 null allele have oral adhesions at the tooth germ. E mbryos heterozygous for Grhl3 have oral adhesions and fusion s posterior to the tooth germ. Double heterozygous embryos had a combination of both phenotypes. Qualitatively, we conclude an additive relationship between Irf6 and Grhl3 in oral epithelium. Considering our sample size of double het embryos was fairly small, we did not quantitatively analyze changes in the extent (anterior -posterior axis) or pervasiveness (length of adhesions relative to free surface) of this phenotype . As such, our conclusion is that we did not detect epistasis and conclude an additive inter action. However, a n additive interaction suggests that Grhl3 and Irf6 have overlapping roles in murine oral epithelium , if not in the same pathway (Chapter 2) . Considering multiple, highly com plex gene regulatory networks, absence of epistasis should not b e miscons trued for proof of absence . In fact, murine epistasis experiments are highly specific but not sensitive. More important ly, we find that Irf6 expression is both necessary and sufficient for Grhl3 in the periderm (Chapter 4) . Consistently, we find t hat Irf6 is required for Grhl3 expression in caudal neurulation (Chapter 3) . In light of this and our previous work (3), it is plausible to conclude that Grhl3 and Irf6 are part of the same gene regulator network in oral epithelium. !"#$! At least t wo additional clues suggest that Grhl3 and Irf6 interact in the mouse . First, we found a significant number of embryonic resorptions. Second, at weaning, we found significantly less Irf6 +/-;Grhl3 +/- mice than expected . While many systemic diseases can cause lethality, prior work shows that both Irf 6 and Grhl3 regulate development of the gastrointestinal system (4, 5). While l oss of Irf6 leads to esophageal adhesions, loss of Grhl3 leads to a shortened digestive track . However, we did not find esophageal adhesions in Irf6 +/-;Grhl3 +/- pups. Analysis of more distal aspects of the digestive track in pups or older mice has not been undertaken . Furthermore, i n the epithelium of both mouse and zebrafish, Grhl1 and Grhl2 interact with Grhl3 , providing redundant function (3, 6, 7). Therefore, in the oral epithelium of Irf6 +/-;Grhl3 +/-embryos, Grhl1 and Grhl2 may be providing p artially redundant, regulatory function and activation of more distal factors . A test for epistasis between Irf6 and Grhl2 is currently underway w ith a collaborator. A similar experiment for Irf6 and Grhl1 is not currently being pursued but this analysis is warranted . Considering redundant roles in epithelium, Grhl1 and Grhl2 may play a role in human orofacial clefting. Sequencing of Grhl1 and Grhl2 in families with VWS is also currently being pursued. Major Themes: Neural Tube Defects In addition to novel insights into orofacial clefting, this work also outlines novel roles for Irf6 in ectoderm development . Importantly, we show that Tfap2a, Irf6 and Grhl3 , human orofacial clefting genes, also regulate epithelial and neural tube development in the mouse via a complex gene regulatory network (Fig. 1) . In addition, o ur preliminary !"#%!sequencing data from individuals with Spina Bifida suggests that IRF6 is also involved in human neurulation. Co nsidering that we found one muta tion in 96 individuals , the impact of IRF6 function in human neurulation is not yet clear. More significant impact from this work comes from clinical and epidemiological research sho wing that orofacial clefting and neural tube defects share common environmental risk (smoking), iatrogenic compounds (valproic acid) and preventative factors (folic acid). Here, we describe a shared molecular network that might be perturbed in both developmental diseases . It is not yet clear if such shared pathways are common in neural tube and orofacial clefting. Examination of orofacial clefting genes in neurulation and neurulation genes i n orofacial clefting may provide novel shared gene regulatory networks. Biologically, few studies have reported over -expression o f a tumor suppresser transcription factor leading to gain -of-function phenotypes (Chapter 3) . One prominent example , although not a transcription factor, is PTEN (8). Importantly, IRF6 , TFAP2A and GRHL3 are tumor suppressor genes and transcription factors. Considering gain -of-function phenotypes for IRF6 , future work may seek to over -expre ss TFAP2A and GRHL3 in orofacial and ne ural tube development. Whatever the mechanism of perturbation, exploring this gene regulatory network may provide additional candidate genes , risk loci and environmental preventative strategies to reduce the risk of two congenital diseases associated with significant morbidity and mortality . !"