Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The mechanism of the chicken cleft primary palate mutation : subtitle altered Fgf8 signalling from frontonasal… Macdonald, Mary 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-750477.pdf [ 9.8MB ]
Metadata
JSON: 831-1.0090674.json
JSON-LD: 831-1.0090674-ld.json
RDF/XML (Pretty): 831-1.0090674-rdf.xml
RDF/JSON: 831-1.0090674-rdf.json
Turtle: 831-1.0090674-turtle.txt
N-Triples: 831-1.0090674-rdf-ntriples.txt
Original Record: 831-1.0090674-source.json
Full Text
831-1.0090674-fulltext.txt
Citation
831-1.0090674.ris

Full Text

The mechanism of the chicken cleft primary palate mutation: altered Fgf8 signalling from frontonasal mass epithelium redirects proximodistal outgrowth of the upper beak by Mary Elizabeth MacDonald B.Sc, Dalhousie University, 1993 M.Sc., Dalhousie University, 1997 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Oral Health Sciences Faculty of Dentistry We accept this thesis as conforming to the required standard The University of British Columbia © Mary Elizabeth MacDonald, 2002 UBC Rare Books and Special Collections - Thesis Authorisation Form 13/02/1905 17:24 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree tha t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r re f e r e n c e and study. I f u r t h e r agree t h a t p ermission f o r ext e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood tha t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r mission. Department of G>f<hX \A-t(S^^\ S c , [ e n c g S The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada http://www.library.ubc.ca/spcoll/thesauth.html Page 1 of 1 ABSTRACT Facial development occurs through the outgrowth and fusion of neural crest-derived facial prominences that surround the primitive oral cavity. Failure of the prominences to grow together properly can result in a number of facial deformities, of which cleft lip with or without cleft palate is the most common. A mutation known as cleft primary palate (cpp) has been discovered in the chicken that causes complete failure of outgrowth of the upper facial prominences with the consequence that a bilateral cleft lip is formed. A series of grafting experiments showed that the mutation affects the frontonasal mass but not other facial prominences. Recombination experiments showed the growth of cpp mutant frontonasal mass mesenchyme is restored by recombination with wild-type epithelium. In contrast, the growth of normal craniofacial mesenchyme is not supported by cpp mutant frontonasal mass epithelium, implying that the mutation primarily affects the epithelium and not the mesenchyme. The failure of outgrowth is not due to a lack of skeletogenesis, since cartilage forms in all grafts containing cpp mutant tissue. Fgf8, which is usually restricted to the epithelium surrounding the nasal pits, was found throughout the frontonasal mass epithelium of stage 24 and 28 cpp mutant embryos. Expression of other genes localized to the frontonasal mass epithelium (Bmp4, Noggin and Shh) or in the frontonasal mass mesenchyme (Patched-1, Msxl, Msx2, AP2, type II collagen) was the same as in normal embryos. Programmed cell death in the cpp mutant frontonasal mass also did not differ from the normal. Cell proliferation, however, was increased in the mesenchyme underlying the ectopic Fg/8-expressing epithelium, leading to growth along the cranio-caudal rather than the proximo-distal axis. These results indicate that the failure of the cpp frontonasal mass to grow along a proximo-distal axis is due to a signalling defect in the epithelium. The cpp mutation is upstream of Fgf8, and ectopic Fgf8 expression may lead to the observed defect in signalling between the epithelium and the mesenchyme in the cpp frontonasal mass. The failure of outgrowth results in loss of the premaxilla and prenasal cartilage, and a bilateral cleft beak. n Table of Contents Abstract . . . . . . . . . ii Table of Contents . . . . . . . . iii List of Tables . . . . . . . . . v List of Figures . . . . : . . . . vi List of Abbreviations . . . . . . . . vii Acknowledgements . . . . . . . . viii Chapter 1: Introduction . . . . . . . 1 Overview of developmental processes . . . . . 1 Normal face development . . . . . . . 2 Outgrowth in face and limb development . . . . . 4 Significance of the cleft primary palate mutation . . . . 6 Chapter 2: Cleft primary palate mutation affects the epithelium and not the mesenchyme in chicken craniofacial development . . . 8 Results . . . . . . . . . 17 Discussion . . . . . . . . . 40 Chapter 3: Morphology and egg tooth formation in grafts of normal and cleft primary palate frontonasal mass tissue . . . . 4 6 Results . . . . . . . . . 49 Discussion . . . . . . . . . 55 Chapter 4: The cleft primary palate mutation lies upstream of Fgf8 and alters epithelial-mesenchymal signalling in the frontonasal mass . 58 Results . . . . . . . . . 64 Discussion . . . . . . . . . 86 iii Chapter 5: General discussion . . . . . . 96 Fgf8 induces cell proliferation during limb bud development . . 96 Ectopic Fgf8 expression in cpp mutant embryos is correlated with increased cell proliferation . . . . . . . 97 Selective restriction of proliferation may be required to direct outgrowth of the facial prominences . . . . . . . 98 Consequences of Fgf8 overexpression: truncations in the limbs and in the face 100 Downregulation of Fgf8 in the frontonasal mass epithelium during normal development may be due to a signal from underlying mesenchyme . 102 Comparison of the cpp mutation with other models of abnormal outgrowth. 103 Future directions . . . . . . . . 106 Concluding remarks . . . . . . . . 107 Literature Cited . . . . 108 iv List of Tables Table 1: Growth of intact grafts , isolated mesenchyme and recombinations from normal embryos at stages 20, 24 and 28 . . . . 24 Table 2: Growth of grafts and recombinations of frontonasal mass tissue from stage 28 embryos affected by the cpp mutation . . . 27 Table 3: Recombination of frontonasal mass mesenchyme from stage 20 and 24 cpp embryos with normal frontonasal mass epithelium . 31 Table 4: Length, morphology and egg tooth formation in frontonasal mass grafts from normal embryos and recombinations of normal and cpp mutant frontonasal mass tissue. . . . . 53 Table 5: In situ hybridization of face signaling genes in normal and cleft primary palate mutant embryos at stages 20, 24 and 28 . 65 Table 6: Expression of frontonasal mass signalling genes in culture after application of exogenous FGF8, SHH protein or Shh antibody . 75 List of Figures Figure 1: Dissection, grafting and tissue recombination of normal and cpp chicken embryos . . . . . . . 15 Figure 2: Skeletal derivatives of normal and cpp mutant embryos . . 18 Figure 3: Grafts of intact facial prominences from stage 28 normal and cpp mutant embryos . . . . . . . 21 Figure 4: Grafts of recombined mesenchyme and epithelium from stage 24 and 28 normal and cpp mutant embryos . . . 28 Figure 5: Cell death in normal and mutant stage 28 embryos from the cpp strain . . . . . . . . 33 Figure 6: Cell proliferation in the stage 28 normal and cpp mutant embryo 36 Figure 7: Cell proliferation differs in superficial, mid and deep sections through the frontonasal mass of stage 28 normal and mutant embryos of the cpp strain . . . . . . . 38 Figure 8: Morphology and egg tooth formation in grafts of normal and cpp mutant frontonasal mass tissue . . . . 51 Figure 9: Expression of Collagen type II, Msxl and -2, Bmp4 and its antagonist noggin is unaffected in cpp mutant embryos . . 66 Figure 10: Shh and its receptor Ptcl are expressed similarly in normal and cpp mutant embryos . . . . . . 69 Figure 11: Fgf8 expression is extended throughout the cpp mutant frontonasal mass epithelium . . . . . 71 Figure 12: Culture of stage 28 normal and cpp mutant frontonasal mass shows that epithelium is not required to maintain AP2 or Msxl, nor does exogenous SHH protein affect their expression . . 77 Figure 13: Absence of a FGF8-SHH feedback loop in the frontonasal mass 82 Figure 14: Exogenous SHH causes truncation of the premaxillary bone . 84 Figure 15: Molecular signalling and cell proliferation in the stage 28 normal and cpp mutant embryo frontonasal mass . . . 