1 MAMMALIAN ANOGENITAL DEVELOPMENT: OF MICE FOR MEN By ASHLEY WINN SEIFERT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Ashley Winn Seifert
3 To my mom and dad
4 ACKNOWLEDGMENTS First and foremost, I would like to thank my advisor Dr. Martin J. Cohn for taking me on as his Ph.D student and for stoking my scientific flame. His encouragement, mentoring, and friendship have transformed me forever. I would also like to thank Dr. Malcolm Maden for our many philosophical discussions and his fearless pursuit of the scientific truth. In this last point I hope to emulate. Thanks to Dr. Scott Stadler for his patient advice and forward-looking approach to helping me better understand development. Many thanks to Dr. Brian Harfe and Dr. Lou Guillette Jr. for agreeing to serve on my committee and offering their expertise to my projects and to Dr. Brandi Ormerod who despite my relentless quest to understand stereology guided me with patience. I would like to thank Dr. Karel Liem for starting my graduate career in his office at the MCZ and Dr. Lauren Chapman for taking a chance on me with little rhyme or reason when I was underqualified for our department. I would like to thank Renata Frietas, and Jun Zhang for imparting their laboratory and intellectual wisdom on me. I would also like to thank all the undergraduate students whom I have mentored over the years and who put up with my absentminded ways. A huge thanks goes out to my peers and the future Nexus Group, Adrian Stier, Matt Smith, Bret Pasch, Francois Michonneau, and C. Rossi for the opportunity to keep these dreams alive. A very special thanks to my family for their constant love and support and for believing in me along the road less traveled. And finally, Megan Gittinger for supporting me and understanding my passion.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4!LIST OF TABLES ................................................................................................................................ 8!LIST OF FIGURES .............................................................................................................................. 9!ABSTRACT ........................................................................................................................................ 11 CHAPTER 1 GENERAL INTRODUCTION .................................................................................................. 13!Anogenital Pathology .................................................................................................................. 13!Descriptive Embryology ............................................................................................................. 15!Molecular Landscape .................................................................................................................. 18!General Overview ....................................................................................................................... 25!2 CELL LINEAGE ANALYSIS DEMONSTRATES AN ENDODERMAL ORIGIN OF THE DISTAL URETHRA AND PERINEUM ......................................................................... 27!Introduction ................................................................................................................................ 27!Materials and Methods ................................................................................................................ 31!Transgenic Mice and Lineage Analysis ............................................................................. 31!Flutamide Administration ................................................................................................... 31!X-gal Staining ...................................................................................................................... 31!Histology .............................................................................................................................. 32!Results .......................................................................................................................................... 32!Cloacal Endoderm Gives Rise to the Entire Urethral Epithelium .................................... 32!Cloacal Endoderm Gives Rise to the Perineum ................................................................ 33!Sexual Differentiation of the Urethra ................................................................................. 34!There is no Ectodermal Contribution to the Glandar Urethra ........................................... 36!Disruption of Androgen Signaling Feminizes Male Genitalia without Affecting Urethral Epithelial Integrity ............................................................................................ 37!Discussion .................................................................................................................................... 38!Endodermal Origin of the Distal Urethra ........................................................................... 38!Cloacal Morphogenesis and Development of the Genital Tubercle ................................ 38!Origin of the Perineum ........................................................................................................ 39!The Role of Stromal Mesoderm in Urethral Tube Development ..................................... 40!Urethral Tubulogenesis and Cloacal Septation are linked by a Common Developmental Mechanism ............................................................................................. 41!A Model for Masculinization of the Urethral Plate ........................................................... 42!
6 3 FUNCTIONAL AND PHYLOGENETIC ANALYSIS OF FGF8 SHOWS THAT FGF8 IS NOT INVOLVED IN EXTERNAL GENITAL DEVELOPMENT ........................ 52!Introduction ................................................................................................................................ 52!Materials and Methods ................................................................................................................ 53!Mice ...................................................................................................................................... 53!In situ Hybridization and Analysis of Apoptosis ............................................................... 54!Results and Discussion ............................................................................................................... 54!Fgf8 Expression is Associated with Genital Tubercle Initiation in Wnt5a Null Mice .... 54!Fgf8 is induced at the Endodermal-Ectodermal Boundary of the Cloacal Membra ne .... 56!Fgf8 is not necessary for Outgrowth of the Genital Tubercle .......................................... 58!Does Fgf8 Regulate Bmp4, Wnt5a, Hoxd13 and Shh Expression in the Genital Tubercle? .......................................................................................................................... 60!Conditional Removal of Fgf8 Results in Normal Development of the Penis .................. 61!Divergent Roles of Fgf Signaling in Limb and Genital Development ............................. 62!Is Fgf8 Signaling Active During Normal Genital Development? .................................... 62!Phylogenetic Distribution of Fgf8 in Amniote Genital Tubercles Shows That it is not Required for External Genital Development ........................................................... 65!Conclusions ................................................................................................................................ 66!4 FIBROBLAST GROWTH FACTOR SIGNALING DURING INITIATION AND OUTGROWTH OF THE EXTERNAL GENITALIA ............................................................. 80!Introduction ................................................................................................................................ 80!Materials and Methods ................................................................................................................ 82!Mice ...................................................................................................................................... 82!In situ hybridization and histology ..................................................................................... 82!Results and Discussion ............................................................................................................... 83!Fgf Expression During Initiation and Outgrowth of the External Genitalia .................... 83!Expression at Stage E10.5 ................................................................................................... 83!Expression at Stage E11.5 ................................................................................................... 84!Expression at Stage E12.5 ................................................................................................... 85!Loss of Fgf9 Results in Female Hypospadias .................................................................... 87!5 SONIC HEDGEHOG CONTROLS GROWTH OF EXTERNAL GENITALIA BY REGULATING CELL CYCLE KINETICS ............................................................................. 96!Introduction ................................................................................................................................ 96!Materials and Methods ................................................................................................................ 97!Animals and Injections ........................................................................................................ 97!Immunohistochemistry and X-Gal Staining ...................................................................... 97!In situ Hybridization ............................................................................................................ 97!Stereological Estimates of Total Cell Numbers ................................................................ 97!Determination of Cell Cycle Kinetics ................................................................................ 98!Statistics ................................................................................................................................ 99!Results and Discussion ............................................................................................................... 99!
7 6 A MULTIPHASIC ROLE FOR SHH DURING CLOACAL SEPTATION AND EXTERNAL GENITALIA DEVELOPMENT ....................................................................... 114!Introduction ............................................................................................................................... 114!Materials and Methods .............................................................................................................. 118!Animals .............................................................................................................................. 118!Histology ............................................................................................................................ 118!In situ Hybridization .......................................................................................................... 118!Proliferation Index ............................................................................................................. 119!Ectodermal Removal ......................................................................................................... 119!Statistics .............................................................................................................................. 119!Results ........................................................................................................................................ 120!Continuous Shh Signaling is required for Patterning and Outgrowth of the External Genitalia .......................................................................................................................... 120!Loss of Shh Signaling Prior to E13.5 Results in Persistent Cloaca ................................ 122!Posterior Location of the Anorectal Opening is Dependent on Sustained Shh Signaling ......................................................................................................................... 122!Malformations of the Caudal Axis Associate with Genital and Cloacal Defects .......... 123!Cell Death Occurs at the Lateral Margins of the GT Following Loss of Shh ............... 124!Shh Regulates Outgrowth by Maintaining Cell Cycle Progression ............................... 125!Shh is Required for Normal Tubulogenesis and Masculinization of the Urethra .......... 127!Discussion .................................................................................................................................. 129!Initial Septation Requires Hh Signaling ........................................................................... 130!Shh Regulates Proliferation during the Perineal Period .................................................. 131!Shh Regulates Epithelial Morphogenesis through an Unknown Mechanism ................ 133!Loss of Shh Pathway Activation in the Genital Ectoderm Results in Hypospadias ...... 135!The Perineal and Preputial Periods as Targets for ARM ................................................ 136!7 GENERAL CONCLUSIONS .................................................................................................. 152!Three-dimensional Visualization of Anogenital Morphogenesis ........................................... 152!Control of Initiation and Outgrowth of the External Genitalia .............................................. 152!Shh Controls Various Aspects of Anogenital Development .................................................. 154!LIST OF REFERENCES ................................................................................................................. 158!BIOGRAPHICAL SKETCH ........................................................................................................... 171!
8 LIST OF TABLES Table page 4-1 Fgf and FgfR expression in the genital tubercle ................................................................... 89!4-2 Genital phenotypes of Fgf and FgfR knockouts in mouse .................................................. 90!
9 LIST OF FIGURES Figure page 2-1. Shhgfpcre-expressing endodermal cells give rise to the epithelium of the urethral plate, perineum, bladder and anorectum ......................................................................................... 45!2-2. Sexual differentiation of the urethral plate ............................................................................. 4 7!2-3. Transformation of the urethral plate to a urethral tube ............................................................ 48!2-4. Genital tubercle ectoderm does not contribute to the distal urethra ....................................... 49!2-5. Feminization of male genitalia by disruption of androgen receptor activity ......................... 50!2-6. Model for masculinization of the urethral plate in the mouse penis ....................................... 51!3-1. Variable penetrance of external genital development in Wnt5a-/mice correlates with Fgf8 expression .................................................................................................................... 6!3-2. Fgf8 expression marks a contiguous AER in Wnt5a-/embryos with medially displaced hindlimb buds ......................................................................................................................... 70!3-3. Fgf8 is normally expressed exclusively in Shh-expressing cloacal endoderm ...................... 71!3-4. Fgf8 is activated following endoderm/ectoderm contact at the cloacal membrane ............... 72!3-5. Normal development of external genitalia in the absence of Fgf8 ....................................... 74!3-6. A mosaic readout of Fgf-pathway activation during initiation and outgrowth of GT ........... 75!3-7. Lateral expression boundaries of Dusp6, Sprouty4, Erm and Pea3 ....................................... 76!3-8. Pea3 and Sprouty4 expression is unaffected following removal of Fgf8 ............................... 77!3-9. Fgf8 protein is undetectable in the GT. .................................................................................... 78!3-10. Phylogenetic characterization of Fgf8 expression in amniote genitalia ............................... 79!4-1. Only Fgf5 and Fgf8 are expressed at E10.5 in the genital field .............................................. 91!4-2. Several Fgfs and Fgf receptors are expressed at E11.5 ......................................................... 93!4-3. Fgf expression at E12.5. In situ hybridization shows expression of Fgf ligands ( 2, 6, 8, 9, 10, 18 ) at E12.5 in the genital tubercle ............................................................................. 94!4-4. Loss of Fgf9 leads to feminization of the external genitalia and hypospadias ...................... 95!5-1. Shh regulates temporal development of the external genitalia and cloaca ........................... 107!
10 5-2. Early and transient Shh expression is sufficient for pattern specification in the genital tubercle .................................................................................................................................. 108!5-3. Stereological estimates reveal decreased cell number ........................................................... 110!5-4. Loss of Shh signaling does not result in widespread cell death in the genital tubercle ....... 111!5-5. Shh controls growth by regulating cell cycle kinetics ........................................................... 112!5-6. A model for Shh-mediated integration of growth and patterning ......................................... 113!6-1. Shh is required until sexual differentiation for anogenital development .............................. 139!6-2. Shh is required for cloacal septation until E13.5 .................................................................... 140!6-3. Loss of Shh signaling results in anorectal malformations. .................................................. 142!6-4. Ectopic cell death following loss of Shh signaling is not widespread in the genital tubercle .................................................................................................................................. 143!6-5. Urorectal septum mesenchyme proliferates during cloacal septation .................................. 144!6-6. Loss of Shh signaling from the cloacal endoderm leads to a decrease in proliferative index at the posterior end of the URSM. .......................................................................... 146!6-7. Endodermal progenitor layer remains intact following loss of Shh ..................................... 147!6-8. Persistent Ptc1 expression in URSM following loss of Shh suggests activation by Ihh ..... 148!6-9. Loss of Shh signaling from the urethral epithelium prior to sexual differentiation leads to hypospadias ...................................................................................................................... 149!6-10. Genital ectoderm helps maintain ventral connection of urethral epithelium ..................... 150!6-11. Inability of genital ectoderm to respond to Shh results in loss of ectodermal integrity and hypospadias ................................................................................................................... 151!
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MAMMALIAN ANOGENITAL DEVELOPMENT: OF MICE FOR MEN BY Ashley Winn Seifert May 2009 Chair: Martin J. Cohn Major: Zoology Early during mammalian gestation the anogenital system begins to undergo development. Coordinated morphogenesis at the posterior of the embryo will transform the primitive cloaca into the separate genitourinary and anorectal portals of the body. Despite the obvious importance of these separate systems, the developmental mechanisms underlying their formation remain poorly understood. Although originally described over a century ago, the cellular origins of the distal urethra, and perineum have remained unresolved. Through construction of a lineage-based fate-map of the anogenital endoderm, I show that these two structures are derived from endoderm. Using this lineage map I three-dimensionsionally reconstruct the processes of urethragenesis and propose a model for masculinization of the urethra that mimics division of the cloaca. This model forms the basis for revisiting the underlying molecular mechanisms driving morphogenesis of the cloacal endoderm. Research over the past 20 years has used the developing limb as a model to investigate and interpret molecular control of the developing external genitalia. Central to this thesis has been the importance of Fibroblast Growth Factor 8 ( Fgf8 ) to control outgrowth and patterning of the
12 genitalia. Here I show that Fgf8 is not required for genital development and propose instead that it marks an epithelial-epithelial interaction at the cloacal membrane during induction of the genitalia. In addition to investigating Fgf signaling, I examine the importance of Sonic hedgehog (Shh) as a master regulator of anogenital development. Through the temporally controlled removal of Shh over the entire course of anogenital development, I demonstrate that Shh is specifically required for cloacal septation, and growth and patterning of the external genitalia during two critical developmental windows. Furthermore, I show that Shh controls the rate of cell proliferation by regulating cell cycle progression and this in turn regulates progenitor pool expansion, which ultimately governs organ size. These findings provide a framework in which to better understand the developmental mechanisms underlying congenital anorectal malformations (ARM) in humans and present an opportunity to dissect the contributions of both heritable genetic mutations and environmental teratogens during anogenital development.
13 CHAPTER 1 GENERAL INTRODUCTION Normal and correct development of the vertebrate anogenital system (anorectal + genitourinary) is necessary for the proper excretion of waste products and for the transmission of gametes to the next generatio n. Despite the physiological necessity of this system, heritable mutations affecting part, or all, of the anogenital system are maintained within animal populations. In addition to genetic mutations, environmental pollution from industrialization creates teratogens that frequently affect developing vertebrates and interact with the underlying molecular pathways that control the developmental program. Thus, there is a pressing need from the research and medical communities to study and understand the deve lopmental biology of both normal and abnormal anogenital development in animal models and in humans. Using the mouse as a model for mammalian development, this dissertation focuses on multiple levels of this system in an attempt to understand the develop mental mechanisms underlying anogenital development at the morphological and molecular level. Anogenital Pathology Congenital anomalies of the anogenital system encompass a broad array of malformations affecting the genitourinary and anorectal systems either in concert or in isolation. The most extreme syndromes, such as persistent cloaca (1 in 50,000 live births), affect the entire anogenital system causing the genitourinary and anorectal tracts to empty into a single outlet and present variable agene sis of the external genitalia (Mo et al., 2001) Complexes such as VACTERL ( V ertebral, A nal, C ardiovascular, T racheosophageal fistula, E sophageal atresia, R enal, L imb), represent a non random association of defects affecting various organ systems including severe defects of the anogenital system (Arsic et al., 2002; Kim et al., 2001) Other malformations are less systemic, but more common, such as imperforate anus (1 in 5000 ) and
14 hypospadias (1 in 250) and affect the anorectal and genitourinary systems separately, although they too can occur in tandem (Mo et al., 2001; Paulozzi et al., 1997) Despite the efforts of scientists and the medical community, the etiology of these malformations remains poorly understood (Stadler, 2003) Undoubtedly, embryological examinations of these pathologies in humans have contributed to our understanding of the defects, however, they do not explain the underlying cellular mechanisms that cause them. A synthetic analysis with the tools of developmental biology has made it possible to conduct genetic screens in human cases and relate this information to studies in animal models that recapitulate abnormal development. It is precisely this approach that has led to our identification of mutations associated with other syndromes affecting the anogenital system such as Curriano ( HLXB9 ), Townes Brocks ( SAL1 ), Pallister Hall ( GLI3 ), and Hand Foot Genital ( HOXA13 ) (Kang et al., 1997; Kohlhase et al., 1998; Mortlock and Innis, 1997; Ross et al., 1998) A knowledge of the defects associated with these syndromes, in light of the developmental pathways affected, provides powerful information to examine anomalie s with unknown etiologies that are likely more complex and possibly multi factorial. While the reported incidences of some congenital malformations remains stable in certain populations, diagnoses of other defects such as hypospadias have been increasing in prevalence over the last thirty years without reasonable explanation (Pau lozzi et al., 1997) This has raised concern among both public health officials and research scientists worried of possible links to environmental contaminants (Kim et al., 2004; Paulozzi et al., 1997) Indeed, e ndocrine disrupting compounds and environmental teratogens along with chemotherapy agents (adriamycin) and retinoid analogues (etretinate) can cause hypospadias, feminization of the
15 genitalia and induce serious congenital malformations during pregnancy (Dawrant et al., 2008; Ioannides et al., 2003; Kim et al., 2004; Paulozzi et al., 1997; Qi et al., 2002) In order to better understand the role of these exogenous agents in causing mal development we must understand the molecular mechanisms that direct normal development of the anogenital system. This begins with a systems level understanding of normal development in both animal models and in humans. Descriptive Embryology Somehow or other we must find how to bring into the story the physical forces which are necessary to push the material about into the appropriate places and mould it into the correct shapes C.H. Waddington Development of the anogenital system remains one of the most poorly understood organ complexes in embryology. Midway through mouse development, approximately embryonic day 10.5 (E10.5), and corresponding to gestational week 4 in humans, the posterior ending of the gut has formed the primitive cloaca. The cloaca will undergo stereotypical morphogenesis such that the ventral part of the cloaca will give rise to the urogenital sinus, bladder, and urethra, while the dorsal portion will give rise to the hindgut (rectum + anus). This division will give rise to the genitourinary and anorectal o utlets, and in association with the genitourinary outlet, the sexually indifferent external genitalia will ultimately differentiate into the male and female forms (see Figure 2 1). Although the basic embryological descriptions of the late 19 th century w ere in general agreement, the precise mechanisms describing how this development proceeded were deeply debated (Nievelstein et al., 1998; Retterer, 1890; Tourneux, 1888; van der Putte, 2005) The process that divid ed the cloaca figured centrally into this debate and led to the identification of a population of mesoderm termed the urorectal septum. Retterer (1890) favored fusion of two
16 lateral ridges of the cloacal wall, while Tourneux (1888) favored descent of a se ptum in the caudalward direction (Retterer, 1890; Tourneux, 1888) Examinations over the next century favored one or both of these theories, until investigations into normal and abnormal development of the pig (van der Putte, 1986; Van d er Putte and Neeteson, 1984) and mouse (Kluth et al., 1995) failed to find evidence supporting either or these theories. Subsequent investigation s have revisited the debate by resurrecting early ideas for cloacal septation by fusion of two lateral folds although the issue to date remains unresolved (Dravis et al., 2004; Hynes and Fraher, 2004c) As the ur orectal septum mesoderm behind the dividing cloaca nears the surface, ectoderm encapsulating the genital tubercle will rupture at the base of the tail to expose the urogenital and anorectal openings (Kluth et al., 19 95; Sasaki et al., 2004) Ectoderm at this position is part of the cloacal membrane and was formed at the boundary between the posterior gut and surface ectoderm at E10.5. While there is disagreement about what causes the cloacal membrane to rupture (e. g. either it's contact with the urorectal septum or independent of the contact), there is general agreement that the rupture occurs prior to any fusion event (Kluth et al., 1995; Sasaki et al., 2004; van der Putte, 1 986; van der Putte and Neeteson, 1983) Once the cloacal membrane has ruptured, the perineum forms between the urogenital opening at the base of the genital tubercle and anorectal opening of the gut. In males, part of this area will ultimately form the scrotum and house the testis as they descend through the body wall. Despite a number of careful embryological descriptions, studies have failed to produce cellular or molecular explanations for the mechanisms driving morphogenesis of cloacal septation. Concurrent with the morphogenesis and division of the cloaca, the external genitalia begins to form at E10.5 as mesoderm lateral to the cloacal membrane emerges from the body wall (Perriton et al., 2002) During initiation and outgrowth, the external genitalia develop as an
17 indifferent anlage termed the genital tubercle. Proximodistal growth of the genital tubercle incorpo rates the distal, and ventral portion of the cloacal endoderm and this forms a simple epithelium on the ventral side of the tubercle termed the urethral plate (Haraguchi et al., 2001; Perriton et al., 2002) The ur ethral plate and urogenital sinus form a continuous structure derived from the cloaca that extends to the distal tip of the glans. Following perineum formation (~E13.5), the urethral plate will undergo epithelial differentiation to form the pseudostratifi ed epithelium of the urethra (Kurzrock et al., 1999; Petiot et al., 2005) The prepuce (foreskin) emerges from the dorsal and lateral margins of the genital tubercle and will develop around the glans until androgen mediated development will transform the indifferent organ into the male penis or the female vagina (see Figure 2 2). Despite an elegant embryological description of this process by Perriton et al. (2002) using a combination of SEM and traditional histo logy, controversy surrounding the cellular origins of the distal urethra and masculinization of the urethra persist (Baskin et al., 2001; Glenister, 1954; Kurzrock et al., 1999; Larson, 2001; Moore, 2007; Perriton et al., 2002; Sadler, 2006; Yamada et al., 2003) Until approximately E16.0 in mouse (8 weeks gestation human), development of the external genitalia has proceeded such that the male and female genitalia are morphologically identical (Asby et al., 2008; Perriton et al., 2002) Following this period, virilization of the genitalia occurs under androgen mediated processes that are poorly understood at the mechanistic level. There is a rich literature on the effects of androgen antagonists and androgen insensitivity in both animals and man, and increasingly toxicologists are findings environmental agents that, following in utero treatment, can cause either virilization or feminization of the external genitalia (Connolly and Resko, 1989; Gehring and Tomkins, 1974; Goldstein and Wilson, 1975; Kim et al., 2004; Kurzrock et al., 1999; Lyon and Hawkes, 1970;
18 Mahendroo et al., 2001; Stadler, 2003; Welsh et al., 2007; Wolf et al., 2002) However, these studies have failed to adequately provide mechanistic explanations for the actions of the compounds at the cellular and genetic level. Classical embryology and current human medical texts purport that the distal urethra forms from an inva gination of ectoderm that joins the underlying endoderm of the urethral plate (Glenister, 1954; Glenister, 1956; Hynes and Fraher, 2004a; Larson, 2001; Moore, 2007; Sadler, 2006) However, recent evidence from SEM and histological data (Perriton et al., 2002) and from immunological staining and sections has suggested that the origin of the distal urethra may lie in the endodermal urethral plate itself (Kurzrock et al., 1999) Additionally, there is currently no cellular mechanism that accounts for urethragenesis during masculinization when the urethral plate becomes i ncorporated within the glans penis. Prior to sexual differentiation, the urethral plate is canalized in its most proximal portion and is identical in males and females. As virilization of the genitalia occurs, the urethra will become incorporated within the glans penis as it canalizes to empty at the urethral meatus (distal opening). Cellular mechanisms explaining the transition from a plate to a tube have been proposed involving apoptosis, an epithelial to mesenchymal transition, fusion of the preputial folds or invasion of deep stromal tissue of the former urorectal septum (Baskin et al., 2001; Dravis et al., 2004; van der Putte, 2007) In the absence of a lineage based fate analysis, or live imaging using fluor escent tracers, these issues have yet to be resolved. Molecular Landscape It is our experience that even detailed knowledge of the normal morphology of embryos sometimes is not sufficient to explain embryological principles. Dietrich Kluth
19 In sp ite of the advances in developmental biology over the past 30 years, applications of molecular techniques towards a more complete and mechanistic understanding of anogenital development have been slow to materialize (Stadler, 2003) With between 2,400 2,900 uniquely targeted gene deletions i n mice now created, it is surprising that so few have been characterized for anogenital anomalies upon first, second, or even third description (Austin et al., 2004; Stadler, 2003) Anogenital development appears t o be well conserved in mammals and the mouse represents a powerful model to understand the molecular genetics underlying its morphogenesis. The rapid rise of genetic screening to identify mutations associated with congenital malformations, coupled with t he rising public specter of environmental contaminants and their effects on sexual development, has renewed interest into the underlying developmental mechanisms controlling morphogenesis of the anogenital system. Indeed, understanding both the developmen tal morphology of a defect, and the specific gene(s) that account for its occurrence, represent entry points into a clearer dissection of how and when teratogens perturb development. One of the first descriptions of severe anogenital defects in a target ed transgenic mouse was the compound deletion of Hoxa13 and Hoxd13 (Warot et al., 1997) Homozygous mutants for copies of both 5 Hox genes displayed a persistent cloaca and complete agenesis of the external genita lia. While a deletion of Hoxd13 alone resulted in defects of the anorectal tract and external genitalia, and a mutation in the human HOXA13 was found to result in Hand Foot Genital syndrome, the compound mutants revealed a level of functional redundancy u nmasking the absolute requirement for 5 Hox genes during anogenital development (Dolle et al., 1991; Kondo et al., 1996; Mortlock and Innis, 1997; Warot et al., 1997) More recent investigations
20 into various Hoxa1 3 loss of function alleles have revealed their requirement during development of the external genitalia for urethral tube closure, os penis formation, and for normal vasculagenesis (Morgan et al., 2003; Post and Inni s, 1999; Shaut et al., 2007) In addition to these anogenital defects, associated defects during limb development and the discovery of a global control region (GCR) driving expression of the 5 Hox genes in both limbs and external genitalia uncovered a f unctional link between these appendages during development (Dolle et al., 1991; Spitz et al., 2003) These findings suggested some measure of evolutionary linkage between the development of limbs and external genit alia. These earlier studies had initiated interest into development of the external genitalia and put forth the limb as a model to investigate genital development. Early investigation into the role of Fibroblast growth factors ( Fgfs ) and their cognate r eceptors found that Fgf8 was expressed at the onset of genital tubercle initiation in the cloacal membrane and that its restricted expression resembled Fgf8 expression localized to the apical ectodermal ridge (AER) during limb bud outgrowth (Haraguchi et al., 2000) Pioneering studies in the limb had shown that removal of the AER resulted in stage dependent limb truncations and thus controlled proximal distal outgrowth (Saunde rs, 1948; Summerbell, 1974) More recent genetic studies have confirmed the role of Fgf8 and other Fgfs in the AER to direct limb outgrowth (Lewandoski et al., 2000; Mariani et al., 2008; Sun et al., 2002) In a series of organ culture experiments Haraguchi et al. (2000) showed that removal of the distal genital tubercle epithelium truncated outgrowth and that growth could be rescued with an Fgf8 bead. Together with another experiment reducing outgrowth after tre atment with an Fgf8 antagonizing antibody, this led to the proposition of a specialized signaling region in the distal urethral plate they termed the distal urethral epithelium (DUE) (Haraguchi et al., 2000; Ogino et al., 2001) Other groups subsequently showed the localization
21 of several other molecules to the DUE, including Bmp7, Bmp2, Tgf and Shh (Morgan et al., 2003; Perriton et al., 2002; Scott et al., 2005) This reg ion remains a prospective signaling center purported to direct outgrowth and patterning of the genital tubercle and awaits further characterization using genetic techniques to remove these genes individually as well as in concert. Following the identi fication of this proposed outgrowth center the anogenital phenotype of the Sonic hedgehog mutant was described along with various single and compound mutations of the Hh pathway transducers Gli2 and Gli3 (Haraguchi e t al., 2001; Mo et al., 2001; Perriton et al., 2002) While various allelic heterozygous and null combinations of Gli2 and Gli3 produced an array of phenotypes, mice homozygous null for Gli2 / ;Gli3 / recapitulated the anogenital phenotype of Shh / mic e (Mo et al., 2001) These findings established the absolute requirement for Hh signaling during anogenital development as both Shh / and Gli2 / ;Gli3 / mice exhibited complete agenesis of the external genitalia and persistent cloaca. Perriton et al. (2002) further characterized the cloacal epithelium as a polarizing region within the genital tubercle, akin to the zone of polarizing activity in the limb where Shh is expressed and functions to coordinate anteropos terior patterning. Furthermore, while organ culture experiments demonstrated that exogenous Shh soaked beads could upregulate Ptc1, Bmp4, Hoxd13, and Fgf10 expression analysis in Shh / embryos suggested that Hoxd13, Bmp4 and Fgf8, did not require Shh fo r their expression (Haraguchi et al., 2001; Perriton et al., 2002) The analysis from Shh / embryos also suggested that Bmp2, Wnt5a, and Fgf10 might require Shh for their induction as well as their maintenance (Perriton et al., 2002) Interpretation of these expression studies was difficult in light of the agenic phenotype. The co nfounding factor for the analyses of both Shh / and Hoxa13 / ;d13 / was the severe nature of the developmental defects which precluded further
22 investigation because no anogenital development occurred. Taken together, these findings underscored the impor tance of conditional transgenic technology for both tissue targeted and temporal control over gene deletion to examine the role these genes play during normal development of the anogenital system. Given the publicly available transgenic null alleles, onl y Wnt5a / mice present an anogenital phenotype of similar severity to Shh / Hoxa13 / ;d13 / and Gli2 / ;Gli3 / mice (Perriton et al., 2002; Suzuki et al., 2003; Yamaguchi et al., 1999) Wnt5a / mice display complete agenesis of the external genitalia and a severe shortening of the gastrointestinal tract (Cervantes et al., 2009; Perriton et al., 2002; Suzuki et al., 2003; Yamaguchi et al., 1999) While the original des cription of the genitalia attributed agenesis partly to an axial elongation defect, a subsequent study found a decrease in mesenchymal proliferation as the underlying mechanism responsible for this phenotype (Suzuki et al., 2003; Yamaguchi et al., 1999) Interestingly, a recent study by Cervantes et al. (2008) showed that truncation of the gastrointestinal tract was due not to a loss in proliferative ability, but to an inability of newly dividing progenitor cells to intercalate back into the epithelium leading to stacking and an overall decrease in length. Only the study by Cervantes et al. (2008) investigated in detail the underlying mechanism responsible for abnormal gastrointestinal development. While the aforem entioned genes may act as master regulators coordinating development of the entire anogenital system, certain genes appear to regulate the compartmentalized development of the hindgut, urogenital sinus, bladder and urethra. Analysis of Fgf10 / FgfR2iiib / ephrin B2 lz/lz Bmp7 / p63 / and Dlx5 / ;6 / mice suggests that although the cloacal epithelium appears homogeneous at E10.5, its dorsal compartment which gives rise to the hindgut and its ventral portion which gives rise to the urogenital sinus, bladder and urethra, may
23 exhibit different developmental properties. Among these knockout lines, only loss of p63 phenocopies the genital defects described above, exhibiting either complete agenesis or bifid genital rudiments (Cheng et al., 2006; Ince et al., 2002) Loss of p63 in the ventral cloacal epithelium leads to massive apoptosis of the epithelium adjacent to the cloacal membrane and an ~80% decrease in proliferation of the genital mesenchyme (Cheng et al., 2006) As p63 indicates an intact progenitor cell layer it is likely that its removal leads to a loss of epithelial integrity and possibly a subsequent loss of Shh signaling from the cloacal epithelium. Interesti ngly, these mice undergo seemingly normal cloacal septation, although this point warrants closer examination and further study. The external genitalia of the other null alleles and allelic combinations mentioned above appear to develop normally until pe rineum formation. At birth, both males and females appear similar with ventral hypoplasia of the external genitalia and a proximal hypospadias characterized by an open urethra. That all these genes are expressed in the urethral epithelium, or in the case of Fgf10 signal through FgfR2iiib in the epithelium, suggests that late stage defects of the external genitalia may all involve perturbations to urethral integrity. Characterization of FgfR2iib / mice during genital tubercle development determined that loss of the receptor in the epithelium leads to degradation of the progenitor cell layer in the urethral plate and a failure to differentiate into a pseudostratified epithelium marked by keratin 14 expression (Petiot et al., 2005) Similar findings were reported in the genitalia for both Bmp7 / and Dlx5 / ;6 / mice (Suzuki et al., 2008) Taken together, these findings suggest that transition from an immature epithelial sta te marked by keratin 8 to a mature, stratified epithelium marked by keratin 14 is a requirement for proper tubulogenesis (Petiot et al., 2005; Suzuki et al., 2008)
24 In addition to the observed genital phenotypes, ephrin B2 lz/lz and Bmp7 / mice appear to phenocopy the anorectal fistulas exhibited by Gli2 / mice where the hindgut communicates with the urogenital sinus and a separate anorectal opening does not develop. While no mechanistic explanation was present ed for the description of the Bmp7 / anorectal fistula, the authors investigating the phenotype in ephrin B2 lz/lz mice suggested that reciprocal forward and reverse signaling between ephrin B2 and its receptors EphB2 and EphB3 are disrupted in epithelial cells (Dravis et al., 2004) Expression analysis of these genes localized them to the cloacal epithelium at the sight of the hindgut/urogenital sinus junction and in the urethra and adjacent mesenchyme at the sight of epithelial rearrangement as the urethra was masculinized. EphrinBs are well known to functionally regulate cell adhesion and maintain cell boundaries and loss of function at these junctions may contribute to epithelial breakdown and rearrangement (Lee et al., 2006) Fgf10 / and FgfR2iiib / mice present a unique hindgut phenotype unlike the null alleles described above (Fairbanks et al., 2004; Fairbanks et al., 2006; Sala et al., 2006) Fgf10 is highly expressed in the rectal mesenchyme with FgfR2iiib expression restricted to the hindgut epithelium. Loss of either gene results in a complete loss of the rectum and anus with no communication found between the colon and urogenital sinus (Sala et al., 2006) Following loss of FgfR2iiib Fairbanks et al. (2006) found a decrease in mesenchymal proliferation and a concomitant increase in epithelial proliferation. This same group observed similar findings in a hypomorphic allele of Fgf10 (Fairbanks et al., 2004) Integration of these findings strongly suggests independent regulation of the ventral and dorsal compartments of the cloaca as they develop into the urogenita l sinus, urethra and hindgut respectively. The advent of conditional knockout technology in the mouse has brought new opportunities to target and remove genes in a tissue specific manner, most often to circumvent
25 perinatal lethality and examine their fun ctions during organogenesis. Recent work to circumvent the severe phenotype of Wnt5a / mice utilized a conditional floxed allele of catenin to investigate canonical Wnt catenin signaling in the genital tubercle (and presumably in the rest of the anog enital system although these results have not been published) (Lin et al., 2008) Ina ctivation of catenin using the Shh gfpcre allele to drive cre expression in Shh expressing endoderm resulted in complete agenesis of the external genitalia and loss of Shh, Fgf8 and Bmp4 expression. Using the Msx2 cre allele to drive cre in the genital ec toderm resulted in a loss of ectodermal integrity and both severe hypoplasia of the genitalia and hypospadias. Lastly, using a tamoxifen inducible Shh creERT2 allele to temporally control removal of catenin in the genital endoderm genital outgrowth was f ound to be dependent on the timing of catenin removal (Lin et al., 2008) Interesti ngly, this study also employed a gain of function form of catenin driven by the Shh gfpcre allele in the endoderm and showed that constitutive activation of this molecule could drive increased expansion of the genital epithelium and concomitant growth of the genital tubercle. The loss of function and gain of function experiments combined with the finding that both Shh and Fgf8 expression are regulated by catenin suggested that the stability and function of catenin is vital to development of the extern al genitalia. Whether this intracellular signaling molecule functions as a master regulator during anogenital development awaits further investigation. General Overview The combination of traditional embryological examination and investigations into the molecular control of anogenital development has left the research community on the precipice of a deeper understanding into the developmental mechanisms of anogenital malformations. While classical embryology in vertebrates has revealed a surprising amou nt of developmental detail, the cellular origin of the cloacal membrane, distal urethra, vaginal opening and perineum remains
26 unknown. In Chapter 2, using a cre mediated reporter line, and both endoderm and ectoderm specific cre alleles, I test the hypot hesis that the distal urethra is derived from endoderm and that masculinization of the urethra is facilitated by an epithelial to mesenchymal transition. Chapters 3 and 4 specifically address the role of Fgf signaling, and in particular, the role of Fgf8 during initiation, outgrowth, and development of the external genitalia. I test the hypotheses that Fgf8 is required for initiation and outgrowth of the genital tubercle, and that it sits atop a molecular cascade controlling genital patterning. Chapter 4 describes the expression patterns of all known Fgf ligands and Fgf receptors during initiation and outgrowth of the genital tubercle to test for functional redundancy between alternate Fgf ligands and Fgf8. Chapters 5 and 6 investigate the role of Shh du ring anogenital development. By using a conditional knockout strategy I circumvent the early and severe phenotype of Shh / mice to test the temporal requirement of Shh as a master regulator of anogenital development. Chapter 5 focuses on a series of exp eriments in the genital tubercle to test the hypothesis that Shh regulates cell cycle progression and thus the rate of cell proliferation to control genital growth. Chapter 6 tests the requirement for Shh during multiple stages of anogenital development i ncluding cloacal septation and late stage maturation of the genital tubercle.