#&!Genetic risk for orofacial and neural tube defects VWS and PPS are monogenic diseases, with as yet inconclusive evidence for genetic modifiers (9). In contrast, nonsyndromic cleft palate, cleft lip and cleft lip and palate constitute t hree different congenital diseases each with multiple genetic and envi ronmental risk factors. Likewise, neural tube defects are a highly heterogeneous cohort of anomalies involving at least three different types of developmental defects, each with an array of presentations. For example, spina b ifida, a type of neural tube defect, has four different presentations, including occult a, closed, meningocele and myelomeningocele . Each of these, in turn, has multiple genetic and environmental modifiers. Therefore, upon examination, spina bifida, a type of neural tube defects, may be as genetically complex as orofacial clefting. This reasoning provides important considerations for i dentifying additional candidate genes in human neural tube defects, including sample size, power, effect size and heterogeneity of populations. Considering this genetic complexity, can spina bifida co -occur with orfacial clefting? From our murine studies , we found that under -expressing Irf6 can lead to a curled tail (Chapter 3) and oral adhesions (Chapter 5). Consistently, a review of the literature reveals that the phenotypic spectrum of Popliteal Ptyergium Syndrome (PPS) can include spina bifida (10). In addition, Multiple P tyergium Sydnrome, which has a phenotypic spectrum highly similar to but more severe then PPS, includes multiple examples of spina bifida (11-13). Considering murine studies and multiple clinical examples of orofacial clefting with spina bifida, families with VWS and PPS may be at increased risk of developing neural tube defects. Understanding and managing risk in these families can !"#'!there fore prevent significant co-morbidity. In that context, murine models with both orofacial clefting and neural tube defect s ( Irf6 neo/ -) provide a tractable system to assess patient risk. This work supports two important questions about risk . First , is there a common cohort of genes that Ôstack the deckÕ for all neural tube defects? Second, what is the architecture of genetic risk, i.e. involving a single or multiple biological pathways ? I would predict that a cohort of genes in a sing le or multiple pathways contribute baseline risk but that a environmental or additional genetic insult tips the balance toward pathology . From that vantage point, it is difficult to contextualize a highly complex disease like spina b ifida in the Òcommon di sease, common alleleÓ model. Similarly, gene discovery efforts in diabetes (14) have shown that common alleles contribute a paucity of the risk toward common disease. Expanding the gene regulatory n etwork : Tfap2 and Grhl paralogs Expanding the Gene Regulatory Network (GRN) is critical in determ ining and then modifying clinical risk for isolated orofacial clef ting and neural tube defects. Initially, we discovered an association between orofacial clefting and a TFAP2 binding site within an IRF6 enhancer sequence (15). Mutations in TFAP2! can lead to Branchio -Oculo -Facial Syndrome, which like VWS, is associat ed with orofacial clefting and lip pits. Significantly, we find that Tfap2 ! regulates Irf6 expression in multiple time points and tissues (Chapter 3). However, we also realize that TFAP2 family members, !, ", #, $ and %, share an identical cis-binding motif . In fact, functional redundancy for Tfap2 paralogs !"#(!is observed in different tissues and species (16, 17). Therefore, are TFAP2 paralogs co-regulating IRF6 in the same or different spatiotemporal contexts? In particular, TFAP2" and TFAP2# are highly expressed in the epidermis, in the same cell types as IRF6 (18). Exploring the relationship between IRF6, TFAP2" and TFAP2# may provide additional nodes in this GRN . In support of additional feedback loops, genome -wide analysis of IRF6 binding sites also shows a signal within TFAP2" (19). Consistently, we fin d that Irf 6 regulates Tfap2 " in the skin (Chapter 3). This interaction seems to be evolutionarily conserved in zebrafish, where irf6 is necessary for tfap2c expression in the periderm (20). Considering this, our collaborators are currently