94 List of Abbreviations AER apical ectodermal ridge A-P anterior-posterior AP2 activating protein 2 Bmp bone morphogenetic protein BrdU bromodeoxyuridine BSA bovine serum albumin cpp cleft primary palate D-V dorsal-ventral d dentary bone e eye eg entoglossal f frontal bone fb forebrain Fgf fibroblast growth factor fnm frontonasal mass Hox homeobox ios interorbital septum j jugal lnp lateral nasal process md mandibular mRNA messenger RNA (ribonucleic acid) mx maxillary bone Msx muscle segment homeobox n nasal bone nc nasal conchae P-D proximal-distal pmx premaxilla pnc prenasal cartilage pp palatine process of the maxillary bone pt pterygoid Ptc patched q quadrate bone qj quadratojugal r rhombencephalon RA retinoic acid s stomodeum Shh sonic hedgehog TUNEL TdT-mediated dUTP-digoxygenin nick end labelling wt wild type vii Acknowledgements I must first thank my supervisor, Dr. Joy Richman, for the opportunity to complete an interesting and worthwhile project during my time as her student. Members of my supervisory committee contributed greatly to my understanding of this work and of the process of research. My sincere thanks to Dr. Don Brunette, Dr. Virginia Diewert and Dr. Tim O'Connor. The graduate coordinators have gone beyond the call of duty to smooth the way for all of their students. I am very grateful to Dr. Doug Waterfield and Ms. Viki Beretanos Koulouris for their friendship, interest and timely aid on many occasions. Dr. Joseph Tonzetich was an excellent example to me of a dedicated career scientist, and I am honoured to have been financially supported by the fellowship that bears his name. Amir Ashique and Andre Wong shared their considerable skills very generously, teaching me many techniques and troubleshooting with me whenever necessary. My thanks as well for the technical assistance provided by Les Grad, Steve Ritchie, and Kathy Fu, who grew up some of the constructs and linearized DNA used in making in situ hybridization probes; Kathy Fu, Paul Plut and Kathy Wong, who carried out some of the in situ hybridizations; Mya Truong and Laurene Yen, who helped clear and stain limb graft specimens; and Jeff Chen, who counted many, many cells. Your help was invaluable. Other researchers in the Richman lab, past and present, have provided me with professional support and friendships that have enriched my life both in the lab and out. Thank you Amir, Heather, Jeff, Jenny, Jessica, Kathy, Laurene, Les, Marnie, Masa, Mya, Paul, Sandra, Sang-Hwy, Sonia and Steve. I also owe a debt to my former supervisors, Dr. Ann Graveson and Dr. Brian Hall, without whose instruction I would not have been prepared to undertake this project. My family, though far away, has been close in my thoughts throughout this work. Mom and Rhett, thank you for everything. Vlll Chapter 1: Introduction Overview of developmental processes The evocation of developmental potential occurs through interactions between cell populations with very different origins, characteristics and fates. These interactions, or inductions, are the means by which one cell or population influences the fate of another. Populations of early embryonic cells are organized into three germ layers: the ectoderm, which will eventually give rise to the integument and nervous system; the endoderm, from which the digestive system is derived; and mesoderm, which will form the connective tissue, blood and some organs. In order to form their derivatives, the germ layers undergo differentiation (the process by which cellullar diversity is established), morphogenesis (the formation of structures), and growth. During the earliest stages of development, cells from all germ layers are organized into two basic arrangements that facilitate movement within the embryo and interactions among cells: epithelium, which consists of sheet-like tissues with close connections, and more loosely ordered mesenchyme. As in much of development, complex reciprocal epithelial-mesenchymal inductions are required for differentiation, patterning and morphogenesis of the face. Study of the molecular events that underlie inductions indicates that many signalling networks are conserved, thus processes such as cell proliferation and morphogenesis which occur repeatedly throughout development may be subject to similar molecular patterning mechanisms. The cleft primary palate {cpp) mutation is localized to the face and first becomes apparent soon after many of these inductions are known to take place (Abbott and MacCabe, 1966), suggesting that the mutation affects epithelial-mesenchymal interactions specific to the face. To introduce this thesis I first present an overview of development in the model system, the chicken embryo face, then outline the process of outgrowth in the face and in the limb, a model analagous to the face in which the molecular signalling of outgrowth has been studied in detail. Molecular signals implicated in face patterning are considered, with respect to the gene expression experiments in this work. The significance of the chicken cpp mutation as a model for outgrowth is introduced, however comparisons with other mutations affecting outgrowth and contributions of this research 1 to understanding clinical conditions such as cleft lip will be addressed in the discussion and concluding remarks. Normal face development \, The vertebrate face is one of the most complex model systems in the field of developmental biology. Contributions are made from all germ layers (ectoderm, endodbrm and mesoderm), however the facial skeleton is unique in that it is comprised primarily of ectodermally-derived neural crest, rather than the mesoderm which forms the body skeleton (Le Lievre and Le Douarin, 1975). During formation of the neural tube, a population of cells segregates from the boundary between surface and neural ectoderm (the neural crest). These neural crest cells transform from epithelium to mesenchyme, proliferate and migrate through the mesoderm (Hall, 1988; Johnston, 1966; Noden, 1975). Specifically, cranial (as opposed to trunk) neural crest cells migrate over and around the developing brain into the presumptive face region, undergoing inductive interactions that help to specify their eventual position and fate (Bee and Thorogood, 1980; Hall and Tremaine, 1979; Noden, 1975; Noden, 1978; Noden, 1983b). Once in the face, these cells proliferate and undergo inductive interactions with the overlying facial epithelium (Hall, 1978; Hall, 1999; Noden, 1983a; Tyler MS, 1977). The basic framework of the face consists of prominences, or buds of epithelium-covered mesenchyme surrounding the stomodeal (oral) cavity. In the chicken, these prominences form by approximately the third day post-fertilization (HH stage 18-20; Hamburger and Hamilton, 1951), grow and differentiate over a two-day period (to stage 28), then fuse and grow together to form the recognizable chicken upper and lower beaks (reviewed in Bellairs and Osmond, 1998). Facial prominences in the chicken are (from cranial to caudal) the frontonasal mass (also called the frontonasal process) a single prominence in the centre of the upper face; on either side, lateral to the nasal pits, are the paired lateral nasal prominences. Below these (in the presumptive cheek region) are the maxillary prominences, and finally the lower face (in the chicken, the lower beak) is formed by the paired mandibular arches. Craniofacial development during growth and convergence of these structures is strongly conserved, as analogous prominences are seen in all of the vertebrate classes (DeBeer, 1985; Langille, 1993a). Investigations of this most crucial 2 stage of facial development, which are technically very difficult to perform in placental mammals, may therefore be carried out in other vertebrates. A literature has been built up using two main animal models for face development; the mouse, which is often used in tissue culture and for genetic studies, and the chicken, which is accessible for in vivo experimentation. The intricate patterning required to form the face is subject to numerous stresses, and the many errors possible during craniofacial development make the face the most common site of congenital abnormalities (Schutte and Murray, 1999). Functional anomalies during face development primarily occur during fusion of the facial prominences. In order for fusion to occur the seam, the double layer of epithelium present at the junction where prominences meet, must be broken down. Failure of epithelial cells to undergo programmed cell death (McGonnell et al., 1998), migration (Chai et al., 1997) or to transform into mesenchyme (Boshart et al., 2000; Fitchett and Hay, 1989; Sun et al., 2000a) will cause the seam to persist, leading to cleft lip. Inadequate mesenchymal cell proliferation (Brown et al., 1993; Minkoff, 1980; Peterka and Jelinek, 1983; Tamarin et al., 1984) or growth (Diewert, 1985; Wang and Diewert, 1992) will prevent facial prominences from reaching the correct size to make contact, leading to clefts. Finally, abnormal vasculogenesis (Poswillo, 1975a; Poswillo, 1975b) or skeletogenesis (Fitch, 1957; Hall and Miyake, 1992) can lead to other facial defects, although not to clefts. Both genetic and non-genetic factors can influence any or all of these processes, which have been extensively studied with respect to morphogenesis and fusion of the mammalian secondary palate, failure of which leads to cleft palate (Johnston and Bronsky, 1991; Schutte and Murray, 1999). The facial prominences, in which failure of fusion leads to cleft lip, have been less studied (Boshart et al., 2000; Chai et al., 1997). Studies of normal development are complemented by 'natural experiments' or mutations which can often help to elucidate a specific process in development; the cleft primary palate mutation, in which the frontonasal mass does not grow sufficiently to fuse with other prominences, appears to be a model for cleft lip. In addition, development in other regions of the embryo can parallel facial development. Although they are derived from different germ layers, there are many similarities between early development of the limb bud (formed from paraxial mesoderm and flank ectoderm) and the facial prominences (neural crest-derived mesenchyme and head ectoderm), therefore a brief account of outgrowth in both facial prominences and limb buds follows. Outgrowth in face and limb development Buds of mesenchyme encased in epithelium surround the oral cavity to form the basic pattern of the face, while in a similar process buds from the flank of the embryo form the limbs. In order for either set of buds to continue to develop, they must undergo outgrowth. This process has been studied extensively in the limb. Outgrowth of a limb bud is associated with a high level of cell