27 CHAPTER 2 CELL LINEAGE ANALYSI S DEMONSTRATES AN EN DODERMAL ORIGIN OF T HE DISTAL URETHRA AND P ERINEUM Introduction Despite the high incidence of congenital malformations of the anorectal and urogenital systems in humans, the mechanisms that govern normal anogenital development are poorly understood. The most common of these malformations is hypospadias, an ectopic opening (or multiple openings) of the urethra on the ventral asp ect of the phallus. Frequently, defects of anorectal and genitourinary organ systems occur together, which raises the possibility that they are linked mechanistically during early development. For example, failure of the embryonic cloaca to subdivide int o separate anorectal and urogenital sinuses (clinically referred to as persistent cloaca) is often associated with ambiguous genitalia, and numerous other syndromes involve associated defects of the bladder, anorectum, and external genitalia (Mo et al., 2001) Insight into how development of the external genitalia, urethra, bladder, rectum and perineum are coordinated at both cellular and molecular levels is necessary for understanding the basis of their association in congenital anomalies. In mammals, the embryonic cloaca undergoes septation to form separate genitourinary and anorectal sinuses, whereas in birds, reptiles and most anamniotes, the cloaca persists as a common outlet for the digestive, urinary and repro ductive tracts. Surface ectoderm and endoderm are in direct contact at only two positions during vertebrate development, posteriorly at the cloacal membrane and anteriorly at the oropharyngeal membrane. Cloacal endoderm lines the posterior most portion o f the gut tube and contacts the overlying ectoderm at the cloacal membrane. The morphogenetic mechanisms that drive division of the cloaca into separate urogenital and anorectal tracts are unclear, although a variety of processes have been proposed, inclu ding descent of a urorectal septum (known as the Tourneaux fold), extension of the Rathke
28 folds from the lateral walls of the cloaca, differential growth of the cloacal mesoderm, and reorganization of the cloacal epithelium (Hynes and Fraher, 2004c; Kluth et al., 1995; Nievelstein et al., 1998; van der Putte, 2005) After division of the cloaca, the anal and the genitourinary outlets are separated by the perineum on the posterior surface of the embryo. Since the mi ddle of the 19 th century, it has been reported that the perineum forms by medial growth and fusion of the cloacal folds, in a manner similar to fusion of the palatal shelves (Larso n, 2001; Nievelstein et al., 1998) Recent work in humans and in mice has challenged this interpretation, suggesting instead that the perineum is derived from the urorectal septum (Dravis et al., 2004; Hynes and Fr aher, 2004c; Sasaki et al., 2004; van der Putte, 2005) Identifying the developmental origin of the perineum should clarify how the posterior most part of the embryonic gut gives rise to anorectal and genitourinary organs. In both male and female mammali an embryos, development of the external genitalia begins with the emergence of the paired genital swellings immediately above the cloaca (see Perriton et al. 2002 for a detailed description of normal external genitalia development in mouse). These swellin gs fuse medially and give rise to the bipotential genital tubercle, which can be masculinized to form the penis or feminized to form the clitoris. As the genital tubercle grows out, the ventral side of the cloacal endoderm forms a bilaminar urethral plate that extends into the genital tubercle, and this structure later cavitates in a proximal to distal direction to form the urethral tube (Hynes and Fraher, 2004a; Perriton et al., 2002) Classical accounts of extern al genital development reported that the urethra has a dual embryonic origin -with the distal (glandar) portion arising from an ectodermal invagination from the distal tip of the genital tubercle and the proximal portion coming from the endodermal urethr al plate -a description that remains in contemporary embryology textbooks (Glenister, 1954; Larson, 2001; Moore, 2007;
29 Sadler, 2006) An alternative model proposes that the entire urethra forms from endoderm, whi ch undergoes differentiation in the glandar portion to form squamous epithelium (Felix, 1912; Kurzrock et al., 1999; Penington and Hutson, 2002a; Penington and Hutson, 2002b; Perriton et al., 2002) however neither model has been tested by direct analysis of cell lineage. During sexual differentiation of the external genitalia in mice, which occurs under the control of androgens, the bilaminar urethral plate is transformed into a central urethral tube along the leng th of the penis (Baskin et al., 2001; Hynes and Fraher, 2004b; Mahendroo et al., 2001; Yamada et al., 2003) Prior to masculinization, the urethral plate extends from the center of the genital tubercle to its ventr al edge, where it contacts the surface ectoderm at the cloacal membrane. The dorsal aspect of the urethral plate is thought to become the definitive urethral tube, while the ventral portion is remodeled such that the urethral tube becomes surrounded by st romal mesenchyme. It is unknown whether removal of the ventral aspect of the urethral plate is accomplished by apoptosis, an epithelial to mesenchymal transition or through the morphogenetic movement of the urethral epithelium. By contrast, female genita l development involves little remodeling of the urethral plate, resulting in a more proximal and ventral urethral opening. While apoptosis has been reported to occur in this region, the possibility of an epithelial to mesenchymal transition has not been e xcluded. Resolution of the cellular origin of the urethra and the fate of ventral urethral plate cells is important for identifying the cell population(s) affected in hypospadias and for investigations of gene function during urethragenesis. Sonic hedgeho g (Shh) is expressed throughout the endodermal epithelium of the gut, where it persists during division of the cloaca and formation of the urethral plate (Bitgood and McMahon, 1995; Echelard et al., 1993; Haraguchi e t al., 2001; Perriton et al., 2002) Shh / mice
30 fail to form a genital tubercle, indicating that Shh is required for development of the external genitalia (Haraguchi et al., 2001; Perriton et al., 2002) In addi tion, loss of Shh function results in a failure of cloacal septation, and mice are born with a persistent cloaca. The finding that Shh / mutants have severe defects of their genital and cloacal derivatives suggests that early Shh signaling from the hind gut endoderm may act to coordinate morphogenesis of the entire anogenital system. Here I investigate the cellular origins of the distal urethra and the perineum, and test the hypothesis that the ventral aspect of the urethral plate is removed during urethr agenesis by an epithelial to mesenchymal transition. I exploited the fact that endodermal cells of these organ systems express Shh during early development, in order to genetically label and fate map the cloacal endoderm during anorectal and genitourinary organogenesis. My lineage analysis provides the first direct evidence that the entire urethra is derived from endoderm, and that the transformation of the solid urethral plate into the definitive male urethra occurs in the absence of an epithelial to mes enchymal transition. The epithelial linings of the bladder, rectum and anterior region of the anus also are derived from Shh expressing endoderm. Moreover, I present the unexpected finding that cloacal endoderm gives rise to the perineum, which is the fi rst demonstration that endoderm differentiates into skin. I also fate mapped the ectoderm of the genital tubercle and show that it does not contribute to the urethral tube. Finally, I followed Shh descendant cells after disruption of androgen signaling a nd show that feminization of the male genitalia results from persistence of the endodermal urethral plate along the ventral margin of the genital tubercle. Taken together, these results reveal the fate of the cloacal endoderm during anorectal and urogenit al organogenesis, and highlight the importance of this cell population in the coordinated formation of the anogenital system.
31 Materials and Methods Transgenic Mice and Lineage Analysis The Shh gfpcre Msx2cre and Rosa26 reporter ( R26R ) mice used in this s tudy have been described previously (Harfe et al., 2004; Soriano, 1999; Sun et al., 2000) The Shh gfpcre allele was generated by knocking a gfpcre fusion cassette into the start site of the Shh locus, placing cre r ecombinase under the control of the endogenous Shh promoter (Harfe et al., 2004) The Shh gfpcre allele is a null allele, however heterozygous animals are phenotypically normal and breed successfully (Harfe et al., 2004) In Msx2cre mice, cre recombinase is driven by the proximal Msx2 promoter (Liu et al., 1994; Sun et al., 2000) To irreversibly label Shh gfpcre or Msx2cre expressing cells, I crossed heterozygous males to females carrying the R26R reporter allele. Females were inspected for vaginal plugs and the morning they were found was determined as embryonic day (E) 0.5. Pregnant dams were sacrificed at specific time points to collect a staged series of embryos with either Shh gfpcre ;R26R or Msx2cre;R26R genotypes. The genitalia and limbs were used to confirm age, and embryos were processed for X gal staining and histological analysis. Flutamide Administration Suspensions of flutamide (Sigma F9397) were prepared daily in corn oil. The corn oil was filtered and heated at ~55C to dissolve the flutamide. Flutamide was administered in 200ul S.C. injections at 100 mg/kg/day. Injections began on E13.5 and continued until pups w ere born. Injection sites were altered each day between shoulder and haunch. Control females were injected daily with the vehicle alone in the same manner. X gal Staining galactosidase activity was detected using X gal. Embryos were harvested in PBS and fixed overnight in 0.2% PFA at 4C. Embryos were washed three times in lacZ rinse buffer (1M
32 sodium phosphate pH 7.4, 0.1% sodium deoxycholate, 1M MgCl 2 0.02% NP40), then stained with X gal overnight rocking at room temperature. Embryos were then r insed, post fixed and stored in 4% PFA at 4C. Histology X gal stained embryos were processed into either paraffin wax or OCT for histological analysis. For wax prepared specimens, samples were dehydrated in a graded ethanol series, taken through a xylen e substitute (XS 3, Statlab) to preserve the galactosidase, and embedded in paraffin. Samples were cut at 10 m thickness and counterstained with Biebrich Scarlet. For cryosectioning, embryos were taken through a graded series of 15% sucrose/PBS, 30% su crose/PBS, and 30% sucrose/50% OCT before being embedded in 100% OCT and cut in 10 m serial sections. Results Cloacal Endoderm Gives Rise to the Entire Urethral Epithelium In order to resolve the origin of the glandar urethra, I first sought to determine the cellular origin of the entire urethral plate. It has been shown previously that Shh is expressed in the developing gut endoderm and is excluded from the surrounding mesoderm and ectoderm during development of the urogenital system (Bitgood and McMahon, 1995; Haraguchi et al., 2001; Perriton et al., 2002) Therefore, in order to fate map the entire cloacal endoderm, I utilized the Shh gfpcre allele to irreversibly activate the Rosa26 reporter ( R26R ) allele in al l Shh expressing cells and their descendants. Prior to initiation of genital tubercle outgrowth at E10.5, all cloacal endoderm expresses Shh, and R26R activity confirmed that Shh gfpcre expression faithfully follows the endogenous Shh expression pattern (F igure 2 1A). Histological sections through the cloacae of Shh gfpcre ;R26R mice at E10.5 showed that labeled cloacal endoderm cells are in contact with surface ectoderm,
33 and the junction of these two cell layers comprises the cloacal membrane (Figure 2 1B). The caudal most junction of these two cell layers, at the base of the tail, occurs at the proctodeum, the future site of the anal opening (Figure 2 1C, white asterisk). Between E10.5 and E11.5, the lateral walls of the lacZ labeled cloacal epithelium co me into apposition at the distal tip of the tubercle to form the beginning of a bi laminar urethral plate (Figure 2 1D). This apposition continues proximally as the tubercle grows out (Figure 2 1D, F). The junction between urethral plate endoderm and sur face ectoderm is maintained, and mesoderm lateral to the cloacal membrane does not invade this morphological boundary (Figure 2 1F). As the genital tubercle grows out, expansion of the anterior mesodermal population on the dorsal aspect of the tubercle (v isible as a dorsal swelling) results in ventral displacement of the urethral plate (Figure 2 1C, E, G ; see also Perriton et al, 2002). The entire urethral plate consisted of Shh gfpcre descendant cells at E13.5, and analysis of labeled cells during urethra l tube formation indicated that this endodermal population gives rise to the entire urethral tube, including the glandar portion (Figure 2 2). Cloacal Endoderm Gives Rise to the Perineum Activation of reporter gene expression in cloacal endoderm also allow ed us to map morphogenesis of the urorectal septum, a population of mesoderm that divides the cloaca into the anorectal and urogenital sinuses. By E10.5, the leading edge of the urorectal septum extended into the labeled endoderm at the anterior side of t he cloaca (Figure 2 1A). Beginning at E11.0, I observed a caudal expansion of the peritoneal cavity into the urorectal septum mesoderm (Figure 2 1C). The urorectal septum mesoderm continues partitioning the cloaca into the urogenital sinus and the hindgu t between stages E11.0 and E13.5 (Figure.1C, E, G). At the cloacal membrane, lacZ labeled endoderm is in contact with surface ectoderm, and the strict boundary maintained between lacZ positive and negative cells suggest that these cell
34 populations do not intermix (Figure 2 1B, D, F). Endoderm and ectoderm abut one another along the ventral side of the genital tubercle and at the proctodeum until the cloacal membrane ruptures at ~E13.0, creating the anus and resulting in transient exposure of the urethra at the base of the tubercle (Figure 2 1G, H). As the cloacal membrane ruptures, ectoderm degenerates between the anus and the base of the genital tubercle and is replaced by lacZ labeled endodermal cells at leading edge of the urorectal septum (Figure 2 1 H). As a result, endodermal cells at the caudal end of the urorectal septum come to lie on the surface of the embryo, where they form the central margin of the perineum (Figure 2 1H, Figure 2 2A, B). Sexual Differentiation of the Urethra I next investi gated the mechanism by which the male forms a tubular urethra within the glans penis, whereas in females, an epithelial cord persists ventrally in the clitoris. At E15.5, the male and female genital tubercles are still morphologically similar, although th e anogenital distance (the perineal area between the urogenital duct and the anus) is shorter in females (Figure 2 2A, B). At E15.5, lacZ labeled endodermal cells that previously covered the urorectal septum are visible along the central seam of the perin eum (Figure 2 2A, B). Beneath the perineum, mesoderm of the scrotal swellings in the male and labial swellings in the female is continuous with the mesoderm of the urorectal septum and with the proximal portion of the emerging preputial swellings (Figure 2 2A, B). Thus, three distinct outgrowths (labioscrotal swellings, preputial swellings and urorectal septum) arise from a continuous population of cloacal mesoderm. This mesoderm then envelops the glans from proximal to distal (Figure 2 2A F). At E15.5, the urethra extends to the distal tip of the genital tubercle and is composed entirely of cells descended from Shh gfpcre expressing endoderm (Figure 2 2A, B and Figure 2 3E, F). The proximal portion of the urethral plate has cavitated by E15.5 to form t he proximal urethral tube in both males and females (Figure 2 3A, B). The urethral duct is open at the
35 proximal end of the phallus in both sexes Within the distal portion of the glans, which is not yet surrounded by the prepuce, the urethral plate remai ns in contact with the overlying ectoderm (Figure 2 3E, F). Preputial glands begin to develop in both sexes at E13.5, when focal spots of Shh gfpcre expressing ectoderm begin to invaginate into preputial mesenchyme (Figure 2 2A, B). Beginning at E15.5, th e distribution of Shh gfpcre descendants reveals the onset of sexual differentiation of the urethra (Figure 2 2C, D and Figure 2 3A F). In males, the urethral plate is septated from proximal to distal to create the definitive urethral tube within the glans whereas in females this septation fails to occur (Figure 2 2C, D). By E17.5, the male urethral plate has been divided into a central urethral tube and a ventral seam (Figure 2 3G), whereas the female urethral plate persists to the ventral edge of the cl itoris (Figure 2 3H). The distribution of Shh gfpcre descendant cells in males shows that septation occurs when mesoderm of the preputial swellings and urorectal septum converges ventral to the definitive urethra (Figure 2 3G). During urethral septation, the preputial swellings continue to envelop the glans (Figure 2 2C, E). As a result of these coordinated movements, the urethral plate is divided dorsoventrally, with the dorsal portion forming the definitive urethra and the ventral portion forming the ur ethral seam along the ventral edge of the penis (Figure 2 3G). The absence of lacZ labeled cells from the genital mesenchyme indicates that the urethral epithelium does not undergo a transition to mesenchyme during septation (Figure 2 3G, I). In females proximal mesoderm fails to invade the genital tubercle and, consequently, the urethral plate is not septated (compare Figure 2 2C with 2D, Figure 2 3G with 3H, and 3I with 3J). As in males, the female preputial swellings continue to grow distally and bo th the cloacal and preputial folds expand medially to envelop the glans clitoris (Figure 2 2D, F and Figure 2 3H, J, L). At the same time, the proximal urethral plate is cavitated centrally to form the female
36 urethra (Figure 2 3J). This results in the fe male urethra remaining ventral to the glans, in contrast to the male condition, in which the urethral tube lies within the glans at E17.5 (Figure 2 3 compare I and J). The urethral duct remains open and will form the posterior portion of the vagina, which we also found to be derived from Shh gfpcre expressing endoderm (Figure 2 3H). In the distal region of the clitoris, the labeled urethral plate cells form an epithelial cord, whereas in males the plate continues to cavitate distally to form the penile ur ethra (Figure 2 3K, L). At birth, the female urethra lies distal to the vaginal opening, and is also derived entirely from endoderm (Figure 2 2F, H). In neonatal (P0) males, the glandar urethra is composed of Shh gfpcre descendants, indicating its endoderm al origin (Figure 2 2E, G). The distal glans is enveloped by the prepuce, and the penile urethra is positioned centrally within the glans (Figure 2 2E). Endodermally derived cells are still visible at the apical opening of the urethra, along the ventral seam of the preputial folds, and on the exterior surface of the perineum (Figure 2 2E, G). The proximal urethral duct in the male has closed, and labeled cells were restricted to the definitive urethral tube, including the distal meatus, and to the remnan t of the ventral preputial seam (Figs. 2 2E, G). There is no Ectodermal Contribution to the Glandar Urethra The fate map of Shh gfpcre expressing cells and their descendants showed an unequivocal contribution of endoderm to the glandar urethra, however th ese experiments could not exclude the possibility that some ectodermal cells are incorporated into the distal region. Therefore, we investigated whether the glandar urethra has an ectodermal component by using another cre allele to activate reporter gene e xpression in the genital ectoderm. We found previously that Msx2 is expressed in the ectoderm overlying the developing genital tubercle at E11.5, and it remains restricted to the dorsal and distal ectoderm of the genital tubercle at later developmental st ages (Figure 2 4A inset). We first determined whether the Msx2cre allele (Sun et al., 2000) is
37 expressed in distal genital tubercle ectoderm by crossing Msx2Cre males to females carrying R26R, and then examining reporter gene expression in embryos at E12. 5. We found that Msx2cre activated lacZ in ectoderm of the developing genital tubercle in a domain that was broader than endogenous Msx2 expression, and included the dorsal, distal and ventral medial ectoderm, but excluded the endoderm and mesoderm (Figur e 2 4A). Having determined this cre line to be an efficient marker of the genital tubercle ectoderm, we then examined the distribution of Msx2cre descendants in P0 males. Msx2cre expressing cells and their descendants were distributed throughout the surf ace ectoderm of the penis, however labeled cells were notably absent from the urethral opening, the seam of preputial fold fusion, and the midline of the perineum (Figure 2 4B). Histological analyses confirmed that blue cells were restricted to the ectode rm and did not invaginate into the glandar urethra (Figure 2 4C). Transverse sections showed a sharp boundary between the ectodermally derived Msx2cre expressing cells and the urethral tube (Figure 2 4C). Taken together, these findings show that the glan dar urethra is derived entirely from endoderm and that ectoderm makes no detectable contribution. Disruption of Androgen Signaling Feminizes Male Genitalia without Affecting Urethral Epithelial Integrity Disruption of androgen signaling during mammalian e xternal genital development causes feminization of the male genitalia and can result in hypospadias. Having identified the cellular basis of masculinization and feminization of the urethra, we next investigated whether antagonism of androgen receptor (AR) activity, using the AR antagonist flutamide, altered the behavior of the male urethral cell lineage. Administration of flutamide (100mg/kg/day) to pregnant dams from E13.5 resulted in complete feminization of the external genitalia and urethras of male p ups (Figure 2 5A D). Analysis of Shh gfpcre descendant cells in flutamide treated males revealed that the urethral plate did not undergo septation by the mesoderm, a
38 process that occurred normally in control males treated with corn oil (Figure 2 5, compare A, A with C, C). Consequently, the urethral endoderm of flutamide treated males failed to form a centralized urethral tube. The distribution of Shh gfpcre descendant cells in flutamide treated mice showed that the urethral plate persisted from the midl ine of the phallus to the ventral margin, indicating that feminization of the male urethral plate by AR antagonism mimics the process that occurs in normal female development (i.e., there is no epithelial to mesenchymal transition and the urethral plate re mains in contact with the surface ectoderm on the ventral edge of the phallus; Figure 2 5A D). Discussion Endodermal Origin of the Distal Urethra My fate map of the cloacal endoderm in mice provides the first direct evidence that the entire urethra is derived from endoderm. This finding challenges the longstanding view that the distal/glandar urethra arises from an invagination of distal ectoderm (Larson, 2001; Moore, 2007; Sadler, 2006) Histological studies and immunohistochemical analysis of cytokeratins had cast doubt on the hypothesis that the urethra has a dual origin (Kurzrock et al., 1999; Penington and Hutson, 2002a; Penington and Hutson, 2 002b) however in the absence of a fate map, the origin of the urethra was unresolved. The cell lineage analysis presented above demonstrates that, in mice, the entire urethra (including the distal most portion) originates from endodermal cells. This ra ises the possibility that a similar embryonic origin exists in human urethral development. Cloacal Morphogenesis and Development of the Genital Tubercle Previously, Perriton et al. (2002) showed that an asymmetric dorsal swelling appears as the genital tubercle begins to emerge from the ventral body wall. Hynes and Fraher (2004b) further clarified the importance of this outgrowth by suggesting that it contributes to the glans of the genital tubercle. By using Shh gfpcre to distinguish endoderm from meso derm during cloacal
39 development, I have been able to visualize the dynamics of these two cell populations relative to one another. The data show that the glans forms from mesoderm situated anterior to the cloaca and ventral to the urogenital sinus, along with mesoderm of the initial genital swellings. Expansion of the dorsal swelling and the urorectal septum mesoderm, respectively, on the dorsal and ventral sides of the urogenital sinus is associated with a dorsoventral compression of the urogenital sinus Thus, morphogenesis of the cloacal mesoderm may result in the distinct shape of the bladder and the ventral position of the urethral plate. After formation of the coelomic cavity, lateral plate mesoderm in contact with the gut epithelium is defined as splanchnic, whereas that in contact with the surface ectoderm is defined as somatic. The mesoderm of the genital tubercle is unique, in that it is sandwiched between endoderm and ectoderm (i.e., it not divided by the peritoneal cavity). This raises the p ossibility that genital tubercle mesoderm may differ from the splanchnic and somatic populations, both in the signals that it receives and in its responses to these signals. The expression patterns of a number of genes, including Ptc1, Fgf10, Hoxd13 and H oxa13, encircle the cloacal endoderm and are reminiscent of the response of gut mesoderm to endodermally derived Shh (Burns et al., 2004; Perriton et al., 2002; Petiot et al., 2005; Roberts et al., 1995) While pre vious work has compared emergence of the genital tubercle to early outgrowth of the limb bud (Haraguchi et al., 2001; Murakami and Mizuno, 1986; Perriton et al., 2002; Suzuki et al., 2003; Yamada et al., 2006) I pr opose that early development of external genitalia may be more similar to formation of the posterior gut tube, in which signaling occurs between endoderm and adjacent mesoderm. Origin of the Perineum My observation of Shh gfpcre descendant cells along the central margin of the perineum reveals an unexpected endodermal origin of perineal skin in newborn mice. Perineal ectoderm does not express Shh (Perriton et al., 2002) confirming that lacZ labeled cells of the perineal
40 seam are derived from endoderm. As the urorectal septum extends towards the site of the future perineum, cloacal endoderm is driven towards the posterior su rface of the embryo, where it ultimately comes to lie between the anus and the base of the genital tubercle. The discrete population of lacZ labeled endodermal cells that I observed along the central margin of the perineum appears to result from caudal mo vement of the hindgut, and marks the terminal point at which the cloacal swellings meet the urorectal septum to form the definitive perineum. This result clarifies the longstanding confusion over how the embryonic cloaca is divided into separate urogenita l and anorectal tracts (Hynes and Fraher, 2004c) The classical view of anogenital septation is that t he perineum forms from fusion of the cloacal shelves, in a manner similar to palatal shelf fusion (Dravis et al., 2004; Larson, 2001; Nievelstein et al., 1998) However, if movement of the mesoderm of the cloacal s wellings and urorectal septum was lateral to medial in an inward direction, then the central seam of the perineum would be expected to move deep within the perineum as these shelves fused in the midline. I found no evidence for this in my lineage map. Ra ther, cloacal endodermal cells were found on the surface along the central margin of the perineum. This supports the hypothesis that the urorectal septum contributes to the perineum (Dravis et al., 2004; Hynes and F raher, 2004c; Sasaki et al., 2004; van der Putte, 2005) Therefore, I propose that morphogenesis of the perineum involves posterior lateral eversion of urorectal septum mesoderm, which results in cloacal endoderm being displaced to the posterior surface of the embryo. The Role of Stromal Mesoderm in Urethral Tube Development The results presented here show that proximal to distal invasion of the urethral plate by the urorectal septum and preputial mesoderm drives masculinization of the urethral plate. This occurs in association with preputial fold fusion along the ventral midline of the tubercle. Moreover, my fate map shows that Shh gfpcre descendants do not contribute to the mesenchyme of
41 the external genitalia, which allows me to reject the hypothesi s that the remodeling of a urethral plate into a centrally positioned urethral tube is due to an epithelial to mesenchymal transition. I t is intriguing that apoptosis was not reported to occur in the epithelium of the urethral plate at E17.5 (when the ure thral plate is undergoing septation), but was restricted to the mesenchyme between the urethra and the ventral ectoderm (Baskin et al., 2001) Taken together these two findings suggest that septation of the urethr al plate results from morphogenetic reorganization of the epithelium which may be a response to signals or mechanical influence from the adjacent mesenchyme, and this does not involve significant apoptosis or an epithelial to mesenchymal transition. It ha s long been appreciated that disruption of androgen signaling (or treatment with estrogens) can lead to hypospadias in male genitalia (Agras et al., 2006; Gehring and Tomkins, 1974; Lyon and Hawkes, 1970) Despite extensive work on these pharmacological effects, the underlying developmental mechanisms responsible for hypospadias have been unclear. Recent work has shown that disruption of androgen signaling can modulate gene expression and alter epithelial organizat ion within urethral plate cells (Dravis et al., 2004; Petiot et al., 2005) My spatio temporal lineage map of endodermal morphogenesis during urethral tube formation suggests that t he timing of such disruptions may determine whether affected individuals have mild, moderate or severe hypospadias (see below). By identifying the cellular differences that occur during sexual differentiation of the genital tubercle my results suggest that hypospadias can be interpreted as a morphogenetic feminization of the male external genitalia. Urethral Tubulogenesis and Cloacal Septation are linked by a Common Developmental Mechanism The data presented above suggest that the cellular processes underlying septation of the cloaca als o underlie septation of the urethral plate to form the definitive male urethra.
