New molecular mechanisms controlling dental epithelial stem cell maintenance, growth and craniofacial morphogenesis

Transcription

1 University of Iowa Iowa Research Online Theses and Dissertations Spring 2016 New molecular mechanisms controlling dental epithelial stem cell maintenance, growth and craniofacial morphogenesis Zhao Sun University of Iowa Copyright 2016 Zhao Sun This dissertation is available at Iowa Research Online: Recommended Citation Sun, Zhao. "New molecular mechanisms controlling dental epithelial stem cell maintenance, growth and craniofacial morphogenesis." PhD (Doctor of Philosophy) thesis, University of Iowa, Follow this and additional works at: Part of the Cell Anatomy Commons, and the Cell Biology Commons

2 NEW MOLECULAR MECHANISMS CONTROLLING DENTAL EPITHELIAL STEM CELL MAINTENANCE, GROWTH AND CRANIOFACIAL MORPHOGENESIS. by Zhao Sun A thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree in Anatomy and Cell Biology in the Graduate College of The University of Iowa May 2016 Thesis Supervisor: Professor Brad A. Amendt

3 Copyright by ZHAO SUN 2016 All Rights Reserved

4 Graduate College The University of Iowa Iowa City, Iowa CERTIFICATE OF APPROVAL PH.D. THESIS This is to certify that the Ph.D. thesis of Zhao Sun has been approved by the Examining Committee for the thesis requirement for the Doctor of Philosophy degree in Anatomy and Cell Biology at the May 2016 graduation. Thesis Committee: Brad A. Amendt, Thesis supervisor John F. Engelhardt Andrew F. Russo Andrew C. Lidral Martine Dunnwald Hank Qi

5 To My Family, My Parents, Zhenghuang Sun and Shangjun Yin, My Wife, Hongjie Gu, and My Son, Joshua S. Sun For their all-enduring and selfless love, powerful support and lasting encouragement during my long journey for Ph.D. ii

6 ACKNOWLEDGMENTS First and foremost, I am sincerely grateful to my mentor and thesis advisor, Dr. Brad A. Amendt, who brought me into the craniofacial development field and guided all my research projects. Dr. Amendt is a dedicated and well-respected professor in our university who heads the Craniofacial Anomalies Center and leads the research in dental school. He is also a very patient and knowledgeable mentor in guiding lab members to conduct research. I am honored to have had a chance to work with him at Texas A&M University and The University of Iowa. Throughout all the years in his lab, I not only learned the basic theory and technology in the field, but also more importantly learned the essential elements of a good scientist, such as critical thinking, generating good ideas and trouble-shooting. So many thanks to you, Brad! Thank you for your trust, time investment, and sharing your experience and knowledge with me. I am also thankful to all my thesis committee members, Drs. Brad A. Amendt, John F. Engelhardt, Andrew F. Russo, Hank Qi, Andrew C. Lidral and Martine Dunnwald, for their time, selfless help and valuable advice, which put me on right track for research. I also want show to my gratitude to two excellent research institutes in the U.S. where I received my Ph.D training. I am grateful to the Institute of Biosciences and Technology at Texas A&M University for accepting me as a graduate student. I want to show my appreciation to Dr. Mingyao Liu for his high recommendation which was very helpful during my time applying to graduate school. Thanks to Drs. Mingyao Liu and James F. Martin for accepting me as a rotation student. After two years in Texas A&M, I moved with my mentor to the Department of Anatomy and Cell Biology (ACB) at The University of Iowa, where I finished most of experimental work. From south to north, iii

7 from an aggie to a hawkeye, I enjoy my new life in Iowa. Here, I wish to thank all the faculty and students in ACB department for providing a friendly environment for me to stay and study. Particularly I want to express my appreciation to department head Dr. John F. Engelhardt for his generous contribution to the mouse lines and research protocols. Thanks to Dr. Robert Cornell for his help as a program director and as a knowledgeable teacher. Thanks to Dr. Hank Qi and his lab members for their valuable comments and help in joint lab meeting, I benefited a lot from communication with you guys. Thanks to Julie Stark, for her hard and patient administrative work which makes graduate student life much easier in the department of ACB. Thanks to our collaborator, Dr. Lina Moreno-Uribe and her student Clarissa S.G. de Fontoura from dental school, for finding the human FoxO6 polymorphisms for us. Thanks to all the members of the weekly Craniofacial Interest Group (CFIG) meeting, which offers a platform and pipeline for craniofacial folks on and out of campus to communicate and collaborate. I also feel grateful and lucky for working with members of the Dr. Amendt s lab. They are like a second family to me, not just simply because I spent most of my time here but because of the fact that most of the time I am the only one of my family who stays in Iowa. Previous lab member, Dr. Jianbo Wang, Dr. Huojun Cao, Dr. Xiao Li, Dr. Shan Gao, Ms. Diana Gutierrez, Mr. Thad Sharp, Ms. Myriam Moreno, and Mr. Brandon Van Cleave, I cannot express how much I miss you and I deeply appreciate all the help and joy you brought to me. I would also like to thank the current lab members, Steve Eliason, Wenjie Yu and Mason Sweat, thank you all three for helping me conduct my experiments and revising our papers. Without you, it would have been much harder for me to finish everything alone. Thanks to all the new lab members, Ryan Ries, Maurisa Aimable and iv

8 Yan Yan, for their help. I wish you an enjoyable life in the new lab and hope you learn as much as I did. For my personal life, I need to say thanks to my many old and new friends met in China and in U.S. It is you who make my life memorable and full of laughter. I also want to show my gratitude to my beloved family. I feel lucky to have had my dear sister, Yuchan, to grow up with me. I am grateful to my dear wife Hongjie Gu for her infinite support, patience and love. I am forever thankful to my beloved parents and parents-inlaw, for bringing me and my wife into the world, raising us and educating us. Moreover, you extend your love from our generation to the third generation--my son, your grandson Joshua, so thank you all again for the babysitting. I am extremely thankful for having the best gift from god, Joshua, who is currently a 2 years old Iowan and growing everyday! v

9 ABSTRACT The regenerative tissues such as hair follicles, intestine and teeth have a particular microenvironment known as stem cell niche which houses stem cells and act as a signaling center to control stem cell fate. The precise and timely regulation of stem cell renewal and differentiation is essential for tissue formation, growth and homeostasis over the course of a lifetime. However, the molecular underpinning to control this regulation is poorly understood. To address this issue, we use the continuously growing mouse incisor as a model to study the gene regulatory network which controls dental epithelial stem cell (DESC) maintenance, growth and craniofacial morphogenesis. We found FoxO6, a transcription factor mainly expressed in the brain and craniofacial region, control DESC proliferation by regulating Hippo signaling. FoxO6 loss-of-function mice undergo increases in cell proliferation which finally leads to lengthening of the incisors, expansion of the face and skull and enlargement of the mandible and maxilla. We have screened three human FOXO6 single nucleotide polymorphisms which are associated with facial morphology ranging from retrognathism to prognathism. Our study also reveals that Sox2 and Lef-1, two markers for early craniofacial development, are regulated by Pitx2 to control DESC maintenance, differentiation and craniofacial development. Conditional Sox2 deletion in the oral and dental epithelia results in severe craniofacial defects, including ankyloglossia, cleft palate, arrested incisor development and abnormal molar development. The loss of Sox2 in DESCs leads to impaired stem cell proliferation, migration and subsequent dissolution of the tooth germ. On the other hand, conditional overexpression of Lef-1 in oral and dental epithelial vi

10 region increases DESC proliferation and creates a new labial cervical loop stem cell compartment in dental epithelial stem cell niche, which produces rapidly growing long tusk-like incisors. Interestingly, Lef-1 overexpression rescues the tooth arrest defects but not the ankyloglossia or cleft palate in Sox2 conditional deletion mice. Our data also reveal that mirna and histone remodeler are involved in regulating DESC proliferation and craniofacial morphogenesis. We describe a mir- 23a/b:Hmgn2:Pitx2 signaling pathway in regulating dental epithelial cell growth and differentiation. Pitx2 activates expression of amelogenin which is the major protein component for enamel deposition. This activation can be repressed by the chromatinassociated factor Hmgn2. mir-23a and mir-23b directly target Hmgn2, leading to the release of the Hmgn2 inhibition of Pitx2 transcriptional activity and thus enhance Amelogenin production. Phenotypically, ablation of Hmgn2 in mice results in an overgrowth of incisors with increased Amelogenin expression. The findings in this study increase our current understanding of the molecular regulation of dental epithelial stem cell fate. It not only highlights new gene regulatory network that controls dental stem cell maintenance, growth and craniofacial morphogenesis, but also sheds new light on developing novel stem cell therapy or gene therapy for tooth regeneration and dental diseases. vii

11 PUBLIC ABSTRACT The precise and timely regulation of stem cell renewal and differentiation is essential for tissue formation, growth and homeostasis. However, the molecular underpinnings to control this regulation is poorly understood. To address this issue, we use the continuously growing mouse incisor as a model to study the gene regulatory network which controls dental epithelial stem cell (DESC) fate and craniofacial morphogenesis. We found a transcription factor FoxO6 controls DESC proliferation by regulating Hippo signaling in craniofacial region. FoxO6 loss-of-function mice undergo increases in stem cell proliferation which finally leads to lengthening of the incisors and expansion of the face and skull. Our study also reveals that Sox2 and Lef-1 are regulated by Pitx2 to control DESC maintenance, differentiation and craniofacial development. Conditional Sox2 deletion in the dental epithelia leads to impaired stem cell proliferation, migration, and arrested incisor development. Conditional overexpression of Lef-1 in dental epithelia increases DESC proliferation and creates a new stem cell compartment in dental epithelial stem cell niche, leading to rapidly growing long tusk-like incisors. Interestingly Lef-1 overexpression rescues the tooth arrest defects in Sox2 conditional deletion mice. We also describe a mir-23a/b:hmgn2:pitx2 signaling pathway in regulating dental epithelial cell growth and differentiation. mir-23a /b directly target Hmgn2, a repressor for Pitx2 transcriptional activation of Amelogenin, leading to enhanced Amelogenin production. Our findings provide new molecular mechanisms for controlling dental stem cell fate and bases for developing novel stem cell therapy or gene therapy for tooth regeneration and dental diseases. viii

12 TABLE OF CONTENS LIST OF TABLES... xv LIST OF FIGURES... xvi LIST OF ABBREVIATIONS AND SYMBOLS... xix CHAPTER I INTRODUCTION... 1 Morphogenesis and tissue interactions during tooth development... 1 Dental stem cell niche and tooth regeneration capacity... 2 Molecular regulation of dental stem cell maintenance, renewal and differentiation... 4 Molecular regulation of facial growth... 6 Focus of this thesis... 7 CHAPTER II FOXO6 REGULATES HIPPO SIGNALING TO CONTROL FACIAL MORPHOLOGY Abstract Introduction Materials and methods Animals Histology and immunofluorescence assay Detection of β-galactosidase (LacZ) activity DNA cloning, shrna, cell culture, transient transfection, luciferase, betagalactosidase assay and western blotting Chromatin immunoprecipitation assay (ChIP) ix

13 BrdU labeling Microarray and quantitative real time PCR gene expression analysis MRI methods Imaging and microcomputed tomography (microct) Geomorphometric analyses and genotype-phenotype correlations in adults with dento-skeletal bite problems Statistical analysis Results FoxO6 is expressed in the brain, craniofacial tissues and somites FoxO6 -/- mouse heads are expanded anteriorly and feature an enlarged submandibular gland and a decrease in ossification Expansion in the FoxO6 -/- head follows a specific pattern FoxO6 activates Lats1 to promote Yap phosphorylation FoxO6 binds directly to the Lats1 promoter and activates Lats1 expression FoxO6 regulates odontogenesis FoxO6 modulates cell proliferation The position of the lower incisor in the jaw correlates with increased anterior growth of the mandible Human FOXO6 variants are associated with particular maxillo-mandibular horizontal discrepancies leading to dento-skeletal bite problems rs alters AP-1 binding and activation of FoxO Discussion x

14 CHAPTER III PITX2:SOX2:LEF-1 NETWORK REGULATES DENTAL STEM CELL MAINTENANCE AND TOOTH DEVELOPMENT Abstract Introduction Materials and methods Mouse lines and embryonic staging Immunohistochemistry, immunofluorescence and histology Chromatin immunoprecipitation assay (ChIP) BrdU labeling IdU/CldU labeling assay Incisor injury and recovery assay TUNEL assay Quantitative real time PCR gene expression analysis Imaging and microcomputed tomography (µct) GST pull-down assays Immunoprecipitation assay D reconstruction of the labial cervical loops Statistical analysis Results Specific ablation of Sox2 in the oral and dental epithelia Sox2 deletion in the oral epithelium causes ankyloglossia and cleft palate Inactivation of Sox2 leads to lower incisor arrest at E16.5 and abnormalities in upper incisor and molar development xi

15 Sox2 regulates incisor growth in adult mice Sox2 ablation in DESCs leads to reduced stem cell renewal Sox2 and Lef-1 epithelial expression domains are juxtaposed in the mouse oral epithelial dental placode Conditional overexpression of Lef-1 creates a new LaCL stem cell niche and abnormal tusk-like incisors Lef-1 overexpression rescues tooth arrest in Sox2 cko embryos Sox2 attenuates Pitx2 transcriptional activation of Lef-1, Sox2 and Pitx2 through direct protein interactions Discussion CHAPTER IV A MIR-23A/B:HMGN2:PITX2 SIGNALING PATHWAY REGULATES CRANIOFACIAL/INCISOR MORPHOGENESIS Abstract Introduction Materials and methods Mouse strain breeding Bimolecular fluorescence complementation (BiFC) assay Immunocytochemistry LacZ staining Histology, immunofluorescence and trichrome staining Expression and luciferase reporter constructs Cell culture, transfections and reporter assays Western blot assays xii

16 Real-time PCR assays Imaging and microcomputed tomography (Micro-CT) Chromatin immunoprecipitation assay (ChIP) Statistical analysis Results Hmgn2 interacts with Pitx2c in the nucleus and Hmgn2 represses Pitx2 transcriptional activity mir-23a and mir-23b repress Hmgn2 in dental epithelial-like cells mir-23a/b indirectly activates Pitx2 and Amelogenin expression by repressing Hmgn2 in dental epithelial cells Pitx2a represses mir-23a and mir-23b expression Hmgn2 -/- incisors undergo abnormal expansion with increased enamel formation Discussion CHAPTER V SUMMARY AND FUTURE DIRECTIONS FoxO6 Regulates Hippo Signaling to Control Facial Morphology Summary Future Directions To carry out functional analysis of 3 human FoxO6 SNPs that are associated with facial morphology ranging from retrognathism to prognathism Strengthen our conclusion that FoxO6 regulates Hippo signaling by generating FoxO6 overexpressing mice xiii

17 3. Targeting FoxO6 in human disease might be beneficial for tissue regeneration Pitx2:Sox2:Lef-1 network regulates dental stem cell maintenance and tooth development Summary Future Directions To explore the molecular mechanism of ectopic expression of amelogenin in lingual epithelium of the rescue mice To identify Sox2 target genes which mediate the cleft palate and ankyloglossia phenotypes in Sox2 cko mice Reprogramming epithelial or mesenchymal cells to dental epithelial stem cell (DESC) by inducing Sox2, Lef1 and Pitx A mir-23a/b:hmgn2:pitx2 signaling pathway regulates craniofacial/incisor morphogenesis Summary Future Directions To explore the genome wide targets of Hmgn2 in mouse embryonic development Potential gene therapy for enamel defects in Axenfeld-Rieger Syndrome (ARS) patients by inducing mir-23a/b and Pitx Working model of this thesis REFERENCES APPENDIX xiv

18 LIST OF TABLES Table II-1 Comparison of sizes of craniofacial structures between WT and FoxO6 -/- mice Table III-1 List of all the primers for genotyping Table III-2 Primer list for ChIP assay and Real-time PCR xv

19 LIST OF FIGURES Figure I-1 The developmental stages of early tooth morphogenesis and formation of different tooth types Figure I-2 Schematic of the mouse incisor and its epithelial stem cell niche Figure I-3 The epithelial transcription factor expressed during tooth development Figure I-4 Contribution of three germ layers during craniofacial development Figure II-1 FoxO6 expression during mouse embryonic development Figure II-2 The heads of FoxO6 -/- mice are larger than those of wild type mice Figure II-3 The heads of FoxO6 -/- mice are expanded in the anterior direction Figure II-4 Gene ontology (GO) analysis of biological process, indicating the cellular processes with FoxO Figure II-5 FoxO6 activates Lats1/2, Runx2, Shh and the Hippo signaling pathway Figure II-6 FoxO6 directly binds to and activates the Lats1 promoter Figure II-7 FoxO6 regulates Lats1 and pyap expression Figure II-8 In FoxO6 -/- mice, the incisors are enlarged, the LaCL is expanded and the polarity of the dental epithelium is abnormal Figure II-9 FoxO6 -/- mice have increased Amelogenin expression Figure II-10 FoxO6 -/- incisors exhibit the increased cell proliferation Figure II-11 FoxO6 -/- mouse incisors are positioned anteriorly and distal to the molars compared to WT mice Figure II-12 FOXO6 SNPs are associated with variation in human facial shapes Figure II-13 A polymorphism in 5.8kb upstream of FoxO6 transcription start site (TSS) alters the AP-1 binding in GMSM-K cells Figure II-14 rs polymorphism affects AP-1 activation Figure II-15 AP-1 binds to the rs derived sequence in HEPM cells xvi

20 Figure II-16 Model for FoxO6 regulation of Hippo signaling in controlling growth of the head and craniofacial structures Figure III-1 Sox2 expression in the dental epithelial stem cell (DESC) niche and Pitx2Cre/Sox2 F/F (Sox2 cko ) ablates Sox2 expression in the dental and oral epithelia Figure III-2 Sox2 cko embryos develop abnormal oral adhesions and present with ankyloglossia and cleft palate Figure III-3 Loss of Sox2 in murine embryos causes tooth arrest Figure III-4 Molar and upper incisor development is impaired in Sox2 cko embryos Figure III-5 Sox2 ablation with the K14 Cre causes abnormal molar formation and changes in Shh and Fgf4 expression Figure III-6 Deletion of Sox2 in adult mice inhibits incisor regeneration Figure III-7 Sox2 regulates dental epithelial stem cell renewal and cell migration Figure III-8 Loss of dental epithelial stem cells in Sox2 cko embryos does not involve increased apoptosis Figure III-9 Sox2 and Lef-1 expression domains are juxtaposed in the dental placode and oral epithelium Figure III-10 Conditional overexpression of Lef-1 (COEL) results in tusk-like incisors, creation of a new stem cell compartment in the LaCL and increased dental epithelial stem cell proliferation Figure III-11 Overexpression of Lef-1 in dental and oral epithelia creates a new LaCL stem cell compartment and increases amelogenin expression Figure III-12 Overexpression of Lef-1 rescues tooth arrest in Sox2 cko embryos Figure III-13 Lef-1 overexpression rescues lower incisor development in the Sox2 cko embryo Figure III-14 Endogenous Pitx2 and Sox2 bind to elements in the Sox2 promoter. 100 Figure III-15 Sox2 interacts with Pitx2 through the HMG domain Figure III-16 Model for the roles of Pitx2, Sox2 and Lef-1 in regulating incisor development, growth and renewal Figure IV-1 PITX2C interacts with Hmgn2 in the nucleus and Hmgn2 represses Pitx2 transcriptional activity xvii

21 Figure IV-2 mir-23a and mir-23b target Hmgn2 in dental epithelial-like cells Figure IV-3 mir-23a/b indirectly activate Pitx2 and Amelogenin expression by repressing Hmgn2 in dental epithelial cells Figure IV-4 PITX2A represses mir-23a and mir-23b expression Figure IV-5 Endogenous Pitx2 binds to the mir-23b flanking regions Figure IV-6 Hmgn2 expression was decreased in early mouse development and Hmgn2 -/- mice exhibit increased size of incisors Figure IV-7 Ameloblastin and E-cadherin expression were not affected in Hmgn2 -/- mice Figure IV-8 Hmgn2 -/- mice exhibit increased amelogenin expression and enamel deposition Figure IV-9 Dental epithelial cell proliferation is increased in Hmgn2 -/- lower incisors Figure V-1 New molecular mechanisms controlling dental epithelial stem cell (DESC) maintenance, growth and craniofacial morphogenesis xviii

22 LIST OF ABBREVIATIONS AND SYMBOLS 3D AI AM ANCOVA ANOVA ARS BiFC ChIP CL CNC cko CMV COE COEL DE DESC DI DL DP DSPP E ES GM GO HERS Hmgn IEE IF IHC LaCL LI LiCL Lnc RNA LV MD mbirn Three-dimensional Amelogenesis Imperfecta Ameloblast Analysis of Covariance Analysis of variance Axenfel-Rieger Syndrom Bimolecular fluorescence complementation assay Chromatin Immunoprecipitation assay Cervical Loop Cranial Neural Crest conditional Knock-Out Cytomegalovirus Conditional Overexpressing Conditional Over Expression of Lef-1 Dental Epithelium Dental Epithelial Stem Cell Distal Dental Lamina Dental Placode DentiSialophosphoprotein Embryonic Embryonic Stem Geomorphometric Gene Ontology Hertwig s Epithelial Root Sheath High-Mobility Group N Inner Enamel Epithelium Immunofluorescence Immunohistochemistry Labial Cervical Loop Lower Incisor Lingual Cervical Loop Long Non-Coding RNA Lateral Ventricle Mandible mouse Biomedical Informatics Research Network xix

23 microct mirna MRI MX NE NVB OD OE OEE P PBS PCA PR PSSA SI SPSS SSA SR TA TAC TF TN TSS TUNEL UI UTR WT microcomputed Tomography microrna Magnetic Resonance Imaging Maxilla Nasal Epithelium Neurovascular Bundle Odontoblast Oral Epithelium Outer Enamel Epithelium Postnatal Phosphate-Buffered Saline Principal Component Analysis Proximal Pre-Secretory Stage Ameloblasts Stratum Intermedium Statistical Package for Social Science Secretory Stage Ameloblasts Stellate Reticulum Transit-Amplifying Transit Amplifying Cells Transcription Factors Tongue Epithelium Transcription Start Site Terminal deoxynucleotidyl transferase dutp Nick End Labeling Upper Incisor Untranslated Region Wild Type xx

24 CHAPTER I INTRODUCTION Morphogenesis and tissue interactions during tooth development For past two decades, teeth have been widely used as a model to study developmental patterning, signaling transduction and tissue regeneration. As teeth are composed of two cell lineages------dental mesenchyme and epithelium, tooth development shares many developmental mechanisms with epithelial organs (e.g. hair and mammary glands) and neural crest derivatives (e.g. palate) (Jernvall and Thesleff, 2012; Soukup et al., 2008). The ectoderm derived dental epithelium gives rise to enamel secreting ameloblasts. The neural crest originated dental mesenchyme generates pulp cells and odontoblasts which produce dentin. The tooth organogenesis requires sequential and reciprocal epithelial-mesenchymal interactions, and this process has to be precisely regulated by signal pathways, signaling molecules and transcriptional factors (Thesleff and Tummers, 2008) The basic steps of tooth organogenesis have been characterized over 100 years ago and are shared by all vetebrates (Thesleff and Tummers, 2008) (Figure I-1). The first step of tooth formation is the establishment of a structure called dental lamina (mouse embryonic day 10.5), which is derived from ectoderm. The dental lamina is featured as a sheet of thickened oral epithelial cells grown on top of neural crest-derived dental mesenchyme. These oral epithelial cells proliferate and elongate to form the dental placode (mouse embryonic day 11.5), which consistes of thickened epithelium and the underlying dental mesenchyme. Next, the dental epithelium invaginates into condensed mesenchyme to form a tooth bud, called bud stage (mouse embryonic day 12.5). The epithelial cells of the tooth bud further grow and invaginate to encompass the condensed 1

25 mesenchymal dental papilla during cap stage (mouse embryonic day ). At cap stage, the epithelial cervical loops are visible on the edge of the epithelium, which offer a niche for dental epithelial stem cell to reside. From this time point, different developmental fates are made which leads to different types of teeth. In the continuously growing incisor, the dental epithelial stem cell niche is maintained through the lifetime of the mouse. However, in rodent molars or human teeth, the dental epithelial stem cell niche undergoes a structural modification to form a structure called Hertwig s epithelial root sheath (HERS), which guides the root formation but possesses a very limited capacity of continuous growth. Another feature of cap stage is the appearance of primary enamel knot, which serves a signaling center to regulate tooth morphogenesis and cusp pattern. The primary enamel knot is a transient structure and it disappears through cell death during the bell stage (E16.5-P0). In mouse molars and human teeth, the secondary enamel knot will take over and guide cusp formation. During bell stage (mouse embryonic day 16.5 and after), the cusping pattern is established, as the epithelium forms multiple folds in the molars, but not in the incisors (Thesleff and Nieminen, 1996; Thesleff and Tummers, 2008). Dental stem cell niche and tooth regeneration capacity The rodent incisors grow continuously throughout the life of the animal and have a fast turnover period of self-renewal, making it an excellent model to study dental stem cells and tooth regeneration (Zhao et al., 2014). The continuous growth of the incisor relies on stem cells in both epithelial and mesenchymal stem cell niches. The neurovascular bundle (NVB) provides a niche for dental mesenchymal stem cells to reside and generate pulp cells and odontoblasts (Thesleff and Tummers, 2008; Zhao et 2

26 al., 2014). The epithelial stem cells reside ina structure called cervical loop (CL), which is located in the posterior end of the incisor. In rodent incisor, the lingual cervical loop (LiCL, lingual side) and Labial cervical loop (LaCL, labial side) are asymmetric in size and possess different cell types. The LaCL is composed of inner enamel epithelium (IEE), outer enamel epithelium (OEE) and stellate reticulum (SR)(Figure I.2). SR is centrally located and composed of loosely arranged cells with mesenchymal appearance. The dental epithelial stem cells (DESCs) are believed to consist of SR and adjacent enamel epithelium at the tip of the LaCL (Harada et al., 1999) (Juuri et al., 2012a; Thesleff and Tummers, 2008). The DESCs in the LaCL proliferate and migrate down towards the anterior region to form transient amplifying cells in the IEE and differentiate into ameloblasts to form enamel. However, Lingual cervical loop (LiCL) is smaller in size compared to LaCL and lack SR cells leading to a lack of capacity to differentiate into ameloblast (Tummers et al., 2007). In fish and reptiles, continuous replacement of teeth occurs throughout life. Mammals differ from all other vertebrates and have limited capacity of tooth renewal. The majority of the mammals replace their teeth once, which is exemplified by human which have a set of milk teeth and a set of permanent teeth. Rodent incisors, the main animal model for studying mammalian tooth development, do not have tooth replacement. This could explain why little is known about the mechanisms of mammalian tooth replacement. However, the continuously growth feature of rodent incisors (e.g. mouse incisors) provides a valuable model to study stem cell regulation during tooth renewal (Cao et al., 2013a; Juuri et al., 2012a) 3

27 Moreover, the dental epithelial stem cell niche resembles that of other stem cell niches like the crypt of the intestine and the bulge of the hair follicle and the developmental mechanisms are similar among those ectodermal organ (Spradling et al., 2001). All those niches are surrounded by mesenchymal tissues, which further illustrate the importance of the reciprocal signals between epithelial and mesenchymal tissues. One of the advantages to study the dental epithelial stem cell niche is its highly tractability. Another advantage is that the niche can be disassociated and cultured in vitro (Thesleff and Tummers, 2008). All these features of dental epithelial stem cell niche make it become an excellent model to study the stem cell regulation of organ renewal. Molecular regulation of dental stem cell maintenance, renewal and differentiation The LaCL was first identified as a dental epithelial stem cell niche when label retaining cells were found to reside in the stellate reticulum (Harada et al., 1999). However, little was known about the cell fate of dental epithelial stem cells in LaCL due to lack of a marker to label those stem cells. More recently, Sox2 was reported as a marker for the dental epithelial stem cells (DESCs). Sox2 + cells contribute not only to ameloblast cells but also to other dental epithelial lineages (Juuri et al., 2012a). Another study shows Bmi1 is also expressed in the dental epithelial stem cell niche and deletion of Bmi1 in teeth leads to fewer stem cells, which finally results in defective enamel production (Biehs et al., 2013). Each of the studies demonstrate the DESCs are located at the tip of LaCL and give rise to the highly proliferative transit-amplifying (TA) cells in the inner enamel epithelium and migrate towards the distal region and differentiate into enamel-secreting ameloblasts. 4

28 Signaling molecules regulate dental stem cell fate at different levels. It has been reported that FGF signaling is involved in dental epithelial stem cell maintenance and proliferation (Harada et al., 2002, Wang et al. 2007; Klein et al.,2008). Seven FGF genes are expressed in the developing tooth (Fgf1,-2,-3,-4,-7,-8,-9) which regulate stem cell proliferation and teeth patterning (Thesleff and Sharpe, 1997). Ablation of Sprouty genes, which encode antagonists of FGF, in mice leads to bilateral enamel deposition and tusklike incisors, suggesting the inhibition of FGF signaling in the lingual side is crucial for blocking lingual dental epithelium from differentiating and secreting enamel (Klein et al., 2008a). BMP4 and Activin were also found to regulate DESCs proliferation and differentiation. Follistatin, a BMP inhibitor, limits the amount of stem cells in LiCL and contributes to the asymmetric deposition of enamel in rodent incisors by antagonizing the effect of Actvin. Overexpression of Follistatin in mice causes hypoplastic teeth with no enamel on either side. On the other hand, deletion of Follistatin leads to an overgrowth of LiCL which resemble LaCL (Wang et al., 2007b). Wnt signaling was also reported to play role in the homeostasis of DESCs. Lef-1 deficiency in mice results in arrested tooth morphogenesis at late bud stage (van Genderen et al., 1994b). Epithelial and mesechymal tissues recombination assays showed that Lef-1 is required only transiently in the dental epithelium and dental lamina (Kratochwil et al., 1996a). Shh was reported to be required for the generation of stem cell progeny but not for the survival of stem cell themselves, since Shh is mainly expressed in pre-ameoblasts not in stem cell niche. Transcription factors (TFs) are also crucial for tight genetic control of dental stem cell proliferation, maintenance and differentiation. Venugopalan et al. have summarized that many epithelial and mesenchymal TFs such as Pitx2, FoxJ1, Lef-1 and Msx2 have 5

29 hierarchical expression in tooth development (Figure I-3) (Venugopalan et al., 2011). Pitx2 is the first transcriptional marker of tooth development and the Pitx2 null mouse shows arrested tooth development at E12.5 (Liu et al., 2003b; Lu et al., 1999a). We have conditionally deleted Pitx2 with the K14 Cre and also demonstrated tooth arrest at the bud stage (unpublished data). FoxJ1 is also reported to play an important role in differentiation of DESCs (Venugopalan et al., 2011). mirnas have been reported to play critical roles in tooth development. Inactivation of Dicer1 in dental epithelium results in severe phenotypes such as loss of enamel, multiple teeth etc (Cao et al., 2010b). Chromatin-remodeling proteins are also important for the regulation of tooth development (Amen et al., 2008a). One example for such activity is Hmgn2, a chromatin-associated high-mobility group protein that binds to histones and affects Pitx2 transcriptional activity (Li et al., 2014). Molecular regulation of facial growth The vertebrate head/face develops from a complex and coordinated movement of epithelial cells originated from ectodermal and mesenchymal cells derived from neural crest and mesoderm (Chai and Maxson, 2006). During craniofacial development, the cranial neural crest (CNC) cells disperse from the dorsal side of the neural tube and migrate ventrally through epithelial-mesenchymal transition to give rise to different cell types in craniofacial region. Orofacial ectoderm, which gives rise to all the epithelial tissues in craniofacial region, interacts with CNC derived mesenchyme to control the position, size and shape of craniofacial organs. The continuous and reciprocal interaction between neural crest-derived mesenchyme and orofacial ectoderm is very common and 6

30 fundamentally required for craniofacial patterning and morphogenesis, such as tooth, palate and mandible (Figure I-4) (Miletich and Sharpe, 2004; Minoux and Rijli, 2010). The molecular mechanisms controlling vertebrate facial growth and head sizes are highly orchestrated by signaling and growth factors in a temporally and spatially specific manner. The Wnt, Fgf, Bmp, Shh, Tgf-β signaling pathways control early patterning and growth of the craniofacial skeleton through the regulation of mesoderm-derived and cranial neural crest-derived cell migration, proliferation, differentiation and transformations (Cox, 2004; Sasaki et al., 2005b). Those factors and pathways interact and intersect to control brain and skull development (Marcucio et al., 2011; Parsons et al., 2011). Furthermore, early signals especially Wnt signaling from the pharyngeal endoderm and epithelium, appears to pattern the developing skeleton (Reid et al., 2011). The orofacial ectoderm interacts with the pharyngeal endoderm and at later stages interacts with the cranial neural crest cells to establish epithelial patterning and palate and tooth development (Hu and Marcucio, 2009). All these early developmental cues drive the morphogenesis and patterning of perinatal craniofacial tissues, however other factors may regulate later craniofacial growth as the head continues to grow after birth. Focus of this thesis Although the locations of the stem cell niches has been well characterized in different regenerative tissues, the molecular regulatory mechanisms of precise and timely controlling stem cell renewal and differentiation are poorly understood. In the current study, we use the continuously growing mouse incisor as a model to study the gene regulatory network which controls dental epithelial stem cell (DESC) maintenance, growth and craniofacial morphogenesis. 7

31 In chapter II, I identified the transcriptional factor FoxO6 as an activator of Hippo signaling, which controls stem cell proliferation and mouse facial growth. These studies are significant for human as we identified (in collaboration with investigators in dental school) human genetic variants in FOXO6 associated with human facial forms and orthodontics disorders. This is the first study to provide evidence of a genetic association between SNPs in FOXO6 and facial shape variation. In chapter III, I determined that both Sox2 and Lef-1 are required for dental epithelial stem cell maintenance and proliferation. Furthermore, I established that Pitx2 regulates the expression of Sox2 and Lef-1 in tooth. These studies were performed using by using Sox2 conditional deletion mouse model, Lef-1 conditional overexpressing mouse model and cell based biochemistry assays. Chapter IV uncovered that mirna-23a/b positively regulate dental epithelial stem cell proliferation and incisor morphogenesis by targeting Hmgn2 (a chromatin remodeler). All work described above will be summarized and discussed in chapter V. In summary, I hope the finding of this study on current understanding of the molecular regulation dental epithelial stem cell fate not only benefits us to have a better understanding of the gene regulatory network that controls dental stem cell maintenance, growth and craniofacial morphogenesis, but also sheds new light on developing novel stem cell therapy or gene therapy for tooth regeneration and dental diseases. 8