proliferation at the base, relative to the surrounding non-bud tissue (Searls and Janners, 1971). The epithelium of the bud may act to initiate and is often required to maintain this high level of proliferation (Crackower et al., 1998; Dealy et al., 1996; Dealy et al., 1997; Sherman et al., 1998; Summerbell, 1974). The mechanism of budding is similar in the face, in that levels of cell proliferation in budding mesenchyme are maintained while proliferation drops in adjacent, non-budding regions (Minkoff, 1991; Minkoff and Kuntz, 1977; Minkoff and Kuntz, 1978). Molecular signalling of limb outgrowth Limb outgrowth is initiated by induction of Fgf8 in the presumptive limb ectoderm by FgflO in the paraxial mesoderm (Crossley et al., 1996b; Ohuchi et al., 1997). Underlying mesodermal cells then proliferate at a higher rate than in the rest of the flank, forming a bud (Searls and Janners, 1971), and the Fg/8-expressing limb ectoderm (Crossley and Martin, 1995; Heikenheimo et al., 1994) thickens to form the morphologically and molecularly distinct patterning centre, the AER (apical ectodermal ridge, (Saunders, 1948). Outgrowth of the limb bud occurs through rapid proliferation of mesenchymal cells at the distal tip of the bud, directly under the Fg/8-expressing AER in a region known as the progress zone (Summerbell and Lewis, 1975). Concurrently, establishment of axes within the limb bud permits growth and patterning in specific regions at specific times. The limb has three distinct axes; proximo-distal (i.e., shoulder to digit), antero-posterior (i.e., thumb to little finger) and dorso-ventral (i.e., palmar to dorsum of the hand). Subsequent patterning along the antero-posterior (AP) axis is under 4 the control of Bmps (Francis et al., 1994; Laufer et al., 1994) and Shh (Chiang et al., 1996;-Riddle et al., 1993) in the zone of polarizing activity (Saunders and Gasseling, 1968),; which may in turn be mediated by endogenous retinoids (Johnson and Tabin, 1997; Thaller and Eichele, 1987; Thaller and Eichele, 1988; Wanek et al., 1991). Dorso-ventral (DV) patterning involves many genes including Wnt7a (Parr and McMahon, 1995), Lmxl (Vogel et al., 1995) and Engrailedl (Logan et al., 1997; Loomis et al., 1996; Loomis et al., 1998). Proximodistal (PD) growth proceeds under the control of the AER and involves Fgf8, which is thought to maintain the progress; zone cells in an undifferentiated, proliferative state (Crossley et al., 1996b; Vogel et al., 1996) as well as several transcription factors expressed in the progress zone, including Lhx2 (Rodriguez-Estebah et al., 1998) and Msxl (Niswander and Martin, 1993; Pizette et al., 2001). Specific molecules that set boundaries, lay down the pattern and induce growth and morphogenesis are therefore associated with each axis of the limb bud. Some of the same genes are found in the face and may play similar roles, but their expression does not necessarily fall into the same axial pattern as in the limb. Molecular signalling of face outgrowth Unlike the limb bud, outgrowth in the face is complicated by the requirement for several buds or prominences to grow together and fuse. In addition, the roles of regions of the limb bud such as the AER and ZPA were first identifed at the tissue level, aiding greatly in later examinations of molecular signalling. Tissue-level interactions have been studied in the face with respect to skeletogenesis (Dunlop and Hall, 1995; Hall, 1978; Hall, 1980; Hall, 1981; Tyler, 1978; Tyler and Hall, 1977) and odontogenesis (Kollar and Baird, 1969; Kollar and Fisher, 1980; Mina and Kollar, 1987; Thesleff et al., 1991) but the only studies to examine outgrowth specifically did not find any morphologically distinct structures or ridges (like the limb AER) in facial epithelium that might be responsible for inducing outgrowth (Richman and Tickle, 1989; Richman and Tickle, 1992). Consequently, examination of molecular signalling in facial outgrowth proceeds without some of the morphological signposts that guided studies of outgrowth in the limb. 5 Data from many experimental models demonstrates that different signals direct outgrowth of different regions of the face. Molecules whose activity and expression in the face make them candidates for roles in facial outgrowth include members of the Bmp (bone morphogenetic protein) family, secreted signalling molecules important in early regional specification as well as skeletogenesis (Barlow and Francis-West, 1997; Francis et al., 1994; Mina et al., 2002; Shigetani, 2000). Msxl (muscle segment homeobox-1) and Msx2 are homeodomain-containing transcriptional repressors often associated with areas of high cell proliferation (Brown et al., 1993, 1997; Catron et al., 1995; McGonnell et al., 1998). The secreted signalling molecule Sonic hedgehog (Shh) and its receptor Patchedl (Helms et al., 1997; Marigo et al., 1996a; Marigo et al., 1996b; Riddle et al., 1993) are expressed in the facial epithelium and establish the midline of the upper face (Chiang et al., 1996; Helms et al., 1997). Fgf2, 4, and 8 (fibroblast growth factors) promote proliferation (Richman and Crosby, 1990) and outgrowth of facial mesenchyme (Richman et al., 1997), and are expressed in the epithelium of the face (Francis-West et al., 1998; Richman et al., 1997; Vogel et al., 1996). Different patterns of gene expression and of cell proliferation, movement and death in each prominence (McGonnell et al., 1998), however, have complicated study of molecular signalling in facial prominence outgrowth. Significance of the cleft primary palate mutation The cleft primary palate mutation affects facial development in a very consistent and specific manner; the upper beak is truncated but the rest of the face appears not to be affected by the mutation (Abbott and MacCabe, 1966). The phenotype first becomes apparent at stage 28 (Yee and Abbott, 1978), long after neural crest cells have migrated into the face, and after the time of peak expansion of the facial prominences (McGonnell et al., 1998). That the cpp phenotype is most likely due to a single gene mutation (Abbott and MacCabe, 1966) and restricted to the face makes cpp embryos amenable to the study of inductive mechanisms of facial outgrowth. This thesis investigates several aspects of the cpp mutation. The first is epithelial-mesenchymal interactions involved in outgrowth and patterning. Hypotheses are that the mutation: 1) primarily affects the epithelium; 2) primarily affects the mesenchyme; or 3) 6 affects both tissue layers. Hypotheses regarding truncation of the upper beak are that it results from 1) increased cell death; 2) decreased cell proliferation, or 3) a disruption of patterning unrelated to cell number. Conclusions are applicable not only to the mutation, but may also contribute to a more general understanding of normal facial outgrowth. 7 Chapter 2: Cleft primary palate mutation affects the epithelium and not the mesenchyme in chicken craniofacial development INTRODUCTION The head comprises tissues of ectodermal, neural crest and mesodermal embryonic origins, which contribute to craniofacial structure in a manner remarkably conserved among vertebrates (DeBeer, 1985; Langille, 1993b). Facial prominences consist of neural crest-derived mesenchymal cells that migrate to the presumptive face, where they underlie the facial epithelium. Differentiation and growth of the craniofacial prominences are preceded by a cascade of gene interactions that we are beginning to elucidate (Couly et al., 2002; Francis-West et al., 1998; Lee et al., 2001; Schneider et al., 2001; Schorle et al., 1996; Shigetani, 2000; Trainor et al., 2002). The complexity of the embryonic face means that minor anomalies or delays during development can result in serious abnormalities, of which cleft lip with or without cleft palate is one of the most common. Tissue interactions in facial prominence growth Facial prominences, though initially similar in appearance and closely coordinated with one another as the face develops, have different fates and different capacities for outgrowth (Richman and Tickle, 1989; Wedden, 1987). Although neural crest-derived facial mesenchyme is capable of forming bone and cartilage, it is not initially competent to pattern specific skeletal elements. In the frontonasal mass, the characteristic rod morphology of the upper beak is dependent on an interaction with facial epithelium. Without this interaction, isolated mesenchyme forms a shapeless mass of bone, cartilage and connective tissue (Wedden, 1987; Richman and Tickle, 1989). Mandibular prominence mesenchyme acts similarly. Maxillary prominence mesenchyme when isolated forms bone but not cartilage (Wedden et al., 1988). When grafted out of the face but with epithelium intact, the maxillary prominence forms a disorganized mass of bone (Richman and Tickle, 1989), and in fact even in vivo requires contact with the frontonasal mass to form morphologically normal derivatives (McCann et al., 1991). Although well established that patterning of facial prominence derivatives resides in the mesenchyme of each prominence by stage 20, a signal intrinsic to facial epithelium is required to support outgrowth and patterning (Richman and Tickle, 1992). At earlier stages some patterning is inherent in the neural crest cells (Couly et al., 1998; Noden, 1983b), although this view has recently been challenged (Trainor et al., 2002). In light of this and other evidence for plasticity of facial patterning during and after migration of