42 Historically, formation of the external genitalia and septation of the cloaca have been considered separate developmental processes. My findings indicate that these two proce sses are coordinated along a spatiotemporal continuum, beginning with formation of the urorectal septum and ending with formation of the urethral meatus. As such, disruption of urorectal septum development during morphogenesis of the cloaca would be expec ted to result in malformations of both the urogenital and anorectal systems. Whole mount and histological data from both male and female mice shows that the primary cellular difference that occurs during masculinization of the urethral epithelium is the d ivision of the urethral plate by the mesenchyme of the urorectal septum and proximal preputial folds. Therefore, disruption of mesodermal septation of the urethral plate at earlier time points would be expected to result in more severe (i.e., proximal) hy pospadias, with severity being classified by proximodistal position of the urethral opening. According to my model, described in detail below, an arrest of urethral plate septation at E15.5 would lead to a complete feminization of the male genitalia, wher eas arrest at later time points would allow formation of a centralized urethra proximally but persistence of a ventrally open urethral plate distally. A Model for Masculinization of the Urethral Plate Based on the above results, I present a new model for morphogenesis and sexual differentiation of the urethra (Figure 2 6). The model suggests that posterior urogenital and anorectal development is divisible operationally into three integrated phases. Firstly, during initiation of external genital outgrowt h, paired genital swellings emerge ventro lateral to the cloaca, which is coordinated temporally with convergence and extension of cloacal mesoderm at the urorectal septum anterior to the cloaca. Disruption of either event will lead to external genital re duction or agenesis and persistent cloaca, consistent with the phenotypes found in Shh / Gli2 / Gli2 / ;Gli3 / p63 / and Hoxa13 / ;Hoxd13 / mutants, and in mice exhibiting caudal
43 regression syndrome (Cheng et al., 2006; Haraguchi et al., 2001; Ince et al., 2002; Kimmel et al., 2000; Mo et al., 2001; Perriton et al., 2002; Warot et al., 1997) The second phase involves cloacal morphogenesis and outgrowth of the genital tubercle, which covers the period from t he end of Phase I through septation of the cloaca into urogenital and anogenital sinuses, and includes formation of the urethral plate and perineum. Disruption in Phase II would lead to associated malformations of both the external genitalia and perineum (e.g., proximal hypospadias, micropenis, imperforate anus, persistent cloaca, etc). Shh has been suggested to act as an organizer during formation of the external genitalia, and my results suggest that its role as an organizing signal from the cloacal end oderm may act also to coordinate morphogenesis of the mesoderm surrounding the cloaca (Perriton et al., 2002) Consistent with my hypothesis, null mutations in the Gli family of proteins, which are key modulators of the Shh pathway, display malformations of this type (Kimmel et al., 2000; Mo et al., 2001) Lastly, the third phase of d evelopment includes the period from the completion of anorectal and urogenital septation (perineum formation) through sexual differentiation of the external genitalia. The behavior of cloacal endoderm in response to AR antagonism shows that, in contrast t o the previous two phases, Phase III is androgen dependent, and thus both genetic and hormonal disruption can affect normal morphogenesis during this time period. Phase III is defined by the invasion of urorectal septum and preputial mesoderm into the gen ital tubercle and a proximal to distal septation of the urethral plate to form a tubular urethra in the male. This is accompanied by growth and fusion of the prepuce along the ventral margin of the genital tubercle. These three phases provide a developme ntal framework for interpretation of congenital malformations and allow for the identification of the precise temporal windows during which morphogenesis has been disrupted in patients with urogenital and anorectal malformations.
44 My finding that septation of the urethral plate involves sustained growth of the mesoderm surrounding the glans and urethra identifies a morphogenetic mechanism for the proximal to distal progression of urethral tubulogenesis. This suggests that disruption of morphogenesis during Phase III will result in hypospadias of varying severity, with earlier perturbations resulting in more proximal hypospadias. Most importantly, unlike Phase I and Phase II morphogenesis, development during Phase III is directed by both local and systemic s ignals. How systemically circulating endocrine signals interact with the gene networks that operate locally within the genital tubercle is only beginning to be understood (Dravis et al., 2004; Petiot et al., 2005) however this dual nature of developmental control during sexual differentiation potentially increases the number of perturbations that can affect urethral tube formation at later stages of development. This chapter is reprinted from Developmental Bio logy, Jun 1;318(1), Seifert, A.W., Harfe, B.H., and Cohn, M.J., Cell lineage analysis demonstrates an endodermal origin to the distal urethra and perineum, 143 52, Copyright (2008), with permission from Elsevier.
45 Figure 2 1. Shh gfpcre expressing en dodermal cells give rise to the epithelium of the urethral plate, perineum, bladder and anorectum. Shh gfpcre ;R26R mouse embryos stained with X Gal to reveal lacZ positive cells (labeled blue). A, C, E and G are lateral views of genital region with the rig ht hindlimb bud removed. B, D and F are sections along the proximodistal axis of the genital tubercle, cut transverse to the trunk. H, Ventral view of the genital tubercle. Stages shown in upper right corners. Schematic diagrams across the bottom refer to stages shown in A, C, E, and G; dotted lines indicate planes of section shown in B, D, and F (red, ectoderm; blue, endoderm, gray; mesoderm). (A) Urorectal septum mesoderm has begun to septate the cloaca. (B) Labeled cloacal endoderm lies in contact with the unlabeled surface ectoderm to form the cloacal membrane. (C) Distribution of labeled endodermal cells in urogenital sinus (ugs), hindgut (hg), cloaca (cl), tailgut (tg). Urorectal septum mesoderm (urs) has extended into the anterior region of th e cloaca. Note position of anterior genital mesoderm (agm) relative to urogenital sinus. Asterisk marks the position of the proctodeum. (D) Labeled cloacal endoderm is beginning to form a bilaminar urethral plate (UP), which extends to distal tip of the genital tubercle where it abuts surface ectoderm (ecto). (E, F) Urethral plate (UP) spans the proximodistal length of the genital tubercle. The URS is approaching the proctodeum (asterisk). (G, H) Surface ectoderm at the base of the tail has ruptured and labeled endodermal cells have formed the central margin of the perineum (per).
Figure 2 2. Sexual differentiation of the urethral plate. Urethral tube development in male (A, C, E, G) and female (B, D, F, H) Shh gfpcre ;R26R mice between E15.5 and birt h (P0). Shh gfpcre expressing cells and their descendants are stained blue. Ventral/posterior surface of the genital tubercle is to the right in A F and to the bottom in G and H. The tail is removed. (A, B) lacZ labeled cells extend to the distal most t ip of the urethral plate (up; arrowheads) in the glans. The urethral duct (ud) is open in both males and females. Labeled cells are visible along the surface of the perineum (per; arrows). Dashed lines mark the position of the proximal urethra (u). Pre putial folds (pf) and preputial glands are visible in A D. (C) Male at E17.5, in which mesoderm of the urorectal septum (urs) and prepuce is seen invading the proximal end of the urethral plate (dashed line and asterisk). (D) Female at E17.5 showing the unseptated urethral plate (asterisk). (E) In neonatal males, the urethral duct has closed (compare with A and F) and endodermal cells along the surface of the perineum are contiguous with the ventral urethral seam. (F) In neonatal females, the urethral d uct, which contains lacZ positive cells, remains open and forms the posterior portion of the vagina. (G, H) The distal urethra in both males and females is derived from Shh gfpcre expressing cells. The female urethral opening lies more proximal and ventr al than the male urethra, which is positioned just beneath the apex of the glans.
48 Figure 2 3. Transformation of the urethral plate to a urethral tube. Comparison of urethral tube development in male (A, C, E, G, I, K) and female (B, D, F, H, J, L ) Shh gfpcre ;R26R mice at E15.5 and E17.5. Ventral is at the bottom. Cells derived from the Shh gfpcre expressing population are stained blue. Sections are transverse to the genital tubercle at proximal, middle and distal levels (shown in schematic diagra ms above each column). (A F) Labeled cells are restricted to the urethral plate epithelium and are absent from the mesenchyme at E15.5. Cavitation of the urethral plate proceeds from proximal to distal and the urethral duct (ud) is open in both males and females (u, urethra). (G) Male urethral plate is septated by urorectal septum mesoderm at the proximal end. Note the remnant of the urethral plate at the ventral edge of the penis. (H) The female urethral plate remains unseptated and the proximal urethr al duct remains open to form the posterior portion of the vagina (v). (I, K) Mesoderm has not yet invaded the middle (I) or distal (K) portions of the male urethral plate, which extends to the ventral edge of the penis. (J, L) Female urethral plate is t ubular at the mid shaft (J), but persists as a cord distally (L). Note the absence of mesoderm ventral to urethral plate in the female (arrow in L).
49 Figure 2 4. Genital tubercle ectoderm does not contribute to the distal urethra. Msx2cre;R26R male m ice stained with X gal showing lacZ expression in ectoderm of the genital tubercle. (A, B) Ventral views of E12.5 (A) and P0 (B) mice showing genital tubercles (gt). Dorsal surface of genital tubercles are towards the top and tails have been removed. In set of (A) shows Msx2 mRNA expression in lateral view of E12.5 genital tubercle. (B) Msx2cre activates lacZ throughout the surface ectoderm of the penis. Arrows mark lacZ negative domains along the central seam of the perineum and the ventral midline of the penis, two areas which contain cells derived from Shh gfpcre expressing population (compare with Figure 1 2E, F). (C) Distal, transverse section through penis shown in B reveals that labeled ectodermal cells contribute to skin but are excluded from the urethral tube (uo, urethral opening).
50 Figure 2 5. Feminization of male genitalia by disruption of androgen receptor activity. Comparison of Shh gfpcre ;R26R male (A, A, C, C) and female (B, B, D, D) mice at P0. (A, A) Control male mice showing S hh gfpcre descendant cells of the definitive urethra (u) enveloped by the mature glans. Septation of the urethral plate has displaced some Shh gfpcre descendant cells to the ventral surface of the penis Bracket marks mesodermal cells between urethral tube and ventral seam. (B, B) Control females show that Shh gfpcre descendant cells of the urethral plate persist in the midline between the preputial folds and glans. (C, C) Flutamide treated males show a feminization of the urethra, with Shh gfpcre descend ant cells persisting from the center of the glans to its ventral surface. The urethral plate has failed to septate and mimics development of control and flutamide treated females (compare C with B and D). Restriction of Shh gfpcre expressing cells to t he urethral epithelium in treated males indicates that flutamide induced feminization does not result in a transition of urethral plate epithelium to mesenchyme. (D, D) Females treated with flutamide showing normal position of urethra.
51 Figure 2 6. Mo del for masculinization of the urethral plate in the mouse penis. Diagrams at top show proximal to distal invasion of urorectal septum and preputial mesoderm (green arrows) into the male genital tubercle during masculinization, between E15.5 and P0. Red lines indicate planes of sections below (A C and A C), which show the spatial relationships of the urethral plate (up), urethra (u), prepuce and glans. The urethral plate is derived from endoderm (blue) and gives rise to the entire urethra. Beginning a round E15.5, this process is mediated by androgen signaling. As the preputial mesoderm grows towards the distal glans, preputial cells move in a ventral direction towards the urethral plate. Simultaneously, urorectal septum mesoderm, which is continuous with proximal preputial mesoderm, grows into the genital tubercle, and together these two continuous populations septate the urethral plate. As this occurs, the urethral tube becomes internalized within the maturing glans. The remaining ventral portion o f the urethral plate begins to disintegrate (B) and will form the ventral seam (raphe) of the penis (C). Absence of Shh gfpcre descendants in mesenchyme indicates that this division does not involve an epithelial to mesenchymal transition.
52 CHAPTER 3 FUNCT IONAL AND PHYLOGENET IC ANALYSIS OF FGF8 SHOWS THAT FGF8 IS NOT INVOLVED IN EXTERNAL GENITAL DEVELOPMENT Introduction The genital tubercle is the developmental precursor to the male and female external genitalia. Its outgrowth and patterning have been like ned to the vertebrate limb bud, but unlike the limb, the genital tubercle is composed of cells from all three germ layers (Chapter 1). I nitiation of genital outgrowth begins at approximately E10.5, when paired genital swellings appear on either side of the cloacal membrane (Perriton et al., 2002) The paired swellings are joined by an anterior swelling a day later and collectiv ely these form the genital tubercle. Signaling between the endoderm of the embryonic cloaca, which gives rise to the urethral epithelium, and the surrounding mesoderm is necessary for outgrowth and patterning of the genital tubercle. Mice lacking both co pies of the Sonic hedgehog ( Shh ) gene initiate outgrowth of the paired swellings, but these fail to form a genital tubercle and the mice exhibit complete agenesis of the external genitalia (Haraguchi et al., 2000; Pe rriton et al., 2002) Although Shh is required for maintenance of genital outgrowth, it is not the early initiation signal. In the vertebrate limb bud, Fibroblast growth factors (Fgf) 8 and 10 mediate the initiation of budding, and sustained outgrowth of the limb is controlled by Fgf signaling from the apical ectodermal ridge (AER) (Mariani and Martin, 2003; Ng et al., 1999) In the genital tubercle, Fgf8 is expressed in distal urethral epithelium (Haraguchi et al., 2000; Perriton et al., 2002) Some of the first molecular studies investigating the development of the mammalian genitalia established Fgf8 as an organizing and outgrowth signal and likened its role to that d uring limb development (Haraguchi et al., 2000; Ogino et al., 2001) These studies reported that removal of the distal urethral epithelium causes arrest of genital outgrowth in organ culture, and that application o f Fgf8 loaded beads can restore (or at least augment) gene expression and
53 outgrowth (Haraguchi et al., 2000) Furthermore, treatment of cultured tubercles with Fgf8 neutralizing antibody may inhibit development (Haraguchi et al., 2000) This and subsequent studies led to the hypothesis that Fgf8 expression in the distal urethral epithelium is required for outgrowth of the genitalia (Ha raguchi et al., 2001; Haraguchi et al., 2007; Haraguchi et al., 2000; Morgan et al., 2003; Ogino et al., 2001; Perriton et al., 2002; Satoh et al., 2004; Suzuki et al., 2003; Suzuki et al., 2008; Yamada et al., 2006) Although Fgf8 is widely held to be t he outgrowth signal, its function in genital development has not been examined genetically, in part because Fgf8 null embryos die prior to genital tubercle initiation (Meyers et al. 1998). Here I provide a direct test of the hypotheses that Fgf8 is requir ed for initiation, outgrowth and normal development of the external genitalia. Materials and Methods Mice Mouse strains used in this study have been previously described; Shh gfpcre (Harfe et al., 2004) Fgf8 fl/fl (Lewandoski et al., 2000) Fgf4 fl/fl (Sun et al., 2000) and Wnt5a / (Yamaguchi et al., 1999) I generated mice lacking Fgf8 in the genital tubercle by crossing Shh gfpcre ;Fgf8 fl/+ males to Fgf8 fl/fl females, and denote these mice as Fgf8 cKO (conditional knockout) mutants in the text. In one instance, a Fgf8 c KO embryo was recovered from a cross of Shh gfpcre ;Fgf8 fl/+ ;Fgf4 fl/+ male to F gf8 fl/fl ;Fgf4 fl/fl female ( Fgf4 is not expressed in the genital tubercle, and therefore its removal had no effect on the Fgf8 phenotype). Shh gfpcre ;Fgf8 fl/+ and Wnt5a +/ heterozygous animals were phenotypically normal and were used as controls. Mice were genotyped as described previously (Harfe et al., 2004; Lewandoski et al., 2000; Yamaguchi et al., 1999)
54 In situ Hybridization and Analysis of Apoptosis In situ hybridization was performed on all vertebrate embry os as previously described (Perriton et al., 2002) In order to detect Fgf8 expression in Sus scrufa we used a pig specific probe cloned and kindly provided by Brooke Warner and Zhengui Zheng ; for detection in Monodelphis domesticus we used a n opossum specific probe kindly provided by Anna Keyte and Katherine Smith ; for detection in both Trachemys scripta and Alligator mississi ppiensis we used an Fgf8 probe cloned from T. scripta and provided by Scott Gilbert. Mouse probes used in this study were Fgf8 (G. Martin), Shh, Wnt5a, (both from A. McMahon), Bmp4, Erm, Pea3 (B. Hogan), Dusp6 (J. C. Izpisua Belmonte), Sprouty 4 (B. Harfe ) and Hoxd13 (D. Duboule). Cell death was assayed using lysotracker red (Invitrogen) according to the manufacturers protocol. Immunohistochemistry Embryos were fixed in 2% paraformaldehyde overnight, washed in PBS, cryoprotected in 30% sucrose and em bedded in OCT. Sections were cut at 10 m. I performed immunohistochemistry as previously described (Thew issen et al., 2006) Briefly, I performed antigen retrieval by autoclav ing in sodium citrate followed by cooling on ice, washing with PBT, quenching of endogenous peroxidase with 2% hydrogen peroxide, blocking in 3% rabbit serum/PBT, and overnight incub ation at 4C with anti fibroblast growth factor 8 (1:1000 Santa Cruz sc 6958 ). Sections were washed in PBT, incubated with ABC Vectastain anti rabbit kit and detected with DAB. In addition, negative control samples (minus primary antibody) were used to de termine the level of background staining for all experiments. Results and Discussion Fgf8 Expression is Associated with Genital Tubercle Initiation in Wnt5a Null Mice In order to address the role of Fgf8 during genital tubercle (GT) outgrowth, I first e xamined Wnt5a / mice in which genital outgrowth is initiated but not maintained (Yamaguchi
55 et al. 1999, Suzuki et al. 2003). Surprisingly, in newborn (P0) male and female Wnt5a / mice, I found that external genitalia can develop, although with variable penetrance (Figure 3 1). Out of 15 Wnt5a / mice examined at birth, 6 had external genitalia with varying degrees of outgrowth (Figure 3 1). I examined m ice maintained on both C57BL/6 inbred and C57BL/6 x CD1 mixed backgrounds ( the original Wnt5a knocko ut study was also conducted on a mixed genetic background ; Yamaguchi et al. 1999) I found both of these genetic backgrounds exhibited a similar range of phenotypes, and the phenotypic spectrum did not vary with the genetic sex of the animal (Figure 3 1 a nd data not shown). In mice that developed external genitalia, the limbs were positioned normally (Figure 3 1C ). In the majority of cases where external genitalia were absent, the hindlimbs were displaced medially (Figure 3 1G), and in a few cases were f used into a single structure complete with digits (Figure 3 1I). Although medial displacement of the limbs was generally associated with reduction of the genitalia, some Wnt5a / embryos exhibited complete agenesis of the external genitalia, despite havin g normally positioned hindlimbs (Figure 3 1E). Given the variable degree of genital outgrowth in mutants, I asked whether the extent of outgrowth correlated with Fgf8 expression in the cloacal epithelium Examination of Fgf8 expression and tubercle morp hology in Wnt5a / embryos at E11.5 (when Fgf8 is expressed in normal embryos) revealed a range of expression patterns consistent with the morphological variation observed in mutants at P0 (postnatal day 0) (Figure 3 1B, D, F, H, J). Fgf8 expression was d etectable in Wnt5a / embryos that had a tubercle centrally positioned between the limb buds (Figure 3 1D and Yamaguchi et al. 1999). In other embryos, the hindlimb buds were displaced medially and the tubercle was either absent or severely reduced (Figur e 3 1F, H, J). In these embryos, Fgf8 expression was undetectable in the genital region, although expression was
56 robust in the AER (Figure 3 1F, H, J). In the most extreme cases, medial displacement of the hindlimb buds was so severe that they appeared t o meet in the midline, and in these cases I saw no evidence of genital outgrowth (Figure 3 1J). Embryos with such medially displaced hindimbs showed a single contiguous AER that expressed Fgf8 and had a stratified or pseudostratified columnar epithelial c haracter, and in these mutants the Shh domains were also contiguous between the two limb buds ( Figure 3 2A C). My finding that Fgf8 is expressed only in those Wnt5a mutants that initiate genital outgrowth raised the possibility that loss of Fgf8 expressio n may underlie the absence of external genitalia in the most severely affected mutants Fgf8 is induced at the Endodermal Ectodermal Boundary of the Cloacal Membrane In order to conditionally remove Fgf8 from the genital tubercle, it was first necessary t o map the origin of Fgf8 expressing cells. Previous reports have localized Fgf8 expression to the cloacal epithelium prior to genital tubercle outgrowth (Gofflot et al., 1997) and to the distal urethral epithelium following emergence of the tubercle (Haraguchi et al., 2001; Haraguchi et al., 2000; Morgan et al., 2003; Ogino et al., 2001; Perriton et al., 2002; Satoh et al., 2004) However, in the absence of cell lineage data it has remained unclear to what extent Fgf8 is regionalized in endoderm, ectoderm, or both. As I reported in Chapter 1, the cloacal membrane consists of two lineage restricted cellular compartments derived from endoderm and ectoderm. To determine the o rigin of the epithelial cells that express Fgf8 I used lacZ expression in Shh gfpcre ;R26R mice to mark the endodermal compartment and in situ hybridization to detect Fgf8 expression (Figure 3 3 A C) During initiation and early outgrowth of the genital tu bercle, Fgf8 is restricted to the distal most endoderm immediately beneath the surface ectoderm. The endodermal expression domain extended up to its boundary with the ectoderm, but the ectodermal cells themselves were negative for Fgf8 (Figure 3 3A, C).
57 I then mapped the dynamics of Fgf8 expression beginning at E10, and found that Fgf8 is not detectable in the cloacal endoderm before the onset of budding, although it was expressed in the endoderm and chordoneural hinge of the tailbud (Figure 3 3D and ins et). From E10.5, Fgf8 is expressed in a broad ventral portion of the cloacal epithelium immediately subjacent to the ectoderm and, as the genital tubercle grows out, the domain becomes restricted to the distal urethral epithelium, disappearing after E14.5 (Figure 3 3A, C, E, F and Perriton et al. 2002) Thus, while Fgf8 is expressed within some primitive streak cells that give rise to the definitive endoderm (Crossley and Martin, 1995) it does not appear to be expressed continuously. Rather, its expression is switched back on in cloacal endodermal cells at the endoderm ectoder m boundary of the cloacal membrane during initiation of genital outgrowth, in the cells fated to form the distal urethra (Chapter 1) This comparison of Fgf8 expression and cell lineage shows that Fgf8 is activated in the cloacal endoderm that makes conta ct with the ectoderm, and it is therefore possible that Fgf8 expression is a response to an inductive signal from the adjacent ectoderm. If endodermal ectodermal signaling at the cloacal membrane is required for activation of Fgf8 then one would expect e mbryos in which cloacal endoderm fails to contact ectoderm also to lack Fgf8 expression. I tested this hypothesis in Wnt5a / embryos in which A P axis elongation is disrupted, which affects the posterior position of the cloaca. In order to determine the posterior limit of the gut tube I used expression of Shh which labels the gut endoderm prior to and during development of the genital tubercle (Figure 3 4 A C ). Shh also marks the zone of polarizing activity (ZPA) in the limb buds, which provides a morp hological landmark for the posterior limit of the hindlimb bud ( Figure 3 4A C). Comparisons of the position of the cloaca and ZPA at E10.5 in wild type and heterozygous embryos (which are phenotypically normal) show that the
58 cloacal endoderm is positione d posterior to the ZPA and that Fgf8 is expressed where the endoderm makes contact with the surface ectoderm (Figure 3 4A, A, D, compare position of green and red lines). Comparatively, in Wnt5a / embryos that initiate genital outgrowth, the cloacal end oderm does not extend as far posterior and comes to lie anterior to the ZPA (Figure 3 4 B B, compare position of green and red lines relative to control embryos in A, A). In these mutants, the cloacal endoderm remains in contact with the ectoderm and Fg f8 is expressed, albeit in a smaller domain (Figure 3 4 B E). In those Wnt5a / embryos that lack a genital tubercle which also show a lack of posterior expansion of the peritoneal cavity ( Figure 3 4 ; compare posterior position of black dotted semicircl es in A C relative to the ZPA Figure 3 4C, C), the gut termi nates anterior to the hindlimbs and does not contact cloacal ectoderm (Figure 3 4C, C) In those Wnt5a / embryos where hindgut endoderm fails to reach the cloacal ectoderm, Fgf8 expression wa s not detected (Figure 3 4F). Thus, as posterior extension of the trunk is disrupted in Wnt5a / mice, the cloaca is displaced anteriorly relative to the hindlimb buds. In some cases this disruption maintains endoderm/ectoderm contact and Fgf8 expression is detected, whereas in the more extreme cases the cloaca terminates anterior to the peritoneal cavity, does not contact surface ectoderm and does not express Fgf8 These findings suggest that during initiation of genital tubercle outgrowth, Fgf8 express ion is switched on in response to the genital initiation signal only in cloacal endoderm that contacts surface ectoderm. Fgf8 is not necessary for Outgrowth of the Genital Tubercle The observations that Fgf8 is expressed in Wnt5a / embryos that develop ex ternal genitalia and that mutants lacking genital tubercles also lack Fgf8 expression led me to ask whether Fgf8 is required for initiation and outgrowth of the external genitalia. In order to test this hypothesis directly, I used a cre/loxP strategy to conditionally remove Fgf8 from the cloacal
59 endoderm prior to initiation of the genital tubercle. Based on my finding that Fgf8 is expressed exclusively in Shh expressing cells and their descendants in the GT, I used Shh gfpcre to inactivate floxed alleles of Fgf8 in the cloacal epithelium and I refer to these mice as Fgf8 c KO ( conditional knockout ; see methods for details) Mice carrying Shh gfpcre and two floxed alleles of Fgf8 underwent initiation of the paired genital swellings and developed a morpholog ically normal genital tubercle (Figure 3 5A). To confirm that this strategy results in complete deletion of Fgf8 from the urethral epithelium, I used a riboprobe designed to detect exons 2 and 3, which are internal to the loxP sites in the conditional Fgf 8 allele. Using this deletion probe, Fgf8 expression is undetectable in the genital tubercles of Fgf8 cKO mice before, during, and after initiation of outgrowth (Figure 3 5A). Fgf8 expression in the AER, which lacks cre, served as a positive internal con trol and the deletion probe revealed strong expression in the AER. In littermates lacking the Shh gfpcre allele, the deletion probe showed the normal pattern of Fgf8 expression in the genital tubercle and in the AER (Figure 3 5A). The finding that Fgf8 cK O mice form a genital tubercle in the absence of Fgf8 demonstrates that Fgf8 is not required for initiation or outgrowth of the genital tubercle. Previous studies reported that surgical removal of distal epithelium from genital tubercles at E11.5 led to a decrease in growth when measured in organ culture, and this can be reversed by addition of Fgf8 loaded beads (Haraguchi et al., 2000) Because the distal urethral endoderm is contiguous with distal ectoderm, surgic al excision of the distal tip may result in removal of both ectoderm and endoderm, each of which expresses numerous signaling molecules and transcription factors that are required for outgrowth. While the published interpretation of those manipulations wa s that Fgf8 produced from the distal tubercle is required for proliferation and
60 outgrowth of the genital mensenchyme, our results indicate that those truncations cannot be explained by loss of Fgf8 alone. Does Fgf8 Regulate Bmp4, Wnt5a, Hoxd13 and Shh Expr ession in the Genital Tubercle? Fgf8 has been placed at top of the genetic cascade that regulates genital tubercle outgrowth, and has been proposed to coordinate gene expression in the mesenchyme flanking the urethral epithelium (Haraguchi et al., 2001; Haraguchi et al., 2000; Morgan et al., 2003; Perriton et al., 2002; Satoh et al., 2004; Suzuki et al., 2003; Suzuki et al., 2008; Yamada et al., 2006) Manipulations of genital tubercles in organ culture suggested tha t expression of Bmp4, Msx1 Fgf10, and Hoxd13 are regulated by Fgf8 from the distal urethral epithelium; removal of distal tissue leads to varying degrees of downregulation, and implantation of Fgf8 beads augments their expression (Haraguchi et al., 2000) Interestingly, genital tubercles cultured with an Fgf8 neutralizing antibody maintained normal expression of Msx1 and Fgf10, and Bmp4 showed only minimal response (Haraguchi et al., 2000) Thus, it is unclear to what extent gene expression in the genital tubercle is regulated by Fgf8. To test this directly, I collected Fgf8 cKO embryos between E11.0 E12.5 to examine whether deletion of Fgf8 affects the expression of Bmp4, W nt5a, Hoxd13 or Shh In Fgf8 cKO embryos, the expression domain of each of these genes in the genitalia was maintained (Figure 3 5B), indicating that Fgf8 is not required for expression of Bmp4, Wnt5a, Hoxd13 or Shh in the genital tubercle. These finding s are consistent with the recent report that removal of catenin in the cloacal endoderm results in loss of Fgf8 but Hoxa13, Hoxd13, Msx2 and Wnt5a expression persists (Lin et al., 2008) Fgf8 functions as a cell survival factor in the limb bud (Sun et al., 2002) and it has been reported that Fgf8 inhibits apoptosis during outgrowth of the genitalia (Suzuki et al., 2003) Cell death is normally detected most prominently in the distal portion of the urethral epithelium and in the adjacent distal mesenchyme ( Figure 3 5C and Morgan et al., 2003; Perriton et al., 2002;
61 Suzuki et al., 2003) In contrast to the situation in the limb, I did not detect an increase in cell death at the distal tip of the genital tubercle follo wing removal of Fgf8 (n=6; Figure 3 5C). Cell death was still observed in the distal urethral epithelium, although the apoptotic domain appeared smaller than those observed in wild type littermates (Figure 3 5C). Previous work in organ culture showed tha t antagonism of Shh results in down regulation of Fgf8 and an increase in cell death (Haraguchi et al., 2000) My finding that genetic removal of Fgf8 in vivo results in neither increased cell death nor altered mor phology of the tubercle indicates that Fgf8 is not required as a cell survival factor during genital development and suggests that diminished Fgf8 expression in Shh / genitalia is unlikely to account for the excess apoptosis in those mutants Why deletio n of Fgf8 would result in a slight decrease in apoptosis is more difficult to explain, however it is interesting that mice lacking Noggin a Bmp antagonist, also show reduction of Fgf8 expression and decreased apoptosis in the distal genital tubercle (Suzuki et al., 2003) Conditional Removal of Fgf8 Results in Normal Development of the Penis During normal development, Fgf8 expression persists in the distal urethral epithelium through E1 4 .5. Therefore, I wanted to determine whether Fgf8 plays a later role in genital development and patterning. To investigate this possibility I generated Fgf8 cKO mice and raised them to adulthood to assess morphology of the penis. I found that outgrowth, axial patterning, and ureth ral tube closure were unaffected by removal of Fgf8 (Figure 3 5D). In addition, I analyzed the bacula ( os penis ) of Fgf8 cKO males and found that they were indistinguishable from wildtype males (data not shown). Moreover, the genitalia of Fgf8 cKO males are functional, as they are able to urinate, copulate and produce a semen plug, although they appear to be sterile. These results show that Fgf8 is not required for outgrowth, patterning or differentiation of the external genital ia.