sequencing TFAP2" in individuals with VWS. In addition, tissue specific deletion of Tfap2 " using Sox2 leads to caudal neural tube defects that are highly analogous to Irf6 compound heterozygous embryos (21). Thus, we ask if Irf6 regulates Tfap2 " in caudal neurulation. Importantly, we find that Irf6 is required for Tfap2 " in murine tail develop ment. Considering that Irf6 , Tfap2 !, Tfap2 # and Grhl 3 seem to be co -regulating mouse neural tube development, sequencing additional individuals with spina bifida is plausible. Similarly, previous work shows that Grhl paralogs , 1, 2 and 3, have both independent and overlapping function in epidermal development (6). Like tfap2a and tfap2c , grhl1 and grhl2 are expressed in zebrafish periderm. Furthermore, injecting dominant negative irf6 also perturbs grhl1 and grhl2 expression . Remarkably, knocking out Grhl2 in the mouse leads to anencephaly and abdominal wall defec ts that are highly analo gous to Tfap2 ! !"#)!knockout embryos. While the phenotypic similarity suggests interaction, a test for epistasis between Grhl2 and Tfap2 ! did not reveal novel phenotypes in eight double heterozygous embryos (22). Our work, with incomplete penetrance ( Tfap2 !-Grhl3 and Tfap2 !-Irf6 ) and variable expressivity ( TgKRT14::Irf6 ), would suggest that a larger sample size is required to examine this gene regulator network (Chapter 3). Furthermore, considering zebrafish work, Tfap2a and Grhl2 may be interacting indirectly through Irf6 . Examining the Grhl2 locu s for Irf6 cis-binding elements would provide additional mechanistic insights . Considering that Irf6 post -translationally regulates Trp63 and Tfap2 !, protein -protei n interactions are also plausible (Chapter 3) . If experi mental work is consistent, these paralogs may be excellent candidate genes for human orofacial clefting and neural tube defects. Novel implications for Irf6 knockout phenotype Currently, knockout of five murine alleles produces a phenotype highly analogous to loss Irf6 . This phenotype include craniofacial defect s, orofacial clefting , a hyperproliferative epidermis, defective permeability barrie r, limb clubbing, syndactyly . In addition to Irf6 (4, 23), these genes are Ikka (24), 14-3-3& (23), Ripk4 (25) and Kdf1 (26). Considering five genes, 10 possible genetic interactions can be done. Currently, we know that 1) Irf6 and 14-3-3& interact, 2) Irf6 and Ikka do not appear to interact, 3 ) 14-3-3& and Kdf1 interact. However, we also know that both Irf6 and Kdf1 interact with Trp63 (26, 27). Furthermore, w hile the nature of these interactions is not yet clear, a genome -wide scan shows that IRF6 binds within 14-3-3& (19). Testing all additional combinations may provide novel phenotypes and expand our understanding of this gene regulator network. !"$*!Further, lack of interaction (e.g. Irf6 and Ikka ) can be queried further by the addition of other null alleles from this cohort ( 14-3-3&) to create triple heterozygous embryos ( Irf6 +/-;Ikka +/-;14-3-3&+/-). Considering that 1) Irf6 interacts with Trp63, Grhl3 and Tfap2! and 2) Kdf1 interacts with Trp63, additional work can delineate how Grhl3 and Tfap2 ! interact with this cohort of alleles. Considering that over -expressing Irf6 partially rescues the Irf6 knockout, another way to pursue epistasis would be through heterologous genetic rescue using the TgKRT14::Irf6 transgene in 14-3-3&, Ikka , Ripk4 and Kdf1 knockout embryos. Rescue, partial or complete, would suggest that Irf6 is downstream, which is plausible considering transcriptional regulation. Considering that we have multiple positive controls for a genetic interaction in this pathway ( Irf6 -14-3-3& (4), 14-3-3&-Kdf1 (26)), negative results in multiple assays are more informative, assuming a sufficiently large sample size. As IRF6 plays an important role in orofacial clefting, the role of Ikka , 14-3-3&, Ripk4 and Kdf1 in isolated and syndromic orofacial clefting should be examined. Similarly , considering phenotypic overlap , our finding that Irf6 regulates murine neurulation suggests that these four genes have analogous role. Consistent with this rationale, a test for e pistasis between Irf6 and 14-3-3& (23) showed a caudal neural tube defect. Further, while 14-3-3& and Kdf1 were epistatic in skin, they did not interact in the neural tube. However, Kdf1 and Trp63 did interact in the caudal neural tube (26). Interestingly, Trp63 interacts with Irf6 in palate development but an impact on t he neural tube was not reported in the pertinent study (27). Together, this data suggests that a novel cohort of !"