cranial neural crest cells (Couly et al., 2002; Lee et al., 2001), it is now more important than ever to determine the nature and scope of epithelial influence during facial development. The cleft primary palate mutation One way to approach the study of craniofacial development is by examining specific defects such as failure of outgrowth. A chicken mutation known as cleft primary palate (cpp) has been discovered that causes complete failure of primary palate formation and consequent bilateral cleft lip (Abbott and MacCabe, 1966). Cpp mutant embryos are indistinguishable from normal chicken embryos (Yee and Abbott, 1978) until stage 26 (Hamburger and Hamilton, 1951). The mutant phenotype is first detected when an ectopic ridge forms at the superior margin of the nasal slits. This ridge remains and continues to enlarge as the embryo develops, however instead of growing along the proximodistal axis into a beak, the frontonasal mass tissue appears to arrest at the facial prominence stage of development. Cpp mutants have normal development of the lower beak, and the mature phenotype closely resembles embryos in which the frontonasal mass has been excised (McCann et al., 1991). The mutation may affect the tissue-level interactions, or inductions, that lead to outgrowth of the frontonasal mass and differentiation of this prominence into the upper beak. The appearance of the phenotype in approximately one quarter of cpp embryos suggests that it is an autosomal recessive mutation (Abbott and MacCabe, 1966; Yee and Abbott, 1978). Study of the cleft primary palate embryos proceeded along several lines of investigation. This chapter addresses epithelial-mesenchymal interactions and tissue-level differences observed in cpp mutant embryos. First, the skeletal morphology was examined at stage 38, to determine which elements were affected. Then, each facial 9 prominence was grafted into a host limb system, to distinguish primary effects of the mutation on the facial prominences from possible secondary effects. Recombination experiments tested whether outgrowth in cpp mutant tissues could be rescued by normal mesenchyme or epithelium. Finally, cell proliferation and cell death were examined in order to determine the mechanism for the stunted upper beak outgrowth. 10 M A T E R I A L S A N D M E T H O D S Embryos I Fertilized eggs from the. cleft primary palate (cpp) line were purchased from the Avian;Sciences Center, University of California at Davis, California. White leghorn eggs were purchased from Coastline Chicks (Abbotsford, B.C.) and from the University of Alberta (Edmonton, A B ) . Eggs were incubated at 3 8 ° C , windowed with Scotch tape and staged; according to the Hamilton-Hamburger chicken staging table (Hamburger and Hamilton, 1951). Throughout, the terms normal and wild-type are used interchangeably to describe embryos from the white leghorn strain, or from the cpp strain but with a normal phenotype. Embryos with the cpp phenotype are referred to as cpp mutants. Clearing and staining of bone and cartilage in the skull Embryos sacrificed at H H stage 38 and examined for skeletal development were fixed in 100% ethanol, treated with acetone, stained for bone with alizarin red and for cartilage with Alcian blue, and cleared in decreasing concentrations of 2% K O H and glycerol (Plant et al., 2000). Dissection of facial prominences at stage 20,24 and 28 The frontonasal mass of stage 20 embryos was divided into quarters (approximately 350 x 300 pm). Stage 24 embryo frontonasal masses were also divided into quarters (approximately 400 x 400 pm; Table 1). At stage 28, medial or lateral thirds were dissected out, and these pieces were divided in half with a proximodistal cut. Each piece was therefore 1/6* of the entire frontonasal mass prominence. Only fractions that included the epithelium bordering the nasal pit were designated lateral; all others were considered medial (Fig. 1). These stage 28 sixths were rectangular, approximately 300x600 pm, and fit well into the graft site of the host limb bud. Recombination experiments at all stages (20, 24 and 28) consisted of the entire frontonasal mass epithelium from the appropriate donor, combined with fractions of mesenchyme (quarters for stage 20 and 24, sixths for stage 28) as described for intact grafts above. Epithelium was taken from the superficial (outer) surface of the frontonasal mass. Neither nasal pit epithelium nor, in the case of cpp mutant donors, the epithelium superior to the nasal pits (ridge epithelium) was included in recombinations (Table 2). 11 Grafting Facial prominences either intact or recombined were grafted to the limb bud of a stage 22 host chick embryo after removal of a rectangle of epithelium and mesenchyme from the centre of the bud (Wedden, 1987). This system supports three dimensional outgrowth and skeletogenesis while removing facial tissues from the environment of the mutant embryo. In addition, dissection of facial prominences permits tissue separation and recombination, which is not possible directly within the developing face. Determination of embryonic phenotype at stages 20 and 24 The mutation is not visible at stages 20 or 24, therefore a graft assay was done to determine phenotype. For each donor embryo, a lateral quarter was dissected from the frontonasal mass and grafted intact to a host limb to identify the phenotype. A medial quarter was used for recombination and grafted to a second host limb (Fig. 1). Only embryos in which the control graft survived were analysed further. Tissue separation and recombination Cpp or normal embryos were incubated to the desired stage (HH20, 24 or 28). Prominences were dissected out and either grafted intact or treated with 2% trypsin to separate the epithelium. Recombinations of epithelium and mesenchyme were done as described (Richman and Tickle, 1989). Host embryos were checked for health and for presence of the graft the day after surgery. Those that were not attached to the host limb were not included in further analyses. Hosts were incubated at 38°C for 7 days, and then sacrificed and the limbs fixed. Clearing and wholemount staining of cartilage in grafts Limbs were recovered after 7 days incubation and rinsed in PBS. The general morphology of the graft and the presence or absence of an egg tooth were noted. Both limbs were then fixed in 5%TCA for several days (up to months) and processed through acetone for 1 day, 100% and 80% ethanol for 1 hour each, stained with Alcian blue (1% Alcian blue 8GX (BDH Biochemicals) in 30% acetic acid in ethanol), rehydrated through ethanol to ddH 2 0 (70%, 50%, 25% 1 hour each; ddH 2 0 2 hours), cleared in a graded series of 0.5% KOH/glycerol (1:3, 1:1, 3:1, 1 hour to overnight each) and stored in 100% glycerol. 12 Statistical analysis of outgrowth Cleared and stained grafts were each assessed for outgrowth. Camera lucida tracings of the cleared grafts including cartilage elements were measured to quantify outgrowth. Length of the cartilage was measured from the insertion of the cartilage rod on the host limb and width was measured at the widest part of the cartilage (Fig. 3G,H). The ratio of length to width was calculated for each specimen. Grafts with a L:W ratio of less than 2:1 were classed as nodules. These parameters ensured grafts that failed to grow along an axis were classed as nodules, although many contained a substantial amount of cartilage. Grafts that exceeded a L:W ratio of 2:1, but were less than 1.5 mm in length were classified as nodules based on the overall lack of outgrowth. Chi-square analysis was used to determine significant differences in outgrowth among experiments, and power of the Chi-square was estimated using power of the binomial test (Zar, 1984). TUNEL Cell death was examined in normal and cpp mutant embryos using the TUNEL (terminal uridine nick-end labelling) technique as described (Shen et al., 1997). Sections were coverslipped and photographed. BrdU Cell proliferation was examined by labelling FfH stage 28 embryos with bromo-deoxyuridine (BrdU). Approximately 50 pi of a 10"2 M solution of BrdU was injected into a vitelline vein. This gave more reproducible uptake than injecting BrdU solution into the amniotic cavity. Embryos were incubated at 38°C for 2 hours, then fixed in 4%PFA overnight, processed and embedded. Wax sections (7pm) were cut and BrdU was detected with an anti-BrdU primary antibody (1:30, Becton-Dickinson), biotinylated with an ABC detection kit (Avidin-Biotin) according to the manufacturer's instructions. The substrate was visualized with diaminobenzidine (DAB). Slides were counterstained with 7pg/ml of Hoechst 33258 (Aldrich) in PBS (pH 7.4) and coverslipped. Counts of labelled cells were made in shallow (approx. 30 pm), middle (100pm) and deep (200pm) sections through the frontonasal mass (n=8 embryos, 4 normal and 4 cpp mutant). Two adjacent sections from each depth were labelled, and as BrdU labelling appeared consistent between sections, cells from one section at each depth were counted. Each section was divided into 9 areas (Fig. 6H) and percent of BrdU-labelled 13 cells was calculated from the total cell number (stained with Hoechst) for each area. This was termed the proliferation index. Statistical analysis began with a three-way analysis of variance followed by Tukey's post-hoc test for multiple comparison (SPSS) to determine which areas of the frontonasal mass exhibited significant differences in proliferation. Based on this analysis, comparisons between proliferation in similar areas of mutant and normal embryos were done using Student's t-test (Fig. 6H). Photography and preparation of figures Images were captured with a Minolta RD175 digital camera and were processed using Adobe Photoshop version 4.0 (Adobe, California). Stage 38 heads were photographed with a macrolens and substage illumination. Grafts were photographed using a Leica Wild M3C dissecting microscope and sections with a Zeiss Axioscope compound microscope. 