62 Divergent Roles of Fgf Signaling in Limb and Genital Development During limb development several Fgfs are expressed in the AER and this redundancy results in compensation following loss of any one Fgf (Boulet et al., 2004; Lewandoski et al., 2000; Mariani et al., 2008; Moon et al., 2000; Sun et al., 2000; Sun et al., 2002; Sun et al., 1999) Therefore, I asked whether such redundancy could account for the ability of Fgf8 cKO mice to form genital tubercles. To determine whether other Fgf ligands are expressed in a pattern similar to (or overlapping with) Fgf8 I examined expression of all mammalian Fgf s during initiation and early outgrowth of the genitalia. A complete analysis by whole mount in situ hybridization determined that no other Fgf family members are expressed in a pattern similar to Fgf8 prior to E12.5, although other Fgfs were detected in the mesenchyme and in the ectoderm (see Chapter 4 for a complete description of all Fgf and FgfR expression patterns in the genitalia) I also examined expression of the four FgfRs during initiation and outgrowth of the GT and was unable to detect mesenchymal expression of any receptors prior to E12.5. This highlights an interesting difference between the distal urethral epithelium of t he genital tubercle and the AER of the limb bud. Whereas the AER is a site of extensive redundancy of Fgfs that synergistically control proximodistal outgrowth of the limb, the distal urethral epithelium shows no such redundancy during early stages of gen ital outgrowth and, therefore, normal development of a genital tubercle in Fgf8 conditional mutants cannot be explained by compensatory Fgf signaling from the endoderm. Is Fgf8 Signaling Active During Normal Genital Development? Although I could not detect alternate Fgf ligands in the cloacal endoderm, or the presence of a mesenchymal Fgf receptor prior to E12.5, as an additional test I explored Fgf pathway activity by monitoring the expression of four proposed Fgf signaling targets during normal genital de velopment Dusp6 ( Mkp3 ) Sprouty4, Erm ( Etv5 ) and Pea3 ( Etv4 ) have been shown to
63 be transcribed in response to Fgf signaling (Kawakami et al., 2003; Mariani et al., 2008; Minowada et al., 1999; Roehl and Nusslein Vo lhard, 2001; Sharrocks, 2001; Taniguchi et al., 2007) Dusp6 is induced through the PI(3)K/Akt pathway in response to Fgfs expressed in the AER ( Fgf4 8, 9 and 17 ) and is particularly sensitive to Fgf8 (Kawakami et al., 2003; Mariani et al., 2008) Dusp6 expression could not be detected during early outgrowth of the genital tubercle ( prior to E12.5), although robust expression was observed in other areas of Fgf signaling, including the distal limb mesenchyme and d ermamyotome (Figure 3 6). I first observed Dusp6 expression in the urethral epithelium at E12.5, two days after initiation of genital outgrowth, in a domain that partially overlapped the Fgf8 domain (Figure 3 6). Sprouty4 negatively regulates MAPK/ERK a ctivation in response to Fgf signaling and has been used as a readout of Fgf8 expression in many cell populations during mouse embryogenesis (Minowada et al., 1999; Sasaki et al., 2003) I monitored expression of S prouty4 and, like Dusp6 it was undetectable in the GT at E10.5 (Figure 3 6). Expression of Sprouty4 was first observed around the dorsal swelling and urethral plate at E11.5, with robust expression detected in the GT at E12.5 lateral to the urethral pla te (Figure 3 6 and Figure 3 7). The absence of Sprouty4 and Dusp6 in early genital tubercles suggests an absence of Fgf8 activity during initiation and outgrowth of the external genitalia. ETS transcription factors are expressed in response to ERK activat ion and both Erm and Pea3 have been shown to mirror Fgf8 in zebrafish (Roehl and Nusslein Volhard, 2001) I detected expression of both Erm and Pea3 at low levels in a broad pat tern surrounding the cloacal membrane at E10.5 (Figure 3 6). By E11.5, both genes were expressed broadly in the mesenchyme lateral to the endoderm and in the dorsal swelling (Figure 3 6 and Figure 3 7). Expression of both ETS factors was strong in the do rsal swelling at E12.5 with Pea3 expression
64 now detected in the urethral endoderm partially overlapping the Fgf8 domain and Erm expressed in the mesenchyme lateral to the urethral plate (Figure 3 6). To determine whether activation of Sprouty4 and Pea3 wa s directly regulated by Fgf8 I assayed for expression of these genes in Fgf8 c KO embryos. Expression of Sprouty 4 and Pea3 was unaffected following removal of Fgf8 in Fgf8 c KO embryos at E11.0 and E11.5 (Figure 3 8). Taken together, the findings that nei ther Dusp6 nor Sprouty4 are activated prior to E11.5, and that Sprouty4 and Pea3 are unresponsive to removal of Fgf8 supports our conclusion that Fgf8 signaling is absent from cloacal endoderm during initiation and outgrowth of the genital tubercle. Giv en that Fgf8 is expressed robustly in the distal urethra but fails to activate genes reported to be Fgf8 targets, I next asked whether Fgf8 protein was present in the GT. I used an antibody against the N terminal domain of Fgf8 used previously to detect protein in the limb and fin of vertebrates (Freitas et al., 2006; Thewissen et al., 2006) Although Fgf8 protein was present in and immediately subjacent to the limb ectoderm, I was unable to detect Fgf8 protein in or around the endodermal domain of Fgf8 expression in the genitalia at either E11.0 or E11.5 (Figure 3 9A D). T hese results strongly suggest that Fgf8 is not involved in development of the external genitalia. The observed expression of Sprouty4 Erm an d Pea3 in the dorsal swelling beginning after E11.5, after outgrowth is initiated, coincides with several Fgfs in that region and low levels of FgfR1 suggesting involvement by other Fgfs or alternate RTK pathway activation (Figure 3 5 insets and see chapt er 4). With respect to the broader patterns of Erm and Pea3 expression, both were described as Fgf readouts based on complete loss of their expression following treatment with SU5042 (Roehl and Nusslein Volhard, 2001) However, SU5042 does not spec ifically inhibit FgfR signaling, as previous studies have shown that it can inhibit both PDGF and VEGF
65 (Mohammadi et al., 1997) In addition, several secreted signaling molecules are expressed in a domain overlapping Fgf8 including, Bmp2 Bmp7 and Tgf all of which are capable of ERK activation (Derynck and Zhang, 2003; Grijelmo et al., 2007; Jin et al. 2006; Scott et al., 2005) Thus it is likely that ERK activation of Erm and Pea3 is in response to these or other factors present in the cloacal endoderm Collectively, these results indicate that Fgf8 does not play a role in development of mouse exter nal genitalia Phylogenetic Distribution of Fgf8 in Amniote Genital Tubercles Shows That it is not Required for External Genital Development The results of my experiments indicate that Fgf8 is transcribed in response to, but is not required for, initiatio n of the genital tubercle in mouse embryos External genitalia are found in every class of amniotes, and comparative developmental data indicate that amniote genitalia share a common embryonic origin and utilize a common suite of molecular developmental m echanisms (unpublished observations). To determine whether Fgf8 expression is a conserved feature of amniote external genitalia development, I examined its expression in the genital tubercles of a diverse array of amniote embryos including a rodent, an a rtiodactyl, a marsupial, an archosaur and a testudine. Along with mouse, I detected Fgf8 transcripts in the genital tubercles of both a marsupial (the possum Monodelphis domestica ) and an artiodactyl (the pig, Sus scrufa ) confirming its conserved expressi on in mammals (Figure 3 10). Fgf8 expression in the pig genital tubercle was confined to a single domain in the urethral plate (Figure 3 10 ). Opossums develop a bifid glans penis in which the distal urethral tube bifurcates into a pair of urethral groove s, and I observed two domains of Fgf8 at the distal end of the opossum genital tubercle (Figure 3 10 ). In contrast to the three mammalian embryos, Fgf8 expression was not detectable in the genital tubercles of either testudine (the red eared slider, Trach ymes scripta ) or archosaur (the alligator Alligator mississippiensis ) embryos, although in both taxa I detected
66 robust expression of Fgf8 in the AER, isthmus and pharyngeal arches (Figure 3 10 and data not shown). The finding that both alligators and turt les develop external genitalia in the absence of Fgf8 expression, along with my observation that Fgf8 is transcribed but not translated in the mouse, strongly supports my findings that Fgf8 is not required for development of the external genitalia. Why Fg f8 would be transcribed in mammalian urethrae is unclear. A major distinction between the genitalia of mammals and other amniotes is that only mammals undergo urethral tubulogenesis; in non mammalian amniotes the urethral epithelium persists as an open su lcus. The phylogenetic distribution of an Fgf8 expression domain in the urethral epithelium suggests that its origin coincided with the evolution of a urethral tube. The mechanism responsible for the evolutionary transition from a sulcus to a tube is unk nown, but it is possible that Fgf8 expression is a reflection of this new developmental process. Whatever the evolutionary cause of this novel domain of expression, our finding that Fgf8 is not translated suggests that it is a consequence rather than a ca use of external genital development, and explains why it is dispensable for external genital development in the mouse. Conclusions I have demonstrated through genetic analysis in the mouse that Fgf8 is not required for any aspect of external genital deve lopment. Furthermore, my finding that Fgf8 protein is undetectable and downstream targets are not activated during initiation and early outgrowth of the genital tubercle suggests possible post transcriptional regulation of Fgf8 Interestingly, analysis o f mFgf8 sequence using M icroinspector software (Rusinov et al., 2005) revealed 155 miRNA bin ding sites with 27 detected in exons 2 and 3 which are present in all eight Fgf8 splice variants. My analysis of Wnt5a / mice suggests that Fgf8 expression may serve as a readout of endodermal ectodermal signaling at the cloacal membrane. Although Fgf8 itself is not necessary to direct outgrowth, Fgf8 serves as a useful marker for genital induction as illustrated by the
67 variably penetrant phenotype that I have identified in Wnt5a / mice Future examinations of Fgf8 in mouse mutants that exhibit cloaca l membrane defects may be useful for illuminat ing the role of cloacal ectoderm during induction of genital outgrowth. The conclusion that Fgf8 is simply a re adout of the initiation signal is consistent with the observation that in Shh / embryos a cloacal membrane forms, Fgf8 expression is activated in the cloacal endoderm and external genital outgrowth is initiated (although not maintained ; Perriton et al. 2002). Thus, both Fgf8 and Shh can be eliminated as candidates for the genital initiation signal. The nature of the initiation factor remains to be discovered, and my findings raise the possibility that cloacal ectoderm may be the source of this signal.
Figure 3 1. Variable penetrance of external genital development in Wnt5a / mice cor relates with Fgf8 expression. Wnt5a +/ (A,B) and Wnt5a / (C J) mutants at P0 (A,C,E,G,I) and E11.5 (B,D,F,H,J). Ventral views; tail is to the bottom of each panel. Red arrows mark genitalia and black arrowheads mark positions of hindlimbs (tail and limb s dissected in A to show genitalia). (A, B) Wnt5a +/ mice showing normal position of external genitalia, hindlimbs, anus, and tail (A), and normal Fgf8 expression in the distal genital tubercle (B). (C, E, G, I) Wnt5a / mice showing the range of phenoty pes affecting the hindlimbs and external genitalia, which includes normally positioned hindlimbs and external genitalia (C), normally positioned hindlimbs and no external genitalia (E), medially displaced hindlimbs and no external genitalia (G), and a sing le fused hindlimb with digits and no external genitalia (I). No anal opening was found in any Wnt5a / mice examined. (D, F, H, J) Examination of Fgf8 expression in Wnt5a / embryos with varying degrees of hindlimb displacement shows Fgf8 expression in the distal genital tubercle (D) and no Fgf8 expression in mutants lacking genital tubercles (F, H, J). (J) Fgf8 expression in center of embryo is consequence of continuous AER like structure across the midline (see Figure 3 3 for further details).
70 Fig ure 3 2. Fgf8 expression marks a contiguous AER in Wnt5a / embryos with medially displaced hindlimb buds. (A) Specimen from Figure 1J showing Fgf8 expression in limb AER and across the midline. (B) Sagital section through the midline showing Fgf8 expres sion in the ectoderm. The ectoderm appears stratified as it does in (C) which shows Fgf8 expression in section through the limb bud and AER.
71 Figure 3 2. Fgf8 is normally expressed exclusively in Shh expressing cloacal endoderm. (A) Fgf8 expression (red arrow) in section of E11.5 embryo is localized exclusively to the cloacal endoderm and is excluded from the ectoderm (dotted lines). (B) X gal stained Shhgfpcre;R26R cells are restricted to the cloacal endoderm. (C) In situ hybridization and X gal st aining of E11.5 genital tubercle shows that Fgf8 expression (red bracket) is restricted to Shhgfpcre expressing cloacal endodermal cells. (D) Fgf8 is not expressed in the cloacal endoderm at E10, before initiation of genital outgrowth. Expression is detec ted in the tailbud (inset), where it localizes to the chordoneural hinge (CNH) and tailgut. (E, F) Fgf8 is expressed during initiation (E) and early outgrowth (F) of the genital swellings (red arrows).
72 Figure 3 4. Fgf8 is activated following end oderm/ectoderm contact at the cloacal membrane. (A C) In situ hybridization for Shh shows the posterior limit of the cloaca/hindgut in Wnt5a +/ and Wnt5a / mice. Dotted semicircles mark posterior boundary of the peritoneal cavity. (A C) Cleared spe cimens reveal posterior limit of cloaca/gut (CL) (red arrows) in relation to the ZPA (green arrows) of the hindlimb bud. The gut normally extends posterior to the ZPA (J), but is anterior to the ZPA is some Wnt5a / mutants (K) and is displaced ante rior to the entire hindlimb buds in others (L; compare position of green and black arrows). The endoderm ectoderm boundary of the cloacal membrane has not formed in (C, C). (D F) Endodermal Fgf8 expression is detected in Wnt5a heterozygous embryos (D; red arrow) and in those Wnt5a null embryos where the cloacal endoderm is in contact with surface ectoderm at the cloacal membrane (E; red arrow). (F) Fgf8 expression is not detected in Wnt5a / embryos where the cloacal endoderm has failed to contact the cloacal ectoderm.
Figure 3 5. Normal development of external genitalia in the absence of Fgf8 (A) Fgf8 probe specific to the floxed exons 2 and 3 shows absence of expression in both Fgf8 cKO and control embryos at E10.0. Expression is absent in Fgf8 c KO embryos during initiation (E11.0) and outgrowth (E11.5) but is present in control embryos. Note that expression in the AER is unaffected. (B) In situ hybridizations of Fgf8 cKO and control genitalia showing normal expression of Shh, Bmp4, Wnt5a and Ho xD13 (C) Analysis of apoptosis in Fgf8 cKO and control littermates at E12.5 shows a small decrease in the distal GT domain. (D) Normal penis morphology in Fgf8 cKO and control adults.
75 Figure 3 6. A mosaic readout of Fgf pathway activation during initiation and outgrowth of GT. Analysis of proposed downstream targets of Fgf signaling, Dusp6 Sprouty4, Erm and Pea3 during initiation (E10.5) and outgrowth (E11.5 E12.5) of the genitalia. Neither Dusp6 nor Sprouty4 was detected during initiation of t he GT at E10.5, although both targets were detected in the limb buds, somites and tailbud between E10.5 and E11.5. In contrast, both Erm and Pea3 were expressed in a broad pattern surrounding the cloacal membrane at both E10.5 and E11.5. At E11.5 Dusp6 e xpression was not detected in the GT and expression of Sprouty4 was now detected in the distal GT and dorsal swelling. At E12.5, Dusp6 and Pea3 were expressed in the urethral plate endoderm while Sprouty4 and Erm were expressed lateral to this domain. In addition Erm and Pea3 expression covered the entire dorsal swelling (insets). Detection of these targets at E12.5 was coincident with expression of other Fgfs
76 Figure 3 7. Lateral expression boundaries of Dusp6, Sprouty4, Erm and Pea3 Expression o f Dusp6, Sprouty4, Erm and Pea3 at E11.5 in Shh gfpre ;Fgf8 +/ embryos in lateral view. These embryos are controls for detection of these genes in Figure 6 and show the anteroposterior expression limits with respect to the Fgf8 domain in Figure 3. Expressio n of Dusp6 is undetectable. Expression of Sprouty4, Erm, and Pea3 extends into the dorsal swelling.
77 Figure 3 8. Pea3 and Sprouty4 expression is unaffected following removal of Fgf8 (A C) Pea3 and Sprouty4 expression at E11.5 in both control (A) and Fgf8 cKO embryos (B, C) following removal of Fgf8 Expression of Pea3 and Sprouty4 is unaffected in both the intensity and domain of expression.
78 Figure 3 9. Fgf8 protein is undetectable in the GT. (A D) Detection of Fgf8 protein during outgrowth between E11.0 (A, B) and E11.5 (C, D) in the genital tubercle and limb. Protein was detected with an antibody to the N terminal region of Fgf8 in the distal limb in and near the AER but was undetectable in the GT from the same specimen. Inset in B, shows staining restricted to AER in posterior section through the AER.
79 Figure 3 10. Phylogenetic characterization of Fgf8 expression in amniote genitalia. Fgf8 expression was detected in vertebrate embryos at similar stages of genital development. Fgf8 was detectable in the genital tubercles of all mammalian embryos examined including Monodelphis domestica (opossum), Sus scrufa (pig), and Mus musculus (mouse). Fgf8 was not detectable in the genital tubercles of Tracheyms scripta (turtle) or Alligator mi ssissippiensis (alligator) although robust expression was detected in the limb bud AER. Phylogenetic relationships for amniotes and reptiles is based on (Hedges and Poling, 1999) and for mammals on (Beck et al., 2006)
80 CHAPTER 4 FIBROBLAST GROWTH FA CTOR SIGNALING DURIN G INITIATION AND OUTGROWTH OF THE EXT ER NAL GENITALIA Introduction During mammalian development the male and female external genitalia (penis and clitoris respectively) originate as a single, indifferent anlage termed the genital tubercle (GT). The GT emerges from the ventral body wall surround ing and encapsulating the primitive cloaca beginning at embryonic day E10.5 and is comprised of all three germ layers (Per riton et al., 2002) Morphogenesis of the external genitalia can be grossly divided into three phases; (1) initiation and outgrowth of the GT including septation of the primitive cloaca into separate urogenital and anorectal compartments; (2) complete di vision of the cloaca and formation of the perineum along with continued outgrowth of the GT and development of the prepuce (the foreskin); (3) a hormonally controlled phase that directs sexually dimorphic development of the male and female forms (Hynes and Fraher, 2004a; Hynes and Fraher, 2004b; Hynes and Fraher, 2004c) Research over the past decade has begun to construct a spatiotemporal model for GT morphogenesis along the proximodistal and dorsoventral axis with in a molecular and genetic framework. This research has shown that several highly conserved signaling pathways, such as the Hedgehog (mediated via Gli1 3 ) (Haraguchi et al., 2001; Kimmel et al., 2000; Mo et al., 200 1; Perriton et al., 2002) Wingless (mediated via catenin) (Lin et al., 2008; Yamaguchi et al., 1999) Bone morphogenetic protein (mediated by BmpR1 ) (Morgan et al., 2003; Suzuki et al., 2003) and Fibroblast growth factor (through FgfR2iiib ) (Haraguchi et al., 2000; Ogino et al., 2001; Petiot et al., 2005; Satoh et al., 2004) are necessary for normal GT organogenesis. Among these pathways, the role of Fgf signaling and Fgf8 in particular during external genital development has drawn considerable attention (Haraguchi et al., 2001; Haraguchi et al., 2000; Lin et al., 2008; Morgan et al., 2003; Perriton et al., 2002; Petiot et al., 2005; Satoh et al.,
81 2004; Suzuki et al., 2003) The mammalian Fgf gene family is comprised of 22 members (Itoh and Ornitz, 2 008) With the exception of 4 family members (Fgf 11 14) all encode secreted proteins which function as extracellular signaling molecules binding to and activating tyrosine kinase receptors (FgfRs) (Itoh and Ornitz, 2008) Mutations in FgfR1 and FgfR2 give rise to a plethora of human disorders that include malformations of the limbs and in the case of FgfR2 the genitalia ( Coumoul and Deng, 2003; Xu et al., 1999) In particular, early research examining external genital development within the comparative framework of limb development drew parallels between proximodistal outgrowth in both systems (Haraguchi et al., 2001; Haraguchi et al., 2000; Perriton et al., 2002) Several Fgfs ( 4, 8, 9, 17 ) are expressed in the apical ectodermal ridge, a stratified layer of epithelium overlying the outgrowing limb bud, and signal to the underlyi ng mesoderm to direct outgrowth (Lewandoski et al., 2000; Mariani et al., 2008; Sun et al., 2002) Knockout analysis showed that loss of FgfR2iiib and Fgf10 results in a failure to develop limbs, but the loss of an y single ectodermal Fgf can be partially compensated for through the signaling action of the other Fgfs present in the apical ectodermal ridge (Mariani et al., 2008; Min et al., 1998; Revest et al., 2001; Sun et al., 2002) Thus, Fgfs can function redundantly during limb development and raise the prospect that promiscuous signaling may exist in other organ systems. Similar to its restricted expression pattern in the apical ectoderm of the limb, Fgf8 is expressed in a lineage restricted endodermal compartment during initiation and outgrowth of the GT ( Chapter 3 ). Until recently it was believed that Fgf8 controlled outgrowth of the external genitalia in a manner similar to its role during limb development (Haraguchi et al., 2001; Haraguchi et al., 2000; Lin et al., 2008; Morgan et al., 2003; Ogino et al., 2001; Perriton et al., 2002; Satoh et al., 2004; Suzuki et al., 2003) However, I demonstrated in Chapter 3 that Fgf8 is
82 di spensable for genital development and that its expression pattern during GT organogenesis appears to be restricted to mammalian amniotes. Limited analyses of a few Fgf ligands, FgfR1, and FgfR2 have been examined during GT development, although a more com plete and thorough investigation of Fgf expression has not been conducted. Given the possibility for functional redundancy as occurs between the four apical ectodermal ridge Fgfs I examined the expression of all known Fgf ligands and their receptors durin g initiation and outgrowth of the GT in order to determine if other Fgfs are expressed within the Fgf8 domain. I find that in early genital development no other Fgfs are expressed within the Fgf8 domain. These results show new expression patterns for sev eral Fgfs and FgfRs during initiation and outgrowth of the external genitalia and supports the finding that redundancy of Fgf signaling in the genitalia does not occur during initiation and outgrowth of the genitalia as occurs during limb outgrowth. In add ition, I examine several germline knockouts and present a survey of the genital phenotypes from available transgenic lines. Materials and Methods Mice The following mouse strains used in this study have been previously described; Fgf9 /+ ; Fgf8 fl/fl (Mar iani et al., 2008) I obtained Fgf9 null embryos by crossing male and female Fgf9 /+ ; Fgf8 fl/fl embryos to get Fgf9 / ; Fgf8 fl/fl I refer to these animals as Fgf9 KO in the text. In situ hybridizations were performed on CD1 mice obtained from Harlan. F emales were timed mated and embryos staged according to their genitalia and limbs. In situ hybridization and histology In situ hybridization was performed as previously described (Perriton et al., 2002) Probes used in this study were kindly provided by Gail Martin. I used probes for FgfR 1 4 that detect both isoforms of the receptor. In those cases when the genital tubercle was negative for a
83 particular Fgf I used alternate sites of expression as positive controls. In some instances expression was not detected at a particular embryonic stage and was not reported in the literature. Specimens examined for histology were fixe d overnight in 4% paraformaldehyde, dehydrated, embedded in paraffin and sectioned at 10um. Results and Discussion Fgf Expression During Initiation and Outgrowth of the External Genitalia During initiation and outgrowth of the genital tubercle Fgf8 is high ly expressed in a restricted endodermal domain of the cloacal membrane (Chapter 3). It was previously held that Fgf8 functioned during development of the external genitalia in a similar fashion to its role during limb outgrowth where it is expressed in th e apical ectodermal ridge. I have demonstrated through conditional removal of Fgf8 that it is not required during normal development of the external genitalia (Chapter 3). In order to determine if other fibroblast growth factor ligands ( Fgfs ) and their r eceptors ( FgfRs ) are expressed in the developing genital tubercle during the embryonic stages covering initiation and outgrowth (E10.5 E12.5), I used whole mount in situ hybridization to detect the spatiotemporal expression of these genes. These data ar e summarized in Table 4 1 and presented in Figures 4 1 -4 3 (see below). Expression at Stage E10.5 Outgrowth of the genital tubercle (GT) is initiated at ~E10.5 following expression of Fgf8 in the endodermal compartment of the cloacal membrane. Survey ing expression of all secreted Fgf ligands ( Fgf 1 10, 15 18, 20 23 ) I could only reliably detect expression of Fgf5 and Fgf8 at E10.5, and I was unable to detect the expression of other Fgf ligands in the endoderm, ectoderm or mesoderm during initiation (F igure 4 1). I observed a small area of Fgf5 expression in the ectoderm overlying the future dorsal swelling, anterior to the cloacal membrane (denoted by Fgf8 expression) (Figure 4 1). Although expression of Fgf10 has been reported in both the
84 cloacal en doderm and surrounding mesoderm at E10.5 my analysis was unable to confirm these results (Haraguchi et al., 2000) I assayed for expression of the four FgfRs and detected both FgfR2, and FgfR4 in the cloacal endode rm (Figure 4 1). In addition, I detected FgfR2 expression in the surrounding genital ectoderm. Expression of FgfR2iiib has been shown previously in the cloacal endoderm and my results suggest that the ectodermal expression is due to FgfRiiic (Petiot et al., 2005; Satoh et al., 2004) Thus, during initiation of the genital tubercle, no Fgf ligands are expressed in an overlapping pattern with the expression domain of Fgf8 This supports the conclusion that alternat e Fgf ligands do not compensate when Fgf8 is removed prior to and during initiation of the external genitalia (Chapter 3). Expression at Stage E11.5 Twenty four hours following initiation, expression of Fgf8 is maintained in the distal endoderm on the v entral surface of the GT ( Chapter 3 and Figure 4 2). As the genital tubercle emerges from the ventral body wall, it grows first as two paired swellings lateral to the cloaca and is then joined by an anterior swelling at E11.5 (Chapter 2 and Figure 4 2). During this phase of GT outgrowth other Fgf ligands can be detected (Figure 4 2). I detected expression of Fgf2 in the mesoderm of the dorsal swelling and this expression was excluded from the two lateral swellings (Figure 4 2). I also detected expressio n of Fgf9 in the mesenchyme of the dorsal swelling in a slightly more restricted pattern than Fgf2 (Figure 4 2). Expression of Fgf5 persisted in the ectoderm overlying the dorsal swelling and I was also able to detect Fgf9 and Fgf16 in this ectoderm (Figu re 4 2). In addition, a domain of Fgf5 expression also appeared on the ventral surface of the GT overlying the urethral endoderm (Figure 4 2). Confirming previous findings, I detected weak Fgf10 expression in two small distal domains lateral to the urethr al epithelium and FgfR2 expression in the urethral epithelium (Figure 4 2).
85 Profiling of receptor expression revealed that both FgfR2 and FgfR4 were still expressed (Figure 4 2). Using an FgfR2 probe recognizing both FgfR2 isoforms, I detected expression in the GT ectoderm and mesoderm along with the endodermal expression in the urethral epithelium (Figure 4 2). Faint expression of FgfR3 could be detected throughout the cloacal endoderm and the hindgut (Figure 4 2). I was unable to detect expression of FgfR1 above background levels at E11.5 although there is evidence that this background level represents real expression (Verheyden et al., 2005) The ectodermal expression domains suggest a potentially important role of the genital ectoderm during external genitalia development. However, an examination of genital phenotypes in Fgf and FgfR knockout mice suggests that no single Fgf ligand plays a major role during genital tubercle outgrowth (Table 4 2). Unlike Fgf8, the expression patterns of several Fgfs appear to overlap in the dorsal swelling ( Fgf2, Fgf9 ) and in the overlying ec toderm ( Fgf5, Fgf9, Fgf16 ). Combinatorial knockouts will be necessary to determine the role of Fgf signaling from these domains. Expression at Stage E12.5 At E12.5 the genital tubercle remains an undifferentiated collection of proliferating mesenchyme surrounding cloacal endoderm that has developed into the urethral plate, the precursor to the mature urethra in males and females (Chapter 2). The domain of Fgf8 expression in the distal urethral epithelium has continued to shrink, but Fgf8 is still stron gly expressed (Chapter 3 and Figure 4 3). For the first time during GT development, another Fgf ligand is now expressed within the expression domain of Fgf8 Fgf9 is weakly expressed in the distal portion of the urethral plate endoderm (Figure 4 3). Fgf 9 expression also remains in the mesenchyme of the dorsal swelling (Figure 4 3, inset). Fgf2 expression has persisted as weak, diffuse expression throughout the mesenchyme, but it is not restricted to the dorsal portion of the GT as it was at E11.5 (Figu re 4 3). I detected Fgf6 expression in the mesenchyme lateral to the
86 urethral plate, although it too was weakly expressed (Figure 4 3). The expression of Fgf10 has become stronger and is expressed within the mesenchyme adjacent to the urethral plate (Fig ure 4 3). Fgf18 which is strongly expressed in the limbs, was detected in deep lateral mesenchyme of the genital tubercle at this stage (Figure 4 3). Interestingly, at this stage I was able to detect all FgfRs in varied and unique expression patterns. FgfR1 was expressed in the ectoderm overlying the distal tip of the urethral plate and this ectodermal expression continued as a strip of expression extending over the dorsal midline of the genital tubercle (Figure 4 3 and inset). Expression of FgfR2 per sisted in the ectoderm, urethral plate, and was also weakly expressed in mesenchyme adjacent to the urethral plate (Figure 4 3). FgfR3, which was weakly expressed in the cloacal endoderm at E11.5, was also weakly expressed within the urethral plate at E12 .5 (Figure 4 3). Finally, I detected expression of FgfR4 in ectoderm overlying the distal portion of the urethral plate (Figure 4 3). In a few specimens this expression was more centrally restricted and in others it appeared in a slightly broader domain (data not shown). The expression of the various receptors suggests that Fgf signaling may become more active beginning at E12.5 and this corresponds to the time frame during which Dusp6 is expressed during genital development (Chapter 3). It is likely t hat Dusp6 expression is in response to Fgf10 / FgfR2iib signaling as both these molecules are required for genital development after E12.5 (Haraguchi et al., 2000; Petiot et al., 2005; Suzuki et al., 2003) It is als o interesting to note that the domain of FgfR4 expression corresponds to a region of active cell death as measured by lysotracker red (Perriton et al., 2002; Suzuki et al., 2003) Knockout analysis of the various F gfR2 receptor isoforms combined with an analysis of null FgfR3 4 mice
87 suggests that FgfR2 (and potentially FgfR1 ) are the relevant receptors through with Fgf signaling occurs in the genital tubercle (Table 4 2). Loss of Fgf9 Results in Female Hypospadias Null alleles and conditional knockouts have been generated for the majority of the Fgf family members (Table 4 2). Genital phenotypes have been reported for only a few of these mutations, and for others they have not been analyzed. Therefore, I conducte d a review of the literature and summarize this information in Table 4 2. Most of the individual knockouts of the Fgf ligands are fertile and do not present a genital phenotype (Table 4 2). Fgf8 germline mutations are lethal, but as I reported in Chapter 3, conditional inactivation of Fgf 8 in the genitalia does not affect development. Since I detected Fgf9 in mesenchyme of the dorsal swelling and also in the distal region of the urethral plate at E12.5 I examined newborn Fgf9 KO mice for subtle phenotypic defects. As Fgf9 KO embryos are sex reversed (Colvin et al., 2001) I compared P0 Fgf9 KO male and female newborns to female Fgf9 /+ heterozygous animals (Figure 4 4). Outgrowth of the genital tubercle was unaffec ted in both males and females (Figure 4 4). In normal males the urethra opens in a more distal position relative to the urethral opening in females (Chapter 2). In Fgf9 /+ heterozygous females the urethra is normally positioned (Figure 4 4). Histologica l sections of Fgf9 KO male and female specimens revealed that the urethra was feminized consistent with sex reversal in these mice (Figure 4 4). However, in addition to the feminization in both male and female null specimens, the urethral opening was shi fted proximally and was larger than normal (Figure 4 4 arrows). This finding shows that while Fgf9 is not required for outgrowth of the genital tubercle, it is involved during late stage ventral patterning of the external genitalia and during urethragenes is.