$+!murine alleles might be playing a role in the pathogenesis of spina bifida . Therefore, a comprehensive examination of gene -gene interactions , as mentioned above, would also be informative for thi s phenotype. Proposed gene regulatory n etwork Considering that we discovered a shared gene regulator network in orofacial and neural tube development by examining regulator y elements in multiple tissues, a comprehensive interactome irrespective of cell type, tissue or timepoint may provide candidate regulatory elements in future studies (Fig. 2). Importantly, this regulatory network is involved in multiple tissues and cell lines, providing multiple orthogonal views of function and regulation. In additio n to the data presented in this thesis, prior work shows that Tfap2 ! regulates p21 (CDKN1A) via co -regulation with Smad2/3 in keratinocytes (28, 29). Likewise, in Medial Edge Epithelium (MEE), Irf6 regulates p21 expression through repression of Trp63 (30). Considering that Irf6 stabilizes Tfap2 ! protein, Irf6 seems to be regulating p21 expression in a t least two different indirect mechanisms. Furthermore, Smad2/3 associates with Ikk ! to regulate Ovol1 and Mad1 (31, 32). Likewise, Irf6 also regulates Ovol1 (19). While Ovol1 transcriptionally represses c -Myc (33), Mad1 antagonizes Myc -Max dimmers (34, 35). Importantly, prior work shows that Trp63 positively regulates IKK! expression (36). IKK! in turn positively regulates 14 -3-3& by inhibiting promoter hypermethylation (37). In addition, Notch signaling regulates Irf6 expression in epithelium (38). Consistently, a test for epistasis in the mouse shows that Irf6 interacts with Jagged2 , a transmembrane receptor that regulates Notch signaling (39). In oral epithelium, Irf6 is required for Mmp13 (39) and in breast epithelium it !"$"!stabilizes Maspin (40). While Irf6 is regulated via the proteasome (41), the E3 ubiquitin ligas e that mediates this interaction has not been identified. Considering the function of these genes, the balance of this signaling cascade, activation of p21 and repression of cMyc , is stopping cellular proliferation and driving differentiation. This molecular rationale provides insight into murine knockout phenotypes of Irf6 , Ikk ! and 14-3-3$. At this point, it is not clear how Ripk4 and Kdf1 regulate proliferation and differentiatio n or how they might interact in this gene regulator network. Therapeutic considerations in congenital disease In order to treat genetic disease, we must first understand what tissues are affected. For VWS and PPS, p revious work shows that Irf6 is expres sed in epithelium and Irf6 -/- mice have an abnormal epidermis and oral epithelium (4, 23, 39). As such, many of the syndromic anomalies seen in VWS and PPS were thought to result from epithelial defects. However, the phenotypic spectrum of VWS and PPS also includes musculoskeletal, digit, limb and genital anomalies. Therefore, we ask if epithelial rescue is sufficient to modify associated developmental defects in the mouse. Strikingly, epithelial rescue enables perinatal survival and the limbs are no longer adherent to the body wall but upper and lower limb ptyergium persist. Similarl y, expression of Irf6 in oral basal epithelium dramatically reduces adhesions around the palatal shelves but does not prevent palatal clefting. Moreover, limb clubbing, craniofacial de fects and a curled tail persist . Therefore, unlike complete rescue of Ikka using the KRT14 promoter (42), Irf6 is cell -autonomously required in epithelial and non -epithelial tissues. Using tissue specific promoters to over -express or delete Irf6 in cartilage, bone, neural tube and !"$#!peride rm (a cell type that is thought to regulate oral adhesions) will allow further dissection of this important molecular pathway. Most importantly, this work will inform which tissues we need to target to reduce disease burden in utero. Importantly, while palate -mandible adhesions are rescued anteriorly, palate -tongue adhesions persist. Technically, this may result from regulating Irf6 expression with the KRT14 promoter, leading to a spatiotemporal expression program that is inconsistent with endogenous exp ression. Biologically, this may suggest that nasal periderm is more sensitive to Irf6 dose. Considering the histology, it appears that palate -tongue oral adhesions play an important role in palatal clefting. Considering attachment of the palatal shelves to