32 Figure I-1 The developmental stages of early tooth morphogenesis and formation of different tooth types. The first step of tooth morphogenesis is the thickening of the oral epithelium to form dental lamina Next, the neural crest derived dental mesenchymal cells condense under the epithelial placode. Then the thickened epithelium buds into condensed mesenchyme to form a tooth bud. As tooth development proceeds to cap stage, the dental epithelium invaginates and encompasses the dental mesenchymal, and labial and lingual cervical loops are apparent at this stage. Different development choices are made at this point leading to formation of different tooth types. ERM, epithelial cell rests of Malassez; HERS, Hertwig s epithelial root sheath. This figure is adapted from (Thesleff and Tummers, 2008). 9

33 Figure I-2 Schematic of the mouse incisor and its epithelial stem cell niche. The mouse incisor is a long tooth growing under the molars. Ameloblasts only appeare in labial side and cause asymmetrical deposition of enamel on labial surface. Dentin(DE), produced by odontoblasts(od), is deposited on both labial and lingual side. The Labial cervical loop(lacl) contains inner and outer enamel epithelia(iee and OEE respectively) which surround the stellate reticulum(sr). The progeny of dental epithelial stem cells (DESCs) in the LaCL proliferate and migrate down towards anterior region to form transient amplifying(ta) cells and differentiate into ameloblasts that form enamel. DM: dental mesenchyme; En: enamel; LaCL: labial cervical loop; LiCL: lingual cervical loop; SI, stratum intermedium; TA: transient amplifying; Am: ameloblast. This picture was modified from (Juuri et al., 2012a). 10

34 Figure I-3 The epithelial transcription factor expressed during tooth development. Top: the timeline and hierarchical expression of dental epithelial transcription factors are shown. Bottom: tooth developmental stages and the transcription factors are shown for each stage. This picture was adapted and modified from (Venugopalan et al., 2011). 11

35 Figure I-4 Contribution of three germ layers during craniofacial development. A. Neural crest cells are formed at the junction of neural and surface ectoderm. B. Side view of E9.5 mouse embryos, red: Mesoderm-derived cells; yellow: Endoderm-derived cells; blue: CNC-derived cells. C. Transverse section of the developing first branchial arch covered by orofacial ectoderm. D. Schematic of an adult mouse skull shows both CNC- and paraxial mesoderm-derived elements. This picture is adapted from (Chai and Maxson, 2006). 12

36 CHAPTER II FOXO6 REGULATES HIPPO SIGNALING TO CONTROL FACIAL MORPHOLOGY Abstract In this report we demonstrate that tissue specific Hippo signaling regulates anterior growth of the face and is a crucial component in determining human facial characteristics. We have identified the transcriptional factor FoxO6 as an activator of Hippo signaling with specific expression in the brain and craniofacial tissues. FoxO6 loss-of-function mice undergo expansion of the face and skull, enlargement of the mandible and maxilla and lengthening of the incisors associated with increases in cell proliferation. FoxO6 activates Lats1 expression, thereby increasing Yap phosphorylation to control Hippo signaling. A phenotype-genotype correlation test identified significant associations (p<0.05) with three human FOXO6 single nucleotide polymorphisms in Caucasian adults with dento-skeletal bite problems ranging from retrognathism to prognathism of both jaws. Together, these results identify a FoxO6-Hippo regulatory pathway that controls skull growth, odontogenesis and facial morphology. These data suggest that human FOXO6 genetic variations explain differences in the human facial form. Introduction How organ size is determined is a long-standing puzzle in developmental biology. Several lines of evidence indicate that developing organs possess intrinsic mechanisms that modulate their final size (Heallen et al., 2011; Stanger, 2008). Genetic studies have established that the Hippo pathway plays a crucial role in organ size, controlling cell number by modulating cell proliferation and apoptosis (Huang et al., 2005; Wu et al., 13

37 2003). This pathway is triggered by the binding of extracellular ligands, which activate Mst1/2. Active Mst1/2 phosphorylates the Lats1/2 kinase, which is in turn activated, and subsequently phosphorylates and inactivates the transcriptional co-activator Yesassociated protein (YAP) 1 transcriptional co-activator, a major downstream effector of the mammalian Hippo pathway and TAZ, causing them to accumulate in the cytoplasm (Lei et al., 2008; Zhao et al., 2010). Therefore, upon activation of Lats1/2, expression of the target genes related to cell survival are inhibited due to the retention of YAP and TAZ in the cytoplasm. In contrast, the unphosphorylated (i.e. active) forms of YAP and TAZ associate with transcription factors (TFs) of the TEAD/TEF family in the nucleus activating the expression of target genes, and thereby promoting cell proliferation and inhibiting apoptosis (Figure A1) (Zhao et al., 2008). Thus, Hippo signaling represses cell proliferation and stimulates apoptosis. The role of Hippo signaling in craniofacial development is largely unknown. More importantly the TFs that regulate Hippo signaling components in craniofacial development are not well understood. FoxO6 is a TF that contains a Forkhead (winged helix) domain and is encoded by one of the FoxO class genes (Hannenhalli and Kaestner, 2009; Kaestner et al., 2000). Mammals have four FoxO members (FoxO1, FoxO3, FoxO4 and FoxO6). In humans single nucleotide polymorphisms in FOXO1 and FOXO3 have been associated with increased longevity and in invertebrates the expression of FoxO proteins can increase their life span (Kleindorp et al., 2011; Li et al., 2009). FoxO6 is the most recently identified FoxO-encoding gene and in mammals it appears to be expressed predominantly in the CNS (Hoekman et al., 2006; Jacobs et al., 2003; Salih et al., 2012). FoxO6 is highly expressed in the hippocampus (Salih et al., 2012), is negatively regulated by 14

38 insulin/igf signaling via the PI3K-Akt pathway, and is phosphorylated in contrast to other FoxO factors. However its nuclear localization is not affected by phosphorylation (Jacobs et al., 2003; van der Heide et al., 2005). Here, we identified FoxO6 in a bioinformatics analysis of transcription factor gene regulatory networks that are active during craniofacial development. Our subsequent analyses of FoxO6 -/- mice revealed enhanced craniofacial growth. This growth largely affected the anterior-posterior axis and depended on a Hippo pathway that controls the sizes of the brain, face, jaw and incisors during late stages of development. We also discovered the genetic associations between variants in and around human FOXO6 and maxillo-mandibular horizontal discrepancies leading to dento-skeletal bite problems in human adults. Taken together, the expression and function of FoxO6 specifically in craniofacial tissues such as the brain and frontonasal process, affects growth of the face and may define its morphology Materials and methods Animals All animals were housed at the University of Iowa, Program of Animal Resources and were handled in accordance with the principles and procedures specified in the Guide for the Care and Use of Laboratory Animals. All experimental procedures were approved by the University of Iowa IACUC guidelines. FoxO6 knockout mice were generated using a gene-trap strategy. In these mice 19 kb of FoxO6 genomic DNA (a stretch that contains the two FoxO6 exons) were replaced with a LacZ cassette and a neomycin cassette surrounded by loxp sites for future excision (Figure A2). Briefly, FoxO6 knockout embryonic stem (ES) cells (derived from C57BL/6 mice) were purchased from 15

39 Knockout Mouse Project (KOMP) (project number: VG12465) and injected into blastocysts (from BALB/cJ) by the Texas A&M University Institute for Genomic Medicine. In two chimeras generated from ES cell injection, the mutant allele was passed through the germ line, and these animals produced heterozygous progeny. Mice were maintained on a C57BL/6 background. Observation of a vaginal plug was counted as embryonic (E) day 0.5, and embryos were collected at E14.5, E16.5, E18.5, P0 and P1. Mice and embryos were genotyped based on PCR carried out on DNA extracted from tail biopsies (WT primers: sense: 5 ACCTCATCACCAAAGCCATC3, antisense: 5 GTCACCCTACCAGACCTCCA3 ; KO primers: sense: 5 CCTGCAGCCCCTAGATAACTT3, antisense: 5 GGTTGCTGGCTTCGTGTGGTG3 ). Histology and immunofluorescence assay Mouse embryos or heads were dissected in phosphate-buffered saline (PBS). Embryos were fixed with 4% paraformaldehyde-pbs solution for hours. Following fixation, samples were dehydrated through graded ethanol, embedded in paraffin wax and sectioned (7 µm). Standard Hematoxylin and Eosin was used to examine tissue morphology as previous described (Cao et al., 2010a). For immunofluorescence (IF) assays, slides were boiled in 10 mm sodium citrate solution (ph 6.0) for 20 minutes for antigen retrieval. They were then incubated with 20% goat serum-pbst for 30 min at room temperature, and then with antibodies against Ki67 (Abcam, 1:200), amelogenin (Santa Cruz, 1:200), Lats1 (Cell signaling, 1:200) and pyap (Cell signaling, 1:200) and Beta galactosidase (Abcam, ab9361, 1:50) at 4 C overnight. The slides were treated with 16

40 FITC (Alexa-488)- or Texas Red (Alexa-555)-conjugated Secondary antibody for 30 minutes at room temperature for detection (Invitrogen, 1:500). Nuclear counterstaining was performed using DAPI-containing mounting solution. Detection of β-galactosidase (LacZ) activity Mouse heads were stained for β-galactosidase activity according to standard procedures. Embryos (from E12.5-E16.5) or embrynic heads (from E17.5 to postnatal stage) were fixed for min at RT in 0.2% glutaraldehyde in PBS. Fixed embryos were washed three times (1M MgCl2, 0.5M NaH2PO4, 0.2% Nonidet P-40 and 0.01% sodium deoxycholate in PBS) and stained hours at 37 C using standard staining solution (5 mm potassiumferricyanide, 5 mm potassium ferrocyanide, 0.1% X-gal in wash buffer). On the next day, the samples were rinsed in PBS, photographed, and postfixed in 4% formaldehyde for hours. The fixed samples were dehydrated in a graded series of ethanol solutions, embedded in paraffin wax and sectioned. Sections were cut at 12 µm thickness and lightly counterstained with eosin. DNA cloning, shrna, cell culture, transient transfection, luciferase, betagalactosidase assay and western blotting FoxO6 expression plasmid was cloned into pcdna-myc 3.1 vector (Invitrogen) using the following primers: 5'- GCCTACATACCTCGCTCTGC -3' and 5'- ATCATAAGCTTGATTGGAGTTGGGTGGCTTA -3'. A1.7kb DNA fragment in which the FoxO6 binding site was incorporated upstream of the Lats1 gene was cloned after the luciferase gene in the ptk-luc vector (Promega) using the following primers: 5'- TCAGTGGATCCAGATCCCCTGAAGCTGGAGT -3' and 5'- TGACTGGATCCCAACATTGGGCACTGACATT -3'. The FoxO6 shrna targets the 17

41 5 -CTCCAATCTGGTTCTCAAATGACAC -3 sequence of FoxO6 mrna was cloned into psilencer 4.1 (Life Technologies). The 5 -CCTAAGGTTAAGTCGCCCTCG-3 sequence was cloned into psilencer 4.1 to generate a scrambled shrna control. CHO, LS-8 cells (oral epithelial-like) and ET-16 cells (dental epithelial) were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin and were transfected by electroporation. Cultured cells were fed 24 h prior to transfection, resuspended in PBS and mixed with 5 µg expression plasmids, 10 µg reporter plasmid and 0.5 µg SV-40 ß-galactosidase plasmid. Electroporation of CHO, LS- 8 cells and ET-16 cells was performed at 400 V and 750 microfarads (µf) (Gene Pulser XL, Bio-Rad). Transfected cells were incubated for h in 60 mm culture dishes and fed with 10% FBS and DMEM and then lysed and assayed for reporter activity and protein content by Bradford assay (Bio-Rad). Luciferase was measured using reagents from Promega. β-galactosidase was measured using the Galacto-Light Plus reagents (Tropix Inc.). All luciferase activities were normalized to β-galactosidase activity. For Western blot assay, cell lysates (10 µg) were separated on a 10% SDS polyacrylamide gel and the proteins were transferred to PVDF filters (Millipore), and immunoblotted using the following antibodies: FoxO6, (Abcam, 1:1000), Lats1 (Cell signaling, 1:1000), Yap (Cell signaling, 1:1000), pyap (Cell signaling, 1:1000), GAPDH (Santa Cruz, 1:1000) or β-tubulin (Santa Cruz, 1:2000). ECL reagents from GE HealthCare were used for detection. Chromatin immunoprecipitation assay (ChIP) The ChIP assays were performed as previously described using the ChIP Assay Kit (Upstate) with the following modifications. LS-8 cells were plated in 60 mm dishes and 18

42 fed 24 h prior to the experiment, harvested and plated in 60 mm dishes. Cells were crosslinked with 1% formaldehyde for 10 min at 37 C. Cross-linked cells were sonicated three times to shear the genomic DNA to DNA fragments average ranged between 200 and 1000 bp. DNA/protein complex were immunoprecipitated with FoxO6 antibody (Abcam, ab48730). DNAs from the precipitants were subject to PCR to evaluate the relative enrichment. The following primers were used to amplify the Lats1 promoter region, which contains a FoxO6 binding site: the sense primer (5 - TTCCCAGCAGGACTCTGTCT -3) and the antisense primer (5- CAAATGCCACTTTCTGGTGA -3). All PCR reactions were done under an annealing temperature of 60 C. All the PCR products were evaluated on a 2% agarose gel for appropriate size and confirmed by sequencing. For controls, we performd PCR with primers but without chromation; normal rabbit IgG was used replacing the specific antibody to reveal nonspecific immunoprecipitation of the chromatin. Three parallel Realtime PCRs were also performed in triplicates using these primers to quantify the enrichment of DNA pulled down by specific antibody over the DNA pulled down by IgG control. Primers located 4.9 kb upstream (Sense: 5- ATGGATCTCTCTGGCATTGG -3, antisense: 5-TCCTAGGCAGAGGCAGGTAA -3) of the Lats1 promoter region containing no FoxO6 binding sites were used as negative controls. BrdU labeling BrdU was injected into the pregnant mice (10 µl/g of body weight, Invitrogen, ) 2 h prior of harvesting E17.5 embryos. Embryos were embedded and sectioned as described previously (Cao et al., 2010a), and sections were mounted and rehydrated with sequential concentrations of alcohol, followed by immersion in 3% H2O2 to block the 19

43 endogenous peroxidase activity. Antigen retrieval was carried out by treating sections with 10 mm Sodium Citrate solution for 15 min at a slow boiling state. Sections were hydrolyzed for 30 min in 2 N HCl, neutralized for 10 min in 0.1 M sodium borate (ph 8.5), rinsed, blocked for 1 h in 10% goat serum, and immunostained with rat anti-brdu antibody (1:250, ab6326, Abcam). FoxO6 -/- mice and WT sections were placed on the same slide and processed together for identical time periods. Cartilage and bone staining For staining and visualization of whole skeletons, mice were dissected and skeletons were stained with alizarin red S and alcian blue 8G (Sigma), as previously described(bourgine et al., 2013). Microarray and quantitative real time PCR gene expression analysis Gene expression microarray analyses were performed by LC Sciences (Houston, TX) using GeneChip Mouse Genome Mouse Arrays. Total RNA was extracted from mandible and maxilla tissue of E18.5 WT and FoxO6 -/- embryos using RNeasy Mini Kit from Qiagen. Reverse transcription was performed according to the manufacturer's instruction (BIO-RAD iscript Select cdna Synthesis Kit) using oligo (dt) primers. cdnas were adjusted to equal levels by PCR amplification with primers to beta-actin (primers for beta-actin were 5'- GCCTTCCTTCTTGGGTATG-3' and 5'- ACCACCAGACAGCACTGTG-3'). Primers for FoxO6 q-pcr were: 5'- AAGAGCTCCCGACGGAAC -3' and 5'- GGGGTCTTGCCTGTCTTTC -3'. Primers for Lats1 q-pcr were: 5'- TAGAATGGGCATCTTTCCTGA-3' 5'- TGCTATCTTGCCGTGGGT-3'. Primers for Lats2 q-pcr were: 5'- GACGATGTTTCCAACTGTCGCTGTG-3 and 5' 20

44 CAACCAGCATCTCAAAGAGAATCACAC-3'. Primers for detecting Runx2, Shh, Amelogenin, CCD1 and Sox2 were previously described(cao et al., 2010a). All of the PCR products were sequenced to verify that the correct band was amplified. MRI methods Magnetic resonance imaging (MRI) was performed on a 4.7-T Varian small-bore scanner. All acquisitions utilized a 25-mm diameter transmit/receive coil for highresolution imaging. Mice were anesthetized with isoflurane (3% induction, 1.5% maintenance) and transferred to the scanner for imaging. After a series of three-localizer scans (each about 5 sec long), a set of T2-weighted fast spin-echo images was acquired in the axial plane. The protocol parameters were TR/TE. 2,100/60 msec, echo train length of eight, 0.5-mm thick contiguous slices with in-plane resolution of 0.16 mm X 0.16 mm over a 256 X 256 matrix using 12 signal averages. The total time for the entire protocol was about 40 min. All MRI data were processed using BRAINS software developed locally at the University of Iowa. The mouse brain atlas used for segmentation purposes was the mouse Biomedical Informatics Research Network (mbirn) atlas, which was constructed using T2- weighted magnetic resonance microscopy (MRM) from 11WTC57BL/6J mice at the University of California, Los Angeles. For our pipeline process we employed a directed acyclic pipeline architecture using Nipype, a Pythonbased wrapping library for neuroimaging applications. The mbrin atlas was registered to the input T1 file using bspline warping within BRAINSFit, a mutual information driven application developed under the ITK framework. The atlas was then resampled using BRAINSResample to match the voxel lattice of the T1 image, thereby allowing one-to-one correspondence between atlas and image. Finally, we 21

45 computed the volume measurements for each desired region of our atlas as the sum of voxels within a given label times the volume of a voxel. The atlas defined 43 regions of interest. The regions were then grouped into the following areas: amygdala, hypothalamus, pituitary, thalamus, total brain volume, basal ganglia, brainstem, cerebrospinal fluid(csf), cerebellum, hippocampus, white matter tracts, anterior cortex, and posterior cortex. The anterior cortex was further subdivided into the olfactory and frontal cortices. The posterior cortex was similarly subdivided into the posterior, entorhinal, and perirhinal cortices. Volumes were reported in mm 3. All analyses were performed using Statistical Package for Social Science (SPSS), version 19.0 for Windows (SPSS, Inc., Chicago, IL). Due to the small sample size, non-parametric analysis using the rank of all measures was utilized in order to minimize any effects of outliers. Analysis of variance (ANOVA) was used to evaluate total brain volume across groups. The remainder of the brain regions was compared across the two groups using Analysis of Covariance(ANCOVA), controlling for total brain volume. Imaging and microcomputed tomography (microct) To analyze variation in gross craniomandibular dimensions, we imaged WT (N=4) and FoxO6 -/- (N=4) mice using a Siemens Inveon Micro-CT/PET scanner. Skulls were scanned at 60kVp and 500 ma with a voxel size of 30 µm and reconstructed images were imported into Osirx DICOM imaging software for morphometric analysis. Using twoand three- dimensional renderings we collected a series of linear anterior-posterior and transverse skeletal and dental measurements defined in Table 1. Hemi-mandibles of littermate WT and mutant specimens were dissected, fixed in 4% PFA overnight at 4 C, stored in 70% ethanol and analyzed in ethanol using a MicroCT 22

46 40 (Scanco Medical, Brüttisellen, Switzerland) at 70 kv, 114 ua, 8 W, and 10 um resolution, with an integration time of 300 ms. Image stacks in dicom format were processed in Fiji ( An arbitrary but identical threshold was applied to all image stacks to remove the background signal. Image stacks were processed and anatomical landmarks were used to obtain the same spatial orientation of all specimens. This allows us to obtain two dimensional renderings in different specimens at the same anatomical position and plane. Two and three dimensional renderings were collected using identical imaging parameters to allow for comparison and the interpretation of grey values as mineral density. Geomorphometric analyses and genotype-phenotype correlations in adults with dento-skeletal bite problems A sample of 271 healthy Caucasian adults with dento-skeletal bite problems who were seeking orthodontic treatment at the University of Iowa and Private Practices within Eastern Iowa participated in this study. The study protocol was reviewed and approved by the institutional review board at the University of Iowa. All participants signed approved consent forms. Exclusion criteria included: previous orthodontic treatment, rd history of facial trauma or facial surgery, missing or impacted teeth other than 3 molars, and radiographs with poor quality or missing landmarks. Pre-treatment two-dimensional lateral cephalometric radiographs were utilized for phenotyping. Images were digitized with 29 landmarks distributed throughout cranial base, maxilla, mandible, upper and lower incisors, upper and lower 1 st permanent molars and soft tissues including the forehead, upper and lower lip, nose and chin. The X and Y coordinates were exported from Dolphin software for Geomorphometric (GM) analyses in the Morpho J. software 23

47 package. GM procedures included a Generalized Procrustes superimposition step for rotation, scaling and translation of the data to remove any variation not due to shape. Subsequently, Procrustes residuals were submitted to a principal components analysis (PCA), to identify the most important aspects of dento-facial variation in the data. Genomic DNA was extracted from saliva samples collected with Oragene kits (DNA Genotek, Ontario, CA, USA). For SNP selection haplotype blocks were reconstructed for human FOXO6, using genotypic data from a Caucasian population (HapMap CEU) available in HapMap (HapMap release# 27 based on February 2009 assembly, phase 2 and 3 versions). Three tagging SNPs across FOXO6 were selected and genotyped as part of a much larger study using a Dynamic Array via competitive allele-specific PCR KASPar chemistry (KBioscience Ltd., Hoddesdon, UK) on a Fluidigm (Fluidigm Corp., South San Francisco, CA) nanofluidic platform. Genotyping quality was assessed by visualization of genotyping plots and Hardy-Weinberg (HW) equilibrium tests. SNPs with poor quality plots or in HW disequilibrium p< 10-4 were excluded. SNPs were coded as 0, 1 or 2 copies of rare alleles present in an individual s SNP genotype. Multivariate linear regressions adjusting for age, gender, ceph source (analog film, digital and CBCT) were performed using Stata, to test for correlations between each SNP (one at a time) and the selected PCs. Statistical analysis For each condition, three experiments were performed and the results are presented as the mean ± SEM. The differences between two groups of conditions were analyzed using an independent, two-tailed t-test. 24

48 Results FoxO6 is expressed in the brain, craniofacial tissues and somites We initially identified FoxO6 expression in our bioinformatics analyses of mouse mandible tissue (Figure II-1A). Total RNA was extracted from P0 wild type (WT) mandibles and DNA microarrays identified FoxO6 expression. In light of this discovery, we generated FoxO6-lacZ knock-in mice (FoxO6 +/- ) and performed X-gal staining on staged embryos (Figure II-1B-F) Analysis of whole-mount embryos revealed the following: at E10.5, FoxO6 was not expressed at detectable levels (Figure II-1B); at E12.5 it was expressed specifically within the brain in the cerebral cortex, cerebellar primordium and trigeminal ganglion, but also in somites and the craniofacial region (Figure II-1C); at E14.5, expression was present in the brain and somites, as well as in the posterior regions of the maxilla and mandible (Figure II-1D); at E16.5 expression was restricted to the head structures and brain, with strongest expression in brain but widespread lower-level expression throughout the craniofacial region (Figure II-1E, Appendix C). At E18.5, expression was continuously high in the brain, and increased in the craniofacial region, including the maxilla, mandible, incisor, molar and palate (Figure II-1F). FoxO6 -/- mouse heads are expanded anteriorly and feature an enlarged submandibular gland and a decrease in ossification The FoxO6 homozygous knockout mice (FoxO6 -/- ) are viable and, at 2 months of age, have a significantly large maxilla and mandible and overall increase in anterior growth of the craniofacial skeleton relative to the WT mouse (FoxO6 +/+ ), (Figure II-2A). The anterior lengthening resulted in a head with a shape resembling that of the rat, leading us 25

49 to assay for FoxO6 expression in P0 rat maxilla/mandible tissue. Real-time PCR revealed that FoxO6 expression in the rat at this time is 80% lower than that in WT mice (Figure II-2B). This difference may be the genetic basis of the difference in head size between these animals. Notably, the submandibular gland was also larger in the FoxO6 -/- mouse (Figure II-2C). To determine if formation of the cranial bones was affected in the FoxO6 -/- mice, we stained embryos with Alcian Blue/Alizarin Red. This staining revealed that in E18.5 FoxO6 -/- embryos the skulls were larger than those of their heterozygous counterparts. Moreover, ossification (red stain) of the occipital bone (OB), temporal bone (TB) and nasal bone (NB) appeared slightly decreased in such embryos (Figure II-2D). At P1, the frontal, parietal and occipital bones were slightly larger in the FoxO6 -/- mice, due to an increase in volume of the skull vault (Figure II-2E). Taken together, these bone staining data indicate that osteogenesis, in particular endochondral ossification, was modestly decreased in the FoxO6 -/- mice allowing for extended growth and a larger skull. Expansion in the FoxO6 -/- head follows a specific pattern Microcomputed tomography (µct) analysis of whole heads of 2 month-old FoxO6 -/- and WT mice identified specific changes in growth patterns in the context of loss of FoxO6 function (Figure II-3). Lengths of the nasal was approximately 6% greater in the mutant animals, N=4 (Figure II-3B). Mandibular length was larger by 8.98, N=4 (Figure II-3B). The lengths of the skull overall was 4.87% greater (N=4) (Figure II-3B). A midsagittal section measuring showed a 6.5% increase of the cranial base length in the FoxO6 -/- mice, N=4, but the cranial base angle showed no change between the FoxO6 -/- and WT mice, (Figure II-3C). Overall the base length of the cranium was ~7% larger, 26

50 with increased lengths in the palate, anterior cranial base and lower incisor (Table 2-1). Magnetic resonance imaging (MRI) analyses of the FoxO6 -/- mice and subsequent volumetric measurement indicated that specific areas of the brain and craniofacial regions differed with respect to growth effects. The pituitary, perirhinal cortex, corpus callosum, fourth ventricle, interpeduncular nucleus, caudoputamen, substantia nigra, cerebral peduncle, spinal cord, midbrain and entorhinal cortex, were all larger in the FoxO6 -/- mice +/- than their FoxO6 littermates (Figure A3). In contrast, the frontal cortex, cerebral cortex, cerebellar cortex, olfactory system, third ventricle, thalamus and pons were all smaller in FoxO6 -/- mice compared to controls (Figure A3). These measurements are consistent with findings from an earlier study on the consequences of reduced hippocampal FoxO6 function for memory and synaptic function. Thus, there is an overall total reduction in brain tissue, with a proportional increase in white matter mainly in the brain stem and a proportional decrease in cortical gray matter (cerebral and cerebellar cortex). The increases in anterior growth became apparent immediately after birth and continued through 2 months of age. Furthermore, FoxO6 is expressed at later embryonic stages and does not appear to affect cell/tissue-specific patterning. These data suggest that FoxO6 functions after neural-crest migration has taken place and tissue identity, including that of epithelial tissues. FoxO6 activates Lats1 to promote Yap phosphorylation In order to find genes downstream of FoxO6 that mediate craniofacial development, we assessed gene expression using arrays generated from the mandibular and maxillary tissue from E18.5 WT and FoxO6 -/- mouse embryos. Subsequent gene ontology (GO) analysis of biological processes indicated that FoxO6 plays a role in regulating 27

51 transcription and development of the central nervous system (Figure II-4). Expression of Lats1, an important component of the Hippo signaling pathway, was ~20-fold lower in FoxO6 -/- mice compared to WT (data not shown). To confirm these array data and characterize other genes involved in regulating cell proliferation and differentiation, we performed real-time PCR using the same tissues (Figure II-5A). This analysis revealed reduced expression of the following genes: Lats1 and Lats2 (~80% and 70% reduced, respectively), Runx2 (85% reduced), and Shh (80% reduced). The observed change in expression in Runx2 is consistent with the roles of the encoded transcription factor in osteoblast differentiation and skeletal morphogenesis, as well as with the decrease in ossification observed at E18.5 in FoxO6 -/- mice (Figure II-2E). Likewise the difference in Shh expression is consistent with the role of its product in regulating the size and shape of the face. Genes whose products were expressed at higher levels were amelogenin and cyclin D1. The former is a marker for the differentiation of ameloblasts and required for the formation of dental enamel, and the latter is a positive regulator of cell proliferation. We next used an in vitro assay to test whether FoxO6 regulates Hippo signaling through Lats1. To this end, we over-expressed FoxO6 in CHO and LS-8 cells (oral epithelial-like) by transient transfection. This led to an increase in Lats1 expression in both cell lines (Figure II-5B, C), and the specificity of the effect was supported by the responsiveness to FoxO6 dosage in the CHO cells (Figure II-5B). Notably, the level of Yap phosphorylation was also higher in both of the transfected cell lines (Figure IIB,C) even though overall levels of Yap expression did not differ (Figure II-5B,C). Subsequent analysis of the effects of FoxO6 knockdown on Lats1 and Lats2 in ET-16 cells (oral 28

52 epithelial cells in which FoxO6 is highly expressed) by real-time PCR revealed that a 70% reduction in FoxO6 was accompanied by a 60% decrease in Lats1 (Figure II-5D). These data suggest that FoxO6 controls Hippo signaling by regulating Lats1 expression. FoxO6 binds directly to the Lats1 promoter and activates Lats1 expression Sequence analysis of the Lats1 5 flanking region identified a consensus FoxO6 binding site approximately 2,471 bp upstream of the Lats1 transcription start site (Figure II-6A). Chromatin immunprecipitation (ChIP) assays demonstrated that endogenous FoxO6 binds to this consensus binding site (Figure II-6B, lane 2). The chromatin input is shown in lane 1, and the control IgG, which failed to pull down the chromatin, in lane 3 (Figure II-6B). As a control the ChIP assay was carried out using primers to an upstream region of the Lats1 promoter that does not contain a FoxO6 binding element. The FoxO6 antibody did not immunoprecipitate chromatin lacking a FoxO6 binding site (Figure II- 6B, lane 7); chromatin input and IgG control are also shown (Figure II-6B, lane 6 and 8, respectively). Quantitative PCR demonstrated an 8-fold enrichment of the FoxO6 ChIP product over that of the IgG control (Figure II-6C). The Lats1 promoter was cloned (1.7Kb) upstream of the luciferase gene to measure promoter activity. Cotransfection of the Lats1 promoter with FoxO6 resulted in 15-fold activation (Figure II-6D; comparison is to transfection with empty vector transfection). Mutation of the FoxO6 binding site abolished this activation (Figure II-6E). These results indicate that FoxO6 directly activates the Lats1 promoter. FoxO6 regulates odontogenesis We used the incisor as a model to determine the role of FoxO6 in craniofacial development, since mechanisms of dental development and overall craniofacial 29

53 development and morphology are intimately related. The mouse incisor is a useful model, as it grows continuously throughout the life-time of the animal, relying on a stem-cell niche in the labial cervical loop (LaCL). Initially, during development, the teeth grow with the mandible and maxilla. However, after eruption and as the incisors are worn down, the tooth structure is regenerated by stem-cell proliferation and differentiation in the posterior region of the tooth resulting in continuously erupting incisors. Figure II-7A depicts the rodent lower incisor, including the structures that give rise to the enamel forming ameloblast, specifically the LaCL, outer enamel epithelium, and inner enamel epithelium. The dental mesenchyme also contains stem cells giving rise to odontoblasts to produce dentin. Analyses of gene expression and epithelial cell proliferation using the X-gal antibody at E16.5, E18.5, and P0 showed high FoxO6 expression at E18.5 in the dental epithelium of both incisors and molars (Figure II-7B-E), the oral epithelium, and craniofacial mesenchyme (Figure II-7F, G) in FoxO6 -/- mice compared to WT littermates. This is consistent with a role for FoxO6 in both odontogenesis and development of the mandible and maxilla. Notably, FoxO6 was detected in transit amplifying cells (TACs) of the dental mesenchyme as well (Figure II-7C, E). Given that our analysis of gene expression arrays had shown Hippo signaling in FoxO6 -/- mice to be defective, and that our bioinformatic analyses demonstrated that Lats1, Yap and pyap were expressed during incisor development (data not shown), we sought to confirm the dental effects of defective Hippo signaling in FoxO6 -/- mice. Immunofluorescence assays demonstrated that Lats1 is normally expressed in the lower incisor of the P1 mouse, and that Yap is phosphorylated at this time. In WT, Lats1 expression is predominantly detected in the LaCL region and to some extent in the dental 30

54 mesenchyme (Figure II-7H,I). As expected, in the FoxO6 -/- P1 mouse this expression was lower (Figure II-7J,K). Examination of pyap staining in sections of WT LIs revealed that Yap was activated in the LaCL, differentiating ameloblasts and mesenchymal cells (Figure II-7L,M). In the FoxO6 -/- P1 mice, this expression was lower (Figure II-7N,O). Together with our ChIP analysis, these experiments suggest that FoxO6 directly activates Lats1 and Yap in vivo to regulate Hippo signaling during incisor development. Consistent with a role for FoxO6 in regulating Hippo signaling, we observed an increase in the size of the lower incisors of E16.5 and E18.5 as well as of the LaCL, in FoxO6 -/- embryos (Figure II-8A-H). These differences persisted in P0 FoxO6 -/- mice (Figure II-8I-L), at which point structural defects in ameloblasts and odontoblasts were also noted in the FoxO6 -/- mice, with neither cell layer as uniform as in WT mice (Figure II-8M,N). Using immunofluorescence, we compared the differentiation of ameloblasts and amelogenin expresson analysis between WT and FoxO6 -/- incisors and show greater overall expression of amelogenin. This is due to an increase in length of the forming tooth structure and a corresponding increase in the number of differentiated ameloblasts -/- (Figure II-9). Thus, whereas the ameloblast layer is less structured in FoxO6 mice, the cells differentiate as ameloblasts. This suggests a FoxO6-mediated, Hippo-dependent regulation of incisor size in coordination with craniofacial growth may be due to the control of LaCL stem-cell proliferation. FoxO6 modulates cell proliferation Hippo signaling is known to regulate organ size by modulating cell proliferation. To -/- determine if the observed increase in incisor length in FoxO6 mice was due to an increase in cell proliferation, E18.5 FoxO6 -/- and WT embryos were sectioned and 31