88 Null mutations in the Fgf receptors can produce defects in development of the external genitalia as have been previously reported for both Fgf10 and FgfR2iiib (Haraguchi et al., 2000; Petiot et al., 2005; Satoh et al., 2004) However, these studies showed that initiation and outgrowth of the GT were unaffected and that these mutations only produced defects during the second and third phases of genital development. The expression profile conducted in this stud y supports the conclusion that active Fgf signaling does not play an early role, although without compound mutant analysis it is difficult to rule out the possibility that several Fgfs signaling from the ectoderm or dorsal swelling do not contribute during GT outgrowth. My data, combined with the results from Chapter 3, support the finding that development of the limb and external genitalia differentially rely on Fgf signaling for initiation and outgrowth.
89 Table 4 1. Fgf and FgfR expression in the geni tal tubercle Ligand E10.5 E11.5 E12.5 Notes Fgf1 Fgf2 M M Fgf3 NA Fgf4 Fgf5 EC EC Dorsal swelling Fgf6 M Weak expression Fgf7 Fgf8 EN EN EN Urethral plate Fgf9 M M, EN Dorsal swelling Fgf10 M M Fgf15 Fgf16 EC Dorsal swelling Fgf17 Fgf18 M Fgf20 Fgf21 Fgf22 Fgf23 FgfR1 M, EN, ECT Dorsal swelling FgfR2 ECT, EN ECT, EN ECT, EN, M FgfR3 EN EN, ECT Weak expression FgfR4 EN M, EN, ECT Expr ession profile for all Fgf ligand and Fgf receptors. For each stage n 3. (M) mesoderm/mesenchyme, (EN) endoderm, (ECT) ectoderm, (NA) no data. Low expression has been reported for all posterior mesenchyme at these stages and appears as diffuse background staining (see Verheyden et al. 2005).
90 Table 4 2. Genital phenotyp es of Fgf and FgfR knockouts in mouse Ligand Phenotype Notes Reference Fgf1 fertile Fgf1/2 Dbl KO fertile (Miller et al., 2000) Fgf2 fertile Fgf1/2 Dbl KO fertile (Miller et al., 2000) Fgf3 fertile Fgf3/Fgf10 reported aberrant cloacal membrane formation (Satoh et al., 2004) Fgf4 ( ) E5.5 lethal, not expressed in GT (Feldman et al., 1995) Fgf5 fertile (Hebert et al., 1994) Fgf6 fertile (Fiore et al., 1997) Fgf7 fertile (Guo et al., 1996) Fgf8 fertile Null E8.5 lethal, Cond. KO sub fertile Chapter 3 Fgf9 Feminized/hypospadias Compound Fgf9/Fgf17 same as Fgf9 this study Fgf10 hypospadias (Haraguchi et al., 2000) Fgf15 ? E9.5 lethal (Ornitz and Itoh, 2001) Fgf16 fertile (Itoh and Ornitz, 2008) Fgf17 fertile normal (Xu et al., 2000) Fgf18 ? P0 lethal (Ornitz and Itoh, 2001) Fgf20 ? no knockout Fgf21 fertile (Itoh and Ornitz, 2008) Fgf22 fertile (Itoh and Ornitz, 2008) Fgf23 ? (Shimada et al., 2004) FgfR1 ? E7.5 E9.5 lethal, Need cond. KO (Deng et al., 1994) FgfR2 ? Null E4.5 lethal (Arman et al., 1999) FgfR2iiic fertile (Eswarakumar et al., 2002) FgfR2iiib hypospadia s (Petiot et al., 2005) FgfR3 fertile (Wang et al., 1999) FgfR4 fertile FgfR3/4 Dbl KO subfertile (Weinstein et al., 1998) Analysis of genital phenotypes was assessed from information reported in the referen ce cited, figures in the paper or our own analysis. In some cases a knockout has not been produced and in other cases due to lethality a conditional knockout mouse is necessary to properly assess the phenotype.
91 Figure 4 1. Only Fgf5 and Fgf8 are expr essed at E10.5 in the genital field. In situ hybridization of all Fgf ligands and receptors shows that Fgf8 is expressed in the cloacal endoderm, Fgf5 is expressed in the cloacal ectoderm, FgfR2 is expressed in the cloacal ectoderm and endoderm and FgfR4 is expressed in the cloacal endoderm.
Figure 4 2. Several Fgfs and Fgf receptors are expressed at E11.5. In situ hybridization shows that expression of Fgf8 persists in the urethral endoderm. Fgf2, and Fgf9 are expressed in the dorsal swelling mesoder m and Fgf5, and Fgf16 are expressed in the genital ectoderm. Fgf10 is expressed in the distal mesoderm lateral to the urethral endoderm. FgfR2 persisted in the genital tubercle and was present in all three germ layers, while FgfR3 was weakly detected in t he urogenital sinus.
94 Figure 4 3. Fgf expression at E12.5. In situ hybridization shows expression of Fgf ligands ( 2, 6, 8, 9, 10, 18 ) at E12.5 in the genital tubercle. Fgf8 persists in the distal urethral endoderm. Fgf2, Fgf6, and Fgf10 are express ed in genital mesoderm lateral to the urethral plate. Fgf9 is weakly expressed overlapping Fgf8 and at the lateral margins of the genital tubercle where Fgf18 is expressed. All four FgfRs are expressed at E12.5. FgfR1 and FgfR4 are expressed in the dist al GT ectoderm and FgfR1 is expressed in the dorsal swelling mesoderm. FgfR2 is expressed in all three germ layers while FgfR3 is weakly expressed in the urethral plate.
95 Figure 4 4. Loss of Fgf9 leads to feminization of the external genitalia and hyp ospadias. Comparative analysis of Fgf +/ (control) females with Fgf9 / males and females. Fgf+/ female showing normal ventral closure and urethral opening (arrow). In both Fgf9 / female and male genitalia, there is hypoplasia of the ventral side and a large, proximal hypospadias. Sections show feminization of the urethra.
96 CHAPTER 5 SONIC HEDGEHOG CONTR OLS GROWTH OF EXTERN AL GENITALIA BY REGULATING CELL CYCL E KINETICS Introduction The secreted signaling molecule Sonic hedgehog ( Shh ) acts to specify positional identities and to promote cell proliferation and survival in a wide range of organ systems (Ahlgren and Bronner Fraser, 1999; Duman Scheel et al., 2002; Rowitch et al., 1999; Roy and Ingham, 2002) For e xample, Shh has been proposed to integrate patterning with growth in the vertebrate limb and brain, and in both systems removal of Shh leads to altered pattern formation due to diminished proliferation of progenitor cells (Komada et al., 2008; Towers et al., 2008; Zhu et al., 2008) Although the Shh pathway can regulate expression of genes that control the cell cycle, including G1/S cyclins (Roy and Ingham, 2002; Towers et al. 2008) t he cellular mechanisms by which Shh influences proliferation are not well understood. Development of external genitalia requires initial formation of a genital tubercle (the precursor of the penis and clitoris), followed by a sustained period o f proximodistal o utgrowth that is coordinated with three dimensional tissue patterning and urethral tubulogenesis D eletion of Shh results in complete absence of external genitalia in Shh knockout mice which showed that Shh is required at early stages fo r establishment of the genital tubercle, however the early arrest of tubercle formation precluded studies of Shh function beyond the stage of initiation (Chiang et al., 1996; Haraguchi et al., 2001; Mo et al., 2001; Perriton et al., 2002) Here I test the hypotheses that Shh plays a later role in the coordination of growth and patterning of the external genitalia by controlling the kinetics of the cell cycle.
97 Materials and Methods Animals and Injections The Shh gfpcr e Shh creERT2 Shh C and R26R have been described previously (Dassule et al., 2000; Harfe et al., 2004; Soriano, 1999) Tamoxifen (6mg dissolved in corn oil) was administered to pregnant females at 10.5 and/or 12.5 dpc to induce recombination of the conditional Shh allele. BrdU was injected 44 hours after tamoxifen to label cells in S phase of the cell cycle, and pups were harvested 4 hours later for immunohistochemistry. Immunohistochemistry and X Gal Staining Im munohistochemistry was performed as described elsewhere (Palmer et al., 2000) Briefly, tissue was incubated in rat anti bromodeoxyuridine (1:500; Accurate, Westbury, New Jersey) and phosphorylated Histone H3 (1:50 0; Upstate) overnight at 4 C, then washed and incubated with donkey anti rat Cy3 and donkey anti mouse Cy5 secondary antibodies (1:500 Jackson Immunoresearch, West Grove Pennsylvania) overnight at 4 C. Live embryos were immersed in Lysotracker Red (1:5000 ; Molecular Probes) at 37 for 30min to label regions of cell death, then washed, dehydrated in methanol and photographed. X Gal staining was performed as described previously in chapter 2. In situ Hybridization Whole mount in situ hybridization was condu cted according to published methods (Perriton et al., 2002) using digoxigenin lableled riboprobes for Shh and Wnt5a, (A. McMa hon), Ptc1 (M. Scott) Hoxd13 (D. Duboule) Hoxa13 (S. Stadler), Bmp4 (B. Hogan) Bmp7 (C. Tabin), Msx1 and Msx2 (R. Hill), Foxf1 and Foxf2 (P. Carlsson), and Cyclin D1 and Cyclin E (C. Hui). Stereological Estimates of Total Cell Numbers Total (DAPI label ed), S phase (BrdU labeled), S and G2/M transition phase (BrdU/PHH3), and G2/M phase (PHH3) cells were estimated stereologically using the optical
98 fractionator method through the mesenchyme by counting target cells on every 12 th section (~ 6 sections in both Shh creERT2/C mutants and control littermates), as described previously (Gundersen et al., 1988; Ormerod et al., 2004) Briefly, cells were counted on images acquired using a 10x objective, N.A. 0.3 on an AxioO bserver Microscope and AxioVision Software (version 4.1; Zeiss). Section thickness was confirmed to be approximately 12 m by focusing through the sample. A counting frame of 45 m x 45 m and a sampling grid of 235 m x 235 m was used. The area of genital m esenchyme on each section was traced and quantified using AxioVision software and the total volume of the structure was estimated using Cavalieris principle (Gundersen et al., 1988) Densitie s were determined by dividing number of cells of interest by the area the cells were counted on (i.e. 45 m x 45 m). Proportions of cells were calculated by counting total cells (DAPI labeled) and S phase cells (BrdU labeled) in 5 separate counting frames per section and G2/M transition phase (BrdU/PHH3) and G2/M phase (PHH3) cells were counted on the entire section. This was done to avoid sub sampling errors due to the low number of PHH3 positive cells per section. Determination of Cell Cycle Kinetics F emales were injected with tamoxifen and BrdU as described above. S phase and total cell cycle length were calculated according to equations in Figure 5 4. BrdU labels cells approximately 30 min after injection (Cam eron and McKay, 2001; Martynoga et al., 2005) and is metabolized in approximately 2 hours (Cameron and McKay, 2001; Nowakowski et al., 1989) Given that S phase in both mouse and chick mesoderm (paraxial and later al plate) in vivo is at least 3 hours (Venters et al., 2008; White et al., 1992) a 4 hour interval between injection and harvest was chosen to allow BrdU labeled cells to transition from S phase to G2/M phase. Alt hough this may result in slight under representation of cells that transitioned from G1 to S
99 phase after BrdU metabolism, and cells in late S phase that acquired BrdU will remain positive in early G2, the relative nature of the proportion calculations allo ws for accurate percentages to be determined for each phase. Statistics All group differences in my dependent variables were revealed using Students t tests (one dependent variable between groups) or ANOVAs (more than one dependent variable between group s) and explored using Newman Kewls post hoc tests. Alpha levels were set at 0.05. Results and Discussion Shh expression in the developing genitalia begins prior to initiation of genital budding and persists through the period of sexual differentiation (Perriton et al., 2002) Transcription of Shh is confined to a lineage restricted compartment of cloacal endoderm that gives r ise to the urethral epithelium (Figure 5 1A and Chapter 2 ) To determine whether Shh plays a later role in growth of the external genitalia, I inactivated Shh after initiation of genital budding using a tamoxifen inducible cre cassette knocked into the Sh h locus ( Shh creERT2 ) (Harfe et al., 2004) to delete a conditionally null (floxed) allele of Shh (Shh C ) (Dassule et al., 2000) in Shh creERT2/C embryos. I first determined th e time required for tamoxifen to induce recombination by monitoring expression of the Rosa26 reporter ( R26R ) allele, which is induced in response to cre recombinase (Soriano, 1999) LacZ expressio n was first detectable 6hrs after tamoxifen injection and strong reporter activity was observed in all sites of endogenous Shh expression 9hrs after injection (Figure 5 1B). To identify when Shh signal transduction was terminated, I monitored expression of Ptc1, which is a transcriptional readout of the Shh effector genes Gli1 and Gli2 Ptc1 transcripts were monitored after tamoxifen injection in the genitalia and were detected at low levels at 12 hours, were barely detectable between 16 and 18, and were undetectable after 18 hours at 24, 48 and 72 hours, demonstrating c omplete and irreversible
100 inactivation of Sh h signaling (Figure 5 1C and data not shown). Thus, robust recombination of floxed alleles nine hours after tamoxifen injection leads to complet e inactivation of the Shh pathway between 18 and 24 hours. Having established the timing of Shh inactivation using the Shh creERT2 allele, I then tested whether Shh is required for growth and/or patterning following emergence of the genital tubercle. Base d on the previous finding that genital budding is initiated at E10.5 (Perriton et al., 2002) two time points were selected f or inactivation of Shh ; tamoxifen injection at E10.5 results in loss of Shh signaling by E11.5 (after the onset of budding), and injection at E12.5 results in loss of Shh signaling by E13.5. I found that removal of Shh at either stage resulted in truncati on of proximodistal outgrowth of the tubercle and in disruption of dorsoventral growth of the prepuce, the foreskin that surrounds the tubercle and encloses the urethra (Figure 5 1D). The extent of outgrowth was correlated with the duration of Shh signali ng Loss of Shh signaling by E11.5 resulted in the most severe truncation of the penis, with only a rudimentary glans developing (Figure 5 1D ). When Shh signaling persisted until E13.5, further outgrowth was observed, however the proximodistal length of the penis was still reduced and ventral growth of the prepuce had arrested before the enclosure of the urethra, resulting in an ectopic ventral opening of the urethra (known as hypospadias) Interestingly, I observed a similar relationship between the dur ation of Shh activity and the extent of cloacal septation, the partitioning of the embryonic cloaca into separate urogenital and anorectal tracts (discussed in Chapter 6). Loss of Shh signaling by E11.5 resulted in disrupted cloacal septation, and mutant pups retained a persistent cloaca (Figure 5 1D). When Shh signaling was inactivated by E13.5, mice developed separate anorectal and urogenital sinuses, and posteriorly the anus and genitalia were separated by a perineum (Figure 5 1D). Taken together wit h the finding that Shh / mice fail to develop a
101 genital tubercle, my discovery that the duration of Shh signaling correlates positively with the extent of genital tubercle outgrowth demonstrates that Shh has a persistent and continuous role during externa l genital development. Thus, initial expression of Shh regulates formation of the genital tubercle and sustained Shh activity is necessary for continued outgrowth and patterning of the phallus. The finding that Shh inactivation in the genital tubercle r esults in temporally sensitive proximodistal truncations and in hypoplasia of the ventral side of the penis suggests that Shh is required for sustained outgrowth and may direct genes involved in patterning the genital tubercle. I therefore asked whether e arly, transient expression of Shh in the genital tubercle is sufficient to establish the normal patterns of gene expression required for external genital development. Wnt5a, Hoxd13, and Hoxa13 are required for proper patterning of the genital tubercle and deletion of any one of these genes leads to severe defects of the genitalia (Morgan et al., 2003; Warot et al., 1997; Yamaguchi et al., 1999) Surprisingly, when the Shh pathway was inactivated as early as E11.5 ( tamoxifen injected at E10.5) shortly after initiation of the tubercle, Wnt5a, Hoxd13, and Hoxa13 continued to be expressed in appropriate patterns at E13.5 (Figure 5 2A). Moreover, Bmp4, Bmp7 Msx1 and Msx2, which mark dorsal, ventral and distal sides of the genital tubercle and regulate several aspects of genital morphogenesis were maintained in normally regionalized patterns 24 and 48 hours after loss of Shh signaling, albeit in reduced domains reflecting a smaller tubercle (Figure 5 2B and data not sh own) (Perriton et al., 2002; Suzuki et al., 2003) By contrast, germline deletion of Shh results in failure to establish or maintain expression of multiple genes in the genital region (Haraguchi et al., 2001; Perriton et al., 2002) Thus, Shh is required only during initiation of genital budding for specification of the molecular polarity of the genital tubercle. Interestingly, although the molecular pattern of the
102 genital tubercle was normal in Shh creERT2/C embryos at E12.5, these mice go on to develop severely truncated genitalia and persistent cloacae, as described above. The results indicate that transient Shh activity at the initiation of outgrowth is sufficient for pa ttern specification, but that sustained Shh activity i s required for the continued growth and elaboration of this pattern. Once a molecular pre pattern has been established in a developing organ, execution of the pattern requires extensive growth. To dete rmine whether loss of Shh signaling at E11.5 affects the expression of genes that mediate its mitogenic effects, I examined expression of Foxf1 and Foxf2, two Shh targets implicated in cell cycle control and proliferation in the embryo (Mahlapuu et al., 2001) Interestingly, both Foxf1 and Foxf2, which in normal mice are expressed in genital tubercle mesenchyme adjacent to the Shh expressing urethral plate, were downregulated following inactivation of Shh (Figure 5 2C). The observed reduction in Foxf1/2 led us to ask whether deletion of Shh alters expression of cyclins, which directly regulate cell cycle progression (Musgrove, 2006) I monitored expression of Cyclin D1 and Cyclin E, which control the G1 to S transition during the cell cycle. Immedi ately following inactivation of the Shh pathway (24 hours after E11.5 injection of tamoxifen, when Ptc1 is no longer detectable), expression of both Cyclin D1 and Cyclin E was diminished in the genital mesenchyme (Figure 5 2C). I also followed the express ion of Cyclin D1 and Cyclin E 48h after Ptc1 was no longer detectable and found their expression was still reduced compared to Shh +/C littermates (Figure 5 2D). These findings demonstrate that downregulation of genes that regulate the G1 to S transition i s an immediate early response to inactivation of the Shh pathway and maintenance of strong expression requires continued exposure to Shh Given that Shh is required to maintain activity of several genes that directly control cell cycle progression and proliferation, I quantified growth changes in the genital tubercle following
103 inactivation of the Shh pathway. First, I calculated total cell number /density and volume in genital tubercles of Shh creERT2/C and control mice (Figure 5 3; see Materials and Me thods). Twenty four hours after Shh signaling was inactivated I found that both volume and total cell number were decreased by approximately 75% ( t (4) =3.64, p=0.01 and t (4) =2.92, p=0.02 respectively; Figure 5 3D, E) in the genital tubercles of Shh creERT 2/C embryos relative to their wild type littermates ( Figure 5 3A C; yellow broken lines in 5 3B and 5 3C outline the region measured) I then investigated whether the gross reduction in total cell number and in the size of the genital tubercle was produce d by changes in cell death or in progenitor cell division. Analysis of apoptosis using lysotracker red showed similar low levels of cell death in both mesoderm and endoderm of m utant and wild type genital tubercles (although two small domains of apoptosis were seen at the proximal lateral edges of mutant tubercle ; Figure 5 4) Consistent with the observation that cell death was not markedly different in mutant and wild type genitalia, cell density did not differ statistically between groups ( t (4) =1.33, p= 0.12 ; Figure 5 3F). These results indicate that when the Shh pathway is inactivated after initiation of the genital tubercle, growth is inhibited by a mechanism that does not involve large scale apoptotic removal of cells. Our finding that loss of Shh s ignaling reduced the total cell number without a marked increase in apoptosis suggested that cell proliferation is altered in the genital mesenchyme. Given that genital outgrowth slows but does not arrest following Shh inactivation, we tested the hypothes is that Shh may regulate the rate of genital outgrowth, and that this may reflect changes to the kinetics of the cell cycle (Komada et al., 2008 ) T o determine whether cell proliferation kinetics are altered in the genitalia of Shh creERT2/C embryos, we first injected bromodeoxyuridine (BrdU) 20h after inactivation of the Shh pathway to label cells in the synthesis (S) phase. To allow sufficient time for BrdU labeled cells to transition from S phase to G2/M phase embryos
104 were allowed to develop in utero for 4 hours, then harvested and double labeled with antibodies against BrdU and phosphorylated Histone H3 (PHH3), a marker for cells in G2/M pha se (Hendzel et al., 1997) We calculated the proportion of mesenchymal cells that were labeled with BrdU alone ( S phase ), PHH3 alone (G2/M phase), double labeled with BrdU and PHH3 (S phase cells that had moved int o G2/M phase within 4hrs) or were unlabeled (G0/G1) (Figure 5 3G J; see Materials and Methods for further details) An ANOVA revealed that differences in the proportion of cells in S phase, M phase and G0/G1 in Shh creERT2/C mutant versus Shh +/C littermate s approached significance (F (2,8) =4; p=0.06) Because previous reports have demonstrated that Shh regulates the cell cycle, we conducted planned comparisons on the proportion of cells in S phase, M phase and G0/G1 phase. Relative to Shh +/C littermates, Shh creERT2/C mutant animals had an 8% decrease in S phase cells and a 7% increase G0/G1 phase cells (p = 0.03 and p = 0.05) (Figure 5 5A). Thus, in Shh creERT2/C mutants, we found a reduction of cells in S phase and a concomitant increase in the proportion of cells in G0/G1, suggesting that progression through G1 or the G1/S checkpoint is disrupted. We next asked how these relatively small proportional shifts in cell cycle phase could relate to the large growth differences we observed. If Shh is involved in regulating G1/S and G2/M transitions, then loss of Shh signaling sh ould arrest cells at specific cycle check points, thereby ce asing growth or inducing apoptosis, however this was not observed (Figure 5 1B and Figure 5 4). Alternatively, inactivation o f S hh could decrease the rate of progression through G1 which would be reflected by increased cycle length. To calculate cell cycle kinetics in developing genital tubercle mesenchyme, we applied the principles developed by Nowakowski to examine cell cycl e kinetics of dividing progenitors in developing brain tissue (Nowakowski et al., 1989) Given that the ratio between the number of cells in two phases of the cell cycle is
105 equal to the ratio of the time spent by these two populations in each phase (Martynoga et al., 2005; Nowakowski et al., 1989) we calculated the relative length of S phase (T S ) and the entire cell cycle (T C ) for both Shh creERT2/C and wild type littermates (see formulas in Figure 5 4B). Inactivation of Shh signaling at E11.5 resulted in lengthening of the entire cycle (T c ) from 8.5 to 14.4 hours (T ( 4) = 2.83; p<0.05 ). Our B rdU analysis suggests that this lengthening is not a result of altered S phase duration ( T ( 4) = 1.62; p>0.05), but more likely reflects a delay in G1 or at the G1/S checkpoint, based on the greater proportion of unlabelled cells in Shh creERT2/C mutants (Fig ure 5 4C). Such a marked increase in cell cycle duration would be expected to r educ e the total cell number, which may account for the ~75% reduction of tubercle volume in Shh creERT2/C mutants Indeed, when the proportion of cells in each phase of the cel l cycle is weighed against total cell number, the data show that loss of Shh signaling decreases the cycling cell population by ~73% ( Figure 5 5D). Thus, loss of Shh activity lengthens the time that cells spend in G1/G0, thereby reducing the number of cel ls in S phase which, in turn, feeds fewer cells into G2/M phase and, ultimately, back into the cell cycle. These results indicate that Shh controls the rate of progenitor cell proliferation, and thus progenitor pool size, by regulating the speed of the cel l cycle, highlighting a novel mechanism for Shh mediated control of organ growth (Figure 5 6). Previous work has shown that Shh and its downstream effectors can interact with specific cell cycle proteins (Roy and Ingham, 2002; Towers et al., 2008) and I have demonstrated in vivo that cyclins which regulate the G1/S transition are downregulated following Shh pathway inactivation. The experiments reported here show that after the early pattern is specified in exter nal genitalia Shh promotes its elaboration and growth by regulating the length of the cell cycle. The control mechanism identified here has implications for other developing organs, and may also be utilized in other signaling pathways. Shh has been prop osed to act as a polarizing
106 signal in both the limb and genitalia (Perriton et al., 2002; Riddle et al., 1993) In the limb, two recent studies reported that Shh specifies digit identity at early stages and that su stained expression is required for proliferation of progenitor cells and normal elaboration of skeletal pattern (Towers et al., 2008; Zhu et al., 2008) One of these papers (Towers et al., 2008) reported a reduction in Cyclins D1, D2 and both reported changes in the proportion of cells in different phases of the cell cycle, although how these changes can lead to a reduction of growth at the cellular level is not well understood. I n light of my findings, o ne possibility is that the loss of digits and the proximodistal truncations associated with reduced Shh activity in the limb may be caused by temporal reduction in the kinetics of the cell cycle, similar to those observed in the ge nitalia. Clearly, factors that alter cell cycle rate s during development c ould profound ly influence t he morphology and size of an organ, and this may reflect the extent to which the early pattern has been amplified during growth (Baker, 2007) Shh mediated modulation of c ell cycle duration may provide a cellular mechanism for heterochrony during morphological evolution. For example, temporal truncation of Shh expression in the limb bud is associated with decreased proliferation and reduction of digit number during skink e volution (Shapiro et al., 2003) My results suggest that such a reduction of Shh activity would lengthen cell cycle duration and thereby decrease the progenitor cell population in the limb. Similarly, the ability of hedgehog to alter cell cyc le length may influence the rate of tumor growth in hedgehog pathway mediated cancers (Yauch et al., 2008) These findings highlight the potential for modulators of cell cycle length to result in phenotypic changes in development, disease and evolution.
107 Figure 5 1. Shh regulates tempo ral development of the external genitalia and cloaca. (A) Lateral view of X gal stained Shh gfpcre ;R26R mouse embryo at E12.5 showing position of lacZ expression in Shh gfpcre descendent cells. Image captured using optical projection tomography. Red box sho ws schematic transverse section through the external genitalia at the level of the hindlimbs and depicts the position of Shh producing cells at the posterior end of the embryo. (B) Comparison of Shh creERT2 ;R26R embryos harvested 6 and 9 hours after inject ion of pregnant dams with tamoxifen. (C) Comparison of Ptc1 expression in ShhcreERT2/C and Shh +/C embryos 24 and 48 hours after tamoxifen injection. (D) Range of anogenital phenotypes produced by loss of Shh function at different developmental stages. Al l mice are males. Left panel shows complete agenesis of external genitalia and persistence of cloaca in Shh / mutant. Middle panels show anogenital regions of Shh creERT2/C mice in which Shh was deleted by tamoxifen injection at E10.5 and E12.5. Right pan el shows normal genitalia of wild type mouse with normal Shh activity.
108 Figure 5 2. Early and transient Shh expression is sufficient for pattern specification in the genital tubercle. (A D) Whole mount in situ hybridization showing gene expression patterns in genital tubercles of Shh creERT2/C and control Shh +/c mice. Genital tubercles are shown in ventral view except where indicated. (A C) Shh was inactivated by tamoxifen injection at E10.5, immediately after initiation of tubercle outgrowth, and embryos were harvested at E12.5 ( Bmp4, Msx2, Foxf1 and Foxf2 ) or E13.5 ( Wnt5a, Hoxd13, Hoxa13, Bmp7 ). Genital tubercles are reduced in size but show normal spatial expression patterns (A, B) except for Foxf1 and Foxf2 (C), two Shh targets implicated in c ell proliferation. (D) Embryos were harvested at E12.5 following injections at either E11.5 or E10.5. Cyclins D1 and E are downregulated following inactivation of Shh signaling.
109 Figure 5 2. Continued
110 Figure 5 3. Stereological estimates revea l decreased cell number. Genital tubercles of Shh creERT2/C and control Shh +/c embryos are shown at E12.5. Pregnant mothers were injected with tamoxifen at E10.5 and BrdU 44hrs later. (A) Lateral view of E12.5 Shhgfpcre;R26R embryo stained with X Gal to show Shh expressing cells. Red box marks region of genital tubercle (gt) and underlying cloacal endoderm shown in (B) and (C). (B, C) Control Shh +/C (B) and Shh creERT2/C (C) embryos showing BrdU labeled cells (red). (D F) Stereological estimates of vol ume (D) total mesenchymal cell number (E) and cell density (F) comparing Shh creERT2/C and Shh +/c and embryos. Data represented as means with +/ SEM. Asterisks denote significant differences at p<0.05. (G) High magnification view of boxed area in (B), s howing cells labeled with BrdU (red), phosphohistone H3 (blue) DAPI (white). Yellow arrow marks example of a cell positive for PHH3 but negative for BrdU (see also H J). (H J) Boxed area in (G) showing single channel exposures of BrdU (H), PHH3 (I) and D API (J).