14 Figure 1: Dissection, grafting and tissue recombination of normal and cleft primary palate mutant chicken embryos Camera lucida tracings of stage 28 cpp strain normal (left), cpp mutant (right) and stage 24 cpp strain, unknown phenotype heads. The frontonasal mass was dissected out of the face, either stripped of epithelium or left intact, and divided into sixths (stage 28) or quarters (stage 24). Colour indicates area of frontonasal mass dissected out. All intact graft phenotype assays were done with lateral intact pieces (D) and all recombinations with medial mesenchyme and undivided epithelium (A,B,C). Figure shows recombinations of epithelium (dark) with mesenchyme (light, stippled; A,B,C) followed by grafting into a stage 22 White leghorn host (E). Key: fnm, frontonasal mass; L=lateral; lnp, lateral nasal process; M=medial; md, mandible; rax, maxillary process 15 Stage 28 wild-type Stage 28 cpp mutant Stage 24 cpp embryo unknown phenotype RESULTS Absence of frontonasal mass derivatives in the cleft primary palate mutant craniofacial skeleton Stage 38 cpp mutant embryos (n=6) were cleared and stained for cartilage and bone, in order to determine which skeletal elements were affected, and the consistency of the phenotype. The skeleton of the body showed no effects of the mutation (data not shown). The most significant finding in the cpp mutant skulls was the complete absence of distal skeletal frontonasal mass derivatives, the premaxilla and prenasal cartilage (Fig. 2B,D). The most distal bony element was a morphologically normal nasal bone, a derivative of the lateral nasal prominence. Other derivatives of the lateral nasal process were all present in mutant embryos. The superior, middle and inferior nasal conchae were symmetrical but broader, and in the absence of the premaxilla extend inferiorly below the palatine bone and spread laterally well beyond the nasal septum. The major skeletal derivative of the maxillary process, the maxillary bone, was present in the mutant and articulated normally with the jugal bone (Fig. 2D). This was the case in all embryos affected by the mutation; that is, absence of frontonasal mass derivatives is a completely penetrant effect of the mutation. The palatine process of the maxilla was also present, but appeared enlarged. This may have been due to the absence of the premaxilla, with which it usually articulates. The palatine bone was also present in the mutant embryos, but was shortened and thickened along its proximodistal axis in most specimens (compare Fig. 2C,D). In contrast to the upper beak, the cranial vault appeared normal (compare Fig. 2A,B). Furthermore, all elements of the lower beak were present and appeared normal, including the articulation of the retroarticular process and the quadrate between the upper and lower beak. Meckel's cartilage and surrounding bones of the lower beak appeared slightly shortened and thickened when compared with stage-matched normal embryos (compare Fig. 2E,F). 17 Figure 2: Skeletal derivatives of normal and cleft primary palate chicken embryos Stage 38 skulls stained with alcian blue (cartilage) and alizarin red (bone). A, C, E: normal, and B, D, F cleft primary palate mutant craniofacial skeleton A, B lateral view; C, D palatal view with lower beak removed; E, F lower beak, superior view of oral surface. Scale bar-= 500 pm for all panels. Key: d, dentary bone; eg, entoglossal; j , jugal; mx, maxillary bone; n, nasal bone; nc, nasal conchae; p, palatine bone; pp, palatine process of the maxillary bone; pnc, prenasal cartilage; pmx, premaxilla; pt, pterygoid; qj, quadratojugal ' 18 wild-type c p p mutant Grafts of stage 28 wild-type and cpp mutant facial prominences confirm that the mutation affects only the frontonasal mass In order to distinguish primary effects of the mutation on outgrowth from secondary effects caused by abnormal growth of the facial prominences, the lateral nasal and maxillary prominences and frontonasal mass were dissected out of cpp mutant, cpp normal and white leghorn embryos and grafted to host limb buds. This was done at stage 28, the earliest stage at which it is possible to visually identify cpp mutant embryos (compare Fig. 6B,G). Results from cpp normal and white leghorn embryos were the same and so were combined. Grafts of the lateral nasal prominences to the limb bud develop primarily into cartilaginous nasal conchae, in chickens the superior, middle and inferior. Grafts from mutant embryos also formed conchae (12 of 12) and were morphologically indistinguishable from wild-type grafts (n=9; Fig. 3A,B). Maxillary prominences from stage 28 wild-type and mutant embryos (n=3 each type; data not shown) formed roughly spherical bone nodules approximately 15 mm in diameter, similar to the results of maxillary prominence grafts reported by Richman and Tickle (1989). Grafts of frontonasal mass tissue from, normal embryos develop into long cartilaginous rods (46 of 49; Fig. 3C,E and (Richman and Tickle, 1989). Cpp mutant frontonasal mass grafts consistently gave rise to cartilage-containing nodules without egg teeth (n=28; Fig. 3D,F). The frontonasal mass was therefore the only prominence that exhibited a mutant phenotype in the graft system. Developmental potential of subregions of the normal stage 28 frontonasal mass The potential for patterning and outgrowth of the frontonasal mass has not been studied at stage 28. Before examining the cpp mutant embryos, therefore, it was necessary to characterize the role of the epithelium and mesenchyme in development of the normal frontonasal mass. Furthermore since the frontonasal mass is a large prominence and could not be grafted intact to the host limb bud, it was also necessary to assess the developmental capabilities of different subregions. In previous studies at earlier stages, only the central third of the frontonasal mass was used 20 Figure 3: Grafts of intact facial prominences from stage 28 normal and cleft primary palate mutant embryos Host wings with grafts of intact facial prominences from stage 28 normal and cpp mutant embryos, cleared and stained for cartilage with alcian blue. Scale bar in A = 1 mm and applies to all. A, C, E Normal embryos, B, D, F cpp mutant embryos. A, B lateral nasal prominence grafts. Both of these grafts are attached to the ulna and contain superior and inferior nasal conchae. C, D central third of frontonasal mass grafts. Graft in C, from a normal embryo, has an egg tooth (not visible). E , F lateral third of frontonasal mass grafts. Note similarity to grafts in C and D. G and H are camera lucida tracings of specimens in E and F. Grafts are coloured grey. Connected dots (dotted line) in G indicate points used to measure length of graft cartilage; dots at base of graft indicate points used to measure width. Note in H that dots denoting length and width are equidistant. Key: white arrows denote graft; h, humerus; r, radius; u, ulna. 21 (Richman et al., 1997; Richman and Tickle, 1989; Richman and Tickle, 1992; Wedden, 1987). Here, central and lateral pieces were used for recombinations of stage 28 frontonasal mass tissue, and therefore the developmental potential for each of these regions was assessed. In addition, grafts of different sizes were examined to ensure that the smallest pieces of frontonasal mass tissue used could still give rise to a substantial outgrowth. Thus, as long as the tissue grafted was at least a sixth of the frontonasal mass, failure of outgrowth could be attributed to the experiment rather than to variability in graft size. Cpp normal as well as White leghorn embryos were used to examine regional potential; the results were the same and were therefore pooled. Grafts of stage 28 normal central thirds each produced a long cartilage rod (6/6; Fig. 3C). Wild-type lateral thirds of the frontonasal mass also produced long cartilage rods (24/24; Fig. 3E). However, grafts of stage 28 frontonasal mass thirds at 600 pm2 were larger than the graft site on the host limb bud, so grafts of a smaller size were also examined. Of grafts of medial sixths, the majority formed outgrowths (10/11) and one formed a nodule. Similarly most grafts of lateral sixths produced long cartilage rods (18/21). Nodular grafts were examined and determined to have attached to the host limb and formed cartilage and soft tissue (data not shown). Thus approximately 6% of grafts from normal stage 28 embryos did not grow out despite thriving in the host limb, and this proportion was unrelated to position in the frontonasal mass. Frontonasal mass sixths therefore have a similar potential for outgrowth as thirds at stage 28, so subsequent recombination experiments were done with sixths, which fit very well into the graft site on the host limb. Use of sixths also permitted several recombinations from a single cpp mutant donor. Since medial and lateral pieces have similar potential for outgrowth, lateral pieces were grafted intact as controls for phenotype (especially important in younger embryos, see below) and medial pieces