the tongue and ensuing physical obstruction, we propose an adhesive Pierre Robin Sequence. As opposed to micrognathia limiting the volume of the oral cavity, this process involves physical restraints on movement of the palatal shelves with a normal oral c avity volume (Chapter 4) . Preventative strategies in orofacial clefting and neural tube defects If GRHL3 is downstream of IRF6 in human neural tube develo pment, can we design rationale preventative strategies for further exploration? Considering current data suggesting that Grhl3 is downstream of Irf6 in periderm (Chapter 2) and caudal neural tube development (Chapter 3) , mutually advantageous translational and clinical modalities seem feasible. Epidemiological data has shown that folate supplement ation reduces 45-70% of neural tube defect risk. However, a significant portion of neural tube defects are resistant to folate supplementation . Wha t accounts, if anything, for this !"$$!unresponsive cohort? Critically, m ouse data suggests that neural tube defects in Grhl3 knockout embryos are not responsive to folate but do respond to inositol (43). If TFAP2!-IRF6 -GRHL3 regulate a pathway in human neural tube, then targeted supplementation with inositol may furt her lower the inc idence of neural tube defects . Determining the distal node by which inositol exerts its function may provide important insight in designing and expanding the pool of patients who are eligible for targeted therapy. Alternatively, like folate, fortification in essential foods may be possible. Furthermore, inositol is used to treat patients w ith depression, suggesting safety or minimal side effects. Can inositol also be a therapy in human orofacia l clefting? In humans, we show that mutations in IRF6 and GRHL3 lead to Van der Woude Syndrome. In the mouse, Irf6 and Grhl3 regulate oral epithelium and palatal development (Chapter 2) . If inositol can rescue Grhl3 function in the oral cavity in a mann er analogous to the neural tube then supplementation may also be indicated to prevent orofacial clefting . However, p rior work examining inositol modulation of neural tube defects did not examine the palate and oral epithelium . As such, examinin g these embryos would provide critical inform ation for feasibility. Ideally, one would pursue a co-clinical trial, i.e. mouse and human, for the prevention orofacial and neural tube defects using inositol and folate relative to historical epidemiological data. While drug targets are not currently b eing explored, an epithelial specific factor that regulates proliferation and differentiation, like Ikka , Kdf1 , Ripk4 14-3-3&, Irf6, Grhl3 , !"$%!may provide a robust clinical application. To test many potential targets, we would need tissue. To test for specifi city, we would need both epithelial and non -epithelial cell types. To test for efficacy, we would need a cell type that is particularly affected. These three requirements are uniquely found in murine skin, which contains both epidermis (affected epithelium ) and dermis (unaffected adipocytes, mesenchyme, vasculature, etc). Putative targets could be tested in murine palate cultures, which would provide a robust ex vivo model for further screening. Highly attractive targets could then be tested in vivo using multiple murine alleles (gene trap, hypomorphic, mutated, etc) for the affected genes. Because clefting also occurs in canine and feline, larger animal models may also be tested. These approaches would require substantial funding and a pipeline for tissue processing. Another approach is gene therapy, that is replacing missing or defective gene in the cell type(s) that contribute(s) to disease. This would truly achieve the pinnacle of personalized therapeutic approaches and limit side effects. Toward that e nd, we have begun to perform in utero gene delivery of Irf6 using an adenoviral vector. In addition to a report showing gene deliver to the oral cavity and developing epidermis (44), we find that a [E1,E3 -]Ad -LacZ vector can transduce oral periderm and developing epidermis of both wildtype and Irf6 knockout embryos when injected at E12.5 (unpublished results). We also find that Ad -LacZ vectors can transduce cells along the medial edge of the palatal shelves. As such, our approach is feasible and w e predict that Irf6 gene delivery to the periderm of Irf6 compound heterozygous (Irf6 neo/ -) and knockout embryos (Irf6 -/-) (4) will reduce the severity of oral adhesion as seen with gen etic rescue. Targeting of the !"