111 Figure 5 4. Loss of Shh signaling does not result in widespread cell death in the genital tubercle. (A D) Comparison of cell death in Shh creERT2/C (A, C) and wildtype littermate (B, D) 48 hours after pregnant mothers were injected with tamoxi fen at E10.5. (A, B) Lateral views of the genital tubercle with the tail to the right. (C, D) Ventral view of genital tubercle
112 Figure 5 5. Shh controls growth by regulating cell cycle kinetics. (A) Proportion of cells in each phase of the cell cycle Labeling scheme groups G2 and M. (B) Summary of cell labeling scheme used to determine lengths of S phase (T s ) and total cell cycle (T c ). The red arcs refer to population of cells labeled with BrdU, the blue arc refers to population labeled with phosphoh istone H3, and green arc refers to population labeled with only DAPI (arc lengths not to scale). (C) Estimated cell cycle times for Shh creERT2/C (mutant) and Shh +/c (control) embryos. (D) Total cell number calculated for each phase of the cell cycle at E1 2.5.
113 Figure 5 6. A model for Shh mediated integration of growth and patterning. Shh activity in wild type (WT) and Shh creERT2/C (mutant) genitalia is shown in top panel, and is based on analysis of Ptc1 expression (see Figure 5 1C). During outgrow th of the genital tubercle, regionalized cell populations (red, green, black colored circles) that define a molecular pattern are exposed to secreted Shh (blue shaded areas) during development. During outgrowth of wild type genitalia, these cells divide ap proximately every 8.5 hrs and, as these progenitor pools double in number, this expands gene expression patterns. Following loss of Shh activity in mutant genitalia, cells continue to divide but cell cycle length increases to 14.4 hrs. This leads to a re duction in both the doubling rate of progenitor pools and the overall size of the genital tubercle. The molecular pattern of the mutant tubercle is retained. Large red circles represent the doubling time of all cells in the genital tubercle. T c total ce ll cycle time.
114 CHAPTER 6 A MULTIPHASIC ROLE F OR SHH DURING CLOACAL SEPT ATION AND EXTERNAL GENITALIA DEVELOPMEN T Introduction Classic embryological investigations have treated development of the mammalian genital tubercle and septation of the embryonic cl oaca as a single process whose disruption during development leads to anorectal malformations (ARM) (reviewed in Nievelstein et al. 1998, van der Putte 2005). The identification and examination of spontaneous animal models of ARM in the pig (van der Putte, 1986) and Sd mouse (Kluth et al., 1995) put forth new developmental ideas based on pathological development. More recently, the pharmalogical induction of ARM in mice with retinoic acid, etretinate, and adriamycin, has been re investigated in the context of emerging transgenic mouse models of ARM (Dawrant et al., 2008; Ioannides et al., 2003; Kubota et al., 1998; Ogino et al., 2001) The finding that ARM can result from disruptions to 5 Hox family members, and to alterations in the Hh and Retinoid pathways, stimulate d hope that an etiology for human syndromes that contain ARM could be elucidated. Indeed, subsequent research has found mutations affecting these pathways or downstream targets in human syndromes containing ARM such as Curriano ( HLXB9 ), Pallister Hall ( GL I1 ), Hand Foot Genital ( HOXA13 ), and Townes Brocks ( SAL1 ) (Kang et al., 1997; Kohlhase et al., 1998; Mortlock and Innis, 1997; Ross et al., 1998) In addition, the etiology underlying other human malformations such as the VACTERL complex and caudal dysgenesis have been postulated to arise from alterations to these pathways (Mo et al., 2001; Padmanabhan et al., 1999) Initiation of genital outgrowth and division of the embry onic cloaca in mice begins at approximately E10.5 (4 th week of human gestation) (Perriton et al., 2002) The mammalian gen ital tubercle will develop as a sexually indifferent appendage comprised of all three germ layers until it undergoes androgen mediated transformation into the sexually dimorphic external
115 genitalia (Chapter 2). The cloaca will divide into the urogenital si nus and hindgut compartments as mesoderm dorsal to the urachus and ventral to the gut ( u ro r ectal s eptum m esoderm, URSM) begins to accumulate at the anterior end of the cloaca. Mesenchyme of the urorectal septum that will occupy the space between the hindg ut and urogenital sinus is bounded dorsally, ventrally and posteriorly by Shh expressing endoderm but is itself continuous with posterior lateral plate mesoderm and as such, is defined wholly by the endoderm that borders it (Chapter 2). While the URSM has long been considered a distinct population of mesoderm driving septation of the cloaca into separate urogenital and anorectal tracts, research in various animal models over the last 25 years has challenged its role as an agent of development in lieu of a more active role for all cloacal mesoderm in concert with morphogenesis of the cloacal epithelium itself (Nievelstein et al., 1998; van der Putte, 2005) Molecular examinations of the underlying mechanism(s) respon sible for cloacal septation have been elusive. The finding that Shh / mice exhibit agenesis of the external genitalia and persistent cloaca and that compound Gli2/3 mutant mice phenocopy these malformations suggested that Hh signaling was absolutely requ ired for formation of the external genitalia and for at least some aspect of cloacal septation (Haraguchi et al., 2001; Mo et al., 2001; Perriton et al., 2002) Investigations into Hox gene function uncovered expre ssion of 5 Hox family members in the posterior anogenital system and later uncovered a common enhancer (the GCR) driving 5 paralogs of the HoxA and HoxD genes in both developing limbs and the genitalia (Dolle et al ., 1991; Lehoczky et al., 2004; Spitz et al., 2003) However, the absence of development prevented investigation of Hh signaling, and much like similar phenotypes in Hoxa13 / d13 / mice, has precluded examination of a requirement for these molecules bey ond
116 establishment of the anogenital program. This has left the treatment of anorectal morphogenesis in developmental limbo. Shh ( Sonic hedgehog ) is one of three vertebrate homologs of the Drosophila hh gene and is a potent developmental signaling molec ule that controls key events during vertebrate development (reviewed (Ingham and Plac zek, 2006) Through its binding to its cognate transmembrane receptor Ptc1 ( Patched ), normal repression of the Hh pathway by Smo ( Smoothened ) is relieved and the three Gli1 3 proteins can be processed to induce target gene transcription (Lipinski et al., 2006) Gli2 serves as the primary positive regulator of vertebrate Hh signaling by the three vertebrate hedgehogs, Shh ( Sonic hedgehog ), Ihh ( Indian hedgehog ) and Dhh ( Desert hedgehog ), of which only two, Shh and Ihh are expressed early during anogenital development (Haraguchi et al., 2001; Lipinski et al., 2006; Perriton et al., 2002) Previous work has shown that during gastrointestinal development in vertebrates, activat ion of target genes by the Shh pathway serves to pattern and direct morphogenesis of the endodermal organs (Ramalho Santos et al., 2000; Roberts et al., 1995) At the posterior terminus of the gut tube in amniotes, the primitive cloaca will give rise to the external genitalia and Shh is required for both patterning and growth of the genital tubercle and septation of the cloaca (Haraguchi et al., 2001; Perriton et al., 2002) Recent work has suggested that Shh serves to promote proliferation of the genital tubercle mesenchyme through interaction with the cell cycle control machinery such that Shh signaling promotes cell proliferation and expansion of progenitor pools expressin g specific patterning genes (Chapter 5). While this and other studies reported a requirement for Shh in order to form the perineum the molecular mechanisms of both cloacal septation and genital tubercle formation are poorly understood.
117 Following divisio n of the cloaca and subsequent formation of the perineum, in males the external genitalia undergo androgen driven morphogenesis to become a male penis with either a urethral groove (reptilia) or an insulated urethral tube (mammalia). In mammals, formation of the urethral tube (urethragenesis) is a complex morphogenetic event that is poorly understood beyond descriptive embryology (Chapter 2). Disruption of urethragenesis can lead to hypospadias, displacement of the urethral opening along the ventral aspec t of the penis. Occurrences of hypospadias have been steadily increasing in developing countries and the etiology cannot be explained by simple exposure to environmental endocrine disrupters or genetic malformations alone (Kim et al., 2004; Paulozzi et al., 1997) Thus there is a pressing need to better understand the molecular control of this program such that the underlying causes of hypospadias can be elucidated. In this study, I conditionally remove Shh from the cloacal endoderm in order to test the temporal requirement for Shh signaling during cloacal septation and development of the external genitalia. I provide evidence that prior to sexual differentiation of the genitalia there are two developmental windo ws in which there is an absolute temporal requirement for Shh signaling for formation of the separate urogenital and anorectal openings and for normal development of the genitalia. I present evidence that dynamic changes in both the cloacal mesenchyme and cloacal epithelium are necessary for normal development of the anogenital system. Furthermore, my findings provide a developmental model that can help predict the temporal occurrence of ARM based on the type and severity of malformation present. This wil l help as a diagnostic tool in investigating the etiology of human ARM.
118 Materials and Methods Animals The Shh creERT2 Shh gfpcre (Harfe et al. 2004), Shh C (Dassule et al. 2000), R26R (Soriano 1999), Msx2cre (Sun et al., 2000) Ptc1lacZ (Goodrich et al., 1997) and Smo C (Long et al., 2001) have been previously described. When analyzing Shh removal from the endoderm I used Shh C/+ litterm ates, and analyzing loss of Shh responsiveness in the ectoderm I used either Msx2cre;Smo C/+ or wt;Smo C/C littermates, all of which were found to be phenotypically normal. For activation of the Shh creERT2 tamoxifen was prepared fresh on the day of injectio n in corn oil (Sigma) at a concentration of 20mg/ml. Pregnant females were injected with tamoxifen at concentrations between 4 7mg. Injection schemes and harvest times varied according to the experiment and are outlined briefly in the text where appropri ate. LacZ staining was performed as previously described (Chapter 2). Histology Harvested embryos and X gal stained embryos were processed into either paraffin wax or OCT for histological analysis. For wax prepared specimens, samples were dehydrated in a graded ethanol series, taken through a xylene substitute (XS 3, Statlab) to preserve the galactosidase, and embedded in paraffin. Samples were cut at 12 m thickness and counterstained with Biebrich Scarlet. For cryosectioning, embryos were taken thr ough a graded series of 15% sucrose/PBS, 30% sucrose/PBS, and 30% sucrose/50% OCT before being embedded in 100% OCT and cut in 12 m serial sections. In situ Hybridization We performed in situ hybridization as previously described with probes for Ihh and P tc1, (Chapter 3).
119 Proliferation Index Mothers were injected with BrdU as described in Chapter 5. Because the urorectal septum mesoderm is not a shape but is a space defined by what borders it, I used three points within the space defined as; (1) the an terior end (caudalmost position within the urorectal septum mesoderm that lies adjacent to where the urogenital sinus and hindgut connect to the cloaca), (2) urogenital sinus position (approximate midpoint position of urorectal septum mesoderm immediately dorsal and adjacent to the urogenital sinus), and (3) hindgut position (mirror image position of the urogenital sinus position immediately ventral and adjacent to the hindgut). Counting frames containing approximately 50 cells were placed at these positio ns and the number of BrdU (S phase) and DAPI labeled cells were counted in each frame. Cells were counted in each frame on every other section for the width of the cloaca, averaged for each frame per specimen and converted to the proliferative index (BrdU cells/DAPI cells). Ectodermal Removal To assess morphology of the genital tubercle beneath the ectoderm in whole mount, I prepared specimens as follows. Newborn mice were dissected at the level of the hindlimbs and fixed overnight in 2% PFA. The specim ens were then washed in PBS and the ectoderm was cut along the dorsal side of the tubercle. Following this cut the ectoderm was removed by pulling it over the tubercle. Statistics A MANOVA was used to explore differences in URSM proliferation following r emoval of Shh signaling with position (i.e. anterior, UGS, hindgut) and proliferative index as the dependent variables. All group differences in my dependent variables were revealed using Newman Keuls Post Hoc comparisons. Alpha levels were set at 0.05.
120 Results Continuous Shh Signaling is required for Patterning and Outgrowth of the External Genitalia Shh / mice display complete agenesis of the external genitalia (Chiang et al., 1996; Haraguchi et al., 2001; Per riton et al., 2002; St Jacques et al., 1998) To circumvent this early and severe phenotype and investigate how Shh may regulate anogenital (genitourinary and anorectal) development I crossed male mice with a tamoxifen inducible cre recombinase knocked i nto the Shh locus ( Shh creERT2 ) (Harfe et al., 2004) to females homozygous for a conditional null allele of Shh (Shh C/C ) (Dassule et al., 2000) to generate embryos that wer e Shh creERT2/C Injection of tamoxifen allowed controlled removal of Shh activity at precise developmental stages. I previously defined the kinetics of the Shh creERT2 recombination at loxP sites in the Shh C/C allele using Ptc1 and found the Shh signaling pathway is functionally disabled approximately 18 to 24 hrs after injecting tamoxifen (Chapter 5). I removed Shh at time points corresponding to each developmental stage between E9.5 and E16.5 and observed the phenotypic consequences of this removal in E18.5 harvested embryos (Figure 6 1A). My injection time points began prior to initiation and outgrowth of the genital tubercle and continued through sexual differentiation of the external genitalia (injection at E9.5 and E15.5 respectively). Following l oss of Shh activity prior to genital budding at E10.5, I observed complete agenesis of the external genitalia except for the presence of two small swellings dorsolateral to the cloaca (Figure 6 1A). In addition, the urogenital and anorectal tracts failed to separate and communicated as a common posterior outlet (persistent cloaca). I observed persistent cloaca in all individuals injected prior to E12.5 (Figure 6 1A). Loss of Shh signaling at E11.5 exposed cloacal mesoderm to 24 hrs of Shh signaling durin g outgrowth and resulted in formation of patterned genital tissue surrounding a partial glans penis (Figure 6 1A).
121 Following 48 hours of Shh signaling, a genital tubercle developed, but the ventral aspect of the phallus was open resulting in exposure of t he urethra beneath the unfused preputial folds (Figure 6 1A). Injections at E12.5 and beyond revealed that Shh is required for proper ventral development of the external genitalia and tubulogenesis of the urethra in both males and females. The perineum forms during normal development between E13.0 E13.5 (Chapter 2) and consequently, loss of Shh activity beyond E13.0 resulted in the complete formation of both an anorectal and urogenital opening (Figure 6 1A). Loss of Shh activity between E13.25 and E15. 5 resulted in severe ventral hypospadias (ventral displacement of the urethral opening) along with ventral hypoplasia of the external genitalia (Figure 6 1A). In some cases, loss of Shh at E15.5 resulted in normal closure of the proximal urethral opening, although distal to this there was a hypospadias (Figure 6 1A). Injections at E15.5 resulting in loss of Shh activity at the onset of sexual differentiation resulted in a varying degree of feminization of the mature external genitalia, most likely dependa nt on the precise time of Shh removal (Figure 6 1A and data not shown). Injections both early and late produced the same phenotype in males and females (Figure 6 1B, C). Taken together, these results demonstrate that following initiation Shh is required during a primary developmental window for proper septation of the cloaca and continued outgrowth of the genital tubercle. My findings suggest that when Shh is deleted during the primary developmental window both the genital tubercle and cloaca are coupled through sensitive modulations in Shh signaling. In addition, these temporal deletions identify a second developmental window during which Shh is required for ventral growth and patterning of the genitalia up until at least the onset of sexual differentia tion into a penis or clitoris. Removal of
122 Shh during this period results in severe ventral hypospadias and feminization of the external genitalia. Loss of Shh Signaling Prior to E13.5 Results in Persistent Cloaca In order to address how removing Shh sign aling during the primary developmental window results in persistent cloaca, I followed the fate of the cloacal endoderm using the ROSA26 reporter allele in Shh creERT2/+ ;R26R embryos (Figure 6 2). I previously showed that the perineum forms from Shh expres sing cloacal endoderm in front of the urorectal septum mesoderm around E13.5 (Chapter 2) and thus hypothesized that persistent cloaca results from a failure of this endoderm to reach the surface ectoderm. I stained for lacZ activity in embryos collected f rom females injected at progressively later time points beginning at E9.5 (as above) and proceeding through E12.5. Examination of these embryos at E18.5 showed that the endoderm that normally forms the perineum does not reach the surface (Figure 6 2 A C, A C). Loss of Shh signaling after E13.5 always resulted in formation of the perineum indicating that rearrangement of the cloacal endoderm to fill the void left by the ruptured ectoderm requires Shh signaling until at least this time. That a longer exp osure to Shh correlated with a greater degree of cloacal septation suggests that the duration of Shh signaling affects either the morphogenesis of the cloacal endoderm itself, the surrounding mesoderm, or both, in a temporally controlled fashion. Posterio r Location of the Anorectal Opening is Dependent on Sustained Shh Signaling In order to partially address whether removing Shh affects the structure of the cloacal endoderm during cloacal septation I injected pregnant females with tamoxifen at 24 hr interv als (E8.5 E10.5) and harvested embryos at E14.5, twenty four hours after the cloaca normally divides into separate urogenital and anorectal outlets. I used OPT imaging to examine the structure of the gut and cloaca in Shh creERT2/C ;R26R embryos (Figure 6 3 A D, A D). Fate
123 mapping Shh expressing cells in E14.5 Shh gfpcre ;R26R embryos showed Shh descendent cells contributing to the perineum, the bladder, hindgut, proximal urethra and the urethral plate (Figure 6 3D). OPT imaging showed that loss of Shh sign aling in the primary developmental window affects the length and spatial arrangement of the cloacal endoderm (Figure 6 3A C). Loss of Shh signaling within this period resulted in a failure to form a well developed urethral plate, the persistence of a larg e, vacuous cloaca and a rudimentary bladder (Figure 6 3A C). Short exposure to Shh resulted in a shorter intestinal tract and cloaca, while a longer exposure increased the length of the gut and urethral plate (Figure 6 3A D). In comparison to the norma lly positioned anorectal opening, the hindgut emptied into the cloaca at varying positions dependant on the duration of Shh signaling (Figure 6 3A D). Close examination revealed that as the duration of Shh signaling increased, the location of the hindgut connection with the cloaca shifted posteriorly towards the surface ectoderm (Figure 6 3A C, red asterisks). Malformations of the Caudal Axis Associate with Genital and Cloacal Defects Malformations of the external genitalia are often found to occur in a ssociation with other defects exhibited in varying complexes such as Curriano, Townes Brocks, Pallister Hall, Hand Foot Genital, VACTERL, OEIS, Potter sequence and caudal dysgenesis. I examined the posterior body axis of Shh creERT2/C ;R26R embryos for asso ciated defects and found a wide range of malformations consistent with these syndromes (Figure 6 3E G). Loss of Shh signaling at E10.5 and E11.5 produced severe caudal truncation, loss of digits in the hindlimbs, tail truncation, loss of caudal vertebrae and intestinal hypoplasia (Figure 6 3A G and data not shown). In embryos injected at E9.5 the notochord atrophied in the posterior body axis and Shh descendant cells were found scattered among the mesoderm (Figure 6 3E). Shh positive floorplate cells wer e also absent in the posterior body axis (Figure 6 3E). Embryos injected at E10.5 exhibited a less severe caudal truncation, more robust growth of the limbs, persistence of a
124 notochord to the caudal somites, and Shh positive descendants in the floor plate of the neural tube anterior to the level of the notochord (Figure 6 3F). Embryos injected at E11.5 exhibited near normal posterior growth, although there was still a truncation of the tail and the genitalia (Figure 6 3G). Next I asked if the length of the urethral plate and outgrowth of the glans increased with exposure to Shh. Comparing the distribution of Shh descendent cells in the genital tubercle showed that sustained Shh activity increased the length of the urethral plate and this was associated with both the size of the genital tubercle and amount of outgrowth of the glans (Figure 6 3H J). These sections also show a decreased distance between the leading edge of the urorectal septum mesoderm and the distal tip of the glans, suggesting that Shh i s required for the posterior ventral expansion of the urorectal septum mesoderm (Figure 6 3H J). In addition, the total amount of mesoderm between the hindgut and neural tube was greater in embryos exposed to Shh for a longer time period suggesting loss o f Shh signaling affects mesoderm production during posterior body axis formation (Figure 6 3E G). Cell Death Occurs at the Lateral Margins of the GT Following Loss of Shh I next asked whether the reduced size of the genital tubercle was the result of cell death in the mesenchyme surrounding the cloacal endoderm. Previous research has reported widespread cell death in Shh / embryos, therefore I assayed for cell death following loss of Shh signaling (Haraguchi et al. 2001; Perriton et al., 2002) I used lysotracker red to analyze cell death in embryos following loss of Shh signaling (Figure 6 4 A D). Cell death was observed in the distal urethral epithelium in Shh creERT2/C embryos and in wild type littermates (Fig ure 6 4A D). Twenty four hours following loss of Shh signaling at initiation, I observed two domains of ectopic cell death in proximo lateral mesenchyme of the genital tubercle (Figure 6 4 compare A and B with C and D). I did not detect a significant inc rease in cell death of the urorectal septum
125 mesenchyme (data not shown). I also detected an increase in cell death within the tail somites and neural tube compared to control littermates (Figure 6 4 A D). These findings suggest that while Shh acts as a c ell survival factor in certain domains within the GT, loss of Shh signaling does not lead to widespread apoptosis throughout the genital mesenchyme. Shh Regulates Outgrowth by Maintaining Cell Cycle Progression Given the relatively minor increase in loca lized apoptosis I asked whether loss of Shh signaling from the cloacal endoderm affected growth and proliferation of the surrounding mesoderm. In Chapter 5 I showed that Shh signaling from the urethral epithelium regulates growth of the genital tubercle m esenchyme via its control over cell cycle kinetics. This analysis demonstrated that Shh controls growth of the genital tubercle by regulating the length of the cell cycle and thus the doubling time of progenitor cell populations. Because I found that gro wth of the genital tubercle and division of the cloaca were both dependent on Shh signaling during outgrowth, I asked whether mesenchymal cells of the urorectal septum are actively proliferating during division of the cloaca and if so, whether this prolife ration was in response to Shh activity (Figure 6 5 and Figure 6 6A, B white outlines). An examination of URSM between 10.75 and E12.5 revealed that these cells were actively proliferating as measured by BrdU incorporation (Figure 6 5 and Figure 6 6A). Co mparing proliferation of this mesenchymal population with that of the genital tubercle mesenchyme I noted that proliferation was lower in the URSM (Figure 6 5, Figure 6 6C and data not shown). In order to examine if proliferation was altered following los s of Shh I injected mice with tamoxifen at E10.5 and 44 hr later injected BrdU to label cells in S phase. I harvested embryos 4hrs later at E12.5, and examined the distribution of BrdU labeled cells (Figure 6 6A D). Mesenchymal cells of the urorectal sep tum are exposed to Shh from the cloacal endoderm which bounds these cells posteriorly, dorsal to the urogenital sinus and ventral to the hindgut.