55 proliferation was assessed using the Ki67 antibody. This analysis revealed an increase in Ki67-positive epithelial cells in mutant vs. WT incisors (Figure II-10A,B), specifically within the LaCL (Figure II-10C,D). Quantitative analysis revealed that this increase applied to the inner enamel epithelium and mesenchyme (Figure II-10E), indicating that dental-cell proliferation is up-regulated in FoxO6 -/- incisors. BrdU labeling in E17.5 -/- FoxO6 incisors confirmed this increase in cell proliferation (Figure II-10F-I). Quantitation of the BrdU-positive cells demonstrated that the proliferation of both epithelial and mesenchyme cells was increased (Figure II-10J). The activation of Hippo signaling inhibits cell proliferation, and we have shown that FoxO6 activates Lats1 expression and increases the phosphorylation of YAP. This modification is known to cause Yap to be sequestered in the cytoplasm and to thereby down-regulate the expression of genes required for both cell proliferation and anti-apoptotic activity. Thus, in the absence of FoxO6 the Hippo pathway is not stimulated, and this leads to specific tissue responses involving cell proliferation. The position of the lower incisor in the jaw correlates with increased anterior growth of the mandible Micro-CT analyses showed increased length, delayed mineralization and larger appearance in the parasaggital plane of the incisors of 1 month-old FoxO6 -/- mice (Table 2-1, Figure II-11A), although the molars of 1 month-old FoxO6 -/- mice were normal (Figure II-11B,C). The secretory stage of enamel development is extended in the FoxO6 -/- mice, which results in a later onset of the maturation stage, where mineral density increases drastically(figure II-11B,C). Accordingly, a comparison of mineral density at the plane through the distal root of the first molar, that is an anatomical position 32

56 corresponding to early maturation stage in WT, shows the delay of enamel maturation in the FoxO6 -/- incisor (Figure II-11D, E). The extended secretory stage results in an apparent displacement of the lower incisor in the anterior direction due to the overall increase in mandible length and the absence of mandibular incisor mineralization at the plane of the distal root of the first molar (Figure II-11F, G). Thus, the lower incisor is longer in the FoxO6 -/- mice, and positioned further anterior relative to the molars, reflecting the increase in anterior growth of the jaw. Human FOXO6 variants are associated with particular maxillo-mandibular horizontal discrepancies leading to dento-skeletal bite problems As part of a larger project aimed at the identification of candidate genes associated with human dento-skeletal bite problems (led by co-author L.M.U), 3 single nucleotide polymorphisms (SNPs) in and around human FOXO6 were genotyped in a cohort of 271 Caucasian adults with moderate or severe bite problems. Subjects dento-facial phenotypes were derived from lateral cephalometric radiographs analyzed via Geometric Morphometric (GM) methods. A principal component analysis of the dento-facial variation, identified principal components (PCs 1-4) explaining 60% of the total variation in the sample. The four PCs were correlated one at a time with the genotypes of the 3 FOXO6 SNPs via multivariate linear regression assuming an additive genetic model. Results showed that the third (PC3) and fourth (PC4) components which independently explained 11% and 5.6 % of the total dento-facial variation were significantly (p<0.05) associated with SNPs in FOXO6 (Figure II-12). PC3, captured variation ranging from a bi-maxillary retrusion and steep anterior cranial base on one end (-β) to bi-maxillary protrusion and a flat anterior cranial base at the opposite end (+β). PC3 association results 33

57 indicate that more copies of the rare allele for the SNP rs in an individual s genotype (p-value = 0.007; β = ) were associated with a tendency towards bimaxillary protrusion (+β), while more copies of the rare allele for the SNP rs (pvalue = 0.022; β = ) were associated with bi-maxillary retrusion (-β) (Figure II- 12A, C). PC4 on the other hand captured variation ranging from a long mandibular body with an anterior incisor cross bite relation (-β) to a shorter mandibular body and adequate incisor relation (+β). PC4 association results indicate that more copies of the rare allele for the SNP rs (p-value = 0.015; β = ) are associated with a larger mandibular body length and an anterior cross bite relation (Figure II-12B, C). rs alters AP-1 binding and activation of FoxO6 To understand the functional effects of these polymorphisms in regulating FoxO6 expression, we screened for a potential conserved transcription factor binding site using consite browser ( The rs polymorphism is located nearly 5.8Kb upstream of the FoxO6 transcription start site (TSS). Interestingly, the rs polymorphism occurs in a predicted cfos/cjun complex(or AP-1) binding site(figure II-13 A and B). To test if the cfos/cjun complex interacts with the FoxO6 promoter through this putative binding site, we performed chromatin immunoprecipitation(chip) assay in GMSM-K cells(human oral epithelial cells) (Gilchrist et al., 2000) using a c-fos antibody. ChIP assay shows cfos/cjun complex bound to the chromatin region which harbors rs polymorphism. As a control, cfos/cjun complex also bound to DNA sequence 9.3kb upstream of FoxO6 TSS where another binding site was predicted, but cfos/cjun complex does not bind to the DNA sequence 7.9kb upstream of FoxO6 TSS as no cfos/cjun complex putative binding site 34

58 was identified(fig. X2 C). ChIP using c-fos antibody showed an ~10 fold enrichment of chromatin containing rs polymorphism (Figure II-13 D). Furthermore, we have sequenced the ChIP-PCR products. Sanger sequencing showed that GMSM-K cells contain both G and A alleles for rs polymorphism. However, c-fos only bound to rs a variants (Figure II-13 E). We also performed ChIP in HEPM cells (Human embryonic palatal mesenchyme stem cells, ATCC CRL-1486) using the same antibodies, and a similar result was also found in HEPM cells (Figure II-15 A-C). These results show cfos/cjun complex binds to the derived sequence (rs a) in the FoxO6 promoter region but not to the ancestral sequence (rs g). To determine if the cfos/cjun complex regulates FoxO6 transcriptional activity by binding to the sequence containing rs a allele, we performed luciferase reporter assays in GMSM-K cells. ~2Kb gene fragment spanning the rs containing either the A or G allele was ligated into ptk-luc vector upstream of the minimal TK promoter. Co-transfection of FoxO6 2Kb(G) with AP-1 does not significantly change the luciferase activity compared to co-transfection with empty vector. However, AP-1 significantly activates the FoxO6 2Kb(A) luciferase reporter (Figure II-14 A). To further validate that the rs polymorphism affects AP-1 activation of FoxO6 enhancer activity, we cloned a tandem sequence containing 3 repeats of 20bp DNA sequence harboring the rs a allele or rs g allele to the ptk-luc and performed luciferase reporter assay in GMSM-K cells. AP-1 significantly enhances FoxO6 20bp(A) x 3 luciferase reporter activity compared with empty vector, but this activation was abolished when transfecting AP-1 with FoxO6 20bp(G) x3 luciferase reporter (Figure II-14 B). To test if AP-1 selectively activates the rs a enhancer in other types of cells, we 35

59 repeated the luciferase assay in HEPM cells and observed similar results (Figure II-15 D). To determine if AP-1 positively regulates FoxO6 protein levels, we overexpressed C-Fos in GMSM-K cells which are heterozygous for the rs polymorphism (Figure II-13 E). Overexpression of C-Fos lead to up-regulation of FoxO6 protein level (Figure II-14 C). Taken together, we conclude that rs affects AP-1 binding and activation of FoxO6. Discussion Craniofacial development in vertebrates is guided by multiple growth factors, signaling complexes, epigenetic modifications and transcription factors specific for developing tissues and processes. The temporal, spatial and tissue-specific timing of expression of these factors drives patterning and growth of the facial prominences. Studies of craniofacial morphology and phenotypes in mice have provided new insights for the evolution and ontogeny of the skull (Hallgrimsson and Lieberman, 2008). In humans, facial characteristics vary by ethnic and biographical background. Individuals can present with maxillary and/or mandibular prognathism (increased growth of the upper and lower jaw, respectively), micrognathia (small jaw) or retrognathia (retracted jaw). Each of these conditions affects the development of the dentition demonstrating that odontogenesis and jaw growth are directly linked. The human face has many complex geometric variations. GM methods applied to both two and three dimensional craniofacial images, allow the exploration of specific patterns of craniofacial shape (Cvijanovic et al., 2014) that could be associated with genetic variation. We have used GM methods combined with principal component analysis (PCA) (Liu et al., 2013) to identify dento-facial variation in Caucasian adults 36

60 presenting with moderate or severe dento-skeletal bite problems. This is the first study to provide evidence of a genetic association between SNPs in FOXO6 and facial shape variation. Our results particularly support a role for FOXO6 in variation related to the anterior cranial base angulation and the anterior projection of maxillary and mandibular structures. These features partly resemble those of the FoxO6 -/- mice. Thus, polymorphisms in FOXO6 may define subtle shape alteration patterns that control the anterior growth of the human face. Interestingly the rat maxilla and mandible showed ~80% decrease in FoxO6 transcript, which correlates with the difference in facial length. These data suggest that FoxO6 controls the anterior-posterior growth of the craniofacial skeleton, and that differences in FoxO6 expression may account for the differences in facial morphology seen among mammals and other vertebrates. A model for the role of FoxO6 is shown (Figure II-16). The evolution of the mammalian skull and its morphologically complex traits has been extensively studied as the development of three distinct modules: the calvarium, the cranial base, and the face (Cheverud et al., 1996; Parsons et al., 2011). The hypothesis that growth of the brain and face are linked is supported by some evidence from mouse models (Marcucio et al., 2011). As the brain and skull grow, the facial prominences also grow and both converge to form the face. The rate of growth of each can influence outgrowth of the midface and produce either prognathism or retrognathism. An elegant study showed the phenotypic variation of forebrain and facial shape between mouse strains (Parsons et al., 2011), although noted variations were limited to developmental stages with complex developmental processes and molecular mechanisms that remain undefined. However, it is clear from that and other studies that epigenetic integration and 37

61 sensitivity to gene dosage can affect facial morphology. Development of the craniofacial complex involves the integration of signaling pathways that induce cell migration, differentiation and tissue-specific patterning of the skull and brain. The molecular mechanisms driving early initial patterning and segregation of tissues within the craniofacial region are well studied, such as Wnt signaling and Shh expression (Eberhart et al., 2006; Hu and Marcucio, 2009; Venugopalan et al., 2011). However, little is known about factors that influence later stages in the development of facial morphology. FoxO6 is a good candidate for the growth regulation of specific craniofacial modules, because it is expressed at later embryonic stages and does not appear to affect the prior cell and tissue-specific patterning. This report describes how restricted expression of FoxO6 in the brain and craniofacial region activates Hippo signaling to regulate growth of the brain, maxilla, mandible and teeth. Because Hippo signaling is ubiquitous in the developing embryo, specific factors must modulate its activity in a tissue- and organ-specific manner at particular time points in development. Our MRI analyses revealed an overall total reduction in brain tissue in FoxO6 -/- mice, with a proportional increase in white matter mainly in the brain stem and a proportional decrease in cortical gray matter (cerebral and cerebellar cortex). However, several areas are increased in volume including the fourth ventricle in these mutant mice. Interestingly, the rat craniofacial tissues express low levels of FoxO6 compared to mice and the species-specific control of FoxO6 expression may regulate rodent craniofacial and face morphology. The unique expression pattern of FoxO6 appears to control the development and size of specific brain regions and the craniofacial skeleton. 38

62 Interestingly, our data indicate also that FoxO6 positively regulates Shh expression. Variation in Shh signaling is linked to shape of the upper jaw, with decreased Shh leading to reduced cell proliferation and progressive narrowing and shortening of the midface skeleton(young et al., 2010). We hypothesize that down-regulation of Hippo signaling activates cell proliferation independent of Shh to drive cell proliferation and anterior growth of the face. Hippo signaling also regulates Runx2 expression, and FoxO6 loss-offunction mice have decreased levels of Runx2, a protein that can interact with TAZ/YAP to activate gene expression. Thus, FoxO6 may directly regulate Runx2 independent of Hippo signaling to promote osteogenesis in relation to skeletal growth. Thus decreased or delayed osteogenesis and the decrease in Lats1 expression in the FoxO6 -/- mouse would lead to activation of cell proliferation (with delayed ossification) by promoting the nuclear accumulation of TAZ/YAP and an increase in the expression of WNT/β-catenin target genes. Furthermore, TAZ/YAP and TGF-β/Smads can interact with β-catenin in the nucleus, thereby activating a gene regulatory network that controls cell proliferation and organ size. Therefore, the direct regulation of Lats1 by FoxO6 is a newly identified mechanism controlling craniofacial growth, by inhibiting cell proliferation controlled by the Hippo signaling pathway. Crosstalk between the molecular components of the TGF-β, WNT, Notch, Hedgehog, Fgf and Hippo pathways influences the control of both development and cell proliferation (Guo and Wang, 2009; MacDonald et al., 2009). Interacting components of one such set of pathways are Smad (of the TGF-β pathway) and β-catenin (of the WNT pathway): both accumulate in the nucleus to activate shared targets, and interact with Lef/TCF transcription factors in inducing gene expression and control of cell fate (Hu 39

63 and Rosenblum, 2005). Hippo is another interface for these two pathways; its activation results in cytoplasmic retention of TAZ/YAP proteins, which can also interact with Smads to inhibit their activity, and thus Hippo activity antagonizes WNT signaling (Varelas et al., 2008). TAZ/YAP can also interact with β-catenin to activate gene expression during development (Rosenbluh et al., 2012). In FoxO6 -/- mice, the incisors had low Lats1/2 expression compared with control mice, and were expanded in length and size. During development of the mouse incisor, stem cells proliferate and populate the transient amplifying zone of the inner enamel epithelium and migrate to differentiate into ameloblasts that secret enamel. We conclude that the increased amelogenin expression in FoxO6 -/- mice may be caused by the observed increase in proliferation of the dental stem cells, as shown by Ki67 staining and BrDU labeling. The increase in incisor size correlates with an increase in the size of the anterior region of the mandible, whereas size of the molar was not affected. Dental patterning and tooth number were also not affected, suggesting that FoxO6-mediated regulation of Hippo signaling occurred later in development, after tooth development was initiated. Because Hippo signaling regulates stem-cell proliferation, self-renewal and differentiation, we speculate that crosstalk between the FoxO6, Hippo and Shh pathways may contribute to regulation of the proliferation and maintenance of dental stem cells (Liu et al., 2012). Disturbances in morphogens, epigenetic factors, transcription factors, and proteinprotein interactions can result in craniofacial anomalies such as cleft lip or palate, craniosynostosis, and hemifacial microsomia and these are often linked to abnormal dental development. Several examples show that formation and function of the human 40

64 craniofacial skeleton and dentition are compromised when defects occur in either of these processes (Jumlongras et al., 2001; Minoux and Rijli, 2010). Therefore the dental development process is directly linked to craniofacial formation. 41

65 Figure II-1 FoxO6 expression during mouse embryonic development. A) DNA microarray analyses was initially performed using P0 wild type (WT) mouse mandibles (WT Mand.) to identify new genes involved in craniofacial development. FoxO6 was identified and mandible tissue was isolated from FoxO6 -/- mice at P0. Gene expression was compared between the WT and FoxO6 -/- mouse mandibles. B-F) FoxO6 expression is shown by X-gal staining in FoxO6 +/- mice. B) FoxO6 expression is not detectable at E10.5. C) At E12.5, FoxO6 expression is predominantly in the brain (cerebral cortex, CC; cerebellar primordium, CP; trigeminal ganglion, TG; somites, S. D) At E14.5, FoxO6 expression is in the brain, craniofacial regions, also in the posterior maxilla (PMx) and posterior mandible (PMn). E) At E16.5, FoxO6 is expressed throughout the craniofacial region and in the brain. F) FoxO6 has strong expression in brain, craniofacial region, such as mandibular bone (MB), incisor (IN), molar (M) and palate (PL). 42

66 Figure II-2 The heads of FoxO6 -/- mice are larger than those of wild type mice. A) Size of head in 2 month-old FoxO6 -/- mice compared to WT mice (N>5). B) FoxO6 expression in P0 mouse and rat mandible tissue, revealed decreased FoxO6 in the rat compared to mouse (N=3). C) The submandibular gland was increased in 6M FoxO6 -/- mice compared to WT. D, E) Alcian Blue/Alizarin Red staining of cartilage and bone skeletal preparations of E18.5 FoxO6 +/- mice (left) and FoxO6- /- littermates (right). D) E18.5 head skeletal preparations show a larger skull and head in FoxO6 -/- embryos. Moreover, a reduced ossification (red stain) of the occipital bone (OB), temporal bone (TB) and nasal bone (NB) were observed in E18.5 FoxO6 -/- embryos compared to their heterozygous littermates. E) Skeletal preparations of P1 FoxO6 +/- mice (left) and FoxO6 -/- littermates (right) show that FoxO6 -/- mice have an expanded head compared with control mice. The frontal bone (FB), parietal bone (PB) and occipital bone (OB) are slightly larger in FoxO6 -/- mice. 43

67 Figure II-3 The heads of FoxO6 -/- mice are expanded in the anterior direction. A-C) Microcomputed tomography (µct) analysis of 2 month FoxO6 -/- and WT mice heads. A) µct of wild type and FoxO6 -/- heads used for morphometric measurements. B) Measurements of the nasal length, skull length, and mandibular length are all increased in the FoxO6 -/- mouse head. C) Midsagittal section measurements showed cranial base length was 6.51% increased in the FoxO6 -/- mouse head, but the cranial base angle showed no change between the FoxO6 -/- and WT mice, N=4. (The micro-ct scanning in this figure was done by Dr. Nathan E. Holton from University of Iowa, College of Dentistry.) 44

68 Figure II-4 Gene ontology (GO) analysis of biological process, indicating the cellular processes with FoxO6. DNA microarray bioinformatics data were processed to understand the different biological processes controlled by FoxO6. (These gene expression array data were performed and analyzed by LC Sciences(Houston, TX)) 45

69 Figure II-5 FoxO6 activates Lats1/2, Runx2, Shh and the Hippo signaling pathway. Genes identified as being differentially expressed between FoxO6 -/- and WT mandibles in microarray screen were assessed. A) Relative expression of genes in mandible and maxilla tissue of E18.5 FoxO6 -/- and WT (FoxO6 +/+ ) was assessed by real-time PCR. Experiments were repeated at least three times each, from multiple biological samples. Error bars indicate S.E. *: p-value<0.01. B) Expression and phosphorylation of gene product in CHO cells overexpressing FoxO6. CHO cells were transfected with 5 µg empty vector (control), 5 µg pcdna-foxo6, or 10 µg pcdna-foxo6. Lysates were collected after 2 days for immunoblotting. C) Expression and phosphorylation of gene product in LS-8 (oral epithelial-like) cells overexpressing FoxO6 overexpression. 5 µg empty vector (control) or 5 µg pcdna-foxo6 were used for electroporation. Lysates were collected after 2 days for immunoblotting. D) Expression and phosphorylation of gene product in ET-16 cells (oral epithelial cells) subjected to FoxO6 knockdown. 10 µg sh Scramble DNA or 10 µg shfoxo6 DNA were transfected and cell lysates were collected after 2 days for RNA extraction and realtime PCR. 46

70 Figure II-6 FoxO6 directly binds to and activates the Lats1 promoter. A) Schematic of the Lats1 1.7kb promoter, with the predicted FoxO6 binding motif (- 2467bp). Primers were designed to flank the predicted FoxO6 binding site (-2362 to bp; ChIP primers) and to an upstream region that lacks a FoxO6 binding site (-4735 to -4913bp ; Control primers). B) PCR products from ChIP assays involving immunoprecipitation of endogenous FoxO6 in LS-8 cells. PCR products were resolved in agarose gels. Input chromatin was used as a control to show the primer product; FoxO6 Ab (antibody) immunoprecipiated (IP) the chromatin containing the FoxO6 binding site (lane2); IgG alone did not IP the chromatin and the no template PCR reaction did not produce a band. Control primers to an upstream region of the Lats1 promoter did not detect an IP product with the FoxO6 Ab or the IgG control. C) Quantitation of the enrichment of binding by endogenous FoxO6 chromatin compared to IgG immunoprecipitated DNA. D) Activation of the Lats1 1.7kb promoter by FoxO6 using a Lats1 luciferase reporter in transfected LS-8 cells compared with empty vector. E) Activation of an Lats1 enhancer luciferase reporter whose expression is driven by a duplicated 80 bp DNA fragment derived from the Lats1 promoter region containing the FoxO6 binding site.this construct was transfected into LS-8 cells with or without FoxO6 expression plasmid. In parallel, a reporter with a mutated FoxO6 binding site (Lats1 80bp Mut enhancer) was transfected as control. 47

71 Figure II-7 FoxO6 regulates Lats1 and pyap expression. A) Schematic illustration of the mouse lower incisor, showing the labial cervical loop (LaCL, stem-cell niche), inner enamel epithelium (IEE), outer enamel epithelium (OEE), and ameloblast layer derived from the LaCl cells. B-G) FoxO6 expression in the E18.5 lower incisor (B, C), molar (D, E) and oral epithelium (F, G) of WT. (B, D, F) and FoxO6 +/- (C, E, G) mice, as indicated by immunostaining using an X-gal antibody (green). Scale bars represent 100µm. H-K) Lats expression in FoxO6 -/- vs WT mice. Series of sagittal sections of lower incisors from P1 animals were examined by immunofluorescence staining for Lats1 protein, using an Alexa-488 labeled antibody. I, K) Magnified views of the boxed regions in (H) and (J), respectively. The LaCL, dental epithelium and ameloblast layer is outlined. L-O) Expression of pyap. Less Yap is phosphorylated (activated) in incisors of FoxO6 -/- vs WT mice. M, O) Magnified views of the boxed regions in (L) and (N), respectively.in all sections, DAPI staining was used to identify nuclei. Scale bars represents 100µm. 48

72 Figure II-8 In FoxO6 -/- mice, the incisors are enlarged, the LaCL is expanded and the polarity of the dental epithelium is abnormal. Sections of incisors from E16.5, E18.5, and P0 WT and FoxO6 -/- mice stained with Hematoxylin and Eosin. A-B) E16.5 mandibles, with lower incisor (LI) framed by box to highlight differences in incisor size. C-D) Higher magnification of the boxed region of the lower incisors in A and B, showing that these are enlarged and that FoxO6 -/- embryos have larger labial cervical loops (LaCLs). E-F) E18.5 mandible with longer lower incisor still observed in FoxO6 -/- embryos. G-H) Higher magnification of E and F (LaCL is outlined in black). I-J) P0 mandible. The lower incisor and LaCL are much larger in the FoxO6 -/- mice. K-L) Higher magnification of the red box of I and J, revealing that the LaCL is larger in FoxO6 -/- mice. M-N) Higher magnification of the ameloblast and odontoblast cell layers (blue boxed region in I, and J) show that the cells are not well polarized in FoxO6 -/- mice compared to WT mice. Scale bar represents 100 µm. Abbreviations: Epi, dental epithelia; Mes, dental mesenchyme; LaCL, labial cervical loop; LiCL, lingual cervical loop; Am, ameloblast; Od; odontoblast; SI; stratum intermedium. 49

73 Figure II-9 FoxO6 -/- mice have increased Amelogenin expression. Amelogenin expression (green) in sagittal sections of lower incisors (LI) of P1 FoxO6 +/- and FoxO6 -/- mice. A-B) Overview in low magnification showing the cervical loop at the left and the anterior end of the incisor to the right. C-D) Higher magnification of the boxed regions in (A, B). Am, ameloblast; Od, odontoblast. In all sections, DAPI staining was used to identify nuclei. 50

74 Figure II-10 FoxO6 -/- incisors exhibit the increased cell proliferation. A-B) Cell proliferation in lower incisors from WT or FoxO6 -/- E18.5 embryos, as assessed by staining withki67 and DAPI. C-D) Higher magnification of the LaCL region from WT or FoxO6 -/- embryos in A and B. The Ki67 positive cells are located mainly in the Transient Amplifying (TA) zone of the IEE and adjacent mesenchymal tissue. E) Quantitation of the Ki67-positive cells in the sections of lower incisors. The number of Ki67-positive cells in FoxO6 -/- epithelial and mesenchymal tissue is increased compared to WT embryos. F-G) Cell proliferation in E17.5 WT and FoxO6 -/- embryos, as assessed by BrdU injection. H-I) Higher magnification of posterior or proximal lower region of incisor in WT and FoxO6 -/- embryos in F and G, demonstrating increased epithelial and mesenchymal cell proliferation compared to the WT. J) Quantitation of the BrdUpositive cells in sections of lower incisors. Scale bar represents 100 µm. Epi, epithelium; Mes, mesenchyme. 51

75 Figure II-11 FoxO6 -/- mouse incisors are positioned anteriorly and distal to the molars compared to WT mice. A-G) Microcomputed tomography (µct) analysis of 1 month old FoxO6 -/- and WT mice heads, mandibles, lower incisors and molars. A) µct images of FoxO6 -/- and WT heads used for analyses. B, C) Hemimandible image showing normal molar size and mineralization compared to WT but decreased mineralization and enamel of the FoxO6 -/- lower incisor at this position (pink arrows). D, E) Image in the coronal plane showing that the FoxO6 -/- lower incisor at the position in plane through the distal root of the first molar has decreased enamel formation compared to the WT incisor (pink arrows). F, G) The position of the FoxO6 -/- lower incisor appears more anteriorly in association with the molars compared to the WT incisor due to delayed mineralization (pink arrows). (The micro-ct analysis in this figure is done by Dr. Felicitas B. Bidlack from The Forsyth Institute, Cambridge, MA) 52

76 Figure II-12 FOXO6 SNPs are associated with variation in human facial shapes. Three human FOXO6 single nucleotide polymorphisms (SNPs) were found associated with specific dento-facial phenotypes. A) Extremes of the PC3, which captures variation ranging from a bi-maxillary retrusion and steep anterior cranial base on one end (-β) to bi-maxillary protrusion and a flat anterior cranial base at the opposite end (+β). Association results indicate that more copies of the rare allele for the SNP rs in an individual s genotype (p-value = 0.007; β = ) were associated with a tendency towards bi-maxillary protrusion while more copies of the rare allele for the SNP rs (p-value = 0.022; β = ) were associated with bi-maxillary retrusion. B) Extremes of the PC4, which captures variation ranging from a long mandibular body with an anterior incisor cross bite relation (-β) to a shorter mandibular body and adequate incisor relation (+β). Association results indicate that more copies of the rare allele for the SNP rs (p value = 0.015; β = ) are associated with a larger mandibular body length and an anterior cross bite relation. C) Location of the genotyped SNPs relative to the human FOXO6 gene according to the NCBI Reference Assembly. (The human genetic studies in this figure were carried out by Dr. Lina Moreno-Uribe and her student Clarissa S.G. de Fontoura from University of Iowa, College of Dentistry.) 53

77 Figure II-13 A polymorphism in 5.8kb upstream of FoxO6 transcription start site (TSS) alters the AP-1 binding in GMSM-K cells. A) The rs polymorphism lies in a potential AP-1 binding sequence. B) Schematic of FoxO6 promoter region. rs is located nearly 5.8Kb upstream of the FoxO6 TSS. Two AP-1 binding sites were identified at 5825bp(S1) and 9396bp(S2) upstream of FoxO6 TSS. Primers were designed to amplify the putative binding site regions as indicated by S1 and S2. As a control for ChIP experiment, primers were designed to amplify the chromatin region(n1) which is located nearly 7900bp upstream of FoxO6 TSS and doesn t have a putative AP-1 binding site. C) ChIP assays were performed in GMSM-K cells and PCR products were resolved on agarose gels. C-Fos antibody (Ab) immunoprecipitated the S1 and S2 region but not a control region(n1). D). ChIP-qPCR analyses to assess the enrichment by comparing the binding of C-Fos antibody to IgG in S1 and S2 regions. E). Sanger sequencing results of ChIP-PCR products using S1 primers to amplify the input DNA and C-Fos antibody immunoprecipitated DNA. Note that the GMSM-K cells are heterozygous for the ancestral allele (G) and derived allele(a) of rs polymorphism (Left), however, C-Fos only bound to the derived rs (a) sequence in FoxO6 promoter region(right). 54

78 Figure II-14 rs polymorphism affects AP-1 activation. A). Luciferase reporter assays comparing luciferase activities of FoxO6 2Kb enhancers containing either the ancestral allele (G) or derived allele(a) of rs GMSM-K cells were used for luciferase assays. Note that AP-1 activates FoxO6 2Kb enhancer harbors derived allele (A) but not the ancestral allele (G). B). Luciferase reporter assay showing luciferase activities of FoxO6 20bp x3 enhancer with ancestral allele (G) or derived allele (A) of rs The FoxO6 20bp(A) x 3 and FoxO6 20bp(G) x 3 luciferase reporters were made by inserting tandem DNA sequences including three repeats of 20bp FoxO6 enhancer contains rs derived allele (A) or ancestral allele (G) into firefly luciferase vector respectively. GMSM-K cells were co-transfected with luciferase reporters and AP-1. C). Western blot analysis shows an increase in FoxO6 protein levels when C-Fos is overexpressed in GMSM-K cells. GAPDH is used as loading control. 55

79 Figure II-15 AP-1 binds to the rs derived sequence in HEPM cells. A). ChIP-PCR showing C-Fos binds to S1 region of FoxO6 promoter in HEPM cells, but not the S2 and N1 regions. The schematic of S1, S2 and N1 regions in FoxO6 promoter are indicated in FigX B. B) ChIP-qPCR analyses to assess the relative binding of C-Fos antibody to IgG at S1 and S2 regions. C). The sanger sequencing results of ChIP- PCR product using S1 primers shows c-fos antibody only bound to rs derived sequence in HEPM cells. D). Luciferase reporter assays were performed in HEPM cells using FoxO6 20x3 luciferase reporters. The FoxO6 20x3 luciferase reporters were made by inserting three repeats of 20bp FoxO6 enhancer including the rs polymorphic sequence. 56

80 Figure II-16 Model for FoxO6 regulation of Hippo signaling in controlling growth of the head and craniofacial structures. Mammalian Ste20 family kinases Mst1 and Mst2 form an active complex with Salvador (Salv), to further phosphorylate large tumor suppressor homolog (Lats1 and Lats2) kinase. Lats1 and Lats2 bind to a scaffold protein, Mob, to further phosphorylate two transcriptional co-activator Yap and Taz. Yap and Taz are downstream Hippo signaling components and modulate the expression of genes involved in cell growth, proliferation and apoptosis. Phosphorylated Yap and Taz are retained in the cytoplasm. Unphosphorylated Yap and Taz can enter the nucleus to activate gene expression in concert with TEAD, P73, Pax3, Runx2, Smad, β-catenin and other factors. In this paper, we conclude that FoxO6 binds and activates Lats1to further inhibit Yap and cell proliferation, thereby controlling head and craniofacial growth regionally. 57

81 Table II-1 Comparison of sizes of craniofacial structures between WT and FoxO6 -/- mice. µct measurements were computed and defined measurements of craniofacial growth were identified, N=4. Grey color highlight the bone lengths which are significant changed in FoxO6 -/- mice. 58

82 CHAPTER III PITX2:SOX2:LEF-1 NETWORK REGULATES DENTAL STEM CELL MAINTENANCE AND TOOTH DEVELOPMENT Abstract Sox2 marks dental epithelial stem cells (DESC) in mammals and reptiles, but its transcriptional function in regulating dental stem cell fate and tooth growth is not well understood. In this report we show that conditional Sox2 deletion in the oral and dental epithelia results in severe craniofacial defects, including ankyloglossia, cleft palate, impaired stem cell self-renewal, arrested incisor development and abnormal molar development. Tamoxifen-induced inactivation of Sox2 demonstrates the requirement of Sox2 for maintenance of the DESCs in adult mice. Conditional overexpression of Lef-1 in mice increases DESC proliferation and creates a new labial cervical loop stem cell compartment, which produces rapidly growing long tusk-like incisors. Epithelial Lef-1 overexpression partially rescues the tooth arrest in Sox2 conditional knockout mice. Mechanistically, Pitx2 and Sox2 interact physically and regulate Lef-1, Pitx2 and Sox2 expression during development. Thus, we report a new Pitx2:Sox2:Lef-1 transcriptional gene regulatory network for DESC homeostasis and dental development. Introduction The regenerative tissues such as peripheral blood, hair follicles, intestine and teeth have a particular microenvironment known as a stem cell niche to house stem cells and act as a signaling center to control stem cell fate (Clavel et al., 2012a; Lane et al., 2014; Spradling et al., 2001) The precise and timely regulation of stem cell renewal and differentiation is essential for tissue formation, growth and homeostasis over the course of a lifetime (Moore and Lemischka, 2006; Moore et al., 2006), but the molecular 59

83 mechanisms underpinning this regulation are variable and dependent on tissue-specific signals. The continuous growth of rodent incisors occurs via the renewal and differentiation of stem cells in both the epithelial and mesenchymal stem cell niches. During mouse incisor development, both dental epithelial and mesenchymal cells are replenished within one month (Smith and Warshawsky, 1975). The labial cervical loop (LaCL), which is located at the proximal end of the labial side of the incisor, is thought to be the stem cell niche for DESCs (Biehs et al., 2013; Juuri et al., 2012a; Thesleff and Tummers, 2008). The neurovascular bundle (NVB) provides a niche for dental mesenchymal stem cells to reside and generates pulp cells and odontoblasts (Kaukua et al., 2014; Zhao et al., 2014) The dental epithelial and mesenchymal components produce signaling factors that promote the differentiation and developmental processes of adjacent tissues. The periderm acts to prevent epithelial adhesions during embryogenesis and is a transient structure that covers the developing epithelia until shortly before birth(holbrook and Odland, 1975; Peyrard-Janvid et al., 2014; Richardson et al., 2014a). Epithelial adhesions cause both cleft palate and ankyloglossia, which restricts elevation and extension of the palatal shelves and attaches the tip of the tongue to the mandible (Pauws et al., 2009b; Peyrard-Janvid et al., 2014). Cleft palate, ankyloglossia and hypodontia are often linked in patients with a variety of syndromic or non-syndromic genetic anomalies. However, the molecular mechanisms for these anomalies are poorly understood. The transcription factor Sox2 is critical for stem cells and progenitor cells to maintain pluripotency (Boyer et al., 2005; Takahashi and Yamanaka, 2006a), and ablation of Sox2 in mice leads to early mortality after implantation (Avilion et al., 2003a). 60

84 Sox2 has important roles in the development of several endodermal tissues, such as the trachea (Xie et al., 2014), stomach and gut (Que et al., 2007), as well as in ectodermal tissues including the anterior pituitary (Jayakody et al., 2012), lens epithelium (Taranova et al., 2006a; Taranova et al., 2006b), tongue epithelium (Arnold, 2011) and hair follicles (Clavel et al., 2012b). Sox2 was recently reported to be a marker for DESCs. Sox2 + cells are located in the LaCL and molar cervical loop regions and give rise to the highly proliferative transient-amplifying (TA) cells which can differentiate into enamelsecreting ameloblasts (Juuri et al., 2012b; Li et al., 2015). However, the molecular role of Sox2 in DESC maintenance and proliferation during tooth initiation and growth remain unknown. Previous studies have shown that Lymphoid enhancer-binding factor (Lef-1) is regulated by FGF signaling and is required for early tooth development, where it plays roles in mediating epithelial-mesenchymal interactions (Kratochwil et al., 1996b; Kratochwil et al., 2002; Sasaki et al., 2005a). Lef-1 deficiency in vivo results in arrested tooth morphogenesis at the late bud stage (van Genderen et al., 1994a). Epithelial and mesenchymal tissue recombination assays showed that Lef-1 is required only transiently in the dental epithelium (Kratochwil et al., 1996b). The majority of Lef-1 expression is shifted to mesenchymal cell/tissues surrounding the epithelium at the bud stage, although Lef-1 expression persists in the basal cells of the dental epithelium immediately adjacent to the mesenchyme (Kratochwil et al., 1996b; Sasaki et al., 2005a). Both Sox2 and Lef-1 are markers of early craniofacial development and are expressed in the oral and dental epithelium (Juuri et al., 2013b; Juuri et al., 2012b; Sasaki et al., 2005a; Zhang et al., 61