$&!developing epidermis may also ameliorate skin anomalies, leading to a reduction in proliferating cells and an increase in terminal differentiation (Chapter 4). Having characterized multiple murine models with varying amounts o f Irf6 expression, we can attempt to complement the endogenous deficit with exogenous vector dose. This approach would provide a unique test of translational feasibility because different Van der Woude families have different types of IRF6 mutations with d ramatically different phenotypic presentations. In that respect, finding the dose -disease ratio relative to mutation type would be extremely valuable. At this point, the feasibility of adenoviral vector transduction during neurulation is unknown . Finally, considering that replacing Irf6 expression in epithelium did not rescue limb, digit and skeletal defects , epithelial rescue will not be a panacea. !"$'! APPENDIX !"$(!APPENDIX Figure 28 : Summery of genetic network for orofacial and neural tube development. We find that three human orofacial clefting genes are involved in murine neu ral tube development . Data from h uman sequencing (green boxes: iCLP, isolated cleft lip and palate; VWS, Van der Woude Sy ndrome; SB, Spina Bifida ) and murine models (black boxes: NTD, Neural Tube Defects ) suggest a shared molecular network. Top : From orofacial and epidermal tissue, we find that Tfap2a regulates Irf6 , which in turn regulates Grhl3 . We also find that Irf6 negatively regulates Tfap2 ! transcriptionally . Middle: In rostral neurulation, Irf6 negatively regulates Tfap2 !. As this point, it is not clear if Grhl3 is downstream of Irf6 in rostral neural tube development . It is also not clear if this molecular netwo rk plays a role in anencephaly . Bottom: In caudal neurulation, we find that Tfap2 ! and Irf6 positively regulate each other and that Irf6 is required for Grhl3 expression. !"$)! Figure 29 : Proposed IRF6 gene regulator n etwork . A proposed gene regulatory network for Irf6 and associated molecules. Proposed interactions, molecular processes or function of related genes that are at this point minimally elucidated are highlighted (?). The balance of this gene regulatory network is ai med at stopping proliferation (repressing c-Myc) , initiating cell cycle arrest (induction of p21) and driving differentiation (Grhl3, which drives Tgm1) . While Smad2/3 is required for Irf6 expression in the medial edge epithelium, it is not yet clear which enhancer mediates this activation. Additional questions include: 1) M echanism by which Irf6 is targeted to the proteasome ; 2) E3 ubiquitin ligase for IRF6; 3) How and under what conditions Irf6 is phosphorylated (PO 4); 4) Function of Ripk4 , Kdf1, Grhl2 and other Tfap2 family members in this gene regulatory network; 5) Mechanism by which Irf6 translocates into the nucleus to me diate transcriptional activity. We propose that 1) Irf6 is bound to 14-3-3& when phosphorylated; 2) Irf6 stabilizes Tfap2 ! by inh ibiting proteasome -mediated degradation. !"%*! BIBLIOGRAPHY !"%+!BIBLIOGRAPHY 1 de Lima, R.L., Hoper, S.A., Ghassibe, M., Cooper, M.E., Rorick, N.K., Kondo, S., Katz, L., Marazita, M.L., Compton, J., Bale, S. et al. (2009) Prevalence and nonrandom distribution of exonic mutations in interferon regulatory factor 6 in 307 families with Van der Woude syndrome and 37 families with popliteal pterygium syndrome. Genet Med , 11, 241-247. 2 Koillinen, H., Wong, F.K., Rautio, J., Ollikainen, V., Karsten, A., Larson, O., Teh, B.T., Huggare, J., Lahermo, P., Larsson, C. et al. (2001) Mapping of the second locus for the Van der Woude syndrome to chromosome 1p34. Eur J Hum Genet , 9, 747 -752. 3 de la Garza, G., Schleiffarth, J.R., Dunnwald, M., Mankad, A., Weirather, J.L., Bonde, G., Butcher, S., Mansour, T.A., Kousa, Y.A., Fukazawa, C.F. et al. (2013) Interferon regulatory factor 6 promotes differentiation of the periderm by activating expression of grainyhead -like 3. J Invest Der matol , 133, 68-77. 4 Ingraham, C.R., Kinoshita, A., Kondo, S., Yang, B., Sajan, S., Trout, K.J., Malik, M.I., Dunnwald, M., Goudy, S.L., Lovett, M. et al. (2006) Abnormal skin, limb and craniofacial morphogenesis in mice deficient for interferon regulator y factor 6 (Irf6). Nat Genet , 38, 1335-1340. 5 Yu, Z., Bhandari, A., Mannik, J., Pham, T., Xu, X. and Andersen, B. (2008) Grainyhead -like factor Get1/Grhl3 regulates formation of the epidermal leading edge during eyelid closure. Dev Biol , 319, 56-67. 