126 Therefore I chose to measure the proliferative index at these three positions (Figure 6 6C, D, see materials and methods for details). I used a MANOVA to test the effect of removing Shh on proliferative index with the position of the measurements as a covariate. The MANOVA revealed a significant effect of proliferative indices between the three measured positio ns (F (2,8) =19.93, p=0.001), a significant treatment effect of removing Shh on proliferative index of the entire URSM (F (1,4) =70.12, p=0.001) and a significant interaction of both loss of Shh signaling and position (F (2,8) =4.30, p=0.05). Given the signific ant findings of the MANOVA I then used Newman Keuls post hoc comparions to examine each URSM position to its corresponding position between control littermates and Shh creERT2/C embryos and I found a significant decrease in proliferative index at the poster ior margin of the urorectal septum mesoderm ( t (4) =7.91, p=0.001 ), but not across the dorsal ventral axis of the URSM (UGS; t (4) =1.85, p=0.139 and hindgut; t (4) =2.36, p=0.078 areas did not differ significantly (Figure 6 5E). These findings show that Shh co ntrols proliferation of URSM mesenchyme and that loss of Shh signaling leads to a lower proliferative index, especially at the leading margin of the URSM and that this potentially contributes to the observed cloacal septation defect. Given the observed de fects of the cloacal endoderm I also asked if the epithelium was still actively dividing following loss of Shh signaling. I used BrdU to investigate if cloacal epithelial cells were dividing following loss of Shh signaling and found that there was widespr ead proliferation throughout the epithelium in both Shh creERT2/C embryos and wildtype littermates (Figure 6 5C,D and Figure 6 6). I also tested whether the progenitor cell layer of the epithelium was present using an antibody to p63. p63 was expressed th roughout the cloacal epithelium in both Shh creERT2/C embryos and wildtype littermates (Figure 6 7A, B). These findings suggest that
127 the epithelium maintains the ability to proliferate and that loss of Shh signaling does not lead to loss of progenitor cell turnover. Interestingly, in both Shh / embryos, and Shh creERT2/C embryos there is an appreciable degree of cloacal septation, compared to the severe truncation in GT outgrowth (Figure 6 6A D). I collected Shh creERT2/C embryos at E10.5 shortly after los s of Shh signaling and assayed for Ptc1 expression (Figure 6 8). Although Ptc1 expression was absent from mesenchyme around the cloacal membrane I observed some Ptc1 expression around the ventral portion of the cloaca and around the urorectal septum (Figu re 6 8). Given the persisten ce of Ptc1 expression in the absence of Shh I assayed for Ihh expression following Shh removal. Ihh transcripts were present in the anterior and ventral portions of the cloaca in control and Shh creERT2/C embryos (Figure 6 8). This expression persists until at least E13.5 but is absent from the genital tubercle (data not shown). In light of the temporal requirement for Shh signaling during cloacal septation, this finding raises the possibility that the small degree of cloacal d ivision observed in both Shh / embryos, and early injected Shh creERT2/C embryos might be driven by low level activation of Ptc1 in the urorectal septum mesoderm by Ihh although its expression is not sufficient to compensate for Shh and form the perineum. Shh is Required for Normal Tubulogenesis and Masculinization of the Urethra Once the perineum has formed, the primary developmental window coupling septation of the cloaca and outgrowth of the genital tubercle ends. After the perineum forms the preputia l swellings emerge at E13.5 and will continue to grow and encapsulate the glans. This marks the transition into the second developmental window. Following the observed hypoplasia of the ventral external genitalia, I investigated the requirement for Shh d uring urethragenesis in males. I found an absolute requirement for Shh until at least the onset of sexual differentiation, as loss of Shh signaling between E14.25 and E15.25 produced severe ventral hypospadias (Figure 6 9A
128 C). Observations in both whole mount and section reveal that the junction between the urethral plate epithelium and the ectoderm on the ventral surface of the genitalia had disintegrated (Figure 6 9A C). This loss of ventral epithelial integrity exposed the underlying urethra, which opened proximally instead of forming a closed tube (Figure 6 9A C). I removed the ectoderm from P0 embryos to examine the underlying whole mount structure of this junction (Figure 6 10). After removing the overlying ectoderm, the junction of the urethr a and prepuce opened up to expose the urethra (Figure 6 10 compare black arrows). The nature of the arrangement of these tissues suggests that the urethral plate, while loosely attached to itself at its ventral margin, is partly secured by the ectoderm at this margin, which serves to help maintain the integrity of this connective interface. Given these two results I hypothesized that the ectoderm might respond to Shh either directly or indirectly to maintain a junction where it contacts the urethral plat e epithelium. I assayed for Ptc1 signaling as an indicator of Shh responsiveness between E14.5 and E16.5 and found that along with mesenchymal domains adjacent to the urethral epithelium it was also expressed in the ectoderm overlying the urethral plate ( Figure 6 9D H). In order to test the requirement for Shh signaling to the ventral ectoderm, I genetically removed the ability of the ectoderm to respond to Shh signaling. Smoothened ( Smo ) is a seven pass transmembrane receptor that transduces Shh activit y and is normally repressed by Ptc1 I previously showed that the Msx2cre transgene is expressed in the GT ectoderm (Chapter 2). Therefore, I crossed Msx2cre;Smo C/+ males to Smo C/C females to produce Msx2cre;Smo C/C embryos and littermate controls (wt; Smo C/C ) (Figure 6 11 A D). Harvesting newborn (P0) mice I found that both males and females exhibited a similar phenotype to Shh creERT2 / C embryos that lost Shh signaling in the GT between E14.25 and E15.25 (Figure 6 9A, B and Figure 6 11B). I also harvested
129 Msx2cre;Smo C/C embryos at E16.5 after the onset of sexual differentiation to examine the ectoderm and developing prepuce (Figure 6 11C, D). Compared to control littermates, the ectoderm of Msx2cre;Smo C/C genital tubercles had separated at the midline and its edges were visible alongside the glans and the proximal portion of the urethra was open at the base (Figure 6 11 compare white arrows in C, D). I tested whether loss of ectodermal integrity across the ventral midline was the result of cell death in t he ectodermal/endodermal junction (Figure 6 11 C D). I did not detect an increase in cell death in the ectodermal cells following loss of Shh signaling (Figure 6 11C, D). These findings suggest that ventral ectoderm cells require Shh to maintain their na tive integrity and the junction with the urethral plate during urethragenesis. Loss of ectodermal integrity and thus, loss of this junction, results in hypospadias. Discussion Using various transgenic deletions in the mouse I have investigated the role o f Shh signaling from the cloacal endoderm during anogenital development. I examined division of the cloaca, outgrowth of the genital tubercle, and subsequent sexual differentiation of the external genitalia following loss of Shh from the endoderm ( Shh creE RT2/C ), and via loss of Shh responsiveness in the ectoderm ( Msx2cre;Smo C/C ). This allowed me to isolate and assess the role of all three germ layers during anogenital development with respect to Shh signaling. The temporally controlled deletions of Shh f rom the endoderm allowed me to circumvent the severe anogenital phenotype of Shh / mice (persistent cloaca, agenesis of external genitalia) and revealed a temporal dependence on Shh signaling during two unique developmental time windows. The first develo pmental window I refer to as the perineal period during which outgrowth of the genital tubercle (GT), rearrangement of the cloacal endoderm, and division of the embryonic cloaca into separate urogenital and anorectal openings are controlled by Shh signalin g from the endoderm. This period culminates with complete septation of the cloacal
130 endoderm by urorectal septum mesoderm, formation of the perineum, full extension of the urethral plate within the glans, and the emergence of the preputial swellings. Foll owing formation of the perineum and emergence of the preputial swellings, I identified a second developmental window I refer to as the preputial period This period extends until the onset of sexual differentiation and loss of Shh signaling during this pe riod results in hypospadias. Initial Septation Requires Hh Signaling I found a necessary requirement for Shh signaling during division of the embryonic cloaca. Despite loss of Shh prior to E10.5 I still observed some degree of septation. This is consist ent with Shh / mice that develop a rudimentary hindgut and hypoplastic bladder (Mo et al., 2001; Perriton et al., 2002) My results suggest that Ihh expressed in the hindgut endoderm and dorsal cloaca may compensa te for loss of Shh and initiation of septation. The finding that some Ptc1 persists in the urorectal mesoderm following loss of Shh supports this conclusion. Although Hh pathway activation via Ihh may partially compensate for this division, it is inadequ ate to fully compensate for the loss of Shh and supports the conclusion that Shh signaling is required for division of the cloaca. Comparing cloacal septation in Gli2 / ;Gli3 / mice to Shh / mice, the division of the rectum and the urogenital sinus appe ars more complete in Shh / mice, supporting a role for low level Hh signaling in the absence of Shh (Cheng et al., 2008) More rigorous comparisons of Gli mutants with both the Shh / and Gli2 / ;Gli3 / mice are necessary to further explore this requirement. Recent examination of the zebrafish anogenital system, which undergoes septation, similarly found a requirement for Hh signaling during anorectal development (Parkin et al., 2009) Using transgenic alleles, the most severe anorectal defects occurred following complete loss of Hh pathway activity in smu ( smoothened ) mutants or those treated with cyclopamine. Embryos null for shha displayed similar d efects to those observed in Shh / mice, with some ptc1
131 expression persistent in the posterior hindgut. Although loss of both ihha and shha did not phenocopy smu mutants, all Hh homologs were not removed in zebrafish. These findings support that initial division of the primitive cloaca is Hh dependent (Parkin et al., 2009) Examination of both Shh and Ihh expression in basal vertebrates and evidence of cloacal septation in conjunction with compl ete Hh inactivation in mice will shed light on the origin of this event. Shh Regulates Proliferation during the Perineal Period During the perineal period, my results show that both division of the cloaca and outgrowth of the genital tubercle exhibit a st age dependent requirement for Shh signaling from the cloacal endoderm. The findings suggest that both septation and outgrowth defects are due primarily to alterations in mesenchymal proliferation, but that defects in epithelial morphogenesis may also cont ribute. The findings presented in Chapter 5 showed that Shh signaling to the genital mesenchyme controls the rate of proliferation and I can now extend this to the urorectal septum mesoderm. Proliferation analysis in genital tubercle mesenchyme suggested that Shh acts to promote a constant proliferative rate via interaction with the cell cycle control system to promote progression through G1. Loss of Shh signaling leads to a reduction in the overall length of the cell cycle leading to a slower expansion of the genital mesenchyme. Here I have shown that loss of Shh signaling from the cloacal endoderm results in a lower proliferative index in the URSM and is most pronounced at the anterior end of the urorectal septum. Thus a lower rate of proliferation an d slower expansion of urorectal septum mesenchyme contributes to the septation defect. This finding is supported by evidence from Gli2 / mice that exhibit an anorectal fistula (Cheng et al., 2008; Mo et al., 2001) Gli2 is the primary transducer of the Shh pathway and loss of Gli2 during early bladder formation at E13.5 results in a decrease in proliferation of the surrounding mesenchyme (Cheng et al., 2008;
132 Lipinski et al., 2006) The effect of Shh emanating from the cloacal endoderm appears to be to promote the division and expansion of mesenchymal cells in the anogenital region while simultaneously promoting outgrowth of the genital mesenchyme. The importance of mesen chymal proliferation contributing to normal development, and the lack thereof contributing to abnormal development of the anogenital system is supported by several studies. p63 / mice undergo cloacal septation but fail to form a genital tubercle. Prolif eration is maintained in the urorectal septum mesenchyme, which continues to divide the cloaca, but proliferation decreases by 80% in the genital mesenchyme leading to complete agenesis of the external genitalia (Che ng et al., 2006) Conditional deletion of catenin from the cloacal endoderm using the Shh creERT2 allele leads to a loss of Shh expression from the endoderm and less proliferation in the genital mesenchyme, thus supporting a requirement for Shh regulate d proliferation of the cloacal mesenchyme (Lin et al., 2008) Hoxa13 / /d13 / mice f ail to septate the cloaca and develop external genitalia and it has been suggested that alterations in the rate of cell proliferation produce the resultant anogenital phenotype in these mice (Warot et al., 1997) A lthough Hox genes have been shown to alter rates of cell proliferation in the limb, it is unclear if this is the case in the anogenital region. Hox genes can regulate Shh expression in other tissues and while it is tempting to speculate that agenesis of t he external genitalia and persistent cloaca in Hoxa13 / /d13 / mice may result from loss of Shh signaling, the expression of Hoxa13, Hoxd13 and Shh in the posterior gut predates the emergence of external genitalia in vertebrates (unpublished. observations ). An alternate hypothesis is that HoxA and HoxD genes serve as permissive cues that facilitate cell cycle progression in conjunction with Shh signaling in the surrounding mesenchyme and thus act as upstream regulators of the Shh pathway following inducti on by the genital initiation factor. Further investigation of Hox interactions
133 with the cell cycle control machinery and Shh signaling in Hoxa13 / /d13 / mice will be necessary to explore these possibilities. Shh Regulates Epithelial Morphogenesis thro ugh an Unknown Mechanism In addition to controlling mesenchymal proliferation, my results also support a role for the cloacal endoderm in shaping the development of the anogenital system. OPT imaging revealed a dependence on the length of Shh signaling f or posterior migration of the junction between the anorectal opening and the cloaca and for lengthening of the urethral plate. The junction between the hindgut and the urogenital sinus represents continuous integration of the cloacal epithelium and sugges ts loss of Shh signaling somehow disrupts epithelial rearrangement as the hindgut migrates towards the surface ectoderm. Controlled deletions of Shh at precise developmental stages produced a range of phenotypes resembling various combinations of Gli2/3 k nockouts (Mo et al., 2001) These data suggest that Gli2 or Gli3 can compensate for outgrowth of the genital tubercle as single mutants display normal GT outgrowth, but that Gli2 is necessary for migration of the a norectal opening to the cloacal membrane (Cheng et al., 2008; Kimmel et al., 2000; Mo et al., 2001) Removal of one or both copies of Gli3 in a Gli2 mutant background does not increase the severity of the septation defect but leads, ultimately, to a near recapitulation of the Shh / condition. This suggests that Gli2 exhibits differential specificity to migration of the anorectal opening, although the mechanism behind this has yet to be determined. In ephrin B2 lz /lz mice the anorectal opening fails to migrate from its initial connection with the urogenital sinus, although the bladder and proximal urethra develop normally (Dravis et al., 2004) Ephrins mediate cell cell att raction and their antagonism in the cloacal epithelium might disrupt the communication between hindgut and urogenital sinus in these mutants. One possibility is that as the cloacal epithelium divides in these mutants, an inability of the cells
134 connecting the hindgut to the urogenital sinus to rearrange their cell cell boundaries using ephrin signals results in this connection remaining fixed. This same finding would be extended to the urethral defect in these animals in which the urethral plate fails to s eptate and the urethral plate does not undergo masculinization to form a tube (Dravis et al., 2004) Wnt5a / mice do not undergo proper cloacal septation and fail to properly elongate their gastrointestinal tract (Cervantes et al., 2009) These defects result from a failure of newbor n epithelial cells to properly intercalate into the proliferating gut tube as they proliferate, although they did not investigate this posterior to the developing colon (Cervantes et al. 2009). We did not specifically observe epithelial stacking in the cl oacal epithelium, and we observed proliferating cells in the epithelium, although it is possible that their rate of division was altered following loss of Shh. Further studies are needed to accurately quantify the rate of proliferation in the cloacal epith elium which is complicated by possible regulation of ventral and dorsal compartments through separate mechanisms. Observations of anorectal malformations in mice, rats and pigs suggests that persistence of the tailgut epithelium and its connection with th e dorsal cloaca can contribute to accumulation of mesoderm at the base of the tail and advancement of the hindgut opening (Qi et al., 2002) I observed persistence of the tailgut at E12.5 following loss of Shh signaling after it normally degenerates and subs equent accumulation of mesoderm at this position, which raises the possibility that Shh signaling to the tailgut mesenchyme activates a signal necessary for induction of apoptosis of the tailgut (Figure 6 6). These findings suggest a complex interplay bet ween mesenchymal proliferation and epithelial morphogenesis and suggest that both processes may contribute to the observed defects in Shh creERT2/C embryos. These findings suggest that disruption at the level of Shh signaling leads to defects of the entire anogenital system, and
135 in light of the findings from Gli2 p63 and ephrin null mice, suggests these processes can be uncoupled in the ventral and dorsal cloacal epithelium. Loss of Shh Pathway Activation in the Genital Ectoderm Results in Hypospadias I n addition to the combined defects observed by disrupting Shh signaling during the perineal period, my results show a later requirement for Shh during genital tubercle development. These results demonstrate a requirement for Shh mediated signaling to the genital ectoderm during the preputial period. When I removed Shh signaling from the urethral epithelium in Shh creERT2/C embryos between E13.25 and E15.25 the result was a severe hypospadias and exposed urethra. Sections of these specimens revealed a thin ning of the ectoderm and loss of continuity across the ventral midline. Loss of Hh responsiveness in the ectoderm of Msx2cre;Smo C/C mice resulted in a phenocopy of the Shh creERT2/C condition. These findings suggest that once the perineal period has ended Shh is necessary for maintenance of ectodermal integrity across the ventral midline. Supporting these findings, Lin et al. (2008) arrived at a similar conclusion following removal of catenin from the genital ectoderm and invoked loss of tensile streng th through reduced cell adhesion forces in the ectoderm. Androgen mediated masculinization of the external genitalia begins at approximately E16.0 as mesenchyme beneath the urethral plate begins to invade the preputial mesenchyme (Chapter 2). Treatment of pregnant females with anti androgens feminizes the urethra and external genitalia without disrupting the ventral ectoderm suggesting that androgen mediated transformation of the genitalia is independent of the Shh pathway (Chapter 2). As the urethral plate becomes canalized within the glans it is unclear whether this deep stromal mesenchyme alone or morphogenesis of the urethral plate itself contributes to urethragenesis. Examination of this process in wildtype embryos showed that neither an epithelia l to mesenchymal transition or apoptosis mediated this transformation (Chapter 2). Interestingly, the process of urethragenesis,
136 when viewed in comparison to cloacal septation, and to the tracheo oesophageal division, is remarkably similar. In all three instances a homogeneous epithelial tube must septate into dorsal and ventral compartments with an intervening mesodermal septum suggesting possible conservation of the underlying mechanism. That both I and others have shown the necessity of Shh for all th ree of these processes further supports this claim (Ioannides et al., 2003) Examination of wildtype embryos with the ectoderm removed suggests that the interface between endoderm and ectoderm is partially held i n place through encasement by the ectoderm. This finding suggests that any disruption of the ectoderm will result in hypospadias. Interestingly, the loss of ectodermal integrity following loss of Shh signaling mimics several other mouse models of hypospa dias including Fgf10 / Fgfr2iiib / Ephrin B2 lz/lz Bmp7 / and Dlx5 / ; /6 / (Dravis et al., 2004; Haraguchi et al., 2000; Petiot et al., 2005; Suzuki et al., 2008) These studies showed that some of these mol ecules are expressed in the epithelium at the junction of the urethral plate and surface ectoderm suggesting an important role in maintaining this connection. The relationship between these molecules and Shh signaling has not been explored although a prel iminary analysis suggests that both Fgfr2iiib and Bmp7 are present in Shh creERT2/C genitalia (data not shown). My results suggest that formation of a centralized urethra is dependent on signaling between all three germ layers, several different molecular pathways and hormonal regulation supporting the complex etiology of hypospadias. This reinforces the idea that embryos exposed to extrinsic environmental factors provide multiple targets for malformation. The Perineal and Preputial Periods as Targets for ARM Understanding the complex etiology of anorectal malformations (ARMs) during normal development requires an understanding of the shared pathways and morphogenetic movements of all three germ layers during anogenital development. There is general agre ement that both
137 genetic and teratogenic factors can contribute to ARMs and the challenge is to dissect and integrate these factors in order to isolate the underlying cause. My findings provide a framework in which to examine how cloacal septation and geni tal outgrowth can be simultaneously or differentially regulated such that non genetic cases can be targeted to a developmental window and/or specific molecular pathway. Interestingly, many of the available animal models of ARM converge on the Shh pathway. The anal atresia in pigs documented by van der Putte (1986) has been traced to a heritable mutation on chromosome 15 and this is in good agreement with the estimated location of Gli2 (Cassini et al., 20 05) Both the Shh and Gli2/3 mutants have been suggested as models of the VACTERL complex and ARMs caused by treatment with either retinoic acid or the vitamin A analogue etrentiate produce ARMs reminiscent of our findings suggesting upstream interaction with the Shh pathway. Further work will be necessary to determine if this is indeed the case and how the 5 Hox genes specifically cause ARMs. Observations made in ephrin B2 p63, Bmp7, and Fgf10 null mutants in the context of our findings demonstrate t hat genetic mutations or environmental agents acting upon these pathways below the level of Shh signaling can differentially affect either the anorectal or urogenital tract. Furthermore, these mice demonstrate how development of the external genitalia and septation of the cloaca can be regulated separately and present interesting cases for future study. Lastly, malformations of the male and female external genitalia present an especially murky etiology as both their late stage development and hormonal r egulation open the possibility to multifactorial causes. Cases presenting as severe with either complete or partial agenesis of the genitalia, as represented by disrupting Shh in the perineal period, or by loss of p63 can be separated from cases of hypos padias and partial feminization which most likely occur as disruptions during the preputial period. Environmental teratogens disrupting the signaling
138 pathways mentioned above would present defects associated with the timing of pathway interaction (perinea l vs. preputial exposure) whereas endocrine disrupting compounds would likely mediate their affects during the preputial period alone during active androgen signaling. These findings reinforce the sensitivity of the anogenital system to both genetic mutat ion and exogenous compounds and underscore the importance of understanding when heritable mutations vs. environmental teratogens are the underlying cause.
139 Figure 6 1. Shh is required until sexual differentiation for anogenital development. (A) T emporally controlled removal of Shh signaling during anogenital development. Shh is required during an early period for both genital tubercle outgrowth and cloacal septation and within a second period for normal urethragenesis. (B C) Defects associated w ith loss of Shh signaling affect males and females equally.
140 Figure 6 2. Shh is required for cloacal septation until E13.5. (A D, A D) Following tamoxifen injections between E9.5 and E12.5 Shh gfpcre ;R26R embryos were harvested at E18.5. (A C) Lo ss of Shh signaling prior to E13.5 results in persistent cloaca. (D) Loss of Shh signaling after E13.5 results in formation of the perineum (PER) and severe proximal hypospadias. All specimens are male. Dorsal swelling (DS), preputial swelling (PS), g lans (G), preputial folds (PF), proximal urethral opening (PUO), urogenital sinus (UGS), urorectal septum (URS) and hindgut (HG).
Figure 6 3. Loss of Shh signaling results in anorectal malformations. (A J) Following injection of tamoxifen embryos were harvested at E14.5, 24 hours after the perineum normally forms. (A D, A D) OPT imaging of the anogenital endoderm in Shh gfpcre ;R26R embryos. Compared to wildtype littermates (D) loss of Shh signaling leads to a stage specific shortening of the hindgu t and urethral plate (A C). (A D) Posterior migration of the hindgut connection with the cloaca is temporally controlled by Shh signaling with a shorter exposure to Shh resulting in anterior displacement of this connection. (E) Loss of Shh signaling at initiation results in caudal truncation, loss of notochordal and floorplate descendant cells, and loss of hindlimb digits. (F G) Progressively longer exposures to Shh result in restoration of caudal axis structures. (H J) Sagital sections of the anog enital system showing progressive movement of urorectal septum mesoderm is dependent on length of exposure to Shh. The length of the urethral plate is also dependent on the timed exposure to Shh. Bladder (BL), genital tubercle (GT), urethral plate (UP), urorectal septum (URS) and hindgut (HG).
143 Figure 6 3. Ectopic cell death following loss of Shh signaling is not widespread in the genital tubercle. (A D) Tamoxifen injection at E9.5 corresponded to loss of Shh signaling at the time of genital ini tiation at E10.5. Embryos were harvested 24 hrs later at E11.5 cell death was measured using Lysotracker Red. (A, C) Control littermates showed cell death in the distal GT mesenchyme and in the urethral plate. (B, D) In Shh creERT2/C embryos, cell deat h was observed in similar domains to control embryos. In addition, two ectopic domains of cell death were observed at the lateral margins of the GT. Compared to control littermates cell death was increased in the tail somites.
144 Figure 6 5. Urorecta l septum mesenchyme proliferates during cloacal septation. BrdU labeled cells (S phase) indicate proliferating cells in the anogenital region at E10.75 and E11.5. Cells of all three germ layers are proliferating during this period. At E10.75 and E11.5 u rorectal septum mesenchyme cells are proliferating, but proportionally the proliferation is lower compared to genital mesenchyme.
Figure 6 6. Loss of Shh signaling from the cloacal endoderm leads to a decrease in proliferative index at the posterior en d of the URSM. (A D) BrdU labeled cells (S phase) in E12.5 harvested Shh +/C (A, C) and Shh creERT2/C (B, D) embryos. White outlines (A, B) indicate URSM. Yellow boxes represent counting frames at three positions, posterior, UGS, and hindgut URSM. (E) Comparing the proliferative index at each of these three positions between Shh +/C (C) and Shh creERT2/C (D) embryos the proliferative index was significantly lower in the posterior URSM while proliferative index did not change significantly between the othe r two positions.
1 47 Figure 6 7. Endodermal progenitor layer remains intact following loss of Shh. p63 staining for the progenitor cell layer in Shh +/C and Shh creERT2/C embryos shows positive staining in the epithelium.
148 Figure 6 8. Persistent Ptc1 expression in URSM following loss of Shh suggests activation by Ihh Ptc1 is expressed in Shh +/C embryos around normal domains of Shh expression. Ihh is highly expressed in the gut endoderm and in the ventral cloaca in these embryos. Following los s of Shh signaling in Shh creERT2/C embryos, Ptc1 expression persists in mesenchyme surrounding the ventral cloaca, but not in mesenchyme surrounding the dorsal cloaca (compare red arrows). This expression is coincident with the persistence of Ihh in Shh cr eERT2/C embryos.
149 Figure 6 9. Loss of Shh signaling from the urethral epithelium prior to sexual differentiation leads to hypospadias. (A B, A B) Loss of Shh signaling between E14.5 and E15.5 in Shh creERT2/C embryos leads to exposure of the urethra proximally, and rupture of the overlying ectoderm. LacZ positive urethral epithelium in Shh creERT2/C embryo (B) injected at E14.5 shows the boundary between endoderm and ectoderm. (C, C) Loss of Shh signaling at E16.5 did not affect ectodermal integr ity. (D E) Detection of Ptc1 transcripts by in situ hybridization at E14.5 and at by lacZ expression at E16.5 in Ptc1lacZ embryos (F H) shows Ptc1 in the mesoderm surrounding the urethral epithelium and in the overlying ectoderm.
150 Figure 6 10. Genit al ectoderm helps maintain ventral connection of urethral epithelium. Red asterisks mark the base of the genital tubercle and distal urethral opening. Following removal of genital ectoderm the connection along the ventral midline between the urethral tub e, preputial folds, and ectoderm is exposed. Black arrows shows points of normally in contact.
151 Figure 6 11. Inability of genital ectoderm to respond to Shh results in loss of ectodermal integrity and hypospadias. (A B) P0 wt;Smo C/C and Msx2cre;S mo C /C embryos. Loss of Smo in the genital ectoderm removes the ability of the ectoderm to transduce Shh signaling and results in hypospadias (B). (C, D) Analysis of cell death at E16.5 in wt;Smo C/C and Msx2cre;Smo C /C embryos does not show increase in ectodermal cell death at the ventral midline but does reveal a failure to maintain ectodermal integrity and exposure of the urethra and glans. Ectopic cell death is visible in the glans of Msx2cre;Smo C /C embryos (D).
152 CHAPTER 7 GENERAL CONCLUSIONS Thi s study and the experiments contained herein have sought to investigate several aspects of anogenital embryology at both a tissue level description of temporal morphogenesis and at the level of molecular control of the underlying developmental mechanisms. Three dimensional Visualization of Anogenital Morphogenesis This study has shown that, contrary to the long standing belief that the distal urethra originates from an ectodermal invagination, it in fact has an endodermal origin. In addition, my work has identified an endodermal origin of the perineum which suggests that this endoderm undergoes a direct transition to epidermis, the first description of its kind. The work in this chapter was published in 2008 and has since helped to revise the embryology of the distal urethra in medical embryology textbooks (Schoenwolf et al., 2008) The fate map I produced helps visualize the process of both cloacal septation and urethral tube formation. Three dimensional visualization of these two processes s hows a striking similarity in their morphogenesis, and in the absence of an epithelial to mesenchymal transition, suggests that both active mesenchymal septation and epithelium rearrangement and reorganization mediate the observed morphogenesis. Control of Initiation and Outgrowth of the External Genitalia What, if any, is the role of Fgf8 during genital tubercle development? In Chapter 3 I demonstrated through a series of experiments that Fgf8 is not required for development of the external genitalia. Genetic removal of Fg8 function in the cloacal endoderm showed that outgrowth proceeded normally and that several genes previously identified as Fgf8 targets do not require Fgf8 for their expression in the genital tubercle. Instead, analysis of Wnt5a / mice suggested that Fgf8 might serve as an accurate readout of genital tubercle induction and switch on in the ventral cloacal endoderm in response to an initiation signal from the cloacal ectoderm.
153 Recent data from Lin et al. (2008) suggests that the ini tiation signal may activate catenin, which in turn was shown to maintain Shh mediated Fgf8 expression. Up regulation of catenin in urethral epithelium greatly enhances Fgf8 expression, and this would support this hypothesis. Localization of Wnt famil y members to the cloacal ectoderm at E10.5 and subsequent conditional removal from this tissue may identify a candidate initiation signal. In addition to excluding Fgf8 as the outgrowth signal, in Chapter 4, I showed that alternate Fgf ligands are not e xpressed in the Fgf8 domain, arguing against functional compensation by other Fgfs following loss of Fgf8 function in the genitalia. This screen did identify several new expression domains for Fgfs 2, 5 9, and 16 which localized, in part, to the dorsal s welling. The roles of these genes in genital tubercle development, if any, is most likely complimentary as our survey of genital phenotypes for Fgf and FgfR knockouts shows that none are required in isolation, except Fgf9 for proper genital development. My findings that proposed targets of Fgf8 signaling ( Dusp6, Sprouty4, Erm, Pea3 ) do not require Fgf8 for their expression and that they are expressed broadly in the genitalia suggests a unique situation in comparison to other tissues. Either these molecu les are direct readouts of Fgf signaling in some tissues, as has been demonstrated for Dusp6, or these genes are specific at the level of ERK activation and in those other tissues examined ERK is only activated by Fgf signaling (Kawakami et al., 2003; Mariani et al., 2008) The fact that ERK activation can be targeted through both extracellular and intracellular crosstalk suggests that during genital evolution selection may have acted at the level of ERK activation and the appearance of molecules that can stimulate ERK is controlled below the level of extracellular signals. The overlapping expression domains of several Bmps, and Tgf in the ventral cloacal epithelium reflect the possibility that multiple pathways m ay compensate at the level of ERK activation in the distal genital
154 mesoderm. My phylogenetic analysis showed that Fgf8 expression is not present in the genitalia of non mammalian amniotes, yet these animals develop external genitalia with several other co nserved signaling molecules present (unpublished observations). It remains to be determined why Fgf8 expression in the distal genitalia appears within the mammalian lineage. While acquisition of this novel expression domain is coincident with the develop ment of a urethral tube, more experiments are necessary to test the functional signifance of this correlation. Analysis of ERK activation in non mammalian amniotes and molecular characterization of the distal urethral epithelium will help address the impo rtance of signaling from the distal endoderm. Additionally, experiments that specifically antagonize ERK or impair its phosphorylation sites in genital mesoderm will help test the requirement for its action during genital outgrowth as will removal of mult iple combinations of these signaling molecules from the urethral epithelium. Shh Controls Various Aspects of Anogenital Development Although it was previously known that Shh was necessary for both cloacal septation and development of the external genita lia, the early and severe phenotype of Shh / mice precluded further investigation of Shh mediated regulation of these processes. The findings presented in Chapters 5 and 6 have shown a stage dependent requirement for Shh signaling during anogenital devel opment. They also reveal that during anogenital development there are two unique developmental windows that I refer to as the perineal period and the preputial period. The perineal period encompasses morphogenesis of the cloaca into separate genitourinar y and anorectal openings and the concomitant formation of the genital tubercle, whereas the preputial period targets development of the external genitalia and urethragenesis. The identification of these periods provides a framework to interpret malformati ons of the anogenital system as it relates to both human ARMs and genetic knockout models in mice. There is no question that the genetic program regulating anogenital development is both complex and regionally
155 compartmentalized. More work is necessary to establish the associated regulatory hierarchy of signaling pathways and transcription factors involved. Slowly, a picture is emerging of this network and the findings here have placed the Hh signaling pathways as one of several master regulators of anoge nital development, along with catenin in the cloacal endoderm, and the 5 Hox genes, Hoxa13 and Hoxd13 throughout the anogenital field (Lin et al., 2008; Warot et al., 1997) Detailed compartment based analyses o f other genes expressed in the anogenital system during both time windows will be required to further establish interactions in this gene regulatory network. As a master regulator of anogenital development, it appears that Shh regulates both mesenchymal proliferation and at least some aspect of epithelial morphogenesis. The finding that Shh regulates the rate of mesenchymal proliferation by partially controlling progression through G1 provides a mechanism for organ size regulation during development. Th is has obvious affects on genital size, and cloacal septation, and although apparently not responsible for establishing local pattern, Shh signaling is absolutely required to expand local progenitor pools. It is possible that transcription factors, such H oxa13 and Hoxd13, and signaling pathways, such the Wnt catenin pathway also interact directly or indirectly with the cell cycle control machinery and provide varying levels of proliferative regulation during development. This provides an interesting de velopmental mechanism with both additive and subtractive properties at the level of the cell cycle and warrants further investigation in both the anogenital system and other organ systems. Analysis of Shh conditional knockout mice showed that epithelial organization was also dependent on Shh, with both a failure of the hindgut to migrate posteriorly, and the urethral plate to lengthen. Epithelial proliferation appeared normal, as did the maintenance of the progenitor
156 cell layer in the cloacal endoderm fo llowing loss of Shh signaling. Observations of mice with Smo removed from the endoderm, which precludes a cell autonomous response of Shh expressing cells to the Shh signal undergo normal development of the anogenital system suggesting any epithelial alt erations are mediated through secondary signals from the mesoderm (K. Choi, pers. comm). A comparative analysis of tracheo oesophageal septation, cloacal septation, and urethral plate septation suggests a similar mechanism of morphogenesis involving the e pithelium (unpublished observations). Detailed experiments comparing these three systems at the site of epithelial septation may shed light on the dynamics of cellular rearrangement either as a function of cyctoskeletal rearrangement or planar cell polari ty. These findings also underscore the importance of a multifaceted approach to understanding anogenital development. In particular, an droge n-mediated signaling is required to masculinize the urethral plate and antagonism of an droge n signaling leads to a feminization of the genitalia. However, disruption of Shh signaling results in a more severe malformation than simple feminzation suggesting a complex interaction with the genetic machinery and endocrine signaling. The role of Shh signaling during sexual differenti aton awaits further investigation, although preliminary data suggests that it is dispensible after the onset of this differentiation. There is a pressing need to identify direct genetic targets of both androgenic and estrogenic compounds in the genit al tubercle in order to dissect the genetic program underlying sexual differentiation. This will facilitate a more complete understanding of the contributing factors underlying malformations of the external genitalia and will undoubtedly shed light on normal developmental processes such as urethragenesis. The findings presented in this thesis provide a framework in which to better understand both normal developmental mechanisms of anogenital development and an orectal malformations
157 (ARM) in humans. The mouse appears to be an accurate model for human anogenital development prior to sexual differentiation, and even aspects of urethagenesis appear similar. The identification of two developmental periods controlled by Shh signaling, and development of a temporally controlled mouse model of anogenital development provides a platform to investigate gene function downstream of Shh in this system.
158 LIST OF REFERENCES Agras, K., Willingham, E., Liu, B. and Baskin, L. S. (2006). Ontogeny of androgen receptor and disruption of its mRNA expression by exogenous estrogens during morphogenesis of the genital tubercle. J Urol 176, 18838. Ahlgren, S. C. and Bronner-Fraser, M. (1999). Inhibition of sonic hedgehog signaling in vivo results in craniofacial neural crest cell death. Curr Biol 9 130414. Arman, E., Haffner-Krausz, R., Gorivodsky, M. and Lonai, P. (1999). Fgfr2 is required for limb outgrowth and lung-branching morphogenesis. Proc Natl Acad Sci U S A 96, 118959. Arsic, D., Qi, B. Q. and Beasley, S. W. (2002). Hedgehog in the human: a possible explanation for the VATER association. J Paediatr Child Health 38, 11721. Asby, D. J., Arlt, W. and Hanley, N. A. (2008). The adrenal cortex and sexual differentiation during early human development. Rev Endocr Metab Disord Austin, C. P., Battey, J. F., Bradley, A., Bucan, M., Capecchi, M., Collins, F. S., Dove, W. F., Duyk, G., Dymecki, S., Eppig, J. T. et al. (2004). The knockout mouse project. Nat Genet 36, 921-4. Baker, N. E. (2007). Patterning signals and proliferation in Drosophila imaginal discs. Curr Opin Genet Dev 17, 28793. Baskin, L. S., Erol, A., Jegatheesan, P., Li, Y., Liu, W. and Cunha, G. R. (2001). Urethral seam formation and hypospadias. Cell Tissue Res 305, 37987. Beck, R. M., Bininda-Emonds, O. R., Cardillo, M., Liu, F. G. and Purvis, A. (2006). A higher-level MRP supertree of placental mammals. BMC Evol Biol 6 93. Bitgood, M. J. and McMahon, A. P. (1995). Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol 172, 12638. Boulet, A. M., Moon, A. M., Arenkiel, B. R. and Capecchi, M. R. (2004). The roles of Fgf4 and Fgf8 in limb bud initiation and outgrowth. Dev Biol 273, 36172. Burns, R. C., Fairbanks, T. J., Sala, F., De Langhe, S., Mailleux, A., Thiery, J. P., Dickson, C., Itoh, N., Warburton, D., Anderson, K. D. et al. (2004). Requirement for fibroblast growth factor 10 or fibroblast growth factor receptor 2-IIIb signaling for cecal development in mouse. Dev Biol 265, 6174. Cameron, H. A. and McKay, R. D. (2001). Adult neurogenesis produces a large pool of new granule cells in the dentate gyrus. J Comp Neurol 435, 40617. Cassini, P., Montironi, A., Botti, S., Hori, T., Okhawa, H., Stella, A., Andersson, L. and Giuffra, E. (2005). Genetic analysis of anal atresia in pigs: evidence for segregation at two main loci. Mamm Genome 16, 16470.
159 Cervantes, S., Yamaguchi, T. P. and Hebrok, M. (2009). Wnt5a is essential for intestinal elongation in mice. Dev Biol 326, 28594. Cheng, W., Jacobs, W. B., Zhang, J. J., Moro, A., Park, J. H., Kushida, M., Qiu, W., Mills, A. A. and Kim, P. C. (2006). DeltaNp63 plays an anti-apoptotic role in ventral bladder development. Development 133, 478392. Cheng, W., Yeung, C. K., Ng, Y. K., Zhang, J. R., Hui, C. C. and Kim, P. C. (2008). Sonic Hedgehog mediator Gli2 regulates bladder mesenchymal patterning. J Urol 180, 154350. Chiang, C., Litingtung, Y., Lee, E., Young, K. E., Corden, J. L., Westphal, H. and Beachy, P. A. (1996). Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature 383, 40713. Colvin, J. S., Green, R. P., Schmahl, J., Capel, B. and Ornitz, D. M. (2001). Maleto -female sex reversal in mice lacking fibroblast growth factor 9. Cell 104, 87589. Connolly, P. B. and Resko, J. A. (1989). Role of steroid 5 alpha-reductase activity in sexual differentiation of the guinea pig. Neuroendocrinology 49, 32430. Coumoul, X. and Deng, C. X. (2003). Roles of FGF receptors in mammalian development and congenital diseases. Birth Defects Res C Embryo Today 69, 286304. Crossley, P. H. and Martin, G. R. (1995). The mouse Fgf8 gene encodes a family of polypeptides and is expressed in regions that direct outgrowth and patterning in the developing embryo. Development 121, 43951. Dassule, H. R., Lewis, P., Bei, M., Maas, R. and McMahon, A. P. (2000). Sonic hedgehog regulates growth and morphogenesis of the tooth. Development 127, 4775-85. Dawrant, M. J., Giles, S., Bannigan, J. and Puri, P. (2008). Adriamycin produces a reproducible teratogenic model of gastrointestinal atresia in the mouse. Pediatr Surg Int 24, 7315. Deng, C. X., Wynshaw-Boris, A., Shen, M. M., Daugherty, C., Ornitz, D. M. and Leder, P. (1994). Murine FGFR-1 is required for early postimplantation growth and axial organization. Genes Dev 8 304557. Derynck, R. and Zhang, Y. E. (2003). Smad-dependent and Smad-independent pathways in TGF-beta family signalling. Nature 425, 57784. Dolle, P., Izpisua-Belmonte, J. C., Brown, J. M., Tickle, C. and Duboule, D. (1991). HOX4 genes and the morphogenesis of mammalian genitalia. Genes Dev 5 17677. Dravis, C., Yokoyama, N., Chumley, M. J., Cowan, C. A., Silvany, R. E., Shay, J., Baker, L. A. and Henkemeyer, M. (2004). Bidirectional signaling mediated by ephrin-B2 and EphB2 controls urorectal development. Dev Biol 271, 27290.