were used for recombinations. In these intact normal grafts presence of an egg tooth, a heavily keratinised epithelial specialization derived from the frontonasal mass, was not strictly correlated with outgrowth, so lack of an egg tooth was not used to identify the mutant phenotype. As noted earlier, the frontonasal mass is the only prominence affected by the cpp mutation. Grafts of intact frontonasal mass pieces from stage 28 cpp mutant embryos form nodules rather than rods, and this is the case for both medial (2/2 Fig. 3D) and 23 Table 1: Growth of intact grafts, isolated mesenchyme and recombinations from normal embryos at stages 20, 24 and 28 TREATMENT ORIGIN stage/position N OUTCOME # w/growth Power Estimate Intact M M M M 20 lat 20 med 24 lat 24 med 28 lat 28 med 11 4 66 5 32 17 11 = 100% 3 = 75% 60 = 91% 5 = 100% 30 = 94% 16 = 94% 1.0 0.74 0.97 1.0 0.96 1.0 Isolated Mesenchyme 28 lat 28 med 10 9 2 = 20% 5 = 56% 0.0001 0.42 Recombined w/epithelium (origin/stage) FNM 20 or 24 FNM 20-28 FNM 24 or 28 FNM 28 Mandibular 28 mesenchyme 20 med 24 med or lat 28 med or lat 24 mandibular 28 med or lat 30 23 10 15 14 30 = 100% 23 = 100% 10 = 100% 15 = 100% 14 = 100% 1.0 1.0 1.0 1.0 1.0 All embryos were cpp strain wild type or White leghorn. Hatching indicates epithelium used for recombinations. L, lat = lateral; M, med = medial 24 lateral (26/26) pieces (Fig. 3F), whether thirds or sixths. After establishing that in stage 28 embryos the entire frontonasal mass is affected by the cpp mutation, the question addressed was whether the defect is in the epithelium, the mesenchyme, or affects both. Recombinations of mutant and normal frontonasal mass tissues at stage 28 show that the epithelium is the target of the cpp mutation Before investigating whether outgrowth in cpp mutant frontonasal mass tissue could be rescued, the capacity for outgrowth of normal stage 28 frontonasal mass mesenchyme was first determined. Facial epithelium is necessary for outgrowth of the mesenchyme of stage 20 and 24 facial prominences (Richman and Tickle, 1989). Even though not directly recombined with epithelium, grafts of stage 24 frontonasal mass mesenchyme placed in the limb bud are covered by dorsal wing epithelium within 24 hours (Matovinovic and Richman, 1997). Such isolated stage 24 mesenchyme grafts did not grow out, forming cartilaginous elements only 1-2 mm in length (Richman and Tickle, 1989). In contrast, isolated frontonasal mass mesenchyme from stage 28 embryos (this study) demonstrated a position-related capability for outgrowth that was expressed even in the presence of dorsal wing epithelium. Medial fractions of mesenchyme grew out in the limb bud graft system in just over half of the cases (5 of 9 or 56%; Table 1). Mesenchyme from the lateral edge of the frontonasal mass grew out less frequently, in 2 of 10 specimens (Table !)• To determine whether recombination with normal facial epithelium rescues outgrowth of medial and lateral frontonasal mass mesenchyme equally, stage 28 normal frontonasal mass mesenchyme was recombined with homotypic wild-type epithelium. All grafts formed long cartilage rods (medial, 9/9; lateral, 5/5; Table 1). Mandibular epithelium from stage 28 embryos also supports growth in all stage 28 frontonasal mass mesenchyme (7/7 medial, 7/7 lateral; Table 1). Recombinations were then done to determine whether either frontonasal mass mesenchyme or epithelium from stage 28 cpp mutant embryos would grow out under the influence of the corresponding normal tissue. Lateral pieces were not used for recombinations but were grafted intact to a host limb bud to confirm the phenotype of the 25 donor. Medial pieces were subject to separation and recombination with donor epithelium (Fig. 1). Rescue of stage 28 cpp mutant mesenchyme by recombination with wild-type epithelium The recombination technique itself was tested to determine whether it could induce outgrowth in the mutant frontonasal mass. Mutant frontonasal mass mesenchyme from stage 28 embryos failed to grow out when recombined with mutant frontonasal mass epithelium (medial 0/10; lateral 0/5; Table 2 and Fig. 4B), showing that recombination and grafting does not restore normal signalling between epithelium and mesenchyme. In contrast to the lack of outgrowth observed in the control (mutant/mutant) recombinations, normal epithelium supported outgrowth in all regions of cpp mutant frontonasal mass mesenchyme. Cpp mutant mesenchyme from the medial region, the lateral edge of the frontonasal mass, or the ectopic ridge all grew out when combined with normal epithelium (medial 21/26 growth, i.e. 80% rescue; lateral 14/20 growth, 70% rescue, ridge 6/6 growth; Table 2). Frontonasal mass epithelium was not the only tissue capable of inducing growth in mutant mesenchyme; recombination with wild-type mandibular epithelium also supported outgrowth (12/13; Table 2). Outgrowth of stage 28 cpp mutant frontonasal mass mesenchyme is therefore rescued by wild-type facial epithelium. Cpp mutant epithelium neither inhibits nor supports growth in wild-type facial mesenchyme Recombinations of wild-type stage 28 frontonasal mass mesenchyme and stage 28 cpp mutant epithelium resulted in outgrowth (8/10 for medial mesenchyme, 2/3 for lateral mesenchyme, Table 2). Cpp epithelium therefore does not have an inhibitory effect on outgrowth of competent mesenchyme; that is, the inherent capacity for outgrowth in isolated stage 28 frontonasal mass mesenchyme is not suppressed by cpp mutant epithelium. These results are not significantly different (p>0.05) from the outgrowth observed in grafts of isolated frontonasal mass mesenchyme (56% of medial pieces, 20% lateral, Table 1), and so do not indicate whether the cpp mutant epithelium actively 26 Table 2: Growth of grafts and recombinations of frontonasal mass tissue from stage 28 embryos affected by the cleft primary palate mutation TREATMENT GRAFT ORIGIN stage/phen*/position N OUTCOME # w/growth Intact 28 cpp medial 28 cpp lateral 2 26 0 0 Recombined phen*/stage/origin of epithelium wild type 28 FNM mesenchyme R R p R R ) ! I My wild type 28 mandibular cpp mutant 28 FNM 28 cpp med / 28 cpp lat I 28 cpp ridge ^ 28 cpp medial 28 cpp lateral 28 cpp ridge 28 cpp medial 28 cpp lateral 28 wild type medial 28 wild type lateral 24 wild type mandibular 26 20 3 5 5 3 10 5 10 3 21 = 81% 14 = 70% 3 = 100% 5 = 100% 4 = 80% 3 = 100% 0 0 8 = 80% 2 = 67% 0 *Phenotype of donor embryos was identified visually by the presence of an ectopic ridge. Hatching indicates epithelium used for recombinations. L = lateral; M = medial; R = ridge 27 Figure 4: Grafts of recombined mesenchyme and epithelium from stage 24 and 28 normal and cleft primary palate mutant embryos Host wings with grafts of recombined tissues, cleared and stained for cartilage with alcian blue. Scale bar in A=lmm and applies to A-D and F; scale bar in E=lmm and applies to E only. A: Recombination of normal stage 28 frontonasal mass medial mesenchyme with normal stage 28 frontonasal mass epithelium produced a long, branched rod. B: Recombination of frontonasal mass medial mesenchyme and frontonasal mass epithelium from a stage 28 embryo affected by the cpp mutation (identified visually) forms a small nodule. The host limb, though deformed, is a similar size as the control limb. C: Mesenchyme from a stage 28 cpp mutant embryo recombined with normal stage 28 frontonasal mass epithelium grows out into a long rod. D: Medial frontonasal mass mesenchyme from a normal stage 28 embryo recombined with frontonasal mass epithelium from a cpp mutant embryo has grown out into a rod. E: Mandibular mesenchyme from a stage 24 normal embryo recombined with frontonasal mass epithelium from a normal stage 28 embryo grows out into a long, angled rod. F: Mandibular mesenchyme from a stage 24 normal embryo recombined with frontonasal mass epithelium from a stage 28 cpp mutant embryo grows into a wide, flat nodule. 28 supported outgrowth. To further clarify whether mutant frontonasal mass epithelium could support outgrowth, this epithelium was recombined with stage 24 mandibular mesenchyme. Mandibular mesenchyme from stage 24 embryos is not capable of outgrowth without an induction from competent epithelium (Richman and Tickle, 1989). When mandibular mesenchyme from stage 24 embryos was recombined with wild-type stage 28 frontonasal mass epithelium, all of the grafts grew out into long cartilage rods, many with associated bone (15/15; Table 2, Fig. 4E). When recombined with cpp mutant frontonasal mass epithelium, none of the mandibular mesenchyme grafts formed rods (0/5, Table 2, Fig. 4F), demonstrating that epithelium from stage 28 cpp mutant embryos is not competent to rescue outgrowth. The cleft primary palate mutation does not affect frontonasal mass mesenchyme at stage 20 or 24 , • • . As the cpp mutation is already morphologically evident by stage 28, it is possible that the failure of mutant epithelium to provide a growth signal to the underlying mesenchyme results from an earlier event. Although the mutation is manifested in the epithelium at stage 28, it could affect either the epithelium or the mesenchyme prior to stage 28. The mesenchyme only was examined, as it was not possible to both assay phenotype (using an intact lateral piece graft) and also recover enough epithelium for recombinations from stage 20 and 24 embryos. To verify that both lateral and medial regions of the frontonasal mass could be used, as in the experiments on stage 28 embryos, the capacity for outgrowth of of normal frontonasal mass fragments was examined at stages 20 and 24. Fractions (quarters and sixths) of normal (White leghorn) tissue from medial and lateral regions of the frontonasal mass were tested, to determine whether outgrowth at stage 20 and 24 is dependent on either graft size or origin. Capacity for outgrowth was the same regardless of the region and whether quarters or sixths were used. Lateral quarters were then used for the phenotype assay and medial quarters in all recombination experiments with stage 20 and 24 tissues (Table 3). 