85 2012), however any potential interaction between Sox2 and Lef-1 has not been clearly identified. In this study, we demonstrate a role for Sox2 in DESC maintenance and proliferation during tooth formation by conditionally ablating Sox2 in the oral and dental epithelia using the Pitx2 Cre system. We found that conditionally inactivation of Sox2 leads to severe craniofacial defects, including ankyloglossia, cleft palate, and arrested incisor development at the early bell stage. The dental defects were due to impaired stem cell renewal and defective dental epithelial cell migration, which caused tooth agenesis. Because Lef-1 is also required for tooth development and potentially stem cell proliferation, we generated a Lef-1 conditional over expression (COEL) mouse and used Pitx2 Cre to overexpress Lef-1 in the oral and dental epithelia. Lef-1 overexpression results in tusk-like incisors generated by formation of a new compartment of dental epithelial stem cell niche. Interestingly, the COEL rescued the incisor phenotype in Sox2 cko mice, but not ankyloglossia or cleft palate. Taken together, these data reveal a Sox2-Lef-1 regulatory mechanism for dental epithelial stem cell maintenance and proliferation. Materials and methods Mouse lines and embryonic staging The Program of Animal Resources at the University of Iowa housed mice. Each procedure complied with the guidelines set by the University of Iowa Institutional Animal Care and Use Committee. Sox2 conditional knockout mice (Sox2 Flox/Flox ) have been previously described (Taranova et al., 2006b), and the ROSA-Cre ERT (B Gt(ROSA)26Sortm1(cre/ERT2)Tyj/J) mice originated from the Jackson Laboratory (stock number ). Each of these strains was a generous gift from Dr. John Engelhardt 62

86 (The University of Iowa). The Lef-1 conditional overexpression (COEL) mouse line was generated by inserting Lef-1 downstream of a CAAG promoter and a floxed transcription stop signal. The Pitx2 Cre mouse has been described previously (Liu, et al. 2003). Each mouse line was derived from a C57BL/6 background. For embryonic staging experiments, the observed vaginal plug date of the female was designated as E0.5. Embryos were collected on the required time point, and genomic DNA was isolated from a portion of the embryonic material (usually the tail) for genotyping. The genotyping primers for all the mouse lines were listed in Table III-1. Immunohistochemistry, immunofluorescence and histology Mouse embryos or heads were dissected in a cold PBS solution, and fixed from one to four hours in 4% para-formaldehyde while protected from light. After fixation, the tissues were dehydrated through an ethanol gradient (70% for over an hour, 80% for over an hour, 95% for over an hour, and 100% overnight), perforated with xylene, embedded in paraffin and with a microtome, sectioned into 7µm thick segments. Tissue morphology was examined by Hematoxylin and Eosin staining (Cao et al., 2013b). Sections used for analysis by immunofluorescence and immunohistochemistry were subjected to antigen retrieval by incubating in a citrate buffer (95 C,10mM, PH 6.0) for 20 minutes. Blocking was performed by incubating the sections with 20% donkey serum in PBS-triton at room temperature for 30 minutes. Primary antibodies against Sox2 (Goat, R&D Systems, AF2018 1:200; Rabbit Abcam, ab97959, 1:200), GFP (Abcam ab290, 1:500), Ki67 (Abcam, ab15580, 1:200), Lef-1 (Cell signaling #2230, 1:200), Keratin 6 (Abcam, ab75703, 1:200), E-Cadherin (BD Bioscience, , 1:200), Cleaved caspase-3 (Cell signaling, #9661, 1:200) and amelogenin (Santa Cruz, L0506, 1:200) were then added to 63

87 the sections. Incubation with primary antibody occurred overnight at 4 C. The slides were treated with FITC (Alexa-488)- or Texas Red (Alexa-555)-conjugated Secondary antibody for 30 minutes at room temperature for detection (Invitrogen,1:400). Nuclear counterstaining was performed using DAPI-containing mounting solution. For immunohistochemistry, standard protocols were followed according to manufacturer s manual (Millipore, IHC select HRP/DAB, DAB150). Images were captured with a Nikon eclipse 80i fluorescence microscope or Zeiss 700 confocal microscope. Chromatin immunoprecipitation assay (ChIP) The procedure for the ChIP assay has been described previously (Wang et al., 2013a). Briefly, the ChIP Assay Kit (Zymo Research, Zymo-Spin ChIP Kit, D5210) was used with a modified protocol. LS-8 cells were seeded in T-75 flasks in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin and fed 24 hours prior to the experiment. The day of the experiment, the adherent cells were harvested and collected in a 1.5 ml tube. The cells were washed twice in cold PBS solution and crosslinked (1% formaldehyde, room temp, 7 minutes). After crosslinking, cells were subjected to three rounds of sonication (6 seconds duration, 25% of maximum amplitude), causing lysis and the shearing of genomic DNA in fragments approximately bp. DNA/protein complexes were immunoprecipitated with 5 µg of Pitx2 antibody (Capra Sciences, PA-1023) or Sox2 antibody (R&D Systems, AF2018). Same amount of normal rabbit IgG was used to replace the specific antibody to assess the nonspecific immunoprecipitation of the chromatin. All the primer sequences used in this assay are listed in Table III-2. Three parallel qpcrs were performed by using ChIP products. Relative enrichment was calculated with the 2 -ΔΔCT. All the PCR products were 64

88 visualized on a 1.5% agarose gel to check the size and conformed to be right product by sequencing. BrdU labeling Two hours prior to sacrifice, pregnant mice were injected with BrdU (10 µl/g body weight, Invitrogen, ), and the embryos were collected and processed as previously described for the immunofluorescence assay. Sections were mounted and rehydrated through a reverse ethanol gradient, and to compensate for endogenous peroxidase activity, immersed in 3% hydrogen peroxide. Antigen retrieval was performed by immersing sections in 10 mm sodium citrate solution for 20 min at a slow boiling state. Sections were perforated by a 30 min incubation in 2 M HCL, followed by a neutralization step (10 minutes 0.1 Na 2 B 4 O 7 ). Sections were subsequently blocked for 1 hr in 10% donkey serum, and labeled with anti-brdu antibody (Abcam ab6326, 1:250). Next, standard immunohistochemistry staining was carried out as previous described in immunohistochemistry assay. Experimental and control sections were processed on the same slide and stained together for identical time periods. IdU/CldU labeling assay Standard immunofluorescent detection of IdU/CldU was performed according to previous report with modifications (Tuttle et al., 2010). 24 hours prior to harvesting E15.5 mouse embryos, pregnant female mice were intraperitoneally injected with a 100 µg per gram of body weight of CldU (Sigma, C6891). One hour before harvesting the embryos, the pregnant female mice were again injected with 100 µg per gram of body weight of IdU (Sigma, 17125). Mouse embryos were then harvested and embedded in paraffin. The blocks were sectioned and subjected to antigen retrieval by boiling in citrate 65

89 buffer (10 mm, ph 6.0) for 20 min. Sections were then permeabilized by incubating in 1.5 M HCl at 37 C for 30 minute, followed by a neutralization step (10 minutes 0.1 Na 2 B 4 O 7 at room temperature). Next, sections were blocked in 10% donkey serum diluted in PBST (PBS with 0.05% Triton-100) for one hour at room temperature. Slides were treated with a primary antibody, mouse anti-brdu/idu (Roche, ,1:250), overnight at 4 C to detect IdU. Slides then were stringently washed by vigorous agitation in a shaker for 20 min with the low salt TBST buffer (36 mm Tris, 50 mm NaCl, 0.5% tween-20; ph 8.0) at the temperature of 37 C, speed of 200 rpm. Sections were washed twice with PBST (10 minutes each), and treated with a primary antibody against CldU (anti BrdU/CldU, Accurate chemical, OBT0030, 1:250,) for 2 hr at room temperature. Slides were washed three times with PBST, and treated with a mix of secondary antibodies (Rhodamine-Red donkey anti-rat, Jackson ImmunoResearch, # ,1:400; Alex Fluor488 donkey anti-mouse IgG, Invitrogen, A21202, 1:400) for 30 min at room temperature, and then washed three times in PBST for 10 minutes. Slides were covered using a mounting solution containing DAPI for nuclear staining (Vector lab, H-1200) and prepared for imaging. Incisor injury and recovery assay Starting at P21, experimental mice (Rosa CreERT /Sox2 F/F ) and control mice (Sox2 F/F ) were fed with 130 µg/g bodyweight tamoxifen daily for nearly a week (Sigma, 75648) using a gavage needle. Bodyweight was monitored each day in order to adjust the dosage. After seven days of tamoxifen treatment (P29), the left lower incisor of the mouse was clipped and its length was recorded using a caliper. The length of the uninjured incisor was also recorded. Tamoxifen was administered daily for another five days, and the 66

90 animals were sacrificed at P34. The length of injured and uninjured incisors in both control and experimental groups were measured and the incisors were analyzed by µct. The growth rate was determined as the by daily length increase of the injured incisor (growth rate= (Length P34 - Length P29 )/ 5 days). The relative growth rate was calculated by normalizing the growth rate of injured incisor of experimental mice to control mice. TUNEL assay Paraffin sections were cut (7 mm) and rehydrated with sequential concentrations of alcohol. The TUNEL assay was carried out with DeadEnd Fluorometric TUNEL System (Promega, G3250) according to the manufacturer s protocol. Cloning, transient transfection and luciferase assay The Lef bp (Lef-1 2.7) promoter luciferase vector was constructed as previous descripted (Amen et al., 2007). Sox2 2.0 luciferase reporter was constructed by inserting ~2.0 Kb Sox2 DNA fragment located in the upstream region of Sox2 containing Pitx2 binding sites into the ptk-luc vector. Similarly, ~5.0 Kb upstream of Pitx2 gene was ligated to ptk-luc to generate Pitx2 5.0 luciferase reporter. Standard transient transfection by electroporation and luciferase assay were carried out in LS-8 cells (oral epithelial-like cell) according to a previous report (Cao et al., 2013b). Quantitative real time PCR gene expression analysis Total RNAs were extracted from LS-8 cells overexpressing pcdna 3.1(empty vector), Pitx2 or Sox2 respectively using RNeasy Mini Kit from Qiagen. Reverse transcription was performed according to the manufacturer's instruction (BIO-RAD iscript Select cdna Synthesis Kit) using oligo (dt) primers. cdnas were adjusted to equal levels by PCR amplification with primers to β-actin. Fold change was calculated 67

91 based on the 2 ΔΔCT method. All real-time PCR primer sequences are listed in Table III-2. Imaging and microcomputed tomography (µct) Mouse skulls from three experimental and control animals were scanned with a Siemens Inveon Micro-CT/PET scanner using 60kVp and 500 ma with a voxel size of 30 µm. Reconstructed images were imported using Osirx DICOM software. Mouse heads were prepared by overnight fixation at 4 C, followed by storage in 70% EtOH. Scans directed along the anterior-posterior plane produced 2-D images, which were matched between animals using topology such as the presence of the molar. GST pull-down assays GST pull-down assays were carried out as previously described (Wang et al., 2013a; Wang et al., 2013b). Briefly, GST-Sox2 full-length and truncated fusion proteins were isolated, purified and suspended in binding buffer (20 mm HEPES buffer, ph = 7.5, 5% glycerol, 50 mm NaCl, 1 mm EDTA, 1 mm DTT, with 1% milk and 400 µg/ml of ethidium bromide). Bacterially overexpressed and purified Pitx2A protein (2µg) was added to 10 µg of immobilized GST fusion protein in a total volume of 100 µl and incubated for 30 minutes at 4 C. The beads were pelleted and washed 5 times with binding buffer. The bound proteins were boiled in SDS loading buffer for 5 min to elute protein from beads. The boiled samples were run on a 12% SDS-PAGE gel. Samples were transferred to a PVDF membrane, immunoblotted and visualized by using Pitx2 antibody (Capra Sciences, PA-1023) and ECL reagents. Immunoprecipitation assay ET16 cells were fed 24h before the experiment and grown to 90% confluent in two T-175 flasks. Cells were collected and washed twice in ice-cold PBS, and lysed using 5x 68

92 Lysis buffer (Promega) in the presence of saturated PMSF. Lysates were repeatedly passed through a 27-gauge needle to disrupt cells. The lysates were then incubated on ice for 30 min and centrifuged at 10,000 g for 10 min at 4 C. An aliquot of the supernatant was saved for input. The rest of supernatant was transferred to a new tube and precleared using the mouse ExactaCruz F IP matrix (ExactaCruz F, Santa Cruz Biotechnology) for 30 min at 4 C. An IP antibody-ip matrix complex was prepared according to manufacturer s instruction using Pitx2 antibody (Capra Sciences, PA-1023) or rabbit IgG. The IP antibody-ip matrix complex was incubated with precleared cell lysate overnight at 4 C. After incubation, samples were centrifuged to pellet the IP matrix and followed by washing 3 times with cold PBS and resuspended in 15 µl ddh 2 O. Samples then were boiled and immunoblotted using anti-sox2 antibody (Abcam, ab97959). 3D reconstruction of the labial cervical loops Three-dimensional (3D) reconstructions of the Labial cervical loop (LaCL) in P2 Lef-1 cki and COEL mice were made from serial sagittal sections (7.0 µm). The labial epithelial tissues were manually traced on consecutive sections and automatically aligned using the StackReg plugin for ImageJ. The final 3D reconstructions were rendered using Imaris software from Bitplane AG. Statistical analysis For each condition, at least three experiments were performed and the results are presented as the mean ± SEM. The differences between two groups of conditions were analyzed using an independent, two-tailed t-test. 69

93 Results Specific ablation of Sox2 in the oral and dental epithelia Consistent with previous reports (Arnold et al., 2011; Avilion et al., 2003b; Ellis et al., 2004), we found that Sox2 is expressed in the lateral ventricle (LV) and epithelial tissues of E18.5 control mice, including the nasal epithelium (NE), oral epithelium (OE), tongue epithelium (TN) and dental epithelium (DE), using immunohistochemistry (IHC) and immunofluorescence on sagittal sections (Figure III-1A, B). We also detected specific expression of Sox2 in the labial cervical loop (LaCL), where the dental epithelial stem cells (DESCs) reside (Figure III-1C). To investigate the function of Sox2 in the oral and dental epithelia, we generated Pitx2 Cre -Sox2 F/F mice, hereafter referred to as Sox2 cko mice. We have previously shown that Pitx2 is expressed in the oral ectoderm at E10.5 and later stages (Cao et al., 2010c; Li et al., 2014). Pitx2 Cre has robust and restricted expression in the dental, oral and tongue epithelia at E11.5 and E14.5 (Figure III-1D,E). Pitx2 Cre mice have normal craniofacial and early tooth development (Cao et al., 2010c). Immunofluorescent staining demonstrates that Sox2 was efficiently ablated in the lower incisor LaCL and oral epithelium of Sox2 cko mice (Figure III-1G, K), however Sox2 expression in the lateral ventricle remained unchanged compared to control embryos (Figure III-1M). These data demonstrate the specificity of the Pitx2 Cre construct. Sox2 deletion in the oral epithelium causes ankyloglossia and cleft palate Following loss of Sox2, Sox2 cko mice exhibit craniofacial defects. We detected an attachment of the anterior tongue to the mandible in E18.5 Sox2 cko embryos (Figure III- 2A-D), similar to the human craniofacial disorder, ankyloglossia. This phenotype was 70

94 consistent in all Sox2 cko embryos (n = 15). As early as E15.5, we detected abnormal adhesion between the tongue epithelium, oral epithelium and dental epithelium in Sox2 cko mice, as visualized by E-cadherin staining (Figure III-2E-H). To investigate whether impaired periderm formation caused this phenotype, we analyzed the expression of keratin 6, a marker for periderm, and found that it was reduced in the oral and tongue epithelia of the Sox2 cko embryos (Figure III-2I-L). In addition, Sox2 cko embryos developed a complete cleft palate, in which the palatal shelves failed to elevate, likely due to oral adhesions (Figure III-2M, arrows). Sox2 cko mice die at P0, likely due to the cleft palate. Inactivation of Sox2 leads to lower incisor arrest at E16.5 and abnormalities in upper incisor and molar development The first step of tooth development is a thickening of ectoderm-derived oral epithelium at E11.5, after which the thickened epithelium invaginates into the underlying cranial neural crest-derived mesenchyme to form a tooth bud at E12.5. At the bud stage (E12.5), Sox2 cko tooth germs were detectable but displayed a slightly delayed invagination compared to those of littermate control embryos (Figure III-3B, C). At E14.5, the tooth epithelium further invaginates to envelope the mesenchymal dental papilla to form a cap stage incisor. The cap stage incisors are longitudinally oriented in the mandible with a cervical loop in the labial (LaCL) and lingual (LiCL) epithelium (Figure III-3A). The LaCL was smaller in Sox2 cko embryos at this stage and the lower incisor was positioned more anteriorly and close to the oral epithelium compared with control incisors (Figure III-3D,E). The oral adhesions appear to fuse the dental lamina to the invaginating dental epithelia (Figure III-2F,H), which may position the lower incisor 71

95 towards the anterior region of the mandible. At E16.5, the Sox2 cko incisors were smaller, invagination was hindered and the LaCL was severely underdeveloped, compared to Sox F/F littermates (Figure III-3F and 3G). At E17.5 and E18.5, lower incisor development regressed in Sox2 cko embryos (Figure III-3H-M), until it was no longer detectable at P0 (Figure III-3N-P). Sox2 is also expressed in upper incisors and molars (Juuri et al., 2013b; Li et al., 2015) or Shh Cre -Sox F/F mice exhibit molar defects (Juuri et al., 2012b). We found that the upper incisors and molars in Sox2 cko embryos were smaller and associated with delayed invagination (Figure III-4A-D). At E16.5, Sox2 cko molars also exhibited an abnormal shape and upper incisors showed a delay in rotation (Figure III-4C). At P0, Sox2 cko molars lacked cusps (Figure III-4D). Taken together, these data indicate a role for Sox2 in stem cell maintenance and loss of tooth morphogenesis may be due to a reduction in stem cells required for tooth growth. Sox2 was also conditionally deleted using either Shh Cre and K14 Cre but we found no obvious incisor defects with either Cre, although K14 Cre -Sox2 F/F mice exhibited a mild molar defect (Figure III-5). This defect included an expanded dental lamina starting at E13.5 (Figure III-5B, 3-5E, 3-5H, 3-5K, 3-5N, 3-5Q) and absence of the 3 rd molar (data not shown). Interestingly, the pattern of Shh and Fgf4 expression was expanded in E14.5 K14 Cre -Sox2 F/F molars (Figure III-5T and 3-5V), suggesting that Sox2 may repress Shh and Fgf4 expression. Sox2 regulates incisor growth in adult mice Sox2 cko mice die at birth, similar to other Sox2 conditional knockout mice (Juuri et al., 2012b; Zhang et al., 2012). To determine whether Sox2 plays a role in adult incisor 72

96 growth, Sox2 F/F mice were crossed with Rosa26 CreERT mice in which Cre expression can be induced by tamoxifen. After treatment with tamoxifen, we injured left lower incisors of control and Rosa26 CreERT -Sox2 F/F mice (Figure III-6A,B). Five days after injury, incisors of tamoxifen treated control mice grew to a length comparable to the uninjured right lower incisor. In contrast, incisors of tamoxifen treated Rosa26 CreERT -Sox2 F/F mice exhibited an approximate 50% decrease (p< 0.01) of tooth growth compared to tamoxifen treated control mice (Figure III-6B, C). We confirmed that Sox2 expression was ablated in tamoxifen-treated mice (Figure III-6B). Sox2 ablation in DESCs leads to reduced stem cell renewal To understand the diminished tooth growth in Sox2 cko embryos, we next examined proliferation of cells in the LaCL. Immunofluorescent staining of Ki67 in E16.5 Sox2 cko embryonic incisors showed decreased stem cell proliferation in the LaCL compared to control mice (Sox2 F/F ), consistent with the small LaCL in these mice (Figure III-7A-D). Quantitative analysis indicated that the percentage of Ki67 positive cells was decreased by 40% in Sox2 cko LaCLs (Figure III-7E), suggesting that stem cell proliferation was inhibited by loss of Sox2. As a control, we detected no change in cell proliferation in Pitx2 Cre /Sox2 F/+ mice (Figure III-8A-C), indicating that the defect is not due to the Pitx2 Cre construct. To determine if altered progenitor cell migration could contribute to reduced incisor growth in Sox2 cko embryos, we performed thymidine analogue double labeling in control and Sox2 cko mouse mandibles. In this experiment, highly proliferative cells were sequentially incorporated two different thymidine analogues (CldU and IdU), which allow us to observe and quantify two successive rounds of cell division and migration. 73

97 One hour after IdU injection, most of the dental epithelial cells in the LaCL and dental mesenchyme cells near the neural vascular bundle, were labeled with IdU in control embryos. However, there were relatively few IdU + cells in Sox2 cko LaCL, indicating a decrease in dental stem cell proliferation in Sox2 cko LaCL (Figure III-7F and G, green color). 24 hr after injection of CldU, both the dental epithelial and mesenchymal cells in the proximal region and distal region were labeled with CldU (Figure III-7F and G, red color), but there were less CldU + cells in distal tip of Sox2 cko incisor compared with WT, suggesting DESC migration was affected by loss of Sox2 (Figure III-7F, G and H). It is possible that reduced growth in the Sox2 cko lower incisor (LI) was caused by increased cell death. TUNEL staining revealed no obvious cellular apoptosis in either Sox2 F/F or Sox2 cko incisors (Figure III-8D-I). Similarly, immunohistochemistry staining for cleaved caspase-3, which labels the cells undergoing early cell death, failed to detect cleaved caspase-3 positive cells in either Sox2 F/F incisors or Sox2 cko incisors (Figure III- 8J and K). Take together, these data suggest that Sox2 primarily regulates DESC cell proliferation and subsequent migration in the LaCL, but loss of Sox2 in DESCs does not affect the rate of cell death Sox2 and Lef-1 epithelial expression domains are juxtaposed in the mouse oral epithelial dental placode The arrested tooth phenotype in the Sox2 cko mice resemble that of Lef-1 null mice, in which dental development is arrested at the late bud stage. To determine if Sox2 and Lef- 1 may interact, we analyzed Sox2 and Lef-1 protein expression in E.11.5 wild type, Sox2 F/F and Sox2 cko incisors at dental placode stage (Figure III-9A). Lef-1 and Sox2 were 74

98 both expressed in the oral epithelium and dental placode of both lower incisors and upper incisors of E11.5 wild-type embryos (Figure III-9B,C). However, Lef-1 expression was detectable in the anterior regions of the UI and LI, whereas Sox2 expression was detectable in the posterior region of the UI and LI incisor (Figure III-9D) (Li et al., 2012b). Both were expressed in the dental incisor placodes but in separate and distinct regions (Figure III-9E). At E12.5, the pattern of Sox2 and Lef-1 expression were similar to E11.5 as the dental epithelium invaginates in the surrounding mesenchyme (Figure III- 9F). As expected, E11.5 Sox2 F/F embryos were indistinguishable from wild type (Figure III-9E and 9H-J). Although Sox2 expression was undetectable in Sox2 cko embryos at E11.5, Lef-1 expression was not affected (Figure III-9L-N). The incisor placodes in Sox2 cko E11.5 embryos also showed delayed epithelial thickening at this stage (Figure III-9K). These results suggest that Sox2 and Lef-1 may act independently of each other to regulate incisor development. Conditional overexpression of Lef-1 creates a new LaCL stem cell niche and abnormal tusk-like incisors To determine the effect of Lef-1 continuous expression during incisor development we generated a Lef-1 conditional overexpression mouse. Lef-1 full-length isoform construct was preceded by a loxp flanked STOP and inserted into the Rosa26 loci, to make a Cre responsive Lef-1 overexpression mouse (Lef-1 cki )(Figure III-10A). Lef-1 conditional knock-in (Lef-1 cki ) mice were crossed with Pitx2 Cre mice to drive the overexpression of Lef-1 in the dental and oral epithelium (Figure III-10A). Lef-1 immunostaining confirmed the overexpression of Lef-1 in the dental epithelium of the Pitx2 Cre -Lef-1 cki overexpression embryos (conditional overexpression Lef-1, COEL) 75

99 (Figure III-11C). The COEL mice developed long, thick tusk-like incisors compared with Lef-1 cki mice (control) (Figure III-10B and C). Microcomputed tomography (µct) analysis of 3-month-old mice revealed that both upper and lower incisors in COEL mice undergo rapid growth compared to Lef-1 cki mice (Figure III-10D and E). The LaCL in the lower incisors of E16.5 COEL embryos was larger than control and included a new cell compartment forming on the labial side (Figure III-11A). At P0, the sizes of control and COEL lower incisors were comparable, but an extra branch of the LaCL was detectable in COEL embryos (Figure III-11B). To determine if the cells in the branched LaCL of COEL mice are dental epithelial stem cells, we analyzed Sox2 expression in P1 control and COEL LI LaCL regions. The LaCl and the branched region both contained Sox2 positive stem cells (Figure III-10F and G), suggesting that overexpression of Lef-1 resulted in a new stem cell niche. 3D reconstruction showed the structure of the expanded LaCL with multiple layers and branched stem cell niche (Figure III-10H-K). Previous studies on tusk-like incisors in Spry4 -/- /Spry2 +/- mice (Klein et al., 2008b) revealed that over growth of the LI could be due to ectopic deposition of lingual enamel. To test if Lef-1 overexpression produced ectopic enamel formation, we performed amelogenin immunostaining in E18.5 embryos and found elevated amelogenin expression in the labial side of COEL LIs, but no ectopic amelogenin expression in lingual side of COEL incisors (Figure III-11D). We next asked whether increased stem cell proliferation in the LaCL could contribute to the increased growth of incisors in COEL mice. BrdU labeling of cells in the E18.5 COEL LaCL showed a 15% increase in progenitor cell proliferation compared to the control LaCL (Figure III-10L and M). However, at P1, proliferation in the LaCL 76

100 regions was similar between control and COEL neonates (Figure III-10N), but the total number of stem cells was increased 20% in the COEL LaCL(Figure III-10O). Most of the cells in the COEL branched LaCL region were not labeled with BrdU, suggesting that these cells are quiescent, not undergoing proliferation (Figure III-10N). These data indicate that overexpression of Lef-1 in the incisor results in increased cell proliferation at embryonic stages and formation of a new stem cell niche in the LaCL, which facilitates incisor growth and the formation of tusk-like incisors. Lef-1 overexpression rescues tooth arrest in Sox2 cko embryos Because both Sox2 and Lef-1 appear to control dental epithelial stem cell renewal, maintenance and proliferation we next asked if Lef-1 overexpression could rescue the tooth arrest in Sox2 cko embryos. To test our hypothesis, we crossed Sox2 cko mice with COEL mice to produce Pitx2 Cre /Sox2 F/F /Lef-1 cki mice. In these mice Sox2 is ablated and Lef-1 is overexpressed using the Pitx2 Cre. At E18.5, the Sox2 F/F /Lef-1 cki control embryos developed well-formed late bell stage incisors (Figure III-12A). The E18.5 Sox2 cko embryos have a remnant of the LI at this stage (Figure III-3K, L). In Pitx2 Cre /Sox2 F/F /Lef- 1 cki mice, the lower incisors were detectable with their dental lamina, however the LI is positioned at the anterior region of the mandible and smaller in size (Figure III-12B). The forward positioning of the LI may be due to oral adhesions that remain present. The LaCL in Pitx2 Cre /Sox2 F/F /Lef-1 cki mice is smaller compared with control embryos (Figure III-12B). In the labial side of control embryos, the lower incisors develop three cell layers: odontoblasts (OD), ameloblasts (AM) and the stratum intermedium (SI). However, there was only one layer of cells in Pitx2 Cre /Sox2 F/F /Lef-1 cki embryos, suggesting that the differentiation was blocked (Figure III-12A, B). We also examined P0 incisors and 77

101 found similar phenotypes (Figure III-13). To confirm the effect on differentiation in Pitx2 Cre /Sox2 F/F /Lef-1 cki embryos, we analyzed ameloblast differentiation by immunofluorescence using an antibody against amelogenin, a marker for differentiated dental epithelial cells in E18.5 lower incisors. In control mice (Sox2 F/F /Lef-1 cki ), amelogenin was expressed on the labial side (Figure III- 12C, C ). However, amelogenin was ectopically expressed on the lingual side of the LI in Pitx2 Cre /Sox2 F/F /Lef-1 cki embryos (Figure III-12D, D ). Lef-1 is highly expressed in the lingual cervical loop (LiCL) at E18.5 due to Pitx2 Cre activity in these mice (Figure III- 12F). In control incisors, Lef-1 expression is not detectable in the lingual dental epithelial cells at E18.5. We found that amelogenin is expressed normally on the labial side in COEL mice, whereas it switched to the lingual side when Sox2 is ablated in COEL embryos. We speculate that the interplay between Sox2 and Lef-1 with other factors specifies asymmetric amelogenin expression. To determine if progenitor cell proliferation is rescued and contributes to the development of the LI in Pitx2 Cre /Sox2 F/F /Lef-1 cki mice, we analyzed proliferation at E16.5. Proliferation was dramatically reduced in the LI LaCL region in Sox2 cko mice, but it was restored in Pitx2 Cre /Sox2 F/F /Lef-1 cki mice (Figure III-12G, H). Thus, Lef-1 overexpression reversed the proliferation defect of Sox2 cko mice and could be the mechanism contributing to the rescue of incisor development. Taken together, overexpression Lef-1 in Sox2 ablated incisors partially rescues the tooth arrest by promoting stem cell maintenance and cell proliferation. However, Lef-1 over expression cannot rescue ankylglossia or cleft palate in the Sox2 cko embryos. 78

102 Sox2 attenuates Pitx2 transcriptional activation of Lef-1, Sox2 and Pitx2 through direct protein interactions. To determine a molecular mechanism for the Sox2-Lef-1 effects on incisor development we focused on the transcriptional activities of these factors in concert with Pitx2. Pitx2 is the first transcriptional marker of tooth development and tooth development is arrested at E12.5 in Pitx2 null mice (Liu et al., 2003b; Lu et al., 1999a). We generated K14 Cre / Pitx2 F/F mice and tooth development was arrested at the bud stage (unpublished data). Also we previously showed that Pitx2 regulates Lef-1 expression during odontogenesis (Amen et al., 2007; Vadlamudi et al., 2005). RNA-sequence data showed that Sox2 expression was upregulated in the mandibles of Pitx2 overexpressing mice and downregulated in the mandibles of Pitx2 null mice (unpublished data). Sox2 expression was increased in oral epithelial cells reprogrammed to dental epithelial cells by overexpressing Pitx2 and mir-200c (Sharp et al., 2014b). Sequence analyses of the Sox2 promoter identified one Pitx2 binding element located in the 5 UTR and another in a distal element 845 bp upstream of the transcription start site (TSS) (Figure III-14A). Chromatin-immunoprecipitation (ChIP) showed Pitx2 binding to both elements but not a sequence upstream of the distal site used as a control, with a 5-fold enrichment of Pitx2 binding at the upstream site (Figure III-14B, C). We also identified a Sox2 putative binding element 1972 bp upstream of the Sox2 TSS and ChIP assay showed Sox2 bound to this element (Figure III-14D-F). The Sox2 promoter (4.0kb) was cloned into a luciferase vector and transfected into LS-8 oral epithelial cells to test for regulation by Pitx2 and Sox2. Pitx2 activated the Sox2 promoter 15-fold compared to empty vector and Sox2 activated its own promoter 3-fold 79

103 (Figure III-14G). However, Sox2 repressed Pitx2 activation of the Sox2 promoter from 15-fold activation to 5-fold (Figure III-14G). This repression of Pitx2 transcriptional activity was not promoter specific as Sox2 repressed Pitx2 activation of the Lef-1 and Pitx2 promoters (Figure III-14G). To confirm this regulation, we measured endogenous levels of Sox2, Lef-1 and Pitx2 in LS-8 cells transfected with empty vector (pcdna3.1), Pitx2 or Sox2. Endogenous expression of Sox2 in differentiated oral epithelial cells was low, as expected (Figure III-14H). However, cells transfected with Pitx2 showed increased levels of Sox2 and Lef-1 transcripts (Figure III-14H). Similarly, cells transfected with Sox2 showed small increases in the level of both Lef-1 and Pitx2 (Figure III-14H). Thus, Sox2 appears to modulate the expression of both Lef-1 and Pitx2. Finally we examined whether Pitx2 and Sox2 interact using immunoprecipitation (IP) assays (Figure III-15A). We showed previously that Pitx2 strongly interacts with HMG (high mobility group) domains (Amen et al., 2008b; Li et al., 2013). Sox2 contains an HMG domain and we hypothesized that Pitx2 might bind to Sox2 through this domain. Several GST-Sox2 truncated constructs were tested for Pitx2 binding in a pull-down assay (Figure III-15B-D). Based on our results, we conclude that there are two Pitx2 binding domains in Sox2, the HMG domain and the C-Terminus (Figure III-15B). This is similar to the two Pitx2 binding domains in the Lef-1 protein, the HMG and C-Terminal activation domain (CAD) (Amen et al., 2007). Taken together, Sox2 protein and Pitx2 protein interact to repress Pitx2 transcriptional activity provides a new mechanism to establish the juxtaposed expression domains of Sox2 and Lef-1 during tooth development. Pitx2 is expressed throughout the oral epithelium and dental placode. In posterior region of dental placode, Pitx2 activates 80

104 Sox2 and the Pitx2/Sox2 complex inhibits Lef-1 expression. In the anterior region of the dental placode, the absence of Sox2 allows Pitx2 activation of Lef-1. A working model demonstrates this new transcriptional regulation and expression of Sox2 and Lef-1 (Figure III-16). At later stages of incisor development in the LaCL, Lef-1 is not expressed in the dental epithelium but shifts the adjacent mesenchyme. We demonstrate a Pitx2- Sox2-Lef-1 regulatory mechanism plays a major role in maintaining the stem cell niche and promoting stem cell proliferation (Figure III-16). Discussion In this study we identified a role for Sox2 in the oral epithelium as a regulator of periderm formation. Two genes commonly associated with ankyloglossia are TBX22 and IRF6 and patients with mutations in these genes also have cleft palate and dental anomalies (Blackburn et al., 2012; Braybrook et al., 2001; Kondo et al., 2002; Pauws et al., 2009a). Previous studies have reported that TBX22 mutations are associated with a human birth defect called X-linked cleft palate and ankyloglossia (CPX; MIM303400), in which patients develop ankyloglossia and cleft palate (Kantaputra et al., 2011; Pauws et al., 2009a). IRF6 is a regulator of periderm formation to that functions to prevent epithelial adhesions during embryogenesis (de la Garza et al., 2013; Peyrard-Janvid et al., 2014; Richardson et al., 2014b). This is elaborated by Irf6 mutant mice which exhibit epithelial adhesions due to a failure of periderm formation, leading to cleft palate and ankyloglossia. We found that Keratin 6, a marker for periderm, was reduced in the oral epithelium in Sox2 cko embryos, causing intraoral fusions and leading to cleft palate and the fusion of the tongue epithelium with the mandibular oral epithelium and dental epithelium. The fusion of the tongue epithelium with the dental epithelium appears to 81