6 Boglev, Y., Wilanowski, T., Caddy, J., Parekh, V., Auden, A., Darido, C., Hislop, N.R., Cangkrama, M., Ting, S.B. and Jane, S.M. (2011) The unique and cooperative roles of the Grainy head -like transcription factors in epidermal development reflect unexpecte d target gene specificity. Dev Biol , 349, 512-522. 7 Auden, A., Caddy, J., Wilanowski, T., Ting, S.B., Cunningham, J.M. and Jane, S.M. (2006) Spatial and temporal expression of the Grainyhead -like transcription factor family during murine development. Gene Expr Patterns , 6, 964-970. 8 Garcia -Cao, I., Song, M.S., Hobbs, R.M., Laurent, G., Giorgi, C., de Boer, V.C., Anastasiou, D., Ito, K., Sasaki, A.T., Rameh, L. et al. (2012) Systemic elevation of PTEN induces a tumor -suppressive metabolic state. Cell , 149, 49-62. 9 Leslie, E.J., Mancuso, J.L., Schutte, B.C., Cooper, M.E., Durda, K.M., L'Heureux, J., Zucchero, T.M., Marazita, M.L. and Murray, J.C. (2013) Search for genetic modifiers of IRF6 and genotype -phenotype correlations in Van der Woude and poplit eal pterygium syndromes. Am J Med Genet A , 161, 2535-2544. !"%"! 10 Froster -Iskenius, U.G. (1990) Popliteal pterygium syndrome. J Med Genet , 27, 320-326. 11 McKeown, C.M. and Harris, R. (1988) An autosomal dominant multiple pterygium syndrome. J Med Genet , 25, 96-103. 12 Srivastava, R.N. (1968) Arthrogryposis multiplex congenita. Case report of two siblings. Clin Pediatr (Phila) , 7, 691-694. 13 Srivastava, R.N. (1969) Arthrogryposis or a new syndrome? J Pediatr , 74, 840 -841. 14 Clee, S.M. and Attie, A.D. (2007) The genetic landscape of type 2 diabetes in mice. Endocr Rev , 28, 48-83. 15 Rahimov, F., Marazita, M.L., Visel, A., Cooper, M.E., Hitchler, M.J., Rubini, M., Domann, F.E., Govil, M., Christensen, K., Bille, C. et al. (200 8) Disruption of an AP -2alpha binding site in an IRF6 enhancer is associated with cleft lip. Nat Genet , 40, 1341 -1347. 16 Wang, X., Pasolli, H.A., Williams, T. and Fuchs, E. (2008) AP -2 factors act in concert with Notch to orchestrate terminal differentia tion in skin epidermis. J Cell Biol , 183, 37-48. 17 Li, W. and Cornell, R.A. (2007) Redundant activities of Tfap2a and Tfap2c are required for neural crest induction and development of other non -neural ectoderm derivatives in zebrafish embryos. Dev Biol , 304, 338-354. 18 Popa, C., Dahler, A.L., Serewko -Auret, M.M., Wong, C.F., Smith, L., Barnes, L.M., Strutton, G.M. and Saunders, N.A. (2004) AP -2 transcription factor family member expression, activity, and regulation in human epidermal keratinocytes in vitro. Differentiati on, 72, 185-197. 19 Botti, E., Spallone, G., Moretti, F., Marinari, B., Pinetti, V., Galanti, S., De Meo, P.D., De Nicola, F., Ganci, F., Castrignano, T. et al. (2011) Developmental factor IRF6 exhibits tumor suppressor activity in squamous cell carcinoma s. Proc Natl Acad Sci U S A, 108, 13710-13715. 20 de la Garza, G., Schleiffarth, J.R., Dunnwald, M., Mankad, A., Weirather, J.L., Bonde, G., Butcher, S., Mansour, T.A., Kousa, Y.A., Fukazawa, C.F. et al. (2013) Interferon regulatory factor 6 promotes diff erentiation of the periderm by activating expression of grainyhead -like 3. J Invest Dermatol , 133, 68-77. !"%#!21 Guttormsen, J., Koster, M.I., Stevens, J.R., Roop, D.R., Williams, T. and Winger, Q.A. (2008) Disruption of epidermal specific gene expression and delayed skin development in AP -2 gamma mutant mice. Dev Biol , 317, 187-195. 22 Pyrgaki, C., Liu, A. and Niswander, L. (2011) Grainyhead -like 2 regulates neural tube closure and adhesion molecule expression during neural fold fusion. Dev Biol , 353, 38-49. 23 Richardson, R.J., Dixon, J., Malhotra, S., Hardman, M.J., Knowles, L., Boot -Handford, R.P., Shore, P., Whitmarsh, A. and Dixon, M.J. (2006) Irf6 is a key determinant of the keratinocyte proliferation -differentiation switch. Nat Genet , 38, 1329 -1334. 24 Takeda, K., Takeuchi, O., Tsujimura, T., Itami, S., Adachi, O., Kawai, T., Sanjo, H., Yoshikawa, K., Terada, N. and Akira, S. (1999) Limb and skin abnormalities in mice lacking IKKalpha. Science , 284, 313-316. 25 Holland, P., Willis, C., Kanaly, S., Gl accum, M., Warren, A., Charrier, K., Murison, J., Derry, J., Virca, G., Bird, T. et al. (2002) RIP4 is an ankyrin repeat -containing kinase essential for keratinocyte differentiation. Curr Biol , 12, 1424-1428. 26 Lee, S., Kong, Y. and Weatherbee, S.D. (201 3) Forward genetics identifies Kdf1/1810019J16Rik as an essential regulator of the proliferation -differentiation decision in epidermal progenitor cells. Dev Biol , 383, 201-213. 