160 Duman-Scheel, M., Weng, L., Xin, S. and Du, W. (2002). Hedgehog regulates cell growth and proliferation by inducing Cyclin D and Cyclin E. Nature 417, 299304. Echelard, Y., Epstein, D. J., St-Jacques, B., Shen, L., Mohler, J., McMahon, J. A. and McMahon, A. P. (1993). Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 75, 141730. Eswarakumar, V. P., Monsonego-Ornan, E., Pines, M., Antonopoulou, I., Morriss-Kay, G. M. and Lonai, P. (2002). The IIIc alternative of Fgfr2 is a positive regulator of bone formation. Development 129 378393. Fairbanks, T. J., De Langhe, S., Sala, F. G., Warburton, D., Anderson, K. D., Bellusci, S. and Burns, R. C. (2004). Fibroblast growth factor 10 (Fgf10) invalidation results in anorectal malformation in mice. J Pediatr Surg 39, 360-5; discussion 3605. Fairbanks, T. J., Sala, F. G., Kanard, R., Curtis, J. L., Del Moral, P. M., De Langhe, S., Warburton, D., Anderson, K. D., Bellusci, S. and Burns, R. C. (2006). The fibroblast growth factor pathway serves a regulatory role in proliferation and apoptosis in the pathogenesis of intestinal atresia. J Pediatr Surg 41, 132-6; discussion 1326. Feldman, B., Poueymirou, W., Papaioannou, V. E., DeChiara, T. M. and Goldfarb, M. (1995). Requirement of FGF-4 for postimplantation mouse development. Science 267, 2469. Felix, W. (1912). The development of the urogenital organs. In Manual of Human Embryology (ed. F. P. Mall), pp. 752-973. Philadelphia: J. B. Lippencott. Fiore, F., Planche, J., Gibier, P., Sebille, A., deLapeyriere, O. and Birnbaum, D. (1997). Apparent normal phenotype of Fgf6-/mice. Int J Dev Biol 41, 63942. Freitas, R., Zhang, G. and Cohn, M. J. (2006). Evidence that mechanisms of fin development evolved in the midline of early vertebrates. Nature 442, 10337. Gehring, U. and Tomkins, G. M. (1974). Characterization of a hormone receptor defect in th e androgen-insensitivity mutant. Cell 3 5964. Glenister, T. W. (1954). The origin and fate of the urethral plate in man. J Anat 88, 41325. Glenister, T. W. (1956). The development of the penile urethra in the pig. J Anat 90, 46177. Gofflot, F., Hall, M. and Morriss-Kay, G. M. (1997). Genetic patterning of the developing mouse tail at the time of posterior neuropore closure. Dev Dyn 210, 43145. Goldstein, J. L. and Wilson, J. D. (1975). Genetic and hormonal control of male sexual differentiation. J Cell Physiol 85, 36577. Goodrich, L. V., Milenkovic, L., Higgins, K. M. and Scott, M. P. (1997). Altered neural cell fates and medulloblastoma in mouse patched mutants. Science 277, 1109-13.
161 Grijelmo, C., Rodrigue, C., Svrcek, M., Bruyneel, E., Hendrix, A., de Wever, O. and Gespach, C. (2007). Proinvasive activity of BMP-7 through SMAD4/src-independent and ERK/Rac/JNK-dependent signaling pathways in colon cancer cells. Cell Signal 19, 172232. Gundersen, H. J., Bagger, P., Bendtsen, T. F., Evans, S. M., Korbo, L., Marcussen, N., Moller, A., Nielsen, K., Nyengaard, J. R., Pakkenberg, B. et al. (1988). The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. Apmis 96, 85781. Guo, L., Degenstein, L. and Fuchs, E. (1996). Keratinocyte growth factor is required for hair development but not for wound healing. Genes Dev 10, 16575. Haraguchi, R., Mo, R., Hui, C., Motoyama, J., Makino, S., Shiroishi, T., Gaffield, W. and Yamada, G. (2001). Unique functions of Sonic hedgehog signaling during external genitalia development. Development 128, 424150. Haraguchi, R., Motoyama, J., Sasaki, H., Satoh, Y., Miyagawa, S., Nakagata, N., Moon, A. and Yamada, G. (2007). Molecular analysis of coordinated bladder and urogenital organ formation by Hedgehog signaling. Development 134, 52533. Haraguchi, R., Suzuki, K., Murakami, R., Sakai, M., Kamikawa, M., Kengaku, M., Sekine, K., Kawano, H., Kato, S., Ueno, N. et al. (2000). Molecular analysis of external genitalia formation: the role of fibroblast growth factor (Fgf) genes during genital tubercle formation. Development 127 24719. Harfe, B. D., Scherz, P. J., Nissim, S., Tian, H., McMahon, A. P. and Tabin, C. J. (2004). Evidence for an expansion-based temporal Shh gradient in specifying vertebrate digit identities. Cell 118, 51728. Hebert, J. M., Rosenquist, T., Gotz, J. and Martin, G. R. (1994). FGF5 as a regulator of the hair growth cycle: evidence from targeted and spontaneous mutations. Cell 78 101725. Hedges, S. B. and Poling, L. L. (1999). A molecular phylogeny of reptiles. Science 283 9981001. Hendzel, M. J., Wei, Y., Mancini, M. A., Van Hooser, A., Ranalli, T., Brinkley, B. R., Bazett-Jones, D. P. and Allis, C. D. (1997). Mitosis-specific phosphorylation of histone H3 initiates primarily within pericentromeric heterochromatin during G2 and spreads in an ordered fashion coincident with mitotic chromosome condensation. Chromosoma 106 34860. Hynes, P. J. and Fraher, J. P. (2004a). The development of the male genitourinary system: II. The origin and formation of the urethral plate. Br J Plast Surg 57, 11221. Hynes, P. J. and Fraher, J. P. (2004b). The development of the male genitourinary system: III. The formation of the spongiose and glandar urethra. Br J Plast Surg 57, 20314. Hynes, P. J. and Fraher, J. P. (2004c). The development of the male genitourinary system. I. The origin of the urorectal septum and the formation of the perineum. Br J Plast Surg 57, 2736.
162 Ince, T. A., Cviko, A. P., Quade, B. J., Yang, A., McKeon, F. D., Mutter, G. L. and Crum, C. P. (2002). p63 Coordinates anogenital modeling and epithelial cell differentiation in the developing female urogenital tract. Am J Pathol 161, 11117. Ingham, P. W. and Placzek, M. (2006). Orchestrating ontogenesis: variations on a theme by sonic hedgehog. Nat Rev Genet 7 84150. Ioannides, A. S., Henderson, D. J., Spitz, L. and Copp, A. J. (2003). Role of Sonic hedgehog in the development of the trachea and oesophagus. J Pediatr Surg 38, 29-36; discussion 2936. Itoh, N. and Ornitz, D. M. (2008). Functional evolutionary history of the mouse Fgf gene family. Dev Dyn 237, 1827. Jin, E. J., Lee, S. Y., Choi, Y. A., Jung, J. C., Bang, O. S. and Kang, S. S. (2006). BMP-2enhanced chondrogenesis involves p38 MAPK-mediated down-regulation of Wnt-7a pathway. Mol Cells 22, 3539. Kang, S., Graham, J. M., Jr., Olney, A. H. and Biesecker, L. G. (1997). GLI3 frameshift mutations cause autosomal dominant Pallister-Hall syndrome. Nat Genet 15, 2668. Kawakami, Y., Rodriguez-Leon, J., Koth, C. M., Buscher, D., Itoh, T., Raya, A., Ng, J. K., Esteban, C. R., Takahashi, S., Henrique, D. et al. (2003). MKP3 mediates the cellular response to FGF8 signalling in the vertebrate limb. Nat Cell Biol 5 5139. Kim, J., Kim, P. and Hui, C. C. (2001). The VACTERL association: lessons from the Sonic hedgehog pathway. Clin Genet 59, 30615. Kim, K. S., Torres, C. R., Jr., Yucel, S., Raimondo, K., Cunha, G. R. and Baskin, L. S. (2004). Induction of hypospadias in a murine model by maternal exposure to synthetic estrogens. Environ Res 94, 26775. Kimmel, S. G., Mo, R., Hui, C. C. and Kim, P. C. (2000). New mouse models of congenital anorectal malformations. J Pediatr Surg 35, 227-30; discussion 2301. Kluth, D., Hillen, M. and Lambrecht, W. (1995). The principles of normal and abnormal hindgut development. J Pediatr Surg 30, 11437. Kohlhase, J., Wischermann, A., Reichenbach, H., Froster, U. and Engel, W. (1998). Mutations in the SALL1 putative transcription factor gene cause Townes-Brocks syndrome. Nat Genet 18, 813. Komada, M., Saitsu, H., Kinboshi, M., Miura, T., Shiota, K. and Ishibashi, M. (2008). Hedgehog signaling is involved in development of the neocortex. Development 135, 271727. Kondo, T., Dolle, P., Zakany, J. and Duboule, D. (1996). Function of posterior HoxD genes in the morphogenesis of the anal sphincter. Development 122, 26519.
163 Kubota, Y., Shimotake, T., Yanagihara, J. and Iwai, N. (1998). Development of anorectal malformations using etretinate. J Pediatr Surg 33, 1279. Kurzrock, E. A., Baskin, L. S. and Cunha, G. R. (1999). Ontogeny of the male urethra: theory of endodermal differentiation. Differentiation 64, 11522. Larson, W. J. (2001). Human Embryology, 3rd edition. New York: Churchill Livingstone. Lee, H. S., Bong, Y. S., Moore, K. B., Soria, K., Moody, S. A. and Daar, I. O. (2006). Dishevelled mediates ephrinB1 signalling in the eye field through the planar cell polarity pathway. Nat Cell Biol 8 5563. Lehoczky, J. A., Williams, M. E. and Innis, J. W. (2004). Conserved expression domains for genes upstream and within the HoxA and HoxD clusters suggests a long-range enhancer existed before cluster duplication. Evol Dev 6 42330. Lewandoski, M., Sun, X. and Martin, G. R. (2000). Fgf8 signalling from the AER is essential for normal limb development. Nat Genet 26, 4603. Lin, C., Yin, Y., Long, F. and Ma, L. (2008). Tissue-specific requirements of beta-catenin in external genitalia development. Development 135, 281525. Lipinski, R. J., Gipp, J. J., Zhang, J., Doles, J. D. and Bushman, W. (2006). Unique and complimentary activities of the Gli transcription factors in Hedgehog signaling. Exp Cell Res 312, 192538. Liu, Y. H., Ma, L., Wu, L. Y., Luo, W., Kundu, R., Sangiorgi, F., Snead, M. L. and Maxson, R. (1994). Regulation of the Msx2 homeobox gene during mouse embryogenesis: a transgene with 439 bp of 5' flanking sequence is expressed exclusively in the apical ectodermal ridge of the developing limb. Mech Dev 48, 18797. Long, F., Zhang, X. M., Karp, S., Yang, Y. and McMahon, A. P. (2001). Genetic manipulation of hedgehog signaling in the endochondral skeleton reveals a direct role in the regulation of chondrocyte proliferation. Development 128, 5099108. Lyon, M. F. and Hawkes, S. G. (1970). X-linked gene for testicular feminization in the mouse. Nature 227, 12179. Mahendroo, M. S., Cala, K. M., Hess, D. L. and Russell, D. W. (2001). Unexpected virilization in male mice lacking steroid 5 alpha-reductase enzymes. Endocrinology 142 465262. Mahlapuu, M., Enerback, S. and Carlsson, P. (2001). Haploinsufficiency of the forkhead gene Foxf1, a target for sonic hedgehog signaling, causes lung and foregut malformations. Development 128 2397406. Mariani, F. V., Ahn, C. P. and Martin, G. R. (2008). Genetic evidence that FGFs have an instructive role in limb proximal-distal patterning. Nature 453, 4015.
164 Mariani, F. V. and Martin, G. R. (2003). Deciphering skeletal patterning: clues from the limb. Nature 423, 31925. Martynoga, B., Morrison, H., Price, D. J. and Mason, J. O. (2005). Foxg1 is required for specification of ventral telencephalon and region-specific regulation of dorsal telencephalic precursor proliferation and apoptosis. Dev Biol 283, 11327. Miller, D. L., Ortega, S., Bashayan, O., Basch, R. and Basilico, C. (2000). Compensation by fibroblast growth factor 1 (FGF1) does not account for the mild phenotypic defects observed in FGF2 null mice. Mol Cell Biol 20, 22608. Min, H., Danilenko, D. M., Scully, S. A., Bolon, B., Ring, B. D., Tarpley, J. E., DeRose, M. and Simonet, W. S. (1998). Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless. Genes Dev 12, 315661. Minowada, G., Jarvis, L. A., Chi, C. L., Neubuser, A., Sun, X., Hacohen, N., Krasnow, M. A. and Martin, G. R. (1999). Vertebrate Sprouty genes are induced by FGF signaling and can cause chondrodysplasia when overexpressed. Development 126, 446575. Mo, R., Kim, J. H., Zhang, J., Chiang, C., Hui, C. C. and Kim, P. C. (2001). Anorectal malformations caused by defects in sonic hedgehog signaling. Am J Pathol 159, 76574. Mohammadi, M., McMahon, G., Sun, L., Tang, C., Hirth, P., Yeh, B. K., Hubbard, S. R. and Schlessinger, J. (1997). Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. Science 276, 95560. Moon, A. M., Boulet, A. M. and Capecchi, M. R. (2000). Normal limb development in conditional mutants of Fgf4. Development 127, 98996. Moore, K. L. (2007). The Developing Human: Clinical Oriented Embryology. Philadelphia: W. B. Saunders. Morgan, E. A., Nguyen, S. B., Scott, V. and Stadler, H. S. (2003). Loss of Bmp7 and Fgf8 signaling in Hoxa13-mutant mice causes hypospadia. Development 130, 3095109. Mortlock, D. P. and Innis, J. W. (1997). Mutation of HOXA13 in hand-foot-genital syndrome. Nat Genet 15, 17980. Murakami, R. and Mizuno, T. (1986). Proximal-distal sequence of development of the skeletal tissues in the penis of rat and the inductive effect of epithelium. J Embryol Exp Morphol 92, 13343. Musgrove, E. A. (2006). Cyclins: roles in mitogenic signaling and oncogenic transformation. Growth Factors 24, 139. Ng, J. K., Tamura, K., Buscher, D. and Izpisua-Belmonte, J. C. (1999). Molecular and cellular basis of pattern formation during vertebrate limb development. Curr Top Dev Biol 41, 3766.
165 Nievelstein, R. A., van der Werff, J. F., Verbeek, F. J., Valk, J. and Vermeij-Keers, C. (1998). Normal and abnormal embryonic development of the anorectum in human embryos. Teratology 57, 708. Nowakowski, R. S., Lewin, S. B. and Miller, M. W. (1989). Bromodeoxyuridine immunohistochemical determination of the lengths of the cell cycle and the DNA-synthetic phase for an anatomically defined population. J Neurocytol 18, 3118. Ogino, Y., Suzuki, K., Haraguchi, R., Satoh, Y., Dolle, P. and Yamada, G. (2001). External genitalia formation: role of fibroblast growth factor, retinoic acid signaling, and distal urethral epithelium. Ann N Y Acad Sci 948, 1331. Ormerod, B. K., Lee, T. T. and Galea, L. A. (2004). Estradiol enhances neurogenesis in the dentate gyri of adult male meadow voles by increasing the survival of young granule neurons. Neuroscience 128, 64554. Ornitz, D. M. and Itoh, N. (2001). Fibroblast growth factors. Genome Biol 2 REVIEWS 3005. Padmanabhan, R., Naruse, I. and Shiota, K. (1999). Caudal dysgenesis in staged human embryos: Carnegie stages 16-23. Am J Med Genet 87, 11527. Palmer, T. D., Willhoite, A. R. and Gage, F. H. (2000). Vascular niche for adult hippocampal neurogenesis. J Comp Neurol 425, 47994. Parkin, C. A., Allen, C. E. and Ingham, P. W. (2009). Hedgehog signalling is required for cloacal development in the zebrafish embryo. Int J Dev Biol 53, 4557. Paulozzi, L. J., Erickson, J. D. and Jackson, R. J. (1997). Hypospadias trends in two US surveillance systems. Pediatrics 100, 8314. Penington, E. C. and Hutson, J. M. (2002a). The cloacal plate: the missing link in anorectal and urogenital development. BJU Int 89, 72632. Penington, E. C. and Hutson, J. M. (2002b). The urethral plate--does it grow into the genital tubercle or within it? BJU Int 89, 7339. Perriton, C. L., Powles, N., Chiang, C., Maconochie, M. K. and Cohn, M. J. (2002). Sonic hedgehog signaling from the urethral epithelium controls external genital development. Dev Biol 247, 2646. Petiot, A., Perriton, C. L., Dickson, C. and Cohn, M. J. (2005). Development of the mammalian urethra is controlled by Fgfr2-IIIb. Development 132, 244150. Post, L. C. and Innis, J. W. (1999). Infertility in adult hypodactyly mice is associated with hypoplasia of distal reproductive structures. Biol Reprod 61, 14028.
166 Qi, B. Q., Beasley, S. W. and Frizelle, F. A. (2002). Clarification of the processes that lead to anorectal malformations in the ETU-induced rat model of imperforate anus. J Pediatr Surg 37, 130512. Ramalho-Santos, M., Melton, D. A. and McMahon, A. P. (2000). Hedgehog signals regulate multiple aspects of gastrointestinal development. Development 127, 276372. Retterer, E. (1890). Sur l'origine et l'volution de la rgion ano-gnitale des mammiferes. Journal of Anatomi and Physiologi 26, 126216. Revest, J. M., Spencer-Dene, B., Kerr, K., De Moerlooze, L., Rosewell, I. and Dickson, C. (2001). Fibroblast growth factor receptor 2-IIIb acts upstream of Shh and Fgf4 and is required for limb bud maintenance but not for the induction of Fgf8, Fgf10, Msx1, or Bmp4. Dev Biol 231, 4762. Riddle, R. D., Johnson, R. L., Laufer, E. and Tabin, C. (1993). Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75, 140116. Roberts, D. J., Johnson, R. L., Burke, A. C., Nelson, C. E., Morgan, B. A. and Tabin, C. (1995). Sonic hedgehog is an endodermal signal inducing Bmp-4 and Hox genes during induction and regionalization of the chick hindgut. Development 121, 316374. Ro ehl, H. and Nusslein-Volhard, C. (2001). Zebrafish pea3 and erm are general targets of FGF8 signaling. Curr Biol 11, 5037. Ross, A. J., Ruiz-Perez, V., Wang, Y., Hagan, D. M., Scherer, S., Lynch, S. A., Lindsay, S., Custard, E., Belloni, E., Wilson, D. I. et al. (1998). A homeobox gene, HLXB9, is the major locus for dominantly inherited sacral agenesis. Nat Genet 20, 35861. Rowitch, D. H., B, S. J., Lee, S. M., Flax, J. D., Snyder, E. Y. and McMahon, A. P. (1999). Sonic hedgehog regulates proliferation and inhibits differentiation of CNS precursor cells. J Neurosci 19, 895465. Roy, S. and Ingham, P. W. (2002). Hedgehogs tryst with the cell cycle. J Cell Sci 115, 43937. Rusinov, V., Baev, V., Minkov, I. N. and Tabler, M. (2005). MicroInspector: a web tool for detection of miRNA binding sites in an RNA sequence. Nucleic Acids Res 33, W696700. Sadler, T. W. (2006). Langman's Medical Embryology: Lippincott Williams and Wilkins. Sala, F. G., Curtis, J. L., Veltmaat, J. M., Del Moral, P. M., Le, L. T., Fairbanks, T. J., Warburton, D., Ford, H., Wang, K., Burns, R. C. et al. (2006). Fibroblast growth factor 10 is required for survival and proliferation but not differentiation of intestinal epithelial progenitor cells during murine colon development. Dev Biol 299, 37385. Sasaki, A., Taketomi, T., Kato, R., Saeki, K., Nonami, A., Sasaki, M., Kuriyama, M., Saito, N., Shibuya, M. and Yoshimura, A. (2003). Mammalian Sprouty4 suppresses Ras-independent ERK activation by binding to Raf1. Nat Cell Biol 5 42732.
167 Sasaki, C., Yamaguchi, K. and Akita, K. (2004). Spatiotemporal distribution of apoptosis during normal cloacal development in mice. Anat Rec A Discov Mol Cell Evol Biol 279, 7617. Satoh, Y., Haraguchi, R., Wright, T. J., Mansour, S. L., Partanen, J., Hajihosseini, M. K., Eswarakumar, V. P., Lonai, P. and Yamada, G. (2004). Regulation of external genitalia development by concerted actions of FGF ligands and FGF receptors. Anat Embryol (Berl) 208, 47986. Saunders, J. W., Jr. (1948). The proximo-distal sequence of origin of the parts of the chick wing and the role of the ectoderm. J Exp Zool 108, 363403. Schoenwolf, G. C., Francis-West, P. H., Brauer, P. R. and Bleyl, S. B. (2008). Larsen's Human Embryology 715. Scott, V., Morgan, E. A. and Stadler, H. S. (2005). Genitourinary functions of Hoxa13 and Hoxd13. J Biochem 137, 6716. Shapiro, M. D., Hanken, J. and Rosenthal, N. (2003). Developmental basis of evolutionary digit loss in the Australian lizard Hemiergis. J Exp Zoolog B Mol Dev Evol 297, 4856. Sharrocks, A. D. (2001). The ETS-domain transcription factor family. Nat Rev Mol Cell Biol 2 82737. Shaut, C. A., Saneyoshi, C., Morgan, E. A., Knosp, W. M., Sexton, D. R. and Stadler, H. S. (2007). HOXA13 directly regulates EphA6 and EphA7 expression in the genital tubercle vascular endothelia. Dev Dyn 236, 95160. Shimada, T., Kakitani, M., Yamazaki, Y., Hasegawa, H., Takeuchi, Y., Fujita, T., Fukumoto, S., Tomizuka, K. and Yamashita, T. (2004). Targeted ablation of Fgf23 demonstrates an essential physiological role of FGF23 in phosphate and vitamin D metabolism. J Clin Invest 113, 5618. Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21, 701. Spitz, F., Gonzalez, F. and Duboule, D. (2003). A global control region defines a chromosomal regulatory landscape containing the HoxD cluster. Cell 113, 40517. St -Jacques, B., Dassule, H. R., Karavanova, I., Botchkarev, V. A., Li, J., Danielian, P. S., McMahon, J. A., Lewis, P. M., Paus, R. and McMahon, A. P. (1998). Sonic hedgehog signaling is essential for hair development. Curr Biol 8 105868. Stadler, H. S. (2003). Modelling genitourinary defects in mice: an emerging genetic and developmental system. Nat Rev Genet 4 47882. Summerbell, D. (1974). A quantitative analysis of the effect of excision of the AER from the chick limb-bud. J Embryol Exp Morphol 32, 65160.
168 Sun, X., Lewandoski, M., Meyers, E. N., Liu, Y. H., Maxson, R. E., Jr. and Martin, G. R. (2000). Conditional inactivation of Fgf4 reveals complexity of signalling during limb bud development. Nat Genet 25, 836. Sun, X., Mariani, F. V. and Martin, G. R. (2002). Functions of FGF signalling from the apical ectodermal ridge in limb development. Nature 418, 5018. Sun, X., Meyers, E. N., Lewandoski, M. and Martin, G. R. (1999). Targeted disruption of Fgf8 causes failure of cell migration in the gastrulating mouse embryo. Genes Dev 13, 183446. Suzuki, K., Bachiller, D., Chen, Y. P., Kamikawa, M., Ogi, H., Haraguchi, R., Ogino, Y., Minami, Y., Mishina, Y., Ahn, K. et al. (2003). Regulation of outgrowth and apoptosis for the terminal appendage: external genitalia development by concerted actions of BMP signaling [corrected]. Development 130, 620920. Suzuki, K., Haraguchi, R., Ogata, T., Barbieri, O., Alegria, O., Vieux-Ro chas, M., Nakagata, N., Ito, M., Mills, A. A., Kurita, T. et al. (2008). Abnormal urethra formation in mouse models of split-hand/split-foot malformation type 1 and type 4. Eur J Hum Genet 16, 3644. Taniguchi, K., Ayada, T., Ichiyama, K., Kohno, R., Yonemitsu, Y., Minami, Y., Kikuchi, A., Maehara, Y. and Yoshimura, A. (2007). Sprouty2 and Sprouty4 are essential for embryonic morphogenesis and regulation of FGF signaling. Biochem Biophys Res Commun 352, 896902. Thewissen, J. G., Cohn, M. J., Stevens, L. S., Bajpai, S., Heyning, J. and Horton, W. E., Jr. (2006). Developmental basis for hind-limb loss in dolphins and origin of the cetacean bodyplan. Proc Natl Acad Sci U S A 103, 84148. Tourneux, F. (1888). Sur les premiers dveloppements du cloaques du tubercule gnital et de l'anus chez l'embryon de mouton. Journal of Anatomi and Physiologi 24, 503517. Towers, M., Mahood, R., Yin, Y. and Tickle, C. (2008). Integration of growth and specification in chick wing digit-patterning. Nature 452, 8826. van der Putte, S. C. (1986). Normal and abnormal development of the anorectum. J Pediatr Surg 21, 43440. van der Putte, S. C. (2005). The devlopment of the perineum in the human. A comprehensive histological study with a special reference to the role of the stromal components. Adv Anat Embryol Cell Biol 177, 1 131. van der Putte, S. C. (2007). Hypospadias and associated penile anomalies: a histopathological study and a reconstruction of the pathogenesis. J Plast Reconstr Aesthet Surg 60, 4860. van der Putte, S. C. and Neeteson, F. A. (1983). The normal development of the anorectum in the pig. Acta Morphol Neerl Scand 21, 10732.
169 Van der Putte, S. C. and Neeteson, F. A. (1984). The pathogenesis of hereditary congenital malformations of the anorectum in the pig. Acta Morphol Neerl Scand 22 1740. Venters, S. J., Hultner, M. L. and Ordahl, C. P. (2008). Somite cell cycle analysis using somite-staging to measure intrinsic developmental time. Dev Dyn 237, 377-92. Verheyden, J. M., Lewandoski, M., Deng, C., Harfe, B. D. and Sun, X. (2005). Conditional inactivation of Fgfr1 in mouse defines its role in limb bud establishment, outgrowth and digit patterning. Development 132, 423545. Wang, Y., Spatz, M. K., Kannan, K., Hayk, H., Avivi, A., Gorivodsky, M., Pines, M., Yayon, A., Lonai, P. and Givol, D. (1999). A mouse model for achondroplasia produced by targeting fibroblast growth factor receptor 3. Proc Natl Acad Sci U S A 96, 445560. Warot, X., Fromental-Ramain, C., Fraulob, V., Chambon, P. and Dolle, P. (1997). Gene dosage-dependent effects of the Hoxa-13 and Hoxd-13 mutations on morphogenesis of the terminal parts of the digestive and urogenital tracts. Development 124, 478191. Weinstein, M., Xu, X., Ohyama, K. and Deng, C. X. (1998). FGFR-3 and FGFR-4 function cooperatively to direct alveogenesis in the murine lung. Development 125, 361523. Welsh, M., Saunders, P. T. and Sharpe, R. M. (2007). The critical time window for androgendependent development of the Wolffian duct in the rat. Endocrinology White, R. A., Fallon, J. F. and Savage, M. P. (1992). On the measurement of cytokinetics by continuous labeling with bromodeoxyuridine with applications to chick wing buds. Cytometry 13, 5536. Wolf, C. J., Hotchkiss, A., Ostby, J. S., LeBlanc, G. A. and Gray, L. E., Jr. (2002). Effects of prenatal testosterone propionate on the sexual development of male and female rats: a doseresponse study. Toxicol Sci 65, 7186. Xu, J., Liu, Z. and Ornitz, D. M. (2000). Temporal and spatial gradients of Fgf8 and Fgf17 regulate proliferation and differentiation of midline cerebellar structures. Development 127 183343. Xu, X., Weinstein, M., Li, C. and Deng, C. (1999). Fibroblast growth factor receptors (FGFRs) and their roles in limb development. Cell Tissue Res 296, 3343. Yamada, G., Satoh, Y., Baskin, L. S. and Cunha, G. R. (2003). Cellular and molecular mechanisms of development of the external genitalia. Differentiation 71, 44560. Yamada, G., Suzuki, K., Haraguchi, R., Miyagawa, S., Satoh, Y., Kamimura, M., Nakagata, N., Kataoka, H., Kuroiwa, A. and Chen, Y. (2006). Molecular genetic cascades for external genitalia formation: an emerging organogenesis program. Dev Dyn 235, 173852. Yamaguchi, T. P., Bradley, A., McMahon, A. P. and Jones, S. (1999). A Wnt5a pathway underlies outgrowth of multiple structures in the vertebrate embryo. Development 126, 121123.
170 Yauch, R. L., Gould, S. E., Scales, S. J., Tang, T., Tian, H., Ahn, C. P., Marshall, D., Fu, L., Januario, T., Kallop, D. et al. (2008). A paracrine requirement for hedgehog signalling in cancer. Nature 455, 40610. Zhu, J., Nakamura, E., Nguyen, M. T., Bao, X., Akiyama, H. and Mackem, S. (2008). Uncoupling Sonic hedgehog control of pattern and expansion of the developing limb bud. Dev Cell 14, 62432.
171 BIOGRAPHICAL SKETCH The author was born in Syosset, NY, and the cells of his body can be traced back directly through the germ cells of his parents and the germ cells of their parents and so on and so forth. Every cell in his body came from just two cells that can be traced to an ancestor at the roots of multicellular life. These cells function together as an animal, a man, and a scientist, and this same man was educated in public school in Huntington, NY. From there he went on to receive an A.B. in Biology from Bowdoin College, and after some years in the wilderness of life, gained an M.S. in Zoology under Dr. Lauren Chapman at the Univeristy of Florida where he later joined the lab of Dr. Martin J. Cohn to receive his Ph.D. His cells are happy and looking towards the future.