105 affect lower incisor development, reducing dental epithelial invagination but not causing evagination as reported for Irf6 mutant mice (Blackburn et al., 2012). Previous studies have shown an early role for Sox2 in the specification of DESCs and Sox2 + cells mark adult DESCs(Juuri et al., 2012b; Li et al., 2015). Furthermore, conditional deletion of Sox2 using Shh Cre results in abnormal epithelial growth in mouse molars (Juuri et al., 2013b), consistent with what we observed in K14 Cre -Sox2 F/F embryos. However, by using the earliest dental epithelium marker, Pitx2 Cre, to ablate Sox2 in the dental and oral epithelia, we found that loss of Sox2 leads to incisor developmental arrest at E16.5 with a complete disintegration at P0 and abnormal molar growth. We propose that this is mainly due to the failure to maintain the DESC niche at an early stage resulting in a lack of proliferative cells and gradual loss of the incisor tooth germ. The disappearance of the tooth germ was not due to increased apoptosis but to a lack of stem cell renewal to replenish proliferating progenitor cells. These data correlate well with the role of Sox2 in the specification of the stem cell niche (Juuri et al., 2012b). Furthermore, loss of Sox2 during adult incisor growth resulted in a reduction of incisor regeneration, renewal and growth, demonstrating an essential role for Sox2 in maintaining the adult DESC niche. We show that Lef-1 can control DESC renewal but also establish a new progenitor cell compartment in the mouse lower incisor LaCL. Lef-1 overexpression enhanced DESC production and promoted stem cell growth but also produced a new region of mitotically inactive stem cells in the LaCL. Lef-1 controls stem cell growth, self-renewal of embryonic stem cells and establishes stem and progenitor cell compartments in mouse epidermis and hair follicles (Lowry et al., 2005; Petersson et al., 2011). In our study, Lef- 82

106 1-induced DESCs also expressed Sox2 and Pitx2 (data not shown for Pitx2). Lef-1 overexpression changed the morphology of the incisor LaCL, dramatically increasing growth of the incisor by stimulating DESCs to proliferate and subsequently differentiate. Lef-1 overexpression did not cause ankyloglossia, epithelial cell adhesion defects or cleft palate and the mice live to adulthood. While some aspects of defective Wnt signaling or conditional deletion of β-catenin leads to impaired palatal development, either over expression or deletion of Lef-1 did not cause cleft palate (He et al., 2011). These data suggest that Lef-1 acts preferentially to instruct odontogenesis and not palatogenesis. Lef-1 is endogenously expressed in the cervical loop of enamel organ at E18.5 and contributes to preameloblast cell differentiation (Sasaki et al., 2005a). Several reports demonstrate a role for Lef-1 in maintaining self-renewal of stem cells in other organs (DasGupta and Fuchs, 1999; Huang and Qin, 2010; Lowry et al., 2005; Petersson et al., 2011; Reya and Clevers, 2005). Consistent with this, we found that Lef-1 overexpression rescued tooth development in Sox2 cko embryos. In these mice, we found amelogenin expression aberrantly expressed in the lingual epithelial cell layer. Normally, epithelial stem cells differentiate into ameloblast cells only on the labial side, which express amelogenin and secrete the organic matrices of enamel (Tummers and Thesleff, 2003a; Wang et al., 2004). This complete switch of amelogenin expression to the lingual epithelial cells is similar to other mouse models. Misregulation of BMP signal regulators can cause ameloblast differentiation defects. For example, overexpression of Noggin or Follistatin using the K14 promoter disrupts ameloblast differentiation on the labial side of the incisor (Wang et al., 2007a; Wang et al., 2004), whereas lack of Follistatin causes both sides of the incisor to develop functional ameloblasts that secrete enamel. K14-83

107 Noggin transgenic mice that over-express Noggin in the dental epithelium have overgrown incisors with repressed ameloblast differentiation and increased epithelial proliferation in both the labial and lingual cervical loops (Wang et al., 2007a). Ectopic FGF expression in the lingual side of incisors resulting from ablation of Sprouty genes leads to ectopic deposition of enamel on the lingual side (Klein et al., 2008b; Klein et al., 2006). Although the COEL embryos do not display this effect, the lack of Sox2 expression in the incisor tooth germ appears to regulate other factors that may interact with Lef-1 to regulate its transcriptional activity. Interestingly, Noggin overexpression inhibits Wnt/β-catenin signaling causing early arrested tooth development due to loss of Pitx2 expression (Yuan et al., 2015). In these mice, Wnt/β-catenin signaling disruption was rescued by transgenic activation of Pitx2 and rescued tooth development. Sox2 can have both inductive and repressive transcriptional effects on Lef-1 promoter activities dependent on other factors to specify progenitor cell populations during early and late submucosal gland development (Xie et al., 2014). In tooth development, we have identified that the juxtaposed expression of Sox2 and Lef-1 in the dental placode may be coordinated by Pitx2 expression. We have identified a molecular mechanism where Pitx2 activates Pitx2, Sox2 and Lef-1 expression in the dental epithelial stem cells and these activations can be abolished by Pitx2/Sox2 protein complex. The identification of Sox2 protein interactions with Pitx2 resulting in the repression of Pitx2 transcriptional activation of many target genes provides a model for the role of Sox2 in maintaining the dental stem cells and inhibiting differentiation of these cells. Collectively, our study reveals a Pitx2:Sox2:Lef-1 pathway in regulating DESC maintenance and proliferation and this finding may provide novel molecular approaches 84

108 for tooth regeneration. 85

109 Figure III-1 Sox2 expression in the dental epithelial stem cell (DESC) niche and Pitx2Cre/Sox2 F/F (Sox2 cko ) ablates Sox2 expression in the dental and oral epithelia. A) Immunohistochemistry staining of sagittal sections of WT E18.5 heads using Sox2 antibody reveals that Sox2 is expressed in the lateral ventricle (LV), nasal epithelium (NE), tongue epithelium (TN), oral epithelium (OE) and lower incisor (Li). Nuclei are counterstained with hematoxylin. B, C) are magnified views of the blue and green-boxed regions, respectively in (A). B) Sox2 expression in the OE and TN, C) shows Sox2 expression is localized in the labial cervical loop (LaCL) of the mouse lower incisor. D, E) expression of the Pitx2 Cre at E11.5 and E14.5, respectively. F-M) immunofluorescent staining of Sox2 in the lower incisor LaCL, oral epithelium and lateral ventricles of Sox2 F/F (Control) and Pitx2 Cre /Sox2 F/F (conditional knock-out, Sox2 cko ) mice verified the specificity of the deletion of Sox2 by the Pitx2 Cre. Nuclei are counterstained with DAPI. MD, mandible; MX, maxilla; DE, dental epithelium; DP, dental placode, Scale bars, 100 µm. 86

110 Figure III-2 Sox2 cko embryos develop abnormal oral adhesions and present with ankyloglossia and cleft palate. A, B) H&E staining of E18.5 Sox2 F/F (control) and Sox2 cko heads shows ablation of Sox2 leads to ankyloglossia in mice. Scale bars, 1000µm. C, D) Higher magnification views of boxed region in (A) and (B). Arrows in C and D highlight the region where the tongue is tethered to the mandible in Sox2 cko mice. Scale bars, 100 µm. E, F) E-Cadherin immunostaining in sagittal sections of E15.5 Sox2 F/F and Sox2 cko heads. G, H) Higher magnification of boxed region in A and B displays abnormal adhesion among tongue epithelium, oral epithelium and dental epithelium in Sox2 cko mice compared with WT. The dental lamina is fused with the lower incisor tooth germ in the Sox2 cko. Nuclei are counterstained with hematoxylin. TN, Tongue; Li, lower incisor; TE, tongue epithelium; OE, oral epithelium; DE, dental epithelium; DL, dental lamina. Scale bars, 100 µm.i, J) Periderm formation was reduced in Sox2 cko mice as shown by immunofluorescent staining of a periderm marker, Keratin 6 in a sagittal plane of E15.5 control and Sox2 cko heads. Nuclei are counterstained with DAPI. Scale bars, 100 µm. K, L) Magnified views of (I) and (J), respectively. Scale bars, 100µm. M) Coronal section of E16.5 Pitx2 Cre and Sox2 cko heads and arrowheads indicate cleft palate in Sox2 cko embryos, Scale bars, 100 µm. Mx, Maxilla; Md, mandible, Tn, Tongue. 87

111 Figure III-3 Loss of Sox2 in murine embryos causes tooth arrest. A) Schematic profile of the adult mouse incisor ((Biehs B et al., 2013) with modifications). The mouse lower incisor comprises a major portion of the mandible. Boxed region depicts the LaCL containing progentitor cells in the stellate reticulum (SR) and the inner and outer enamel epithelium (IEE and OEE). Ameloblasts (Am) only appear in the labial side and cause asymmetrical deposition of enamel on labial surface. Dentin (De), produced by odontoblasts (Od), is deposited on both labial and lingual side. B-G) H&E staining of E12.5, E14.5 and E16.5 embryos (sagittal sections). At E12.5, the tooth bud in Sox2 cko embryos (C) is smaller compared with control embryos (B). At E14.5 and E16.5, the incisors in Sox2 cko embryos (E, G) are smaller in size with an underdeveloped LaCL and the incisors position towards the anterior region of the mandible, compared with control littermates (D, F). H-P) H&E staining of E17.5, E18.5 and P0 embryos (sagittal sections). At E17.5 (H), E18.5 (K), and P0 (N) control embryos developed well-formed late bell stage incisors (dotted box). However, Sox2 cko embryos (I, L) only have a remnant of the lower incisor. At P0 the lower incisor is completely absent in Sox2 cko mice (M). J,M,P) higher magnification of boxed regions in (I,L,O) shows the remnant of incisors. DM: dental mesenchyme; En: enamel; LaCL, labial cervical loop; LiCL: lingual cervical loop; SI, stratum intermedium. TA: transient amplifying. Scale bars in B-G, J, M and P represent 100µm. Scale bars in H. I. K. L M and O represent 1000µm. 88

112 Figure III-4 Molar and upper incisor development is impaired in Sox2 cko embryos. A) H&E staining of E14.5 sagittal section of upper incisors (UIs) shows that Sox2 cko UIs are smaller with incomplete invagination compared with control (Sox2 F/F ). Boxed areas are shown as magnified images on the right. B) E14.5 lower molars (LMs) (sagittal section) stained by H&E indicate Sox2 cko embryos have smaller LMs with incomplete invagination compared with controls. Boxed areas are shown as magnified images on the right. C) H&E staining of E16.5 UIs and molars show a smaller size and abnormal shape in Sox2 cko embryos compared with controls. D) P0 molars and upper incisors stained with H&E. Sox2 cko molars are smaller and hypoplastic compared with control molars. Upper incisors are also smaller in Sox2 cko mice. Scale bars, 100µm. 89

113 Figure III-5 Sox2 ablation with the K14 Cre causes abnormal molar formation and changes in Shh and Fgf4 expression. A-F) At E13.5, invagination of the lower molar tooth bud was displaced with a wide dental lamina and tooth bud structure (red line) in the K14 Cre /Sox2 F/F (E, F), compared to controls (B, C). G-L) At E14.5, invagination was affected and the dental lamina was significantly larger in K14 Cre /Sox2 F/F embryos compared to controls. M-R) At E15.5, the invagination defect persisted with abnormal formation of the molar in the K14 Cre /Sox2 F/F embryos. Interestingly, as in the incisors, Sox2 was asymmetrically expressed (C, I, O). S, T) Shh transcripts were detected in the WT enamel knot structure of the E14.5 lower molar, however the Shh expression domain was expanded in the K14 Cre /Sox2 F/F lower incisor, as the Shh domain moved towards the presumptive Sox2 region. U, V) Fgf4 expression (transcripts) was slightly increased in the K14 Cre /Sox2 F/F lower molars. (This figure was done by our collaborator Dr. Zhi Chen and his students Dr. Huan Liu and Li Zhang from Wuhan University, Wuhan, China) 90

114 Figure III-6 Deletion of Sox2 in adult mice inhibits incisor regeneration. A) Tamoxifen treatment, tooth length reduction and histological analysis time line in Sox2 F/F control and Rosa26 Cre-ERT /Sox2 F/F mice. B) Images of P29 and P34 mouse incisors after tamoxifen treatment and cutting of the left lower incisor. Left: Mouse incisors after cutting half of the left lower incisor. Middle: Mouse incisors five days after shortening. Right: Microcomputed tomography (µct) analysis of shortened incisors after 5 days of recovery. Sox2 expression visualized by immunofluorescence in P0 mice after tamoxifen treatment in pregnant females C) Quantitation data of the relative growth rate of the shortened incisor in Sox2 F/F and Rosa26 Cre ERT /Sox2 F/F mice after treatment with tamoxifen. Scale bars in µct pictures represent 5 mm. 91

115 Figure III-7 Sox2 regulates dental epithelial stem cell renewal and cell migration. A, B) Immunofluorescence staining of the proliferation marker, Ki67 in sagittal sections of E16.5 Sox2 F/F and Sox2 cko mouse incisors. C, D) High magnification view of boxed regions in A and B. E) Quantitation of the ratio of Ki67 positive cells to total cells, N=3. F-G) DESC cell migration measured using two different labels (CldU & IdU), were injected and measured after 24 and 1 hour, respectively. The red label-retaining cells (CldU) mark migrating cells and the green label-retaining cells (IdU) mark recently mitotic cells. The arrows indicate the region of neural vascular bundle. F -F ) and G - G ) Higher magnification of the white boxed area in F and G respectively to highlight the progenitor cell migration at the distal tip of the incisors. H). Quantification of CldU + cells in the epithelial tissue of boxed region of F and G. PR: proximal; DI: distal; LaCL: labial cervical loop. Scale bars, 100 µm. 92

116 Figure III-8 Loss of dental epithelial stem cells in Sox2 cko embryos does not involve increased apoptosis. A-B) Control experiments show that proliferation of the dental epithelial cells in the LaCL was not affected in the P0 Sox2 F/+ or Pitx2 Cre /Sox2 F/+ embryos. C) Quantitation of the Ki67 stained cells between the Pitx2 Cre /Sox2 F/+ embryos and controls shows no difference in proliferation. D-I) TUNEL staining in E15.5 (sagittal section) mouse lower incisors reveal no changes in apoptosis in Sox2 cko embryos. Magnified view of boxed regions in D, E and F were shown in G, H and I respectively. No apparent apoptosis signal was detected in control and Sox2 cko incisors. (F, I) are positive controls for TUNEL staining. N=3. J, K) Immunohistochemistry staining of the early apoptosis marker, cleaved Caspase 3 in sagittal sections of E15.5 mouse incisors showing no detectable level of apoptosis in control and Sox2 cko incisors, N=3. Nuclei are counterstained with DAPI. Scale bars, 100 µm. 93

117 Figure III-9 Sox2 and Lef-1 expression domains are juxtaposed in the dental placode and oral epithelium. A) Schematic diagram of E11.5 mouse head shows upper and lower dental placode. B, C) Sox2 and Lef-1 immunofluorescence staining in E11.5 WT embryos, respectively. Arrows indicate position of the lower incisor dental placode and upper incisor dental placode. D) Merged photo of Sox2 and Lef-1 staining in E11.5 WT showing the juxtaposed expression domains of these two factors. E) Higher magnification of panel C. F) Double immunofluorescence staining of Sox2 (red) and Lef-1 (green) in E12.5 WT embryos, showing specific expression domains of each factor in the developing incisor tooth buds. G and K) H &E staining of E11.5 Sox2 F/F and Sox2 cko embryos, arrows point to the LI and UI dental placodes. H-J) and L-N) Sox2 expression is absent in the Sox2 cko embryos, however Lef-1 expression is not affected compared to controls (high magnification of boxed region in G and J). Arrows in H and L highlight the Lef-1 expression. Mx, Maxilla; Md, mandible; UI, upper incisor; LI, lower incisor; DL, dental lamina. Scale bars, 100 µm. 94

118 Figure III-10 Conditional overexpression of Lef-1 (COEL) results in tusk-like incisors, creation of a new stem cell compartment in the LaCL and increased dental epithelial stem cell proliferation. A) Schematic of the Lef-1 conditional overexpression cassette used to generate the COEL mouse. B, C) Heads showing overgrown incisors from 3 month-old Lef-1 cki (Lef-1 conditional knockin, control) and Pitx2 Cre / Lef-1 cki/cki (Lef-1 overexpression, COEL) mice. D, E) µct analysis of the mouse heads in B and C. Both the upper and lower incisors are overgrown. F, G) Immunofluorescent staining of Sox2 in the LaCL of P1 control and COEL incisors. Nuclei are counterstained with DAPI. Note the extra stem cell compartment in the LaCL of COEL mice. H-K) A 3D composite of the LI LaCL is shown and compares WT (Lef-1 cki ) to the COEL LI at P2. The branched stem cell compartment is shown and the LaCL of the COEL embryos appears to contain multiple layers of cells. L) BrdU labeling of the E18.5 control and COEL incisors. Nuclei are counterstained with hematoxylin. M) Quantitation of the percentage of BrdU + cells in the E18.5 LaCL of control and COEL incisors, N=3. Note, a 15% increase of the ratio of BrdU + /total cells in COEL LaCL compared with control. N) BrdU labeling of P1 control and COEL lower incisors. O) Quantification of BrdU positive compared to BrdU negative cells as a comparison of all cells in the LaCL. Note that the ratio BrdU + / total cells are similar between control and COEL incisors (~60%), but the total stem cell count in the COEL LaCL are increased 20% compared to control incisors, N=3. Nuclei are counterstained with hematoxylin. LaCL: labial cervical loop. Scale bars, 100 µm, *p<0.05. (The Lef-1 conditional overexpressing mice were generated by Dr. Adam 95

119 Dupuy, Dr. Michael J. Goodheart and Traci Neff from University of Iowa, Department of Anatomy and Cell Biolgoy.) 96

120 Figure III-11 Overexpression of Lef-1 in dental and oral epithelia creates a new LaCL stem cell compartment and increases amelogenin expression. A, B) H&E staining of lower incisors (sagittal sections) shows an enlarged LaCL in E16.5 COEL incisors compared with control (Lef-1 cki ). The arrowhead denotes the separation of a new stem cell compartment from the normal cervical loop structure in the COEL LI. A newly formed stem cell compartment (arrowhead) that has branched from the LaCL was observed in P0 COEL incisors (N=3). C) Lef-1 immunofluorescence staining in E18.5 lower incisors shows Lef-1 was overexpressed in the LaCL and LiCL of the COEL embryos. D) Amelogenin immunofluorescence staining in E18.5 lower incisors indicates increased amelogenin expression in COEL incisors compared with control incisors. Nuclei are counterstained with DAPI. Tn, tongue. Scale bars, 100µm. 97

121 Figure III-12 Overexpression of Lef-1 rescues tooth arrest in Sox2 cko embryos. A, B) H&E staining of sagittal sections of E18.5 Sox2 F/F /Lef-1 cki (control) and Pitx2 Cre /Sox2 F/F /Lef-1 cki (rescue) mandible and maxilla. A defect in the differentiation of the labial dental epithelial cells (AM and SI) and adjacent odontoblast (OD) cells was observed in the rescue incisors as highlighted in the blue boxes. Red boxes show LaCL is present in rescue embryos but much smaller compared to control incisors. C, D) Amelogenin immunofluorescent staining in E18.5 control and rescue incisors revealed that amelogenin expression was switched to the lingual side in the rescue incisor compared with labial expression in the control. C, D ) Higher magnification of the boxed region in C and D to show the expression pattern of amelogenin in control and rescue incisors. E, F) Sox2 and Lef-1 double immunostaining in E18.5 control and rescue incisors. G) Proliferation as determined by the Ki67 marker was reduced in the Sox2 cko E16.5 LaCL but was activated in the rescue embryos compared to controls. H) Proliferation of the cells in the LaCL was quantitated by comparison of Ki67 stained cells vs. DAPI stained cells. Nuclei are counterstained with DAPI. AM, ameloblast; OD, odontoblast; LaCL, labial cervical loop; SI, stratum intermedium. Scale bars, 100 µm. 98

122 Figure III-13 Lef-1 overexpression rescues lower incisor development in the Sox2 cko embryo. A-C) H&E staining (sagittal sections) of P0 Sox2 F/F (control), Sox2 cko and Pitx2 Cre /Sox2 F/F /Lef-1 cki (Lef-1 rescue of Sox2 cko ) embryos. D-F) High magnification of black boxed regions in A, B and C show their respective lower incisor. Note the absence of a lower incisor in the Sox2 cko mice. Lower incisors were observed in rescue mice, albeit they were smaller. D, F ) Higher magnification of the blue-boxed LaCL (D) shows cells in the stellate reticulum (stem cells) surrounded by the polarized outer and inner enamel epithelium in the control (D ). However, in the rescue mice a LaCL is formed but contains few stem cells and the outer and inner enamel epithelial cells are not well polarized (blue box, F ). D, F ) Higher magnification of the green box in D and F. The rescue LI (F ) show lack of differentiation on the labial side and do not contain the differentiated odontoblasts, ameloblasts or stratum intermedium observed in the control incisors (D ). AM, ameloblast; OD, odontoblast; LaCL, labial cervical loop; SI, stratum intermedium; OEE, outer enamel epithelium; IEE, inner enamel epithelium; DL, dental lamina. Scale bars, 100 µm. 99

123 Figure III-14 Endogenous Pitx2 and Sox2 bind to elements in the Sox2 promoter. A) Schematic of the Sox2 promoter region displaying two putative Pitx2 binding sites (216bp downstream transcriptional start site (TSS) and 845bp upstream of the TSS), the consensus sequences are underlined. B) ChIP assays were performed in LS-8 cells and PCR products were resolved on agarose gels. Pitx2 antibody (Ab) immunoprecipitated (IP) the chromatin containing the Pitx2 binding sites but not a control site which does not contain a putative Pitx2 binding site. C) ChIP-qPCR analyses to assess the enrichment of the binding by Pitx2 Ab compared to IgG using Pitx2 primer 2 probes. D) A putative Sox2 binding site was identified at position 1972 upstream of Sox2 TSS. Primers used to amplify the putative binding site and control site, which doesn t have a putative Sox2 binding are indicated. E) ChIP assays were carried out in LS-8 cells by using a Sox2 Ab and the PCR results demonstrate endogenous Sox2 binds to this Sox2 binding site in the Sox2 promoter region. F) ChIP-qPCR shows an 8-fold enrichment by using Sox2 Ab to pull-down this chromatin region compared with using IgG. G) Expression plasmids containing the Pitx2, Sox2, and vector only cdnas were co-transfected into LS-8 cells with a luciferase reporter plasmid whose expression is driven by the Sox2, Lef-1 or Pitx2 promoter. Luciferase activity is shown as mean-fold activation compared with that in the presence of empty mock expression plasmid. H) Sox2, Lef-1 and Pitx2 transcripts from LS-8 cells transfected with vector only, Pitx2 or Sox2 were assessed by real-time PCR. Pitx2 activates endogenous Sox2 and Lef-1 expression while overexpression of Sox2 shows small increases in endogenous levels of both Lef-1 and Pitx2 (N=3). All data normalized to β-actin transcripts. All the transcripts in Sox2 overexpressing and Pitx2 100

124 overexpressing groups are compared with pcdna3.1(empty vector) group. *, p<0.05. **, p<

125 Figure III-15 Sox2 interacts with Pitx2 through the HMG domain. A) Immunoprecipitation (IP) of endogenous Sox2 using the Pitx2 Ab in LS-8 cells. The input was 5% of the lysate used in the IP assays. IgG used, as a control did not IP the Sox2 protein, whereas the Pitx2 Ab did IP the Sox2 protein. B) Schematic of the GST- Sox2 truncated constructs used to determine the protein interaction domain of Sox2. The two Pitx2 binding domains are shown. C) GST-Sox2 constructs were bacteria expressed, purified and used in the GST-pull down assay. The fusion proteins are shown on a coomassie blue stained SDS PAGE gel. D) Purified Pitx2 protein was incubated with the GST-Sox2 constructs to determine which regions of Sox2 bound Pitx2. Pitx2 bound to the HMG and C-terminal domains of Sox2. 102

126 Figure III-16 Model for the roles of Pitx2, Sox2 and Lef-1 in regulating incisor development, growth and renewal. A) Pitx2 (blue) is expressed throughout the dental placode and oral epithelium at E11.5, while Sox2 (red) and Lef-1 (green) expression occurs in separate domains, Sox2 is posterior to Lef-1 in the dental placode. B) Pitx2 is expressed with Sox2 in the LI LaCL at E16.5. However, Pitx2 is also highly expressed in the transient amplifying cells of the lower incisor that give rise to the ameloblasts. Lef-1 is mostly expressed in the dental mesenchyme adjacent to the dental epithelium at this stage C) In the Pitx2 Cre /Sox2 F/F embryos at E16.5, Pitx2 is present in the LaCL but not Sox2 or Lef1, leading to the failure of maintain stem cell niche. D) In the Pitx2 Cre /Sox2 F/F /Lef-1 cki rescue embryos after E14.5, Pitx2 and Lef-1 can function to produce dental epithelial stem cells that generate an incisor. E) It has been shown that Fgf can activate both Pitx2 and Sox2 expression in early stages of tooth development. Biochemical assays in this report show that Pitx2 activates Sox2, Lef-1 and Pitx2 to maintain their expression. However, Sox2 directly interacts with Pitx2 to repress Pitx2 transcriptional activity and modulates Sox2, Lef-1 and Pitx2 expression levels to coordinate dental epithelial stem cell renewal, proliferation and differentiation of progenitor cells. LI: lower incisor. LaCL, labial cervical loop; 103

127 Mouse lines Pitx2-Cre K14-Cre Shh-Cre Rosa26- Cre ERT Sox2F/F Lef1cKI/cKI Rosa26 Tomato- GFP Genotyping primers Cre-Forward: GCATTACCGGTCGATGCAACGAGTGATG Cre-Reverse: GAGTGAACGAACCTGGTCGAAATCAGTGC Same as Pitx2-Cre Shh-Cre Forward: TGCCAGGATCAGGGTTTAAG Shh-Cre Reverse: GCTTGCATGATCTCCGGTAT Rosa26-Cre Forward: AAAGTCGCTCTGAGTTGTTAT Rosa26-Cre Reverse: CCTGATCCTGGCAATTTCG Sox2Flox mut F:CAGCAGCCTCTGTTCCACATACAC Sox2Flox mut R: CAACGCATTTCAGTTCCCCG Sox2Flox WT F: GCTCTGTTATTGGAATCAGGCTGC Sox2Flox WT R: CTGCTCAGGGAAGGAGGGG Lef-1cKI mut F: TGAGGCGGAAGTTCCTATTCT Lef-1cKI mut R: GGCGGATCACAAGCAATAAT Lef-1cKI WT F: TCCCAAAGTCGCTCTGAGTT Lef-1cKI WT R: GGCGGATCACAAGCAATAAT Tomato-GFP mut F:CTCTGCTGCCTCCTGGCTTCT Tomato-GFP mut R:CGAGGCGGATCACAAGCAATA Tomato-GFP WT F: CTCTGCTGCCTCCTGGCTTCT Tomato-GFP WT R:TCAATGGGCGGGGGTCGTT Table III-1 List of all the primers for genotyping. 104

128 Primer name Forward primer (5-3 ) Reverse primer (5-3 ) Pitx2 ChIP 1 AGGGCTGGGAGAAAGAAG AG ATCTGGCGGAGAATAGTTG G Pitx2 ChIP 2 GAGCTTCTTTCCGTTGATGC TTCCCTACTCCACCAACCTG Pitx2 ChIP control GGCAGAGTTGGGGTAGATG A CCCCGTCTAAGTTTCCTTCC Sox2 ChIP GCTCAACCTTTGCTCTGGTC TAGTCCACCCCTCTCACTGC Sox2 ChIP control GCCTGGCTCCAATGTAATG T CATTCCGAGGAAGAGCAGA C Lef-1 TCACTGTCAGGCGACACTT ATGAGGTCTTTTGGGCTCCT C Sox2 ATGCACAACTCGGAGATCA TGAGCGTCTTGGTTTTCCG G Pitx2 CTGGAAGCCACTTTCCAGA G AAGCCATTCTTGCACAGCT C β - Actin GCCTTCCTTCTTGGGTATG ACCACCAGACAGCACTGTG Table III-2 Primer list for ChIP assay and Real-time PCR. 105

129 CHAPTER IV A MIR-23A/B:HMGN2:PITX2 SIGNALING PATHWAY REGULATES CRANIOFACIAL/INCISOR MORPHOGENESIS Abstract The rodent incisors grow throughout the life of the animal due to the proliferation of the stem cells in both mesenchymal and epithelial compartment, and this continuous growth is counterbalanced by constant abrasion to prevent incisors from becoming excessively long. Previously we have shown that Pitx2, the earliest transcription factor and master regulator for tooth initiation and differentiation, activates expression of amelogenin which is the major protein component for enamel deposition. This activation can be repressed by the chromatin-associated factor Hmgn2. In this report, we describe a mir-23a/b:hmgn2:pitx2 signaling pathway in regulating dental epithelial cell growth and differentiation. mir-23a and mir-23b directly target Hmgn2, leading to release of the Hmgn2 inhibition of Pitx2 transcriptional activity and thus enhancing amelogenin production. Our in vivo studies also showed a negative correlation between mir-23a/b and Hmgn2 mrna expression in craniofacial region. We also identified that Pitx2 represses mir-23a/b expression, forming a negative-feedback loop and allowing a proper amount of amelogenin to be expressed. Ablation of Hmgn2 in mice results in an overgrowth of incisor with increased amelogenin expression. Taken together, our results unveil a mir-23a/b:hmgn2:pitx2 signaling regulatory pathway that functions to fine-tune tooth growth and differentiation. Introduction Unlike other mammalian teeth, mouse incisors grow continuously throughout life. This growth is facilitated by dental stem cells housed in both epithelial and mesenchymal 106

130 compartments of the incisor (Cao et al., 2013a; Juuri et al., 2013a; Zhao et al., 2014). However, the continuous growth of the mouse incisors is counterbalanced by constant abrasion to prevent incisors to become too long. Another unique feature of mouse incisor is the asymmetrical deposition of enamel, the hardest tissue in the body, which is found exclusively on the labial side (i.e towards the lip) of the teeth but not the lingual side (i.e towards the tongue), enabling the abrasion to occurs preferentially on the lingual side and making a sharp tip of the incisors (Klein et al., 2008a; Seidel K et al., 2010). The epithelial stem cells are housed in a micro-environment called labial cervical loop(lacl) at proximal end of the incisor. The stem cells in LaCL proliferate and move distally to form transit amplifying (TA) cells, which can differentiate into ameloblasts and generate enamel (Harada et al., 2002; Thesleff and Tummers, 2008; Tummers and Thesleff, 2003b). However, the molecular mechanisms for controlling the dental epithelial stem cell proliferation and differentiation are still largely unknown. During tooth development, a variety of transcription factors and chromatin remodelers play roles in the regulation of tooth organogenesis. Pitx2, the earliest transcription factor observed in tooth development and expressed exclusively in the dental epithelium, directly activates amelogenin expression, whose product is the major protein component for enamel formation (Amen et al., 2008a; Li et al., 2014). HMGN2 belongs to chromatin-associated high mobility group protein, which impart structural and functional plasticity to the chromatin fibers and thus modulate transcriptional activity (Hock et al., 2007). Previous studies from our group reported that Hmgn2 acts as a molecular switch to regulate Pitx2 transcriptional activity in a β-catenin dependent manner. In the context of absence of β-catenin, Hmgn2 inhibits Pitx2 from binding 107

131 amelogenin or other targets, thus impeding amelogenesis (Amen et al., 2008a; Li et al., 2014). MicroRNAs(miRNAs) mediated gene silencing has been shown to play important roles in many biological processes such as cell proliferation, differentiation and death (Bartel, 2004; Zhao and Srivastava, 2007). mirnas are ~22nt endogenous RNAs that pair with 3 UTR of protein-coding mrnas to post-transcriptionally regulate gene function (Grimson et al., 2007). mirnas have been reported to play critical roles in tooth development. Inactivation of Dicer1 in dental epithelium results in severe phenotypes such as loss of enamel, multiple teeth etc.(cao et al., 2010b). mirna-23a and mirna- 23b are encoded by the mir-23a-27a-24 (mouse chromosome 8) and mir-23b-27b-24 clusters (mouse chromosome 13) respectively and they share the same seed sequence. mirna-23a /b expression has been reported to play important role in endocrine homeostasis (Shen et al., 2013), cell death (Chen et al., 2014), glutamine metabolism (Gao et al., 2009) and a key regulator for cancer development (Viswanathan et al., 2014). A large-scale screen for mirna expression profiles in the tooth germs of Miniature pigs reveals that mirna-23a/b are highly expressed during tooth development (Li et al., 2012a). We found high expression of mirna-23a/b in the mouse dental epithelial cells using microarray analysis by comparing the mirna expression profiles between differentiated and undifferentiated dental epithelial tissues (data not shown), but the potential roles of mirna-23a/b in tooth development have yet to be characterized. Here, we describe a mir-23a/b:hmgn2:pitx2 signaling pathway in tooth development. Hmgn2, a chromatin-associated factor and a repressor for Pitx2 transcriptional activation of Amelogenin, is directly targeted by mir23a/b. Interestingly, 108

132 as development proceeds, the expression of Hmgn2 in mouse craniofacial region was found to decrease gradually in a manner that was inversely correlated with mir-23a/b and amelogenin expression. We also found that Pitx2 directly represses mir-23a/b expression to form a negative-feedback loop. Lastly, deletion of Hmgn2 in mice resulted in an uncontrolled growth of incisor with increased amelogenin expression. In conclusion, our in vitro and in vivo studies demonstrated the central role of mir- 23a/b:Hmgn2:Pitx2 signaling pathway in modulating dental epithelial cell proliferation and differentiation. Materials and methods Mouse strain breeding All animals were housed in the Program of Animal Resources of the University of Iowa, and were handled in accordance with the principles and procedure of the Office of Animal Care. All experimental procedures were approved in accordance with the University of Iowa IACUC guidelines. The Hmgn2 -/- knockout embryonic stem (ES) cells (derived from C57BL/6 mice) were purchased from Knockout Mouse Project (KOMP) (Hmgn2tm1b(KOMP)Wtsi) and injected into blastocysts (from BALB/cJ) by the Texas A&M University Institute for Genomic Medicine. In two chimeras generated from ES cell injection, the mutant allele was passed through the germ line, and these animals produced heterozygous progeny. Mice were maintained on a C57BL/6 background. Bimolecular fluorescence complementation (BiFC) assay The BiFC assay was performed as previously reported (Wang et al., 2013b; Zhang et al., 2013). Pitx2C cdna was cloned into the pflag-cmv-2plasmid (Sigma) containing 109