27 Thomason, H.A., Zhou, H., Kouwenhoven, E.N., Dotto, G.P., Restivo, G., Nguyen, B.C., Little, H., Dixon, M.J., van Bokhoven, H. and Dixon, J. (2010) Cooperation between the transcription factors p63 and IRF6 is essential to prevent cleft palate in mice. J C lin Invest , 120, 1561-1569. 28 Scibetta, A.G., Wong, P.P., Chan, K.V., Canosa, M. and Hurst, H.C. (2010) Dual association by TFAP2A during activation of the p21cip/CDKN1A promoter. Cell Cycle , 9, 4525-4532. 29 Koinuma, D., Tsutsumi, S., Kamimura, N., Tan iguchi, H., Miyazawa, K., Sunamura, M., Imamura, T., Miyazono, K. and Aburatani, H. (2009) Chromatin immunoprecipitation on microarray analysis of Smad2/3 binding sites reveals roles of ETS1 and TFAP2A in transforming growth factor beta signaling. Mol Cell Biol , 29, 172 -186. 30 Iwata, J., Suzuki, A., Pelikan, R.C., Ho, T.V., Sanchez -Lara, P.A., Urata, M., Dixon, M.J. and Chai, Y. (2013) Smad4 -Irf6 genetic interaction and TGFbeta -mediated IRF6 signaling cascade are crucial for palatal fusion in mice. Development , 140, 1220 -1230. !"%$!31 Descargues, P., Sil, A.K., Sano, Y., Korchynskyi, O., Han, G., Owens, P., Wang, X.J. and Karin, M. (2008) IKKalpha is a critical coregulator of a Smad4 -independent TGFbeta -Smad2/3 signaling pathway that controls keratinocyte diff erentiation. Proc Natl Acad Sci U S A , 105, 2487-2492. 32 Marinari, B., Moretti, F., Botti, E., Giustizieri, M.L., Descargues, P., Giunta, A., Stolfi, C., Ballaro, C., Papoutsaki, M., Alema, S. et al. (2008) The tumor suppressor activity of IKKalpha in st ratified epithelia is exerted in part via the TGF -beta antiproliferative pathway. Proc Natl Acad Sci U S A , 105, 17091-17096. 33 Nair, M., Teng, A., Bilanchone, V., Agrawal, A., Li, B. and Dai, X. (2006) Ovol1 regulates the growth arrest of embryonic epid ermal progenitor cells and represses c -myc transcription. J Cell Biol , 173, 253-264. 34 Grandori, C., Cowley, S.M., James, L.P. and Eisenman, R.N. (2000) The Myc/Max/Mad network and the transcriptional control of cell behavior. Annu Rev Cell Dev Biol , 16, 653-699. 35 Zervos, A.S., Gyuris, J. and Brent, R. (1993) Mxi1, a protein that specifically interacts with Max to bind Myc -Max recognition sites. Cell , 72, 223-232. 36 Marinari, B., Ballaro, C., Koster, M.I., Giustizieri, M.L., Moretti, F., Crosti, F., Papoutsaki, M., Karin, M., Alema, S., Chimenti, S. et al. (2009) IKKalpha is a p63 transcriptional target involved in the pathogenesis of ectodermal dysplasias. J Invest Dermatol , 129, 60-69. 37 Zhu, F., Xia, X., Liu, B., Shen, J., Hu, Y. and Person, M. (2007) IKKalpha shields 14-3-3sigma, a G(2)/M cell cycle checkpoint gene, from hypermethylation, preventing its silencing. Mol Cell , 27, 214-227. 38 Restivo, G., Nguyen, B.C., Dziunycz, P., Ristor celli, E., Ryan, R.J., Ozuysal, O.Y., Di Piazza, M., Radtke, F., Dixon, M.J., Hofbauer, G.F. et al. (2011) IRF6 is a mediator of Notch pro -differentiation and tumour suppressive function in keratinocytes. EMBO J . 39 Richardson, R.J., Dixon, J., Jiang, R. and Dixon, M.J. (2009) Integration of IRF6 and Jagged2 signalling is essential for controlling palatal adhesion and fusion competence. Hum Mol Genet , 18, 2632-2642. 40 Bailey, C.M., Khalkhali -Ellis, Z., Kondo, S., Margaryan, N.V., Seftor, R.E., Wheaton, W .W., Amir, S., Pins, M.R., Schutte, B.C. and Hendrix, M.J. (2005) Mammary serine protease inhibitor (Maspin) binds directly to interferon regulatory factor 6: identification of a novel serpin partnership. J Biol Chem , 280, 34210-34217. !"%%!41 Bailey, C.M., Abbott, D.E., Margaryan, N.V., Khalkhali -Ellis, Z. and Hendrix, M.J. (2008) Interferon regulatory factor 6 promotes cell cycle arrest and is regulated by the proteasome in a cell cycle -dependent manner. Mol Cell Biol , 28, 2235-2243. 42 Sil, A.K., Maeda, S., Sano, Y., Roop, D.R. and Karin, M. (2004) IkappaB kinase -alpha acts in the epidermis to control skeletal and craniofacial morphogenesis. Nature , 428, 660-664. 43 Ting, S.B., Wilanowski, T., Auden, A., Hall, M., Voss, A.K., Thomas, T., Par ekh, V., Cunningham, J.M. and Jane, S.M. (2003) Inositol - and folate -resistant neural tube defects in mice lacking the epithelial -specific factor Grhl -3. Nat Med , 9, 1513-1519. 44 Wu, C., Endo, M., Yang, B.H., Radecki, M.A., Davis, P.F., Zoltick, P.W., Sp ivak, R.M., Flake, A.W., Kirschner, R.E. and Nah, H.D. (2013) Intra -amniotic transient transduction of the periderm with a viral vector encoding TGFbeta3 prevents cleft palate in Tgfbeta3( -/-) mouse embryos. Mol Ther , 21, 8-17.