30 Table 3: Recombination of frontonasal mass mesenchyme from stage 20 and 24 cpp embryos with wild type frontonasal mass epithelium EPITHELIUM MEDIAL N OUTCOME OF RECOMBINATION MESENCHYME stage phenotype*/stage outgrowth (normal) nodule (mutant) 20, 24 or 28 wild type 24 23 23 0 20,24 or 28 cpp mutant 24 7 6 1 20 or 24 wild type 20 29 29 0 20 or 24 cpp mutant 20 17 11 6 Donor phenotype was determined using the grafting assay (intact grafts taken from the same donor). Wild type epithelium for recombination was taken only from White leghorn strain embryos. Medial mesenchyme only was used in recombinations. 31 Cpp frontonasal mass mesenchyme from stage 20 and 24 embryos is capable of outgrowth when recombined with normal epithelium from any stage (20-28) Medial frontonasal mass mesenchyme dissected from cpp mutant embryos (stage 20 or 24; phenotype identified retrospectively from matching intact graft) was recombined with epithelium from normal (White leghorn) embryos at stages 20, 24 or 28. The majority of these mutant mesenchyme/normal epithelium recombinations grew out into rods. There were no differences related to stage of donor epithelium, so results were pooled (Table 3). In comparing rescue of stage 20 and stage 24 cpp mutant mesenchyme, it was noted that fewer stage 20 recombinations grew out, but this difference was not found to be significant (stage 20: 11/ 17; Stage 24: 6/7; Table 3; p=0.32). Control recombinations of normal (White leghorn) epithelium and mesenchyme showed that outgrowth of all normal mesenchyme (stage 20 or 24) is supported by normal epithelium from any stage (20, 24 or 28; 53/53 total; Table 1). These results indicate that outgrowth of cpp mutant mesenchyme from the frontonasal mass prior to expression of the phenotype can be rescued by recombining with normal frontonasal mass epithelium. Further, this capacity for outgrowth in response to normal epithelium is not position-dependent at any of the stages studied. Increased cell proliferation in the stage 28 cpp mutant frontonasal mass Failure of skeletal derivatives of the frontonasal mass to form in cpp mutant embryos could be due not only to a defect in signalling, but to changes within the mesenchymal cell population of the frontonasal mass. Cell death and cell proliferation were therefore examined. Stage 28 was chosen because at this stage the cpp mutation is just beginning to influence frontonasal mass morphology. Cell death was examined in frontonasal mass sections from cpp normal (n=4) and cpp mutant (n=3) embryos at stage 28. All embryos exhibit some apoptotic cell death as detected by the TUNEL method, particularly in the mesenchyme of the maxillary processes and above the nasal pits (Fig. 5). Mesenchymal cell labelling in the frontonasal mass is sparse, and no differences were noted among normal and mutant stage 28 embryos from the cpp strain (Fig. 5). Celldeath in the epithelium was not examined. 3 2 Figure 5: Cell death in normal and mutant stage 28 embryos from the cpp strain Scale bar = 0.5 mm A: Frontal section through a normal stage 28 embryo upper face (approx 100pm depth). TUNEL-positive cells are sparsely distributed throughout the mesenchyme of all prominences, but are apparent in the mesenchyme of the maxillary processes and lateral nasal processes above the nasal pits, as well as in the globular processes of the frontonasal mass (arrowheads). Epithelial cell death is difficult to see due to separation of the epithelium in this section and high background staining. B: Frontal section through the upper face of a stage 28 cpp mutant embryo, at a similar depth to the normal section in A. TUNEL-positive cells are present in the globular processes of the frontonasal mass and in the mesenchyme of the lateral nasal and maxillary processes (arrowheads) but as above, labelled cells are sparsely distributed. Key: fnm, frontonasal mass; lnp, lateral nasal process; mxp, maxillary process 33 f n m 34 Cell proliferation is lowest in the central regions of the normal but not the cpp mutant frontonasal mass at stage 28 Cell proliferation was examined in 8 embryos (4 cpp normal and 4 cpp mutant), the same specimens as were used for the TUNEL assay. Despite the eventual lack of outgrowth in mutant embryos, the mean proliferation of the entire frontonasal mass at stage 28 is higher in the mutants than in the normal embryos (10% normal, 16% mutant, p=0.05; Fig. 6H,I). Differences in cell proliferation among areas within each embryo was then examined. In each of the 4 normal embryos, there are significant differences in proliferation among the 9 areas (ANOVA, p<0.05). Areas bordering the nasal pits (areas 1,3,4,6,7,9) exhibit significantly more proliferation in normal stage 28 embryos than does the centre of the frontonasal mass (areas 2,5). This is true for all sections taken through the frontonasal mass of normal embryos; shallow, middle and deep (Fig. 7). In the 4 cpp mutant embryos, however, there was no overall significant difference in the amount of cell proliferation among areas, indicating that cell proliferation is more uniform in mutant embryos. The proliferation indices of subregions (areas) of the cpp mutant embryo frontonasal mass were examined to determine significant differences in proliferation from the same anatomical regions of the wild-type frontonasal mass. Proliferation in the centre of the frontonasal mass of cpp mutant embryos (areas 2,5) was found to be significantly higher than in the same areas of normal embryos (Student's t-test; p<0.01). Proliferation along the cranial ridge (areas 1,2,3) of the frontonasal mass in mutant embryos is similar to the level observed in the caudal (area 8) and lateral frontonasal mass (areas 4,6,7,9); the only significantly lower proliferation index was for area 5 of middle sections, a region that is not in contact with frontonasal mass epithelium (Fig. 61, 7F). 35 Figure 6: Cell proliferation in the stage 28 normal and cleft primary palate mutant embryo Scale bar in A=l mm and is for A, B, F, G. Scale bar in E=0.5mm and is for D, E and I. A, B: Normal stage 28 embryo. Lateral (A) and frontal (B) views show prominences. Arrow in A indicates the proximal part of the frontonasal mass; note the lack of outgrowth in the normal embryo. C: Camera lucida tracing of a stage 28 cpp mutant embryo, lateral view with plane of section indicated. D: Montage of a section through the frontonasal mass of a normal stage 28 embryo stained with Hoechst nuclear stain to show uniform cell density. E: Montage of a frontal section through the frontonasal mass, lateral nasal processes and maxillary processes of a normal stage 28 embryo, labelled with BrdU showing brown DAB precipitate over proliferating cells. F, G: Stage 28 cpp mutant embryo. Lateral (F) and frontal (G) views show the ectopic ridge at the cranial margin of the frontonasal mass (white arrow in F, arrowheads in G). H: Grid shows division of frontonasal mass into areas for quantification of cell proliferation, as described in the Methods. Areas 1, 2, and 3 are bordered superiorly by the cranial margin of the frontonasal mass (includes the ectopic ridge of cpp mutant embryos), and inferiorly at the top of the nasal slits. Areas 4, 5 and 6 are divided from areas 7, 8 and 9 by a horizontal line halfway between the stomodeal margin and the top of the nasal slits, and vertical lines divide the frontonasal mass between the nasal slits into equal thirds. Within each area the mean percent proliferation (standard deviation in brackets) is given for normal (blue) and cpp mutant (black, italicized) specimens. Asterisks in areas 2 and 5 indicate a significant difference in values between normal and cpp mutant. I: Montage of a frontal section through the face of a stage 28 cpp mutant embryo, labelled with BrdU as in E to show proliferating cells. The ectopic ridge is evident, as is the uniformly high cell proliferation along the cranial margin. . 36 Figure 7: Cell proliferation differs in superficial, mid and deep sections through the frontonasal mass of stage 28 normal and mutant embryos of the cpp strain Scale bar = 1 mm and applies to C-H. A , B: Lateral views of stage 28 normal (A) and cpp mutant (B) embryos of the cpp strain with depth and orientation of sections indicated (white lines). Note that the mutant frontonasal mass is slightly thicker than the normal at this stage due to formation of the ectopic ridge. C,D: Superficial (approx 30 pm depth) frontal sections through the upper face of a normal (C) and mutant (D) embryo. Note that the centre of the frontonasal mass (C, areas 2 and 5) of the normal embryo shows low proliferation, while proliferation is uniformly high in D, a section from a mutant embryo. E,F: Mid (approx 100 pm depth) frontal sections through the upper face of a normal (E) and mutant (F) embryo. Note that in E, the centre of the frontonasal mass exhibits low proliferation extending to the cranial margin (corresponding to areas 2 and 5) and closely resembles the pattern seen in superficial sections, while in the mutant (F), only the very centre of the frontonasal mass (area 5) maintains a low level of proliferation. G,H: Deep (approx 200 pm depth) frontal sections through the upper face of a normal (G) and mutant (H) embryo. The pattern of cell proliferation is very similar to that seen in mid sections, with an area of high cell proliferation along the cranial margin of the frontonasal mass (area 2) evident in the cpp mutant but not the normal embryo. 