133 N-terminal fragment of YFP, and Hmgn2 was ligated to C-terminal fragment of YFP in the pflag-cmv-2 plasmid. YN or YC fragments was ligated into the vector were used as controls. 1 µg of each construct was transfected into HEK-293 cells. After 24 h, Zeiss LSM510 confocal microscope was used to detect the fluorescence. Immunocytochemistry Nearly 5000 cells were seeded on glass slides 24 h prior to fixation. The slides were washed in 1xPBS and then incubated in ice-cold acetone for 5 min at 4 C. Fixed cells were washed twice with PBST (5 min each). Subsequently, the slides were incubated in 10% normal goat serum-pbst for 30 min at room temperature for blocking. Slides were then incubated with either Amelogenin antibody (Santa Cruz, 1:500) or Hmgn2 antibody (Cell Signaling, 1:5000) at 4 C overnight. Cells were rinsed with PBST for three times, 10 min each, and were incubated with goat anti-rabbit Alexa-488-labeled secondary antibody (Invitrogen) for 30 min at 37 C. Finally, the cells were washed with PBST for three times, 10 min each, and counter stained using mounting solution containing DAPI. LacZ staining Hmgn2 +/- embryos or postnatal pups at different stages were fixed for mins at room temperature in the fix solution (0.2% glutaraldehyde, 2% formaldehyde, 2 mm MgCl 2, 5 mm EDTA ph 8.0 and 100 mm NaH 2 PO 4 ph 7.3) and washed three times in rinse solution (0.2% Nonidet P-40 and 0.1% sodium deoxycholate, 100 mm NaH 2 PO 4 ph 7.3 and 2 mm MgCl 2 ). Embryos were stained for hours at 37 C in staining solution (1.65 mg/ml potassium ferricyanide, 1.84 mg/ml potassium ferrocyanide, 2 mm MgCl 2, 1 mg/ml X-gal in rinse solution), rinsed in PBS and post-fixed in 4% paraformaldehyde. 110

134 Histology, immunofluorescence and trichrome staining Murine embryos or postnatal pups were used for histology and fluorescence immunohistochemistry (FIHC). Samples were fixed in 4% paraformaldehyde, dehydrated and embedded in paraffin. Sections were cut to 7 µm thickness, and standard Hematoxylin and Eosin staining to assess tissue morphology. Sections used for immunofluorescent assay were rehydrated and treated with 10 mm Sodium Citrate solution for 20 min at a slow boil for antigen retrieval. These sections were incubated with 10% goat serum-pbst for 30 min at room temperature, followed by overnight incubation at 4 C with an antibody against one of the following proteins: Amelogenin (Santa Cruz,1:200), Hmgn2 (Millipore 1:200), Ameb (Santa Cruz1:200), Enml (Santa Cruz, 1:200), Ki67 (Abcam,1:250), E-cadherin (BD Bioscience, 1:200) or Dspp (Santa Cruz, 1:200). After the incubation, the slides were treated with Alexa-488 (FITC channel) - or Alexa-555 (Cy3 channel)-labeled secondary antibody (Invitrogen) at a concentration of 1:500 for 30 min. Each antibody incubation was followed by 3 6 PBST (phosphatebuffered saline with 0.05% Trtion-100) washes. Nuclear counter staining was performed by applying DAPI-containing mounting solution after the final wash (Vector Laboratories). The trichrome staining was carried out as previously described (Cao et al., 2013a). Samples were stained with Azocarmin for 1 h at 50 C, then stained with aniline to differentiate nuclei. Last, samples were stained with Orange G and Aniline blue for 2 h. Expression and luciferase reporter constructs Mouse Pitx2A, Pitx2B, Pitx2C and Hmgn2 cdna were cloned into pcdna-3.1- MycHisC (Invitrogen) using cytomegalovirus(cmv) promoter to allow expression in 111

135 eukaryotics. mir-23a and mir-23b were cloned into psilenser 4.1 (Life Technologies). A 5.3Kb upstream of Pitx2 transcription start site (TSS) and 2.2Kb upstream of Amelogenin TSS were cloned into ptk-luc vectors to generate the promoter luciferase reporters. The Hmgn2 3 UTR was ligated into downstream of a luciferase gene in pgl3 reporter vector(promega). PCR-driven overlap extension method was used to mutate the mir23-a/b binding site in Hmgn2 3 UTR. All the cloned constructs were confirmed by DNA sequencing. All plasmids used for transfection were purified by double-banding in CsCL. Cell culture, transfections and reporter assays LS-8 and CHO cells were cultured in DMEM supplemented with 5% FBS, 5% BGS and 1% penicillin/streptomycin, and transfected by electroporation. Cells were fed 24 hours prior to transfection. Cells were resuspended in PBS and mixed with 2.5 µg of expression plasmid, 5 µg of reporter plasmid and 0.2 µg of SV-40 β-galactosidase plasmid. Transfection was performed by electroporation at 380 v and 950 mf (Gene Pulser XL, Bio-Rad), or using the Lipofectamine 2000 (Life Technologies) transfection reagent. Transfected cells were incubated in 60 mm culture dishes for 24 h, and fed with 10% FBS and DMEM. Following lysis, assays for reporter activity (luciferase assay, Promega) as well as for protein content (Bradford assay, Bio-Rad) were carried out. β- galactosidase was measured using the Galacto-Light Plus reagents (Tropix Inc.) as an internal normalizer. For each assay, all luciferase activity was normalized to the mean value of the first experimental group, and is shown as mean ± SEM. Western blot assays 25 µg of cell lysate were analyzed on 12% SDS-PAGE gels. Following 112

136 electrophoresis, the proteins were transferred to PVDF membrane (Millipore), immunoblotted and detected with an HRP conjugated secondary antibody and ECL reagents from Amersham Biosciences. The following polyclonal antibodies were used to detect the proteins: anti-β-tubulin (1:1000, Santa Cruz Biotechnology), anti-pitx2 (1:500, Capra Science), and Hmgn2 (1:500, Millipore). Real-time PCR assays Total RNA was isolated from cells or mouse mandible and maxilla tissues using mirneasy Mini Kit (Qiagen). Reverse transcription and quantitative real-time PCR were carried out with miscript PCR system(qiagen) according to the manufacturer s protocol. β-actin served as a loading control for mrnas. (primer sequences: forward, 5'- GCCTTCCTTCTTGGGTATG-3'; reverse, 5'- ACCACCAGACAGCACTGTG-3'). U6 probes were used as a reference for normalization of mirnas (manufacturer provided). Primer sequences for detecting amelogenin are: forward, TACCACCTCATCCTGGAAGC; reverse, GTGATGAGGCTGAAGGGTGT. Primers for detecting Pitx2 are forward, CTGGAAGCCACTTTCCAGAG; reverse, AAGCCATTCTTGCACAGCTC. Primers for detecting Hmgn2 are: forward, AAAACCAAGGTGAAGGACGA; reverse,tctgtgcctggtctgttttg. PCR products were examined by melting curve analysis and the sequences were confirmed. Fold changes was calculated base on the 2 ΔΔCT method. Imaging and microcomputed tomography (Micro-CT) Mouse heads of WT and Hmgn2 -/- littermate mice were dissected, fixed with 4% PFA overnight and stored in 70% ethanol for imaging. Micro-CT was conducted according to previous description (Gao et al., 2015). 113

137 Chromatin immunoprecipitation assay (ChIP) ChIP assays were performed as previously described (Cao, et al. 2013) using the ChIP Assay Kit (Zymo research). LS-8 cells were cross-linked in 1% formaldehyde at room temperature for 7 minutes. Crossed linked cells were sonicated three times (6 second duration for each round, 25% of maximum amplitude) to shear the genomic DNA into bp fragments. Then the DNA/protein complexes were immnoprecipitated with 5µg Pitx2 antibody (Capra Science) or 5 µg Rabbit IgG as control. Precipitated DNAs were subjected to PCR to evaluate the enrichment of Pitx2 binding. The following primers were used for PCR. Primers for amplifying the Pitx2 binding site in pre-mir- 23a-27a-24-2 promoter are: forward, TCCTGCCCTAACCTGTCAGA; reverse, AGCTAAGGACCCAACCGACT. Primers for amplifying the control region in which does not have a putative Pitx2 bindings site in pre-mir-23a-27a-24-2 promoter are: forward, GCCTCCCTGTTTGATGTCTC; reverse, CAGCTGGTTCTGTCATGCTC. Primers for amplifying the Pitx2 binding site in pre-mir-23b-27b-24-1 promoter are: forward, GGTGACTGACTGTCCTGTGC; reverse, AGGCTGCCAGATAATGGACA. Primers for amplifying the control region that does not have a putative Pitx2 bindings site in pre-mir-23b-27b-24-1promoter are: forward, TGTGTGTGTGTGATGTTTAAGGA; reverse, CAGCTTTCTTTCTGTGTCAATGAT. All the PCR were performed under an annealing temperature of 60 C. All the PCR products were analyzed on a 1.5% agarose gel for the correct size and confirmed by sequencing. Statistical analysis All quantified results are presented as mean ± SEM, and with an n value indicating the number of biological repeats. A two-tailed unpaired Student's t test and either one- or 114

138 two-way ANOVA were used to determine statistical significance. Results Hmgn2 interacts with Pitx2c in the nucleus and Hmgn2 represses Pitx2 transcriptional activity We have previously reported that Pitx2 interacts with Hmgn2 through homeodomain (HD) using pull-down assay and gel shifting assays (Amen et al., 2007). Here, we used a cell model to further demonstrate the interaction between Pitx2 and Hmgn2 and their cellular localization by performing the Bimolecular fluorescence complementation (BiFC) assay. Pitx2 was ligated to C- terminal of EYFP-N fragment and Hmgn2 was inserted to N- terminal of EYFP-C fragement (Figure IV-1A). Complexing between YN- Pitx2 and Hmgn2-YC results in fluorescence due to complementation between N and C fragments of the EYFP protein (Hu et al., 2002; Zhang et al., 2013). As shown in Figure I-B, YN-Pitx2+Hmgn2-YC complex produces fluorescence that is confined to the nucleus. As control, transfecting either YN-Pitx2 with the Flag-YC or Hmgn2-YC with Flag-YN did not emit fluorescence (Figure IV-1B). These results demonstrate an interaction between Pitx2 and Hmgn2 in the nucleus of living cells. Previous studies have demonstrated that Hmgn2 interacts with Pitx2 to form an inactive complex to inhibit Pitx2 DNA-binding activity (Amen et al., 2008a; Li et al., 2014). To ask if Hmgn2 regulates the Pitx2 activation of targets including Pitx2 promoter, we co-transfected 5.3Kb Pitx2 promoter luciferase reporter and different Pitx2 isoforms with/without Hmgn2 in LS-8 cells. The luciferase results show all the three Pitx2 isoforms activate Pitx2 promoter luciferase activity, but these activation were diminished 115

139 when Hmgn2 was overexpressed (Figure IV-1C), suggesting Hmgn2 inhibits Pitx2 transactivation of its own promoter. mir-23a and mir-23b repress Hmgn2 in dental epithelial-like cells Hmgn2 is required for tight regulation of Pitx2 transcriptional activities to allow for normal tooth development to proceed (Amen et al., 2008a; Li et al., 2014) and therefore its expression must be regulated. micrornas have been reported to play a critical role in regulating of normal tooth development (Cao et al., 2010b). Analyzing Hmgn2 3 UTR sequence, we found a highly conserved mir-23a/b binding element (Figure IV-2A). To determine if mir-23a and mir-23b target Hmgn2 in oral epithelial cells, we cloned the Hmgn2 3 UTR containing the mir-23a and mir-23b binding site into a luciferase reporter and transfected it into LS-8 cells with mir-23a and/or mir-23b. The luciferase activity of wildtype (WT) Hmgn2 3 UTR was significantly repressed by the presence of mir23-a or/and mir23-b (Figure IV-2B). To demonstrate that the regulation was dependent on the putative mir-23a/b binding site in Hmgn2 3 UTR, we mutated the conserved mir23a/b binding site in Hmgn2 3 UTR. The subsequent luciferase assay showed the previous repression by mir23a/b being abolished (Figure IV-2C). These regulations were also confirmed at the protein level as overexpressing mir-23a and mir- 23b separately and together led to reduced endogenous Hmgn2 expression in LS-8 cells (Figure IV-2D). mir-23a/b indirectly activates Pitx2 and Amelogenin expression by repressing Hmgn2 in dental epithelial cells During normal tooth development, the mature ameloblast start expressing amelogenin at late embryonic stages, and Pitx2 is a major activator of amelogenin 116

140 expression (Li et al. 2014). To further understand how the regulation of Hmgn2 by mir- 23a/b contributes to tooth organogenesis, we assayed two known Hmgn2 downstream targets, Pitx2 and Amelogenin, which are critical for enamel development. Consistent with Figure IV-C, Pitx2A activates Pitx2 promoter luciferase activity and Hmgn2 inhibits this activation. Overexpression of mir-23a or mir-23b increases Pitx2 promoter luciferase activity by ~5 fold. Overexpression of mir-23a or mir-23b with Pitx2 can synergistically increase Pitx2 promoter luciferase activity by more than 15 fold (Figure IV-3A). These additive effects further validate the idea that mir-23a and mir-23b derepress Pitx2 promoter luciferase activity by repressing Hmgn2. In accord with the results of luciferase assays, we also observed increased endogenous Pitx2 protein expression following overexpression of mir-23a or/and mir-23b in LS-8 cells, further indirectly illustrating mir-23a and mir-23b targets Hmgn2 to allow Pitx2 activation to its own expression (Figure IV-3B). Previous studies have shown that Pitx2 activates amelogenin expression (Venugopalan et al., 2011, Li et al., 2014). The luciferase activity following co-transfection of mir-23a or mir-23b and Pitx2 with amelogenin promoter luciferase vector followed the same trend as Pitx2 promoter activity. Overexpressing mir-23a or mir-23b with Pitx2 increased amelogenin luciferase activity more than Pitx2 itself (Figure IV-3C). To test if mir-23a/b activate amelogenin expression in a physiological context, we stably overexpressed mir-23b using lentivirus in LS-8 cells culturing in osteogenic differentiation medium for 4 days (Thesleff, 1976), and then we tested both amelogenin mrna and protein level. The level of endogenous Hmng2 mrna was reduced to 40% compared to scramble control, while amelogenin transcripts was increased ~5 folds when mir-23b was overexpressed (Figure IV-3D). We also observed 117

141 an increase of amelogenin protein expression in mir-23b overexpressing cells compared with control as tested by immunofluorescence. (Figure IV-3E). Taking together, our data suggested that mir-23a/b activates Pitx2 and Amelogenin expression by targeting Hmgn2. Pitx2a represses mir-23a and mir-23b expression Interestingly, our microrna array data comparing mirna expression in Pitx2 overexpressing incisors and WT incisors revealed a decrease of mir23a/b expression in Pitx2 overexpression incisors, indicating the presence of a negative feedback loop in the mir-23a/b:hmgn2:pitx2 signaling pathway (Figure IV-4A). To ask if Pitx2 directly regulates mir-23a/b, we compared the mir-23a/b expression in LS-8 cells overexpressing Pitx2 and WT LS-8 cells. Both mir-23a and mir-23b were significantly (p<0.05) down-regulated in Pitx2 overexpressing LS-8 cells compared with control (Figure IV-4B). Analysis of the 5 flanking region of mir-23a-27a-24-2 identified a conserved Pitx2 binding site 8669bp upstream of the Transcription Start site (TSS) with a high degree of conservation among mouse, rhesus monkey, dog, rat and human (Figure IV-4C and 4-4D). To test if Pitx2 regulates mir-23a through this putative binding site, chromatin immunoprecipitation (ChIP) assays were carried out in LS-8 cells, which express Pitx2 (Wang et al. 2013). Endogenous Pitx2 was found to bind mir-23a-27a-24-2 promoter region by this putative binding site, while anti-igg antisera failed to pull down this chromatin region (Figure IV-4E). Additional control assays were performed to show that the Pitx2 antibody and IgG cannot immunoprecipitate the chromatin ~10Kb upstream of mir-23a-27a-24-2 TSS where no putative Pitx2 binding site was found (Figure IV-4F). 118

142 To confirm that the binding of Pitx2 to this region confers a regulatory effect, we cloned 1Kb mir-23a-27a-24-2 enhancer fragment which contains this binding element into the ptk-luc reporter vector. Co-transfection of this reporter and Pitx2 together in LS-8 cells showed significantly repression (p<0.01) of luciferase activity (Figure IV-4G). Similar to the mir-23a-27a-24-2 promoter region, the pre-mir-23b-27b-24-1 promoter also contains a conserved Pitx2 binding site in ~15Kb upstream of pre-mir- 23b-27b-24-1 TSS (Figure IV-5A and Figure IV-5B). ChIP assay showed endogenous Pitx2 bound to this region of the chromatin (Figure IV-5C), but not the region which doesn t have a Pitx2 binding site (Figure IV-5D). Hmgn2 -/- incisors undergo abnormal expansion with increased enamel formation To further validate our in vitro finding, we generated Hmgn2 knock-out mice (Hmgn2 -/- ) by replacing Hmgn2 genomic DNA with a LacZ reporter. Whole mount X-gal staining displayed a robust and wide expression of Hmgn2 in whole body at early mouse embryonic stages (from E10.5 to E14.5) (Figure IV-6A). After E14.5, Hmgn2 expression decreased and was restricted to the craniofacial region. Analyses of Hmgn2 expression in E14.5 WT mouse craniofacial region revealed that Hmng2 is highly expressed in the dental epithelium with relatively less expression in the dental mesenchyme (Figure IV- 6B). The loss of Hmgn2 expression in Hmgn2 -/- embryos indicates successful ablation of Hmgn2 by gene trap strategy (Figure IV-6C). The Hmgn2 -/- mice are viable and fertile. At 1 month of age, Hmgn2 -/- mice develop long and thick incisors (Figure IV-6D and 4-6E). Microcomputed tomography (µct) analysis showed the length of upper incisor and lower incisor in Hmgn2 -/- mice was 59.17% and 40.08% greater in length compared to WT(Figure IV-6F). µct analysis of WT and 119

143 Hmgn2 -/- mice also shows that the Hmgn2 inactivated incisors are thicker than those in WT mice, indicating enamel formation Our in vitro data suggested that mir-23a/b target Hmgn2 in teeth to relieve its inhibition to Pitx2 and allow for the activation of amelogenin. To test if this phenomenon also occurrs in vivo, we examined the mrna expression profile of mir-23a/b, Hmgn2 and amelogenin in the wild-type mouse craniofacial region at different embryonic and postnatal stages. mir-23a/b mrna level increases as development progresses, while Hmgn2 expression decreases which is also consistent with Figure IV-5A. This negative correlation supports the idea that mir23a/b inhibits Hmng2 in teeth. Moreover, we observed a fast and robust increase of amelogenin expression after E16.5, which is negatively correlated with the down-regulation of Hmng2 expression (Figure IV-6G). To investigate if the expression profile was affected in Hmgn2 -/- incisors, we analyzed some markers expressed in dental epithelium. E-cadherin, which is highly expressed in undifferentiated dental epithelium (Biehs et al., 2013), is remained unchanged between WT and Hmgn2 -/- incisors (Figure IV-7A-D). Ameloblastin, accounting for 5%-10% of all enamel protein (Poulter et al., 2014), was normally expressed in Hmgn2 -/- incisors (Figure IV-7E-H). Amelogenin, which is the most abundant (90%) enamel matrix protein and critical for enamel organization, is normally expressed in the secretory stage ameloblast (SSA) located in the distal side of WT incisor at P0 (Figure IV-8A and 4-8C). However, amelogenin expression is expanded to presecretory stage ameloblasts (PSSA) which are located towards the posterior of the incisors in Hmgn2 mutant mice. This suggests that ablation of Hmgn2 in mice results in an increase of amelogenin secretion in teeth (Figure IV-8B and 4-8D). Trichrome- 120

144 staining at P4 stage to mark the enamel and dentin in incisors also revealed that the enamel layer in Hmgn2 -/- mice was extended proximally compared with WT (Figure IV- 8E-H, red). The increase in amelogenin expression and enamel formation in Hmgn2 -/- mice is consistent with the role of Hmgn2 acting as a repressor of Pitx2 activation of amelogenin. Dentin sialophosphoprotein (DSPP), mainly expressed in young odontoblasts and transiently expressed in secretory stage ameloblasts (SSA) of the teeth (D souza et al. 1997), was ectopically expressed in the pre-secretory stage ameloblasts (PSSA) of Hmgn2 -/- mice, indicating a early maturation of ameloblast in the Hmgn2 ablated incisors(figure IV-8I. and 4-8J.). Loss of Hmgn2 leads to increase in dental epithelial cell proliferation in mice Next, we determined if an increase in stem cell proliferation in the LaCl could contribute to this unchecked growth in Hmgn2 -/- incisors. Immunofluorescent staining of a cell proliferation marker, Ki67, was performed in the sagittal sections of P0 Hmng2 -/- and WT lower incisors. This experiment revealed an increase of Ki67-postive epithelial cells in Hmgn2 -/- incisors compared to WT (Figure IV-9A-D). Quantitation data showed a ~10% increase of Ki67-postive epithelial cells/total epithelial cells in the Hmgn2 -/- incisors compared to WT littermates (Figure IV-9E), indicating that the dental epithelial cell proliferation is up-regulated in Hmgn2 -/- mice. Discussion Hmgn2 belongs to a member of the high-mobility group N (Hmgn) family, which encodes a small, chromatin-associated protein. Hmgn2 does not have intrinsic transcriptional activity but can modulate transcription activity by altering chromatin structure (Hock et al., 2007; Nishino et al., 2008). Hmgn2 binds specifically to 121

145 nucleosome core particles mediated by HMGN nucleosomal binding domain(nbd), which is the hallmark of HMG proteins (Gerlitz, 2010). Hmgn2 is ubiquitously and highly expressed in all embryonic tissues at early stages, but its expression downregulated as mouse embryogenesis progresses (Hock et al., 2007). Previous studies from other groups and our group have shown that Hmgn2 expression is correlate with Pitx2 (Amen et al., 2008a; Lehtonen et al., 1998; Li et al., 2014). We have shown that in the context of absence of nuclear β-catenin, Hmgn2 binds to and inactivates Pitx2 transcriptional activity. This tight control of Pitx2 transcriptional activity would allow proper activation of Pitx2 downstream target genes during early mouse development. Amelogenin, one of the direct targets of Pitx2 and major protein for enamel deposition, is increased in Hmgn2 ablated mice. Here, we further show that Hmgn2 expression is modulated by mir-23a/b. All these fine-tunings ascertained the appropriate amount of expression of Pitx2 targets and allow normal tooth development to proceed. Our data show that Hmgn2 expression is negatively correlated with expression of amelogenin and mir-23a/b during early tooth development. This negatively correlation supports our idea that mir-23a/b-hmgn2 regulatory pathway modulates amelogenin expression during tooth development. Tooth development is tightly controlled by micrornas (Cao et al., 2010b; Michon, 2011). Inactivating Dicer1 in mouse teeth leads to multiple and branched enamel-free incisors and attenuated ameloblast differentiation (Cao et al., 2010b). We previously demonstrated the roles of specific microrna in tooth development. For example, knockout mir-200c/141 in mice results in defects of enamel formation, with decreased E- cadherin and amelogenin expression and increased noggin expression (Cao et al., 2013a). 122

146 Pitx2 and mir-200a-3p can converts mesenchymal cells to amelogenin-expressing dental epithelial cells (Sharp et al., 2014a). Our previous data also showed that mir23a/b are highly expressed in mouse dental epithelial tissues (Cao et al., 2010a). In this study, we demonstrate mir23a/b directly regulates Pitx2 transcriptional activity by targeting Hmgn2. This regulation would modulate Pitx2 transcriptional activation of downstream targeted genes (e.g amelogenin) which allow normal craniofacial development to occur. We also demonstrate that mir23a/b expression increases as mouse embryonic development progress, while Hmgn2 expression decreases. This inverse correlation further verified that mir-23a/b target Hmgn2 which allows Pitx2 transcriptional activation of differential related genes in late mouse embryonic development. This is a great example of how micrornas regulate differentiation during development. Pitx2 is a major regulator of early tooth development and its expression can be detected as early as E8.5 of mouse embryonic development (St Amand et al., 2000). Pitx2 null mice have arrested tooth development at the bud stage (Liu et al., 2003a; Lu et al., 1999). PITX2 mutations are also associated with Axenfel-Rieger Syndrom (ARS). Patients with ARS display various dental defects, such as hypoplasia of enamel, Amelogenesis Imperfecta (AI) (Murray et al., 1992; O'Dwyer and Jones, 2005). Pitx2 directly activates Amelogenin,which is the major protein for enamel deposition(li et al., 2014). Pitx2 is the earliest transcription factor detected in oral epithelium. Its expression is tightly controlled by histone remodeler Hmgn2. Hmgn2 is highly and ubiquitously expressed in early mouse embryos. However, Hmgn2 expression decreases after E14.5 and limited to craniofacial region. By P0, Hmgn2 expression is low in all tissues. Interestingly, Pitx2 expression remain relatively stable as development progresses(li et 123

147 al., 2014). However, Amelogein is first detected at E16.5 at low level, and its expression dramatically increases later in development. Pitx2 activates Amelogenin expression in late embryonic stages could be due to decreased expression of Hmgn2, and increased mir-23a/b could contribute to the down-regulation of Hmgn2. In summary, our data identified a mir-23a/b:hmgn2:pitx2 signaling pathway that modulates dental epithelial cell proliferation and differentiation(figure IV-9F). mir- 23a/b are highly expressed in the later stage of dental epithelium and they repress Hmng2 expression. Hmgn2 represses Pitx2 transcriptional activation of downstream genes including Amelogenin. Pitx2 also feeds back to repress mir-23a/b expression. In late mouse embryonic stages, mir-23a/b:hmgn2:pitx2 pathway activates amelogenin expression and promote amelogenesis during tooth development. 124

148 Figure IV-1 PITX2C interacts with Hmgn2 in the nucleus and Hmgn2 represses Pitx2 transcriptional activity. A). The schematic of the constructs used in Bimolecular fluorescence complementation (BiFC) assay. Pitx2c was cloned into the pflag-cmv-2 plasmid (Sigma) containing N- terminal fragment of YFP, and Hmgn2 was ligated to C-terminal fragment of YFP in the pflag-cmv-2 plasmid. YFP-N terminal only or YFP-C terminal only were ligated into the vector and used as controls. Interaction between the proteins facilitates the association of the YN and YC fragments to produce fluorescence. B). The constructs were transfected into HEK-293 cells. After 24 h, a Zeiss LSM510 confocal microscope was used to detect the fluorescence. C). Hmgn2 represses the self-activation of Pitx2. LS-8 cells were transfected with Pitx2 5.3kb luciferase reporter vector and three different Pitx2 isoforms (Pitx2A, B and C) with/without Hmgn2 respectively. 48 h later, the luciferase and β-galactosidase activities were measured. The luciferase activities were shown as mean fold activation compared with the normalized luciferase activity in empty vector (pcdna 3.1) with Pitx2 5.3kb promoter reporter. 125

149 Figure IV-2 mir-23a and mir-23b target Hmgn2 in dental epithelial-like cells. A). The seed region of mir-23a and mir-23b are evolutionarily conserved among several vertebrate species and the potential mir-23a and mir-23b binding site in the Hmgn2 3 - UTR is highly conserved among different species. B). mir-23a and mir-23b directly target the Hmgn2 3 -UTR. 5 µg Hmgn2 3 -UTR pgl3 reporter and 2.5µg pre-mir-23a or mir-23b or empty vector (psil) were transfected into LS-8 cells. Cells were incubated for 48 h and then assayed for luciferase and β-galactosidase activities. C). mir-23a and mir-23b do not target the mutant Hmgn2 3 UTR with mutations engineered in the region complementary to the mir-23a and mir-23b seed region (AAUGUGA to CCGAGAC), were inserted into the pgl3 vector. The activities are shown in B) and C) as mean fold activation compared with the luciferase in empty vector (psil) and normalized to β-galactosidase activity. ±S.E. from three independent experiments. p values are shown. D). mir-23a and mir-23b repress endogenous Hmgn2 expression. Western blot of Hmgn2 protein in control or mir-23a or/and mir-23b precursor transfected LS-8 cells 48 h post-transfection. β-tubulin is shown as a loading control. psi empty vector (Mock) and psi-neg vector served as controls. 126

150 Figure IV-3 mir-23a/b indirectly activate Pitx2 and Amelogenin expression by repressing Hmgn2 in dental epithelial cells. A). Expression plasmids containing the mir-23a, mir-23b, Pitx2A and Hmgn2 were cotransfected into LS-8 cells with a promoter luciferase reporter plasmid---pitx2 5.3Kb-TKluc. Luciferase activities are shown as mean-fold activation compared with that in the presence of an empty mock expression plasmid. B). mir-23a/b indirectly activate Pitx2A, Pitx2B and Pitx2C expression. Western blot of Pitx2 protein in control or mir-23a or mir-23b precursor transfected LS-8 cells 48 h post-transfection. β-tubulin is shown as a loading control. Mock, psi-neg vector served as controls. C). mir-23a/b targets endogenous Hmgn2, which repress Pitx2 activation of Amelogenin 2.2Kb-TK-Luc. Note that mir-23a/b and Pitx2 synergistically activate Amelogenin promoter by repressing Hmgn2. D) Stable overexpression of mir-23b in LS-8 cells increases Amelogenin expression by inhibiting Hmgn2. mrna levels of Hmgn2, Amelogenin and mir-23b in mir-23b overexpressing cells were compared to the pll-scramble overexpressing cells. All cells were cultured in osteogenic medium to induce differentiation. E). Immunocytochemistry staining of Hmgn2 (Top) and Amelogenin(Bottom) in LS-8 cells stably transfected with pll-scramble or pll-mir-23b and cultured in osteogenic medium. Nuclei were stained with DAPI. Scale bar represent 10 µm. 127

151 Figure IV-4 PITX2A represses mir-23a and mir-23b expression. A). mir-23a-27a-24-2 and 23b-27b-24-1 cluster expression are down-regulated in PITX2c transgenic mice. microrna array (heat map) shows the expression levels of mir-23a/b families in PITX2c-Transgenic and wild type P0 mice. B). PITX2A represses endogenous mir-23a and mir-23b in epithelial cells. LS-8 cells were co-transfected with 2.5 µg of either the pcdna-pitx2a, or the empty plasmid pcdna3.1(control). Cells were incubated for 48 hours, and then total RNAs were isolated for q-pcr assay. The mir-23a or mir-23b expression level in cells transfected with PITX2A were normalized to cells transfected with empty vector. C). Schematic representation and location of the Pitx2 binding site in the pre-mir-23a-27a-24-2 promoter. Red arrow indicates the region containing a conserved Pitx2 binding element (TCATCC). The blue arrow indicates a region which lacks Pitx2 consensus binding motif and was used as negative control as shown in Figure 4F. D). The PITX2 binding element of the mouse pre-mir-23a-27a-24-2 promoter was mapped to a highly conserved region among mouse, monkey, dog, human, and rat. The red box indicates the PCR amplified region on pre-mir-23a-27a-24-2 promoter in (C). E). ChIP of endogenous Pitx2 binding to the chromatin region approximately 8600bp upstream of pre-mir-23a-27a-24-2 transcript in LS-8 cells. F). Control ChIP using the Pitx2 antibody and primers to a 9.9kb upstream region of the premir-23a-27a-24-2 transcript. This chromatin does not contain a Pitx2 binding site G). Inhibition of the mir-23a-27a-24-2 cluster promoter by PITX2. 1Kb mir-23a-27a-24-2 enhancer fragment which contains this binding element was cloned into the ptk-luc reporter vector.ls-8 cells were transfected with 5 µg mir-23a-27a-24-2 cluster promoter luciferase reporter constructs. The cells were co-transfected with 2.5 µg of either the 128

152 pcdna-pitx2a, or the empty plasmid as a control (pcdna3.1). 129

153 Figure IV-5 Endogenous Pitx2 binds to the mir-23b flanking regions. A). Schematic representation and location of the Pitx2 binding site in the pre-mir-23b- 27b-24-1 promoter. Red arrow indicates the Pitx2 binding element (AAATCC). The blue arrow indicates a region which doesn t have a conserved Pitx2 binding motif as a negative control shown in D. B). The PITX2 binding element of the mouse pre-mir-23b- 27b-24-1 promoter was mapped to a highly conserved region among mouse, human, monkey, dog, and rat. The red box indicates the PCR amplified region on pre-mir-23b- 27b-24-1 promoter in (A) C). ChIP of endogenous Pitx2 binding to the Pitx2 element 15358bp upstream of pre-mir-23b-27b-24-1 transcript in LS-8 cells. Rabbit IgG was used as a control IP and Pitx2 Ab was used to IP Pitx2 binding to the chromatin. The input chromatin is shown as a positive control for the ChIP. D). Control ChIP using the Pitx2 antibody and primers to nearly 21Kb upstream region of the pre-mir-23b-27b-24-1 transcript. This chromatin does not contain a Pitx2 binding site and was not immunoprecipitated using Pitx2 antibody, the primers did amplify the input chromatin. 130

154 Figure IV-6 Hmgn2 expression was decreased in early mouse development and Hmgn2 -/- mice exhibit increased size of incisors. A). The Hmgn2-LacZ knock-in mice were used to confirm the Hmgn2 expression during mouse embryonic and neonatal development. Whole mount LacZ staining was performed at various embryonic and neonatal stages. Note that the decreased level of Hmgn2 expression in later stages of development. B-C). Hmgn2 expression pattern in the E14.5 mandible of WT and Hmgn2 -/- mice. Note that Hmgn2 is highly expressed in the dental epithelium with relatively less expression in dental mesenchyme, also Hmgn2 is successfully deleted in Hmgn2 -/- mice. Scale bars, 50 µm. D-E). Microcomputed tomography (µct) analysis of 1 month old WT and Hmgn2 -/- mouse heads shows the size and enamel in the incisors of Hmgn2 -/- mice are increased compared with WT. F). Comparison of sizes of incisors and incisor roots between WT and Hmgn2 -/- mice by µct measurements. N=3. G). mrna levels of Hmgn2, mir-23a,mir-23b and amelogenin at different embryonic and neonatal time points in WT embryos/mice. Hmgn2, mir-23a and mir-23b expression were normalized to those at E14.5. Amelogenin expression was normalized to those at E

155 Figure IV-7 Ameloblastin and E-cadherin expression were not affected in Hmgn2 -/- mice A-B). Immunofluorescence staining of E-cadherin in P2 WT and Hmgn2 -/- mouse incisor. C-D). Higher magnification of the boxed regions in A and B. Note the E-cadherin expression was not affected by loss of Hmgn2 in teeth. E-F). Ameloblastin immunostaining in P2 WT and Hmgn2 -/- mouse lower incisor. G-H). Higher magnification picture of the boxed regions in A and B. The ameloblastin expression level are similar in both genotypes. Nuclei are counterstained with DAPI. Scale bars, 100 µm. 132

156 Figure IV-8 Hmgn2 -/- mice exhibit increased amelogenin expression and enamel deposition. A-B). Amelogenin immunofluorescent staining in E18.5 lower incisors from P0 WT and Hmgn2 -/- mice. C-D). Magnified images from boxed region in A-B highlight the amelogenin expression in pre-secretory stage ameloblast (PSSA) and secretory stage ameloblast(ssa). Note that amelogenin expression moved further towards PSSA in the Hmgn2 -/- incisors. E-F). Images of trichrome-stained lower incisors from WT and Hmgn2 -/- mice, respectively. Note the enamel was stained as dark red and dentin was stained as blue. G-H). Higher magnified pictures of boxed regions in E and F. Note in P4 Hmgn2 -/- mice, enamel deposition move further toward proximal region of incisor. I-J). Immunofluorescence staining of DSPP in P2 WT and Hmgn2 -/- mice. Nuclei are counterstained with DAPI. Scale bars, 100 µm. Pro: Proximal. Dis: Distal. En: Enamel. PSSA: Presecretory stage ameloblasts. SSA: Secretory stage ameloblasts Od: Odontoblasts. De: Dentin. 133

157 Figure IV-9 Dental epithelial cell proliferation is increased in Hmgn2 -/- lower incisors. A-B). Immunofluorescence staining of a proliferation marker- Ki67 in P0 WT and Hmgn2 -/- incisors. C-D). Higher magnification of the LaCL region from WT or Hmgn2 -/- incisors in A and B. Nuclei are counterstained with DAPI. Scale bars, 100 µm. E). Quantitation of the Ki67-positive cells in the sections of lower incisors. The number of Ki67-positive cells in Hmgn2 -/- epithelial tissue is increased compared to WT embryos. F). Working model of a mir-23a/b:hmgn2:pitx2 signaling pathway in regulating craniofacial/incisor morphogenesis. Hmgn2 represses Pitx2 activation of amelogenin. mir-23a/b target Hmgn2 expression. Pitx2 feeds back to repress 23a/b expression. All these factors contribute to tight modulation of amelogenin expression for enamel deposition. 134

158 CHAPTER V SUMMARY AND FUTURE DIRECTIONS FoxO6 Regulates Hippo Signaling to Control Facial Morphology Summary Precise control of organ size is a key feature during animal development and tissue regeneration. In mammals, both extrinsic and intrinsic factors influence organ size. For example, extrinsic circulating factors such as hormones and insulin-like growth factors (IGF) promote organ size. Whereas, physiological perturbations, such as prolonged starvation, cause profound reduction of organ size. Other lines of evidences indicate that developing organs possess intrinsic mechanisms to modulate their final size (Heallen et al., 2011; Stanger, 2008). Extensive research led to the discovery of the Hippo signaling pathway as a key regulator of intrinsic mechanisms for organ size control (Buttitta and Edgar, 2007; Zhao et al., 2010). The activation of Hippo signaling by phosphorylation of the effector gene Yap leads to inhibited growth and reduction of organ size. Because Hippo signaling is ubiquitous in the developing embryo, specific factors must be present to modulate its activity to regulate tissue and organ specific growth throughout development. We have characterized FoxO6 expression during craniofacial developmental in our bioinformatics analyses of transcription factor gene regulatory networks. The FoxO6 knock-out mouse was generated and ablation of FoxO6 leads to expanded craniofacial and head size. Our gene expression array data compare gene expression between WT and FoxO6 -/- craniofacial region and shows Lats1, an important component of Hippo signaling, is significantly down-regulated in FoxO6 -/- embryos, leading us to hypothesize that FoxO6 may regulate Hippo signaling to control head/face morphology. 135

159 In this project, we used mouse models and cell-based assays to address how FoxO6 regulates Hippo signaling to control craniofacial growth. We have demonstrated that 1) FoxO6 expression is mainly observed in the brain, craniofacial region and somites during mouse embryogenesis. 2) The FoxO6 -/- head undergoes anterior expansion with defects in endochondral ossification. 3) FoxO6 regulates Hippo signaling by directly binding and activating Lats1 to control head shape and growth. 4) FoxO6 regulates Runx2 expression to control endochondral ossification. 5) FoxO6 regulates odontogenesis and lack of FoxO6 leads to larger incisors with increased amelogenin expression. 6) FoxO6 negatively regulates cell proliferation. 7) Human FoxO6 genetic variants are associated with facial phenotype ranging from retrognathism to prognathism. A schematic summary of this project is shown in Figure II-13. Signaling factors that modulate early face patterning and the segregation of tissues have been well characterized. Both Wnt and Shh signaling have been reported to regulate cranial neural crest migration to affect craniofacial patterning (Chai and Maxson, 2006). We think the facial morphology also has to be regulated in later stages after patterning. According to our data, FoxO6, which starts to be expressed at E12.5 in the mouse head when neural crest migration is finished, and is a good candidate for meditating this growth regulation. Our genetic study from human who have dento-skeletal bite problems has identified 3 SNPs in and around human FoxO6 that are associated with human facial morphology. This provides a genetic link between FoxO6 and variance of human facial morphology. Human carrying the rs (located in the first intron of FoxO6) rare allele tend to have bi-maxillary protrusion, and this feature partially resembles the facial morphology of FoxO6 -/- mice. These findings make FoxO6 a good candidate for 136

160 controlling anterior growth of the human face after patterning. Future Directions 1. To carry out functional analysis of 3 human FoxO6 SNPs that are associated with facial morphology ranging from retrognathism to prognathism. We have identified 3 SNPs in and around FoxO6 that are associated with dentoskeletal bite problems ranging from retrognathism to prognathism of both jaws. Rs is located 5.7Kb upstream of the FoxO6 TSS and more copies of the rare allele of this SNP were associated with bi-maxillary retrusion. Rs is located in the only intron of FoxO6 and the rare allele of this SNP is associated with bi-maxillary protrusion. Rs lies in 5.8kb downstream of FoxO6 3 UTR and the rare allele of this SNP is associated with a larger mandibular body length and an anterior cross bite relation. In order to understand the functional significance of these genetic variations in FoxO6, we could first take advantage of the encode data in UCSC genome browser to see if the SNPs are located in an enhancer site. We would check enhancer markers, such as DNase I sensitive sites, p300 binding sites and H3K4me1binding sites. Next we would analyze the DNA sequences around each SNPs (usually within 10bp) for potential transcription factor binding sites. If the polymorphism abolishes or creates a transcription factor binding site, then the SNP lies in a potential cis-regulatory element. If the SNP is located in a potential regulatory region, enhancer luciferase vectors will be generated by insertion of either the ancestral DNA sequences or the derived DNA sequences and co-transfected with the transcription factor in human oral epithelial or mesenchymal cells. Luciferase assays will be carried out to see if significant difference of 137

161 activities between two different enhancer luciferase reporter upon overexpressing of the transcription factor. If we do see a difference in luciferase activity, we will carry out gelshift assay to see if change from ancestral allele to derived allele affects the transcription factor binding. If we do see significant differences in both luciferase assays and gel-shift assays, we will feel more confident in claiming that the SNP is located in an enhancer region where the polymorphysm could affect transcription factor binding. To confirm if the transcription factor binds to FoxO6 through this enhancer region and regulates FoxO6 expression, we will either overexpress or knockdown this transcription factor in cells and examine the endogenous FoxO6 mrna and protein level. None of these 3 SNPs lie in the characterized long non-coding RNA (Lnc RNA) regions, making it unlikely that those SNPs affect Lnc RNA expression. Rs , located in the intron of FoxO6, does not appear to change an RNA splicing site according to GU-AG rule. 2. Strengthen our conclusion that FoxO6 regulates Hippo signaling by generating FoxO6 overexpressing mice We are currently generating FoxO6 conditional overexpressing (FoxO6 COE) mice controlled by a tissue specific Cre. Given the fact that most craniofacial tissues are mesenchymal tissues that are derived from the cranial neural crest, we will cross Wnt1- Cre mice with FoxO6 COE mice to drive the overexpression of FoxO6 in neural crest derived mesenchymal tissues in craniofacial region. Because overexpression of FoxO6 leads to activation of Hippo signaling component Lats1 and acts to enhance Hippo signaling, we expect to see a reduction in cell proliferation and craniofacial size in FoxO6 overexpressing mice. 138

162 It is known that ablation of some Hippo components in mice lead to increase organ size. For example, conditional inactivation of Mst1 and Mst2, the upstream kinase of Hippo signaling, leads to an enlarged liver (Lee et al., 2010). Deletion of Salvador (salv), which is a Mst kinase scaffolding protein, or Lats1 in the heart, leads to heart enlargement and cardiomegaly (Heallen et al., 2011). According to our data, FoxO6 activates Lats1 expression. Here I propose to use Lats1 ablated mice to rescue the smaller organ size phenotypes in FoxO6 overexpressing mice, or use Lats1 overexpressing mice to rescue the bigger organ size phenotypes in FoxO6 knock-out mice. If we can partially rescue the reduced/increased organ size phenotype, it will further support our conclusion that FoxO6 is a regulator of Hippo signaling. 3. Targeting FoxO6 in human disease might be beneficial for tissue regeneration. An increasing number of studies have shown Hippo signaling is associated with stem cell or progenitor cell expansion (Camargo et al., 2007; Gregorieff et al., 2015; Johnson and Halder, 2014). In general, enhanced YAP/TAZ activity is associated with stem cell and progenitor cell expansion that is coupled with an inhibition of differentiation, whereas inhibition of YAP/TAZ activity tends to have the opposite effects. On the other hand, Lats1/2 kinases that phosphorylates and inactivate Yap, function as negative regulator of reprogramming (Qin et al., 2012). Heallen et al. have reported that Hippo deficiency promotes efficient heart regeneration of the injured heart in both cardiac apex resection and adult myocardial infarction models(heallen et al., 2013). Another recent study shows Yap reprograms Lgr5 + stem cells to enhance tissue regeneration upon intestine injury (Gregorieff et al., 2015). 139

163 According to our research, FoxO6 activates Hippo signaling and causes YAP phosphorylation and inactivation, indicating that FoxO6 is a potential negative regulator of stem cell expansion and tissue regeneration. Thus, attenuation of FoxO6 expression may inhibit Hippo signaling and lead to increase of active nuclear Yap, resulting in reprogramming of differentiated cells to undifferentiated cells and expansion of stem and progenitor cell population. When animals or humans reach adulthood, most of their organs acquire a certain size and lose their regeneration capacity. But these organs still possess quiescent stem cells. These quiescent stem cells can re-enter the cell cycle upon injury. By targeting FoxO6 in tissues, it might be possible to facilitate the re-entry of quiescent stem cells into the cell cycle program and expand the numbers of stem cells, leading to the formation of new tissues. Since FoxO6 is mainly expressed in the brain and craniofacial region, gene or cell therapy by targeting FoxO6 may benefit tissue damage repair in these regions like human teeth. Pitx2:Sox2:Lef-1 network regulates dental stem cell maintenance and tooth development Summary Sox2 is a key transcription factor required for the development of multiple organs and tissues where it functions as a stem cell factor (Brazel et al., 2005; Ellis et al., 2004; Jayakody et al., 2012; Takahashi and Yamanaka, 2006b; Taranova et al., 2006b). Sox2 was recently reported to be a marker for dental epithelial stem cells (DESCs) in the mouse incisor. Sox2 + cells are located in labial cervical loop (LaCL) of the mouse incisor which gives rise to all the dental epithelial cell lineages including enamel- 140

164 secreting ameloblasts (Juuri et al., 2013b; Juuri et al., 2012b; Zhang et al., 2012). However, it remains unknown whether Sox2 is required for dental epithelial stem cell maintenance or proliferation during tooth initiation and growth. Lef-1 is required at early stages for tooth initiation as Lef-1 null mice have arrested tooth development at the bud stage (E12.5) (van Genderen et al., 1994a). Epithelial and mesenchymal tissue recombination assays showed that Lef-1 is required only transiently in the dental epithelium and dental lamina (Kratochwil et al., 1996b). The majority of Lef-1 expression is shifted to mesenchymal cells surrounding the epithelium at the bud stage, although Lef-1 expression persists in the basal cells of the epithelium immediately adjacent to the mesenchyme (Kratochwil et al., 1996b; Sasaki et al., 2005a). Both Sox2 and Lef-1 are markers of early craniofacial development and are expressed in the oral and dental epithelium (Juuri et al., 2013; Juuri et al., 2012; Sasaki et al., 2005; Zhang et al., 2012), however any potential Sox2-Lef-1 genetic interaction remains unclear. In this study, by using Sox2 loss of function and Lef-1 gain of function mouse models and cell-based in vitro assays, we demonstrated a new Pitx2:Sox2:Lef-1 transcriptional mechanism for dental epithelial stem cell maintenance, self-renewal, and dental development (Figure III-16). Specifically, we show that 1) Sox2 regulates periderm formation and that lack of Sox2 results in oral adhesions, ankyloglossia and cleft palate; 2) Sox2 conditional deletion results in incisor arrest at E16.5 due to lack of DESC self-renewal and abnormal molar formation; 3) Postnatal inducible deletion of Sox2 results in a reduction of incisor growth; 4) Sox2 and Lef-1 expression domains define the epithelial component of the initial dental placode; 5) Lef-1 activity controls establishment of stem and progenitor cell compartments in the mouse incisor; 6) Lef-1 141

165 overexpression partially rescues the Sox2 arrested tooth phenotype; 7) Sox2 represses Pitx2 activation of Lef-1 through an interaction with the HMG domain of Sox2. Future Directions 1. To explore the molecular mechanism of ectopic expression of amelogenin in lingual epithelium of the rescue mice. One of the unique findings from this project is the switched amelogenin expression from labial to lingual dental epithelium in Pitx2 Cre /Sox2 F/F /Lef-1 cki (rescue) lower incisors. In wild-type or Lef-1 overexpressing mice, only labial epithelium has stem cells which can differentiate to ameloblast cells to secret amelogenin, leading to asymmetrical deposition of enamel on the labial side and create a sharp tip of incisor (Tummers and Thesleff, 2003a; Wang et al., 2004). In the lingual side, the lingual cervical loop is thin and lacks active stem cells. The epithelial cells on the lingual side fail to differentiate and therefore do not generate ameloblasts (Tummers et al., 2007). Suppression of BMP and FGF on the lingual side of the dental epithelium has been reported to be necessary to prevent ameloblast formation. Ablation of Sprouty genes, which encodes antagonists of FGF, leads to bilateral enamel deposition and tusk-like incisors in mice (Klein et al., 2008a). BMP4 and Activin were also found to regulate DESCs proliferation and differentiation. Follistatin, a BMP inhibitor, limits the amount of stem cells in LiCL and contributes to the asymmetric deposition of enamel in rodent incisors by antagonizing the effect of Activin. Overexpressing Follistatin in mice cause hypoplastic teeth with no enamel on either side. However deletion of Follistatin leads to the presence of ameloblast on both sides of the dental epithelium (Wang et al., 2007b). 142

166 Overexpressing Noggin, another BMP inhibitor, disrupts ameloblast differentiation on the labial side of the incisor (Plikus et al., 2005). Here we hypothesize that Sox2 may regulate the expression of BMP and FGF inhibitors, and ablation of Sox2 leads to abolished/attenuated expression of BMP and FGF inhibitors in lingual dental epithelium and therefore allowing the generation of amelogenin-secreting ameloblasts. Because the dental epithelial cells on the labial side lack differentiation, the rescue mice do not produce amelogenin on labial side. Due to a lack of good antibodies for immunostaining, we propose to check the Spry4, Spry2, Follistatin and Noggin expression in rescue mice by in situ hybridization. By doing this, we will have better understanding of how Sox2 regulates asymmetrical patterning of ameloblasts and finally contributes to asymmetrical deposition of enamel. 2. To identify Sox2 target genes which mediate the cleft palate and ankyloglossia phenotypes in Sox2 cko mice. Overexpressing Lef-1 in Sox2 cko mice rescues the tooth agenesis phenotype but not the cleft palate or ankyloglossia. To ask which factors are affected by the ablation of Sox2 and contribute the cleft palate and ankyloglossia phenotypes in Sox2 cko mice, I propose to carry out RNA-sequencing to compare genome wide RNA expression level between WT and Sox2 cko palates. Among all the candidate genes which show significant difference by RNA-sequencing, we will select the genes which expression is changed more than 2 fold between the two genotypes. We will carry out literature searches and further narrow down our candidate genes by focusing on known genes which have been reported associated with palate or ankyloglossia (e.g. Irf6, Tbx22). We will perform realtime PCR and immunostaining to confirm the difference of expression level in the palate 143

167 tissue. Ultimately, if the mutant mouse models (either knock-out or overexpression) are available for those candidate genes, we will cross those mice with Sox2 cko mice to see if cleft palate and ankyloglossia can be rescued in Sox2 cko mice. 3. Reprogramming epithelial or mesenchymal cells to dental epithelial stem cell (DESC) by inducing Sox2, Lef-1 and Pitx2. Pitx2 is the earliest epithelial transcriptional marker of tooth development and the Pitx2 null mouse has arrested tooth development at E12.5 (Liu et al., 2003a; Lu et al., 1999). Lef-1 is expressed in the dental epithelium at dental placode stages (E11.5) and later shifts to dental mesenchyme where it is adjacent to the dental epithelium. The transient expression of Lef-1 in dental epithelium at E11.5 is a critical epithelial survival factor during tooth morphogenesis, as Lef-1 deficiency in mice results in tooth arrest in this stage (Sasaki et al., 2005b). Epithelial and mesenchymal tissue recombination assays showed that only the Lef-1 positive dental epithelium but not Lef-1 negative dental epithelium which is expressed in the later stage, can form teeth with any other kind of normal mesenchymal cells (Kratochwil et al., 1996a). Sox2 is a dental epithelial marker and Sox2 + dental epithelial stem cells can give rise to all epithelial cell lineages in the teeth (Juuri et al., 2012a). According to our data, Sox2 in a critical factor for dental epithelial stem cells maintenance and lack of Sox2 leads to tooth arrest at E16.5. Taken together, these observations provide evidence that Pitx2, Sox2 and Lef1 are the major regulators of dental epithelial stem cells and tooth initiation. A previous study from our lab has shown that oral epithelial and odontoblast mesenchymal cells can be reprogrammed to amelogenin-expressing dental epithelial cells by a two-step induction method using Pitx2 and mir-200a (Sharp et al., 2014a). The 144

168 most famous example of cell reprogramming is the induced Pluripotent Stem (ips) cells in which four transcription factors (Oct3/4, Sox2, c-myc and Klf4) are sufficient to reprogram fibroblast or keratinocytes to pluripotent stem cells (Takahashi and Yamanaka, 2006a). Here we hypothesize that reprogramming dental/oral epithelial cells or dental/oral mesenchymal cells to dental epithelial stem cells could be achieved by inducing Sox2, Lef-1 and Pitx2. We will overexpress Sox2, Lef-1 and Pitx2 in LS-8 (oral epithelial cells) and MDPC (dental mesenchymal cells) cells and culture them in ES cell medium. We will stain for some dental epithelial stem cell marker such as Bmi1, Gli1 and Lunatic fringe (Lfng) to see if Sox2, Lef-1 and Pitx2 overexpressing cells express those stem cell markers. If the those cells do express or partially express those stem cell markers, I propose to carry out the tissue recombination assay, which we can take isolated dental mesenchymal tissue or any other kinds of mesenchymal tissues and culture them with reprogrammed cells for several days, and then transplant under the kidney capsule of adult mice, to see if they can form fully developed teeth. A mir-23a/b:hmgn2:pitx2 signaling pathway regulates craniofacial/incisor morphogenesis Summary Dental enamel, the hardest tissue of the body, is an epithelial-derived tissue comprised of highly organized hydroxyapatite crystals covering part or all of the crown in the mammalian tooth (Gadhia et al., 2012). In the developing enamel, 30% of the components are proteins, of which 90% is amelogenin (Cobourne and Sharpe, 2013). Amelogenin is secreted by dental epithelial cells ameloblasts. Too much or too little secretion of amelogenin affects enamel deposition, so its expression must be tightly 145

169 controlled. Defects in controlling amelogenin expression lead to dental diseases, such as Amelogenesis Imperfect (AI) (Gadhia et al., 2012). Previous findings from our lab revealed that Pitx2, the earliest transcription factor observed in tooth development and expressed exclusively in the dental epithelium, directly activates amelogenin expression (Li et al., 2014; Venugopalan et al., 2011). Pitx2 Mutations are associated with Axenfeld-Rieger Sydrome (ARS) which display severe craniofacial abnormalities including enamel defects (Murray et al., 1992). Hmgn2 belongs to chromatin-associated high mobility group protein and acts as a negative regulator for Pitx2-mediated activation of Amelogenin (Amen et al., 2008a; Li et al., 2014). In this project, we discovered that mirna-23a/b target Hmgn2 expression to modulate amelogenin expression. By using Hmgn2 loss of function mouse model, we characterized the roles of Hmgn2 as a negative regulator of odontogenesis. We demonstrated the central role of mir-23a/b:hmgn2:pitx2 signaling pathway in regulating odontogenesis and craniofacial morphogenesis (Figure IV-9). In summary, we have shown that 1) Hmgn2 physically interacts with PITX2 in the nucleus and Hmgn2 represses Pitx2 transcriptional activity. 2) mir-23a/b repress Hmgn2 expression in oral epithelial cells. 3) mir-23a/b indirectly activates Pitx2 and amelogenin expression by repressing Hmgn2. 4) mir-23a/b and amelogenin mrna expression is negatively correlated with Hmgn2 mrna expression in mouse embryonic craniofacial development. 5) Pitx2 represses mir-23a/b expression. 6) Hmgn2 -/- incisors undergo abnormal expansion with increased dental epithelial cell proliferation and enamel formation. 146

170 Future Directions 1. To explore the genome wide targets of Hmgn2 in mouse embryonic development. Hmgn2 imparts structural and functional plasticity to the chromatin fibers and thus modulate transcriptional activity (Hock et al., 2007; Nishino et al., 2008). According to previously studies (Furusawa et al., 2006; Li et al., 2014) and the finding in this paper, both Hmgn1 and Hmgn2 are ubiquitously and highly expressed in all embryonic tissues at early stages, but their expression were down-regulated as mouse embryogenesis progressed. Ablation of Hmgn1 in mice leads to a wide range of phenotypes, including defects in DNA repair upon UV or ionizing radiation, defective stress responses and abnormal development of corneal epithelium (Birger et al., 2005; Lim et al., 2004). There is one study showing that Hmgn2 is deleted in human uterine leiomyoma (Polito et al., 1999). With the availability of Hmgn1 knock-out mice, the role of Hmgn1 in early mouse embryonic development has been well characterized. To date, no publication has been reported a Hmgn2 knock-out mouse. This probably why very little is known about the roles of Hmgn2 in early mouse embryonic development. With the ubiquitous and wide expression pattern of Hmgn2 in early embryonic development, we hypothesize that Hmgn2 also plays roles in a wide range of activities in development and disease. To broaden our understanding of the gene regulatory networks that require Hmgn2 in early mouse embryonic development, we propose to identify the genome-wide target genes of Hmgn2 by performing ChIP-sequencing in E14.5 WT mouse embryonic tissues followed with RNA-Sequencing to compare the gene expression between E14.5 WT and 147

171 Hmgn2 -/- embryonic tissues. We will carry out bioinformatics analysis and focus on the genes which are overlapping in both sequencing results. After we identify these target genes, we hope to have a better understanding of the roles of Hmgn2 in development and disease. 2. Potential gene therapy for enamel defects in Axenfeld-Rieger Syndrome (ARS) patients by inducing mir-23a/b and Pitx2 Axenfeld-Rieger Syndrome (ARS) is a rare, autosomal-dominant human genetic disorder. The phenotypes of ARS exhibit a wide spectrum of developmental defects, including umbilical anomalies, eye defects and craniofacial abnormalities (Amendt, 2005). Among all these defects, PITX2 mutations are associated with tooth defects including enamel hypoplasia (Li et al., 2014). Both our finding and a previous paper from our lab (Li et al., 2014) demonstrate that Pitx2 is a major activator of amelogenesis, whose product is important for enamel deposition. In this paper we identified Hmgn2 as a negative regulator for this process and Hmgn2 expression can be target by mir-23a/b. K14-Hmgn2 transgenic mouse line, which phenocopies the enamel hypoplasia observed in ARS patients, is a great mouse model for studying ARS (Li et al., 2014). Here, we propose to use gene therapies to treat the enamels defects of K14-Hmgn2 transgenic mice by introducing Pitx2 and mir-23a/b. The hypothesis here is: by overexpressing Pitx2, we expect to see Pitx2 activating amelogenin expression and produce more enamel to compensate the loss of enamel phenotype in ARS patient; by overexpressing mir-23a/b, we can abolish Hmgn2 expression and thus relieve Hmgn2 repression of Pitx2 transcriptional activation of amelogenin. There are several ways we can design delivery vectors specific for teeth. We can use adenoviruses encoding Pitx2 148

172 and mir-23a/b on a collagen matrix and put it on the root of teeth, or we can use nanoparticles to wrap Pitx2 and mir-23a/b expression plasmids and put the nanoparticles in the matrix for delivery. One challenge here is to figure out the optimum dose for therapy. Another challenge is the potential inflammatory response after gene delivery. A blood test would be highly recommended for detecting any potential histo-pathological changes. By using ARS mouse model, we hope to see that overexpressing mir-23a/b and Pitx2 rescues the enamel defects. Our ultimate goal is to translate our findings to dental clinics to benefit ARS patients and other patients who have enamel defects. Working model of this thesis We have identified the transcriptional factor FoxO6 as an activator of Hippo signaling with specific expression in the brain and craniofacial tissues. FoxO6 loss-offunction mice undergo increases in cell proliferation which finally leads to lengthening of the incisors, expansion of the face and skull and enlargement of the mandible and maxilla. We have screened three human FOXO6 single nucleotide polymorphisms which are associated with facial morphology ranging from retrognathism to prognathism. rs alters AP-1 binding and activation of FoxO6 and more copies of the rare allele for this SNP were associated with bi-maxillary retrusion. Additionally, we also characterized that Pitx2 directly binds and activates FoxO6. Our study also reveals that Sox2 and Lef-1, two markers for early craniofacial development, are regulated by Pitx2 to control DESC maintenance, differentiation and craniofacial development. The loss of Sox2 in DESCs leads to impaired stem cell proliferation, migration and subsequent dissolution of the tooth germ. On the other hand, conditional overexpression of Lef-1 in 149

173 oral and dental epithelial region increases DESC proliferation and creates a new labial cervical loop stem cell compartment in dental epithelial stem cell niche, which produces rapidly growing long tusk-like incisors. Interestingly, Lef-1 overexpression rescues the tooth arrest defects in Sox2 conditional deletion mice. Our data also reveal that mirna and histone remodeler are involved in regulating DESC proliferation and craniofacial morphogenesis. We also describe a mir-23a/b:hmgn2:pitx2 signaling pathway in regulating dental epithelial cell growth and differentiation. mir-23a /b directly target Hmgn2, a repressor for Pitx2 transcriptional activation of Amelogenin, leading to enhanced Amelogenin production. Phenotypically, ablation of Hmgn2 in mice results in an overgrowth of incisors with increased Amelogenin expression. Our findings not only provide new molecular mechanisms for controlling dental stem cell fate and bases for developing novel stem cell therapy or gene therapy for tooth regeneration and dental diseases. 150

174 Figure V-1 New molecular mechanisms controlling dental epithelial stem cell (DESC) maintenance, growth and craniofacial morphogenesis. FoxO6, a direct target of Pitx2, controls DESC proliferation by regulating Hippo signaling in craniofacial region. rs , a polymorphism located nearly 5.8Kb upstream of the FoxO6 transcription start site (TSS), alters AP-1 binding and activation of FoxO6. Sox2 and Lef-1, two markers for early craniofacial development, are regulated by Pitx2 to control DESC maintenance, differentiation and craniofacial development. Sox2 also feedbacks and represses Pitx2 transcriptional activities. mir-23a /b directly target Hmgn2, a repressor for Pitx2 transcriptional activation of Amelogenin, leading to enhanced Amelogenin production. Pitx2 represses mir-23a/b expression, forming a negative feedback loop to allow a proper amount of Amelogenin to be expressed in teeth. 151

175 REFERENCES Amen, M., Espinoza, H., Cox, C., Liang, X., Wang, J., Link, T., Brennan, R., Martin, J., and Amendt, B. (2008a). Chromatin-associated HMG-17 is a major regulator of homeodomain transcription factor activity modulated by Wnt/beta-catenin signaling. Nucleic Acids Res 36, Amen, M., Espinoza, H.M., Cox, C., Liang, X., Wang, J., Link, T.M., Brennan, R.G., Martin, J.F., and Amendt, B.A. (2008b). Chromatin-associated HMG-17 is a major regulator of homeodomain transcription factor activity modulated by Wnt/beta-catenin signaling. Nuc Acids Res 36, Amen, M., Liu, X., Vadlamudi, U., Elizondo, G., Diamond, E., Engelhardt, J.F., and Amendt, B.A. (2007). PITX2 and b-catenin Interactions Regulate Lef-1 Isoform Expression. Mol Cell Biol 27, Amendt, B.A. (2005). The Molecular Mechanisms of Axenfeld Rieger Syndrome (New York, NY, USA: Springer). Arnold, K., Sarkar, A., Yram, M., Polo, J., Bronson, R., Sengupta, S., Seandel, M., Geijsen, N., and Hochedlinger, K. (2011). Sox2(+) adult stem and progenitor cells are important for tissue regeneration and survival of mice. Cell Stem Cell 9, Avilion, A., Nicolis, S., Pevny, L., Perez, L., Vivian, N., and Lovell-Badge, R. (2003a). Multipotent cell lineages in early mouse development depend on SOX2 function. Genes & development 17, Avilion, A.A., Nicolis, S.K., Pevny, L.H., Perez, L., Vivian, N., and Lovell-Badge, R. (2003b). Multipotent cell lineages in early mouse development depend on SOX2 function. Genes Dev 17, Bartel, D. (2004). MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, Biehs B, Hu, J.K., Strauli, N.B., Sangiorgi, E., Jung, H., Heber, R.P., Ho, S., Goodwin, A.F., Dasen, J.S., Capecchi, M.R., et al. (2013). BMI1 represses Ink4a/Arf and Hox genes to regulate stem cells in the rodent incisor. Nat Cell Biol 15, Biehs, B., Hu, J., Strauli, N., Sangiorgi, E., Jung, H., Heber, R., Ho, S., Goodwin, A., Dasen, J., Capecchi, M., et al. (2013). BMI1 represses Ink4a/Arf and Hox genes to regulate stem cells in the rodent incisor. Nat Cell Biol 15, Birger, Y., Catez, F., Furusawa, T., Lim, J.H., Prymakowska-Bosak, M., West, K.L., Postnikov, Y.V., Haines, D.C., and Bustin, M. (2005). Increased tumorigenicity and sensitivity to ionizing radiation upon loss of chromosomal protein HMGN1. Cancer research 65, Blackburn, J., Ohazama, A., Kawasaki, K., Otsuka-Tanaka, Y., Liu, B., Honda, K., Rountree, R.B., Hu, Y., Kawasaki, M., Birchmeier, W., et al. (2012). The role of Irf6 in tooth epithelial invagination. Dev Biol 365, Bourgine, A., Pilet, P., Diouani, S., Sourice, S., Lesoeur, J., Beck-Cormier, S., Khoshniat, S., Weiss, P., Friedlander, G., Guicheux, J., et al. (2013). Mice with hypomorphic expression of the sodium-phosphate cotransporter PiT1/Slc20a1 have an unexpected normal bone mineralization. PloS one 8, e

176 Boyer, L.A., Lee, T.I., Cole, M.F., Johnstone, S.E., Levine, S.S., Zucker, J.P., Guenther, M.G., Kumar, R.M., Murray, H.L., Jenner, R.G., et al. (2005). Core transcriptional regulatory circuitry in human embryonic stem cells. Cell 122, Braybrook, C., Doudney, K., Marçano, A.C., Arnason, A., Bjornsson, A., Patton, M.A., Goodfellow, P.J., Moore, G.E., and Stanier, P. (2001). The T-box transcription factor gene TBX22 is mutated in X-linked cleft palate and ankyloglossia. Nat Genet 29, Brazel, C.Y., Limke, T.L., Osborne, J.K., Miura, T., Cai, J., Pevny, L., and Rao, M.S. (2005). Sox2 expression defines a heterogeneous population of neurosphere-forming cells in the adult murine brain. Aging Cell 4, Buttitta, L.A., and Edgar, B.A. (2007). How size is controlled: from Hippos to Yorkies. Nature cell biology 9, Camargo, F.D., Gokhale, S., Johnnidis, J.B., Fu, D., Bell, G.W., Jaenisch, R., and Brummelkamp, T.R. (2007). YAP1 increases organ size and expands undifferentiated progenitor cells. Current biology : CB 17, Cao, H., Florez, S., Amen, M., Huynh, T., Skobe, Z., Baldini, A., and Amendt, B.A. (2010a). Tbx1 regulates progenitor cell proliferation in the dental epithelium by modulating Pitx2 activation of p21. Dev Biol 347, Cao, H., Jheon, A., Li, X., Sun, Z., Wang, J., Florez, S., Zhang, Z., McManus, M., Klein, O., and Amendt, B. (2013a). The Pitx2:miR-200c/141:noggin pathway regulates Bmp signaling and ameloblast differentiation. Development 140, Cao, H., Jheon, A., Li, X., Sun, Z., Wang, J., Florez, S., Zhang, Z., McManus, M.T., Klein, O.D., and Amendt, B.A. (2013b). The Pitx2:miR-200c/141:noggin pathway regulates Bmp signaling and ameloblast differentiation. Development 140, Cao, H., Wang, J., Li, X., Florez, S., Huang, Z., Venugopalan, S., Elangovan, S., Skobe, Z., Margolis, H., Martin, J., et al. (2010b). MicroRNAs play a critical role in tooth development. Journal of dental research 89, Cao, H., Wang, J., Li, X., Florez, S., Huang, Z., Venugopalan, S.R., Elangovan, S., Skobe, Z., Margolis, H.C., Martin, J.F., et al. (2010c). MicroRNAs play a critical role in tooth development. J Dent Res 89, Chai, Y., and Maxson, R.E., Jr. (2006). Recent advances in craniofacial morphogenesis. Developmental dynamics : an official publication of the American Association of Anatomists 235, Chen, Q., Xu, J., Li, L., Li, H., Mao, S., Zhang, F., Zen, K., Zhang, C., and Zhang, Q. (2014). MicroRNA-23a/b and microrna-27a/b suppress Apaf-1 protein and alleviate hypoxia-induced neuronal apoptosis. Cell Death Dis 5, e1132. Cheverud, J.M., Routman, E.J., Duarte, F.A., van Swinderen, B., Cothran, K., and Perel, C. (1996). Quantitative trait loci for murine growth. Genetics 142, Clavel, C., Grisanti, L., Zemla, R., Rezza, A., Barros, R., Sennett, R., Mazloom, A., Chung, C., Cai, X., Cai, C., et al. (2012a). Sox2 in the dermal papilla niche controls hair growth by fine-tuning BMP signaling in differentiating hair shaft progenitors. Developmental cell 23, Clavel, C., Grisanti, L., Zemla, R., Rezza, A., Barros, R., Sennett, R., Mazloom, A.R., Chung, C.Y., Cai, X., Cai, C.L., et al. (2012b). Sox2 in the dermal papilla niche controls hair growth by fine-tuning BMP signaling in differentiating hair shaft progenitors. Dev Cell 23,

177 Cobourne, M.T., and Sharpe, P.T. (2013). Diseases of the tooth: the genetic and molecular basis of inherited anomalies affecting the dentition. Wiley interdisciplinary reviews Developmental biology 2, Cox, T.C. (2004). Taking it to the max: the genetic and developmental mechanisms coordinating midfacial morphogenesis and dysmorphology. Clinical genetics 65, Cvijanovic, M., Ivanovic, A., Kalezic, M.L., and Zelditch, M.L. (2014). The ontogenetic origins of skull shape disparity in the Triturus cristatus group. Evolution & development 16, DasGupta, R., and Fuchs, E. (1999). Multiple roles for activated LEF/TCF transcription complexes during hair follicle development and differentiation. Development 126, 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. The Journal of investigative dermatology 133, Eberhart, J.K., Swartz, M.E., Crump, J.G., and Kimmel, C.B. (2006). Early Hedgehog signaling from neural to oral epithelium organizes anterior craniofacial development. Development 133, Ellis, P., Fagan, B.M., Magness, S.T., Hutton, S., Taranova, O., Hayashi, S., McMahon, A., Rao, M., and Pevny, L. (2004). SOX2, a persistent marker for multipotential neural stem cells derived from embryonic stem cells, the embryo or the adult. Dev Neurosci 26, Furusawa, T., Lim, J.H., Catez, F., Birger, Y., Mackem, S., and Bustin, M. (2006). Down-regulation of nucleosomal binding protein HMGN1 expression during embryogenesis modulates Sox9 expression in chondrocytes. Molecular and cellular biology 26, Gadhia, K., McDonald, S., Arkutu, N., and Malik, K. (2012). Amelogenesis imperfecta: an introduction. British dental journal 212, Gao, P., Tchernyshyov, I., Chang, T., Lee, Y., Kita, K., Ochi, T., Zeller, K., De Marzo, A., Van Eyk, J., Mendell, J., et al. (2009). c-myc suppression of mir-23a/b enhances mitochondrial glutaminase expression and glutamine metabolism. Nature 458, Gao, S., Moreno, M., Eliason, S., Cao, H., Li, X., Yu, W., Bidlack, F.B., Margolis, H.C., Baldini, A., and Amendt, B.A. (2015). TBX1 protein interactions and microrna-96-5p regulation controls cell proliferation during craniofacial and dental development: implications for 22q11.2 deletion syndrome. Hum Mol Genet 24, Gerlitz, G. (2010). HMGNs, DNA repair and cancer. Biochimica et biophysica acta 1799, Gilchrist, E.P., Moyer, M.P., Shillitoe, E.J., Clare, N., and Murrah, V.A. (2000). Establishment of a human polyclonal oral epithelial cell line. Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics 90, Gregorieff, A., Liu, Y., Inanlou, M.R., Khomchuk, Y., and Wrana, J.L. (2015). Yapdependent reprogramming of Lgr5(+) stem cells drives intestinal regeneration and cancer. Nature 526,

178 Grimson, A., Farh, K., Johnston, W., Garrett-Engele, P., Lim, L., and Bartel, D. (2007). MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell 27, Guo, X., and Wang, X.F. (2009). Signaling cross-talk between TGF-beta/BMP and other pathways. Cell research 19, Hallgrimsson, B., and Lieberman, D.E. (2008). Mouse models and the evolutionary developmental biology of the skull. Integrative and comparative biology 48, Hannenhalli, S., and Kaestner, K.H. (2009). The evolution of Fox genes and their role in development and disease. Nature reviews Genetics 10, Harada, H., Kettunen, P., Jung, H.S., Mustonen, T., Wang, Y.A., and Thesleff, I. (1999). Localization of putative stem cells in dental epithelium and their association with Notch and FGF signaling. The Journal of cell biology 147, Harada, H., Toyono, T., Toyoshima, K., and Ohuchi, H. (2002). FGF10 maintains stem cell population during mouse incisor development. Connective tissue research 43, He, F., Xiong, W., Wang, Y., Li, L., Liu, C., Yamagami, T., Taketo, M.M., Zhou, C., and Chen, Y. (2011). Epithelial Wnt/β-catenin signaling regulates palatal shelf fusion through regulation of Tgfβ3 expression. Dev Biol 350, Heallen, T., Morikawa, Y., Leach, J., Tao, G., Willerson, J.T., Johnson, R.L., and Martin, J.F. (2013). Hippo signaling impedes adult heart regeneration. Development 140, Heallen, T., Zhang, M., Wang, J., Bonilla-Claudio, M., Klysik, E., Johnson, R.L., and Martin, J.F. (2011). Hippo pathway inhibits Wnt signaling to restrain cardiomyocyte proliferation and heart size. Science 332, Hock, R., Furusawa, T., Ueda, T., and Bustin, M. (2007). HMG Chromosomal proteins in development and disease. Trends Cell Biol 17, Hoekman, M.F., Jacobs, F.M., Smidt, M.P., and Burbach, J.P. (2006). Spatial and temporal expression of FoxO transcription factors in the developing and adult murine brain. Gene expression patterns : GEP 6, Holbrook, K.A., and Odland, G.F. (1975). The fine structure of developing human epidermis: light, scanning, and transmission electron microscopy of the periderm. The Journal of investigative dermatology 65, Hu, C.-D., Chinenov, Y., and Kerppola, T.K. (2002). Visualization of molecular interactions by fluorescence complementation. Nat Rev 9, Hu, D., and Marcucio, R.S. (2009). A SHH-responsive signaling center in the forebrain regulates craniofacial morphogenesis via the facial ectoderm. Development 136, Hu, M.C., and Rosenblum, N.D. (2005). Smad1, beta-catenin and Tcf4 associate in a molecular complex with the Myc promoter in dysplastic renal tissue and cooperate to control Myc transcription. Development 132, Huang, C., and Qin, D. (2010). Role of Lef1 in sustaining self-renewal in mouse embryonic stem cells. J Genet Genomics 37, Huang, J., Wu, S., Barrera, J., Matthews, K., and Pan, D. (2005). The Hippo signaling pathway coordinately regulates cell proliferation and apoptosis by inactivating Yorkie, the Drosophila Homolog of YAP. Cell 122,

179 Jacobs, F.M., van der Heide, L.P., Wijchers, P.J., Burbach, J.P., Hoekman, M.F., and Smidt, M.P. (2003). FoxO6, a novel member of the FoxO class of transcription factors with distinct shuttling dynamics. The Journal of biological chemistry 278, Jayakody, S.A., Andoniadou, C.L., Gaston-Massuet, C., Signore, M., Cariboni, A., Bouloux, P.M., Le Tissier, P., Pevny, L.H., Dattani, M.T., and Martinez-Barbera, J.P. (2012). SOX2 regulates the hypothalamic-pituitary axis at multiple levels. J Clin Invest 122, Jernvall, J., and Thesleff, I. (2012). Tooth shape formation and tooth renewal: evolving with the same signals. Development 139, Johnson, R., and Halder, G. (2014). The two faces of Hippo: targeting the Hippo pathway for regenerative medicine and cancer treatment. Nature reviews Drug discovery 13, Jumlongras, D., Bei, M., Stimson, J.M., Wang, W.F., DePalma, S.R., Seidman, C.E., Felbor, U., Maas, R., Seidman, J.G., and Olsen, B.R. (2001). A nonsense mutation in MSX1 causes Witkop syndrome. American journal of human genetics 69, Juuri, E., Jussila, M., Seidel, K., Holmes, S., Wu, P., Richman, J., Heikinheimo, K., Chuong, C., Arnold, K., Hochedlinger, K., et al. (2013a). Sox2 marks epithelial competence to generate teeth in mammals and reptiles. Development 140, Juuri, E., Jussila, M., Seidel, K., Holmes, S., Wu, P., Richman, J., Heikinheimo, K., Chuong, C.-M., Arnold, K., Hochedlinger, K., et al. (2013b). Sox2 marks epithelial competence to generate teeth in mammals and reptiles. Development 140, Juuri, E., Saito, K., Ahtiainen, L., Seidel, K., Tummers, M., Hochedlinger, K., Klein, O.D., Thesleff, I., and Michon, F. (2012a). Sox2+ stem cells contribute to all epithelial lineages of the tooth via Sfrp5+ progenitors. Developmental cell 23, Juuri, E., Saito, K., Ahtiainen, L., Seidel, K., Tummers, M., Hochedlinger, K., Klein, O.D., Thesleff, I., and Michon, F. (2012b). Sox2+ Stem Cells Contribute to All Epithelial Lineages of the Tooth via Sfrp5+ Progenitors. Dev Cell 23, Kaestner, K.H., Knochel, W., and Martinez, D.E. (2000). Unified nomenclature for the winged helix/forkhead transcription factors. Genes & development 14, Kantaputra, P.N., Paramee, M., Kaewkhampa, A., Hoshino, A., Lees, M., McEntagart, M., Masrour, N., Moore, G.E., Pauws, E., and Stanier, P. (2011). Cleft lip with cleft palate, ankyloglossia, and hypodontia are associated with TBX22 mutations. J Dent Res 90, Kaukua, N., Shahidi, M.K., Konstantinidou, C., Dyachuk, V., Kaucka, M., Furlan, A., An, Z., Wang, L., Hultman, I., Ahrlund-Richter, L., et al. (2014). Glial origin of mesenchymal stem cells in a tooth model system. Nature 513, Klein, O., Lyons, D., Balooch, G., Marshall, G., Basson, M., Peterka, M., Boran, T., Peterkova, R., and Martin, G. (2008a). An FGF signaling loop sustains the generation of differentiated progeny from stem cells in mouse incisors. Development 135, Klein, O.D., Lyons, D.B., Balooch, G., Marshall, G.W., Basson, M.A., Peterka, M., Boran, T., Peterkova, R., and Martin, G.R. (2008b). An FGF signaling loop sustains the generation of differentiated progeny from stem cells in mouse incisors. Development 135,

180 Klein, O.D., Minowada, G., Peterkova, R., Kangas, A., Yu, B.D., Lesot, H., Peterka, M., Jernvall, J., and Martin, G.R. (2006). Sprouty genes control diastema tooth development via bidirectional antagonism of epithelial-mesenchymal FGF signaling. Dev Cell 11, Kleindorp, R., Flachsbart, F., Puca, A.A., Malovini, A., Schreiber, S., and Nebel, A. (2011). Candidate gene study of FOXO1, FOXO4, and FOXO6 reveals no association with human longevity in Germans. Aging cell 10, Kondo, S., Schutte, B.C., Richardson, R.J., Bjork, B.C., Knight, A.S., Watanabe, Y., Howard, E., Ferreira de Lima, R.L.L., Daack-Hirsch, S., Sander, A., et al. (2002). Mutations in IRF6 cause Van der Woude and popliteal pterygium syndromes. Nat Genet 32, Kratochwil, K., Dull, M., Farinas, I., Galceran, J., and Grosschedl, R. (1996a). Lef1 expression is activated by BMP-4 and regulates inductive tissue interactions in tooth and hair development. Genes Dev 10, Kratochwil, K., Dull, M., Farinas, I., Galceran, J., and Grosschedl, R. (1996b). Lef1 expression is activated by BMP-4 and regulates inductive tissue interactions in tooth and hair development. Genes Dev 10, Kratochwil, K., Galceran, J., Tontsch, S., Roth, W., and Grosschedl, R. (2002). FGF4, a direct target of LEF1 and Wnt signaling, can rescue the arrest of tooth organogenesis in Lef1(-/-) mice. Genes Dev 16, Lane, S., Williams, D., and Watt, F. (2014). Modulating the stem cell niche for tissue regeneration. Nat Biotechnol 32. Lee, K.P., Lee, J.H., Kim, T.S., Kim, T.H., Park, H.D., Byun, J.S., Kim, M.C., Jeong, W.I., Calvisi, D.F., Kim, J.M., et al. (2010). The Hippo-Salvador pathway restrains hepatic oval cell proliferation, liver size, and liver tumorigenesis. Proceedings of the National Academy of Sciences of the United States of America 107, Lehtonen, S., Olkkonen, V.M., Stapleton, M., Zerial, M., and Lehtonen, E. (1998). HMG- 17, a chromosomal non-histone protein, shows developmental regulation during organogenesis. The International journal of developmental biology 42, Lei, Q.Y., Zhang, H., Zhao, B., Zha, Z.Y., Bai, F., Pei, X.H., Zhao, S., Xiong, Y., and Guan, K.L. (2008). TAZ promotes cell proliferation and epithelial-mesenchymal transition and is inhibited by the hippo pathway. Molecular and cellular biology 28, Li, A., Song, T., Wang, F., Liu, D., Fan, Z., Zhang, C., He, J., and Wang, S. (2012a). MicroRNAome and expression profile of developing tooth germ in miniature pigs. PloS one 7, e Li, C.Y., Cha, W.H., Luder, H.U., Charles, R.P., McMahon, M., Mitsiadis, T.A., and Klein, O.D. (2012b). E-cadherin regulates the behavior and fate of epithelial stem cells and their progeny in the mouse incisor. Developmental Biology 366, Li, J., Feng, J., Liu, Y., Ho, T.V., Grimes, W., Ho, H.A., Park, S., Wang, S., and Chai, Y. (2015). BMP-SHH signaling network controls epithelial stem cell fate via regulation of its niche in the developing tooth. Dev Cell 33, Li, Q., Zhang, N., Jia, Z., Le, X., Dai, B., Wei, D., Huang, S., Tan, D., and Xie, K. (2009). Critical role and regulation of transcription factor FoxM1 in human gastric cancer angiogenesis and progression. Cancer research 69,

181 Li, X., Venugopalan, S., Cao, H., Pinho, F., Paine, M.L., Snead, M.L., Semina, E.V., and Amendt, B.A. (2013). A Model for the Molecular Underpinnings of Tooth Defects in Axenfeld-Rieger Syndrome.. Hum Mol Genet 23, Li, X., Venugopalan, S.R., Cao, H., Pinho, F.O., Paine, M.L., Snead, M.L., Semina, E.V., and Amendt, B.A. (2014). A model for the molecular underpinnings of tooth defects in Axenfeld-Rieger syndrome. Hum Mol Genet 23, Lim, J.H., Catez, F., Birger, Y., West, K.L., Prymakowska-Bosak, M., Postnikov, Y.V., and Bustin, M. (2004). Chromosomal protein HMGN1 modulates histone H3 phosphorylation. Mol Cell 15, Liu, H., Jiang, D., Chi, F., and Zhao, B. (2012). The Hippo pathway regulates stem cell proliferation, self-renewal, and differentiation. Protein & cell 3, Liu, W., Selever, J., Lu, M.F., and Martin, J.F. (2003a). Genetic dissection of Pitx2 in craniofacial development uncovers new functions in branchial arch morphogenesis, late aspects of tooth morphogenesis and cell migration. Development 130, Liu, W., Selever, J., Lu, M.F., and Martin, J.F. (2003b). Genetic dissection of Pitx2 in craniofacial development uncovers new functions in branchial arch morphogenesis, late aspects of tooth morphogenesis and cell migration. Development 130, Liu, Y., Miao, W., Wang, J., Gao, P., Yin, G., Zhang, L., Lv, C., Ji, Z., Yu, T., Sabel, B.A., et al. (2013). Structural abnormalities in early Tourette syndrome children: a combined voxel-based morphometry and tract-based spatial statistics study. PloS one 8, e Lowry, W.E., Blanpain, C., Nowak, J.A., Guasch, G., Lewis, L., and Fuchs, E. (2005). Defining the impact of beta-catenin/tcf transactivation on epithelial stem cells. Genes Dev 19, Lu, M., Pressman, C., Dyer, R., Johnson, R.L., and Martin, J.F. (1999a). Function of Rieger syndrome gene in left-right asymmetry and craniofacial development. Nature 401, Lu, M.F., Pressman, C., Dyer, R., Johnson, R.L., and Martin, J.F. (1999). Function of Rieger syndrome gene in left-right asymmetry and craniofacial development. Nature 401, MacDonald, B.T., Tamai, K., and He, X. (2009). Wnt/beta-catenin signaling: components, mechanisms, and diseases. Developmental cell 17, Marcucio, R.S., Young, N.M., Hu, D., and Hallgrimsson, B. (2011). Mechanisms that underlie co-variation of the brain and face. Genesis 49, Michon, F. (2011). Tooth evolution and dental defects: from genetic regulation network to micro-rna fine-tuning. Birth defects research Part A, Clinical and molecular teratology 91, Miletich, I., and Sharpe, P.T. (2004). Neural crest contribution to mammalian tooth formation. Birth defects research Part C, Embryo today : reviews 72, Minoux, M., and Rijli, F.M. (2010). Molecular mechanisms of cranial neural crest cell migration and patterning in craniofacial development. Development 137, Moore, K., and Lemischka, I. (2006). Stem cells and their niches. Science 311, Moore, K.E., Mills, J.F., and Thornton, M.M. (2006). Alternative sources of adult stem cells: a possible solution to the embryonic stem cell debate. Gend Med 3,

182 Murray, J.C., Bennett, S.R., Kwitek, A.E., Small, K.W., Schinzel, A., Alward, W.L., Weber, J.L., Bell, G.I., and Buetow, K.H. (1992). Linkage of Rieger syndrome to the region of the epidermal growth factor gene on chromosome 4. Nature genetics 2, Nishino, J., Kim, I., Chada, K., and Morrison, S.J. (2008). Hmga2 promotes neural stem cell self-renewal in young but not old mice by reducing p16ink4a and p19arf Expression. Cell 135, O'Dwyer, E.M., and Jones, D.C. (2005). Dental anomalies in Axenfeld-Rieger syndrome. International journal of paediatric dentistry / the British Paedodontic Society [and] the International Association of Dentistry for Children 15, Parsons, T.E., Schmidt, E.J., Boughner, J.C., Jamniczky, H.A., Marcucio, R.S., and Hallgrimsson, B. (2011). Epigenetic integration of the developing brain and face. Developmental dynamics : an official publication of the American Association of Anatomists 240, Pauws, E., Hoshino, A., Bentley, L., Prajapati, S., Keller, C., Hammond, P., Martinez- Barbera, J.P., Moore, G.E., and Stanier, P. (2009a). Tbx22null mice have a submucous cleft palate due to reduced palatal bone formation and also display ankyloglossia and choanal atresia phenotypes. Hum Mol Genet 18, Pauws, E., Moore, G.E., and Stanier, P. (2009b). A functional haplotype variant in the TBX22 promoter is associated with cleft palate and ankyloglossia. Journal of medical genetics 46, Petersson, M., Brylka, H., Kraus, A., John, S., Rappl, G., Schettina, P., and Niemann, C. (2011). TCF/Lef1 activity controls establishment of diverse stem and progenitor cell compartments in mouse epidermis. EMBO J 30, 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. American journal of human genetics 94, Plikus, M.V., Zeichner-David, M., Mayer, J.A., Reyna, J., Bringas, P., Thewissen, J.G., Snead, M.L., Chai, Y., and Chuong, C.M. (2005). Morphoregulation of teeth: modulating the number, size, shape and differentiation by tuning Bmp activity. Evolution & development 7, Polito, P., Dal Cin, P., Kazmierczak, B., Rogalla, P., Bullerdiek, J., and Van den Berghe, H. (1999). Deletion of HMG17 in uterine leiomyomas with ring chromosome 1. Cancer genetics and cytogenetics 108, Poulter, J.A., Murillo, G., Brookes, S.J., Smith, C.E., Parry, D.A., Silva, S., Kirkham, J., Inglehearn, C.F., and Mighell, A.J. (2014). Deletion of ameloblastin exon 6 is associated with amelogenesis imperfecta. Hum Mol Genet 23, Qin, H., Blaschke, K., Wei, G., Ohi, Y., Blouin, L., Qi, Z., Yu, J., Yeh, R.F., Hebrok, M., and Ramalho-Santos, M. (2012). Transcriptional analysis of pluripotency reveals the Hippo pathway as a barrier to reprogramming. Hum Mol Genet 21, Que, J., Okubo, T., Goldenring, J.R., Nam, K.T., Kurotani, R., Morrisey, E.E., Taranova, O., Pevny, L.H., and Hogan, B.L. (2007). Multiple dose-dependent roles for Sox2 in the patterning and differentiation of anterior foregut endoderm. Development 134, Reid, B.S., Yang, H., Melvin, V.S., Taketo, M.M., and Williams, T. (2011). Ectodermal Wnt/beta-catenin signaling shapes the mouse face. Dev Biol 349,

183 Reya, T., and Clevers, H. (2005). Wnt signalling in stem cells and cancer. Nature 434, Richardson, R., Hammond, N., Coulombe, P., Saloranta, C., Nousiainen, H., Salonen, R., Berry, A., Hanley, N., Headon, D., Karikoski, R., et al. (2014a). Periderm prevents pathological epithelial adhesions during embryogenesis. J Clin Invest 124, Richardson, R.J., Hammond, N.L., Coulombe, P.A., Saloranta, C., Nousiainen, H.O., Salonen, R., Berry, A., Hanley, N., Headon, D., Karikoski, R., et al. (2014b). Periderm prevents pathological epithelial adhesions during embryogenesis. J Clin Invest 124, Rosenbluh, J., Nijhawan, D., Cox, A.G., Li, X., Neal, J.T., Schafer, E.J., Zack, T.I., Wang, X., Tsherniak, A., Schinzel, A.C., et al. (2012). beta-catenin-driven cancers require a YAP1 transcriptional complex for survival and tumorigenesis. Cell 151, Salih, D.A., Rashid, A.J., Colas, D., de la Torre-Ubieta, L., Zhu, R.P., Morgan, A.A., Santo, E.E., Ucar, D., Devarajan, K., Cole, C.J., et al. (2012). FoxO6 regulates memory consolidation and synaptic function. Genes & development 26, Sasaki, T., Ito, Y., Xu, X., Han, J., Bringas, J., P.,, Maeda, T., Slavkin, H.C., Grosschedl, R., and Chai, Y. (2005a). LEF1 is a critical epithelial survival factor during tooth morphogenesis. Dev Biol 278, Sasaki, T., Ito, Y., Xu, X., Han, J., Bringas, P.J., Maeda, T., Slavkin, H., Grosschedl, R., and Chai, Y. (2005b). LEF1 is a critical epithelial survival factor during tooth morphogenesis. Dev Biol 278, Seidel K, Ahn CP, Lyons D, Nee A, Ting K, Brownell I, Cao T, Carano RA, Curran T, Schober M, et al. (2010). Hedgehog signaling regulates the generation of ameloblast progenitors in the continuously growing mouse incisor. Development 137, Sharp, T., Wang, J., Li, X., Cao, H., Gao, S., Moreno, M., and Amendt, B.A. (2014a). A pituitary homeobox 2 (Pitx2):microRNA-200a-3p:beta-catenin pathway converts mesenchymal cells to amelogenin-expressing dental epithelial cells. The Journal of biological chemistry 289, Sharp, T., Wang, J., Li, X., Cao, H., Gao, S., Moreno, M., and Amendt, B.A. (2014b). A Pituitary Homeobox 2 (Pitx2):microRNA-200a-3p:Beta-catenin Pathway Converts Mesenchyme Cells to Amelogenin-Expressing Dental Epithelial Cells. J Biol Chem 289, Shen, L., Yang, S., Huang, W., Xu, W., Wang, Q., Song, Y., and Liu, Y. (2013). MicroRNA23a and microrna23b deregulation derepresses SF-1 and upregulates estrogen signaling in ovarian endometriosis. J Clin Endocrinol Metab 98, Smith, C.E., and Warshawsky, H. (1975). Cellular renewal in the enamel organ and the odontoblast layer of the rat incisor as followed by radioautography using 3H-thymidine. The Anatomical record 183, Soukup, V., Epperlein, H.H., Horacek, I., and Cerny, R. (2008). Dual epithelial origin of vertebrate oral teeth. Nature 455, Spradling, A., Drummond-Barbosa, D., and Kai, T. (2001). Stem cells find their niche. Nature 414,

184 St Amand, T.R., Zhang, Y., Semina, E.V., Zhao, X., Hu, Y., Nguyen, L., Murray, J.C., and Chen, Y. (2000). Antagonistic signals between BMP4 and FGF8 define the expression of Pitx1 and Pitx2 in mouse tooth-forming anlage. Dev Biol 217, Stanger, B.Z. (2008). Organ size determination and the limits of regulation. Cell cycle 7, Takahashi, K., and Yamanaka, S. (2006a). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, Takahashi, K., and Yamanaka, S. (2006b). Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors. Cell 126, 663. Taranova, O., Magness, S., Fagan, B., Wu, Y., Surzenko, N., Hutton, S., and Pevny, L. (2006a). SOX2 is a dose-dependent regulator of retinal neural progenitor competence. Genes Dev 20, Taranova, O.V., Magness, S.T., Fagan, B.M., Wu, Y., Surzenko, N., Hutton, S.R., and Pevny, L.H. (2006b). SOX2 is a dose-dependent regualtor of retinal neural progenitor competence. Genes Dev 20, Thesleff, I. (1976). Differentiation of odontogenic tissues in organ culture. Scand J Dent Res 84, Thesleff, I., and Nieminen, P. (1996). Tooth morphogenesis and cell differentiation. Current opinion in cell biology 8, Thesleff, I., and Sharpe, P. (1997). Signalling networks regulating dental development. Mechanisms of development 67, Thesleff, I., and Tummers, M. (2008). Tooth organogenesis and regeneration. In StemBook (Cambridge (MA)). Tummers, M., and Thesleff, I. (2003a). Root or crown: a developmental choice orchestrated by the differential regulation of the epithelial stem cell niche in the tooth of two rodent species. Development 130, Tummers, M., and Thesleff, I. (2003b). Root or crown: a developmental choice orchestrated by the differential regulation of the epithelial stem cell niche in the tooth of two rodent species. Development 130, Tummers, M., Yamashiro, T., and Thesleff, I. (2007). Modulation of epithelial cell fate of the root in vitro. Journal of dental research 86, Tuttle, A., Rankin, M., Teta, M., Sartori, D., Stein, G., Kim, G., Virgilio, C., Granger, A., Zhou, D., Long, S., et al. (2010). Immunofluorescent detection of two thymidine analogues (CldU and IdU) in primary tissue. J Vis Exp 46. Vadlamudi, U., Espinoza, H.M.,, Ganga, M., Martin, D.M.,, Liu, X., Engelhardt, J.F., and Amendt, B.A. (2005). PITX2, b-catenin, and LEF-1 Interact to Synergistically Regulate the LEF-1 promoter. J Cell Sci 118, van der Heide, L.P., Jacobs, F.M., Burbach, J.P., Hoekman, M.F., and Smidt, M.P. (2005). FoxO6 transcriptional activity is regulated by Thr26 and Ser184, independent of nucleo-cytoplasmic shuttling. The Biochemical journal 391, van Genderen, C., Okamura, R.M., Farinas, I., Quo, R.-G., Parslow, T.G., Bruhn, L., and Grosschedl, R. (1994a). Development of several organs that require inductive epithelialmesenchymal interactions is impared in LEF-1-deficient mice. Genes Dev 8,

185 van Genderen, C., Okamura, R.M., Farinas, I., Quo, R.G., Parslow, T.G., Bruhn, L., and Grosschedl, R. (1994b). Development of several organs that require inductive epithelialmesenchymal interactions is impaired in LEF-1-deficient mice. Genes & development 8, Varelas, X., Sakuma, R., Samavarchi-Tehrani, P., Peerani, R., Rao, B.M., Dembowy, J., Yaffe, M.B., Zandstra, P.W., and Wrana, J.L. (2008). TAZ controls Smad nucleocytoplasmic shuttling and regulates human embryonic stem-cell self-renewal. Nature cell biology 10, Venugopalan, S.R., Li, X., Amen, M.A., Florez, S., Gutierrez, D., Cao, H., Wang, J., and Amendt, B.A. (2011). Hierarchical interactions of homeodomain and forkhead transcription factors in regulating odontogenic gene expression. The Journal of biological chemistry 286, Viswanathan, V., Fields, J., and Boman, B. (2014). The mirna23b-regulated signaling network as a key to cancer development-implications for translational research and therapeutics. Journal of Molecular Medicine 92, Wang, J., Saadi, I., Wang, J., Engel, J.J., Kaburas, A., Russo A.F., and Amendt, B.A. (2013a). PIAS1 and PIASy Differentially Regulate PITX2 Transcriptional Activities.. J Biol Chem 288, Wang, J., Sun, Z., Zhang, Z., Saad, i.i., Wang, J., Li, X., Gao, S., Engle, J., Kuburas, A., Fu, X., et al. (2013b). Protein inhibitors of activated STAT (Pias1 and Piasy) differentially regulate pituitary homeobox 2 (PITX2) transcriptional activity. The Journal of biological chemistry 288, Wang, X.-P., Suomalainen, M., Felszeghy, S., Zelarayan, L.C., Alonso, M.T., Plikus, M.V., Maas, R.L., Chuong, C.-M., Schimmang, T., and Thesleff, I. (2007a). An Integrated Gene Regulatory Network Controls Stem Cell Proliferation in Teeth. Plos Biol 5, e159. Wang, X.-P., Suomalainen, M., Jorgez, C.J., Matzuk, M.M., Werner, S., and Thesleff, I. (2004). Follistatin Regulates Enamel Patterning in Mouse Incisors by Asymmetrically Inhibiting BMP Signaling and Ameloblast Differentiation. Dev Cell 7, Wang, X.P., Suomalainen, M., Felszeghy, S., Zelarayan, L.C., Alonso, M.T., Plikus, M.V., Maas, R.L., Chuong, C.M., Schimmang, T., and Thesleff, I. (2007b). An integrated gene regulatory network controls stem cell proliferation in teeth. PLoS biology 5, e159. Wu, S., Huang, J., Dong, J., and Pan, D. (2003). hippo encodes a Ste-20 family protein kinase that restricts cell proliferation and promotes apoptosis in conjunction with salvador and warts. Cell 114, Xie, W., Lynch, T., Liu, X., Tyler, S., Yu, S., Zhou, X., Luo, M., Kusner, D., Sun, X., Yi, Y., et al. (2014). Sox2 modulates Lef-1 expression during airway submucosal gland development. Am J Physiol Lung Cell Mol Physiol 306, L Young, N.M., Chong, H.J., Hu, D., Hallgrimsson, B., and Marcucio, R.S. (2010). Quantitative analyses link modulation of sonic hedgehog signaling to continuous variation in facial growth and shape. Development 137, Yuan, G., Yang, G., Zheng, Y., Zhu, X., Chen, Z., Zhang, Z., and Chen, Y. (2015). The non-canonical BMP and Wnt/β-catenin signaling pathways orchestrate early tooth development. Development 142, Zhang, L., Yuan, G., Liu, H., Lin, H., Wan, C., and Chen, Z. (2012). Expression pattern of Sox2 during mouse tooth development. Gene Expr Patterns 12,

186 Zhang, Z., Gutierrez, D., Li, X., Bidlack, F., Cao, H., Wang, J., Andrade, K., Margolis, H., and Amendt, B. (2013). The LIM homeodomain transcription factor LHX6: a transcriptional repressor that interacts with pituitary homeobox 2 (PITX2) to regulate odontogenesis. The Journal of biological chemistry 288, Zhao, B., Li, L., Lei, Q., and Guan, K.L. (2010). The Hippo-YAP pathway in organ size control and tumorigenesis: an updated version. Genes & development 24, Zhao, B., Ye, X., Yu, J., Li, L., Li, W., Li, S., Yu, J., Lin, J.D., Wang, C.Y., Chinnaiyan, A.M., et al. (2008). TEAD mediates YAP-dependent gene induction and growth control. Genes & development 22, Zhao, H., Feng, J., Seidel, K., Shi, S., Klein, O., Sharpe, P., and Chai, Y. (2014). Secretion of shh by a neurovascular bundle niche supports mesenchymal stem cell homeostasis in the adult mouse incisor. Cell stem cell 14, Zhao, Y., and Srivastava, D. (2007). A developmental view of microrna function. Trends Biochem Sci 32,

187 APPENDIX Figure A1 The Hippo pathway in mammals. Mammalian Ste20 family kinases Mst1 and Mst2 form an active complex with Salvador (Salv), to further phosphorylate large tumor suppressor homolog (Lats1 and Lats2) kinase. Lats1 and Lats2 bind to a scaffold protein, Mob, to further phosphorylate two transcriptional co-activator Yap and Taz. Yap and Taz are most downstream Hippo signaling components and modulate the expression of genes involved in cell growth, proliferation and apoptosis. 164

188 Figure A2 Generation of FoxO6 Knock-out mouse. Schematic of the recombination strategy used to generate the FoxO6 knockout mice. Using homologous recombination in embryonic stem (ES) cells, 19 kb of FoxO6 genomic DNA containing the entire FoxO6 two exons was replace by a LacZ cassette and with a 'floxed' neomycin cassette (in solid triangle). 165

189 Figure A3 Select regions of the FoxO6 null mouse brain were increased in volumetric measurements. A) LacZ expression in E16.5 FoxO6+/- mouse brain. FoxO6 is expressed in the frontal cortex, midbrain and medulla. B) Brain MRI scans of 2 month WT and FoxO6-/- mice were conducted to compare the size of the different compartments of WT and FoxO6-/- mice brains. The percent increase in brain regions is shown on top with decreased regions on the bottom. (The MRI analysis was done by Drs. Peggy Nopoulos and Daniel R. Thedens in Department of Psychiatry, The University of Iowa) 166