38 Shallow Mid Deep DISCUSSION By stage 28, frontonasal mass epithelium from cpp mutant embryos is incapable of supporting outgrowth, although mutant mesenchyme remains capable of responding to epithelial signalling. A ridge forms along the cranial margin of the frontonasal mass between the nasal pits. Cell proliferation in this ridge is higher than cell proliferation in the same area of normal embryos. The frontonasal mass does not grow out along a proximodistal axis, and frontonasal mass derivatives including the prenasal cartilage, the premaxilla and the egg tooth fail to form in mutant embryos. There has been considerable study of epithelial-mesenchymal interactions in control of patterning in the vertebrate face. Connective tissues such as bone and cartilage are formed from mesenchymal cells, yet a specific epithelial signal is required to initiate osteogenesis (Hall, 1980; Tyler and Hall, 1977). It is also known that the identity of skeletal derivatives resides in the mesenchyme of each facial prominence, but for their formation an induction from competent (facial) epithelium is necessary (Richman and Tickle, 1989; Richman and Tickle, 1992; Wedden, 1987). A requirement for epithelial signalling to evoke potential from craniofacial mesenchyme has thus been shown in many ways, but the mechanism of such inductions in the face is not yet as clearly understood as in the limb. The cpp mutation offers an opportunity to examine a failure in the outgrowth of a single facial prominence. This defect manifests only in the frontonasal mass epithelium, and at a time when previous studies have shown the facial epithelia to be functionally interchangeable (Richman and Tickle, 1989). The cpp mutation provides a system within which control of outgrowth is distinct among the facial prominences. Such a model could simplify study of the interactions necessary for outgrowth. Cleft primary palate mutant phenotype: loss of the upper beak proximodistal axis Examination of the mutant craniofacial skeleton shows that frontonasal mass derivatives are the only skeletal elements that fail to form. Grafts of intact facial prominences demonstrated that this defect is confined to the frontonasal mass. The formation of cartilage nodules (often associated with bone) in all of the grafts indicates 4 0 that proximodistal growth rather than skeletogenesis is impaired. As well, derivatives from the other facial prominences form normally in cpp mutant embryos, in contrast to the lack of maxillary-derived bones seen in normal embryos from which the frontonasal mass has been completely removed at stage 24 (McCann et al., 1991). In these frontonasal mass-excised embryos, the maxillary process remains small and nodular to day 19 of incubation, which the authors propose may be because the maxillary process requires the frontonasal mass to draw it forward during outgrowth. In the cpp mutant, however, the maxillary, jugal, and palatine bones are all present and articulate normally with one another. The palatine and maxillary bone are slightly shorter and thicker in the cpp mutant embryos, which is likely secondary to the loss of a proximodistal axis rather than to any effects of the mutation on the maxillary prominence. The enlargement of the nasal conchae in cpp mutant embryos, which are symmetrical and very similar in all the embryos examined, is likely due to the absence of bordering structures, distally the premaxilla and ventromedially the prenasal cartilage. The cpp mutant frontonasal mass therefore still indirectly supports skeletogenesis in the maxillary process; it is only derivatives requiring frontonasal mass proximo-distal patterning that fail to form. Isolated stage 28 frontonasal mass mesenchyme has a position-dependent capacity for outgrowth In the chicken, proximodistal growth is a significant element of skeletal patterning. Mesenchyme from prominences that ordinarily form long rods (the frontonasal mass and mandibular arch) will do so under the influence of any facial epithelium; mesenchyrne from other prominences cannot be induced to exhibit this pattern (Richman and Tickle, 1989; Richman and Tickle, 1992; Wedden, 1987). Evidence from this study indicates that at a later stage, this epithelial induction is no longer required. Stage 28 isolated frontonasal mass mesenchyme consistently produced a long cartilage rod in the limb graft system. Dorsal wing epithelium, the only possible source for an epithelial signal in this system, has previously been shown to be incompetent to rescue outgrowth in stage 24 facial mesenchyme (Richman and Tickle, 1992). The outgrowth of the stage 28 grafts, therefore, clearly indicates that frontonasal mass mesenchyme gains patterning ability as it grows older. Further, the differences in 41 capability of medial and lateral mesenchyme for outgrowth (56% versus 20%) indicate that at this stage, patterning is at least somewhat position-dependent. The growth potential may be correlated with differential cell proliferation within the frontonasal mass, although these results imply a negative correlation at stage 28. The lateral region of frontonasal mass mesenchyme, bordering the nasal pits, exhibits higher cell proliferation than the medial (McGonnell et al., 1998; Peterka and Jelinek, 1983; and these results), yet when isolated the lateral mesenchyme grows out less frequently than the medial. Differential expression of several patterning genes within the frontonasal mass demonstrates that there are other distinctions between lateral and medial regions: for example, Msxl and Msx2 (laterally expressed; (Nishikawa et al., 1994); AP2 (lateral, Shen et al., 1997); and Fhf4 (central; (Munoz-Sanjuan et al., 2001). Response of stage 28 normal mesenchyme to normal epithelium is not position-dependent Stage 28 normal mesenchyme always exhibited outgrowth when recombined with normal facial (frontonasal mass or mandibular) epithelium, and this response was equal for medial and lateral regions, implying that facial, epithelium is required to direct outgrowth even in highly-proliferating (lateral) populations of mesenchyme. The role of the frontonasal mass epithelium, at least by stage 28, may therefore be to signal an axis or midline rather than simply supporting growth. That midline would already be established for the most part within medial mesenchyme by stage 28, explaining why isolated stage 28 medial mesenchyme grows out more frequently (56%), and why mesenchyme from younger embryos requires facial epithelium to grow out at all (Richman and Tickle, 1989). The cpp mutation manifests in the frontonasal mass epithelium Grafting of intact stage 28 cpp mutant frontonasal mass tissue to a host limb bud produced a small nodule containing bone and cartilage but no egg tooth. The defect therefore persists outside of the face, in an environment known to support normal frontonasal mass outgrowth (Richman and Tickle, 1989 and these results). Grafts from mutant embryos resemble grafts of isolated mesenchyme, producing unpolarized nodules 42 of cartilage. This suggests that the epithelial induction necessary for proximodistal outgrowth is absent in the cpp mutant frontonasal mass, although the epithelium itself is present. • • ' All of the mesenchyme tested (all stages, normal or cpp, any region of the frontonasal mass and the tip of the mandibular arch) grew out when recombined with normal epithelium. This demonstrates that regardless of origin, facial mesenchyme responds to patterning cues from the epithelium. Epithelium from cpp mutants, conversely, supported cell proliferation but not outgrowth, demonstrating not only that the epithelium plays a role in controlling the pattern of mesenchymal cell proliferation, but that control of proliferation and of outgrowth into a rod (i.e., patterning) may be separate, at least in the stage 28 frontonasal mass. Previous studies of proliferation in the frontonasal mass of normal (White leghorn) chicken embryos have identified patterns similar to the one shown in this thesis for normal embryos of the cpp strain. Peterka and Jelinek (1983) counted cells arrested in mitosis after injection of colchicine, and identifed the globular processes (the distal corners of the frontonasal rnass, corresponding to areas 7 and 9 in the present study) as 'centres of proliferative activity' in the 5-day (stage 28) normal chicken embryo. Labelling of the normal chick face with 3H-thymidine (Minkoff and Kuntz, 1977) also demonstrated that mesenchymal proliferation is highest in the globular processes; the authors calculated a proliferative index of 35% for the globular processes at stage 28 and approximately 20-25% elsewhere in the frontonasal mass, a difference that is comparable to the results obtained here for normal embryos using BrdU label (Fig. 6,7). The division of the frontonasal mass into shallow, middle and deep sections in the present study permits a closer examination of the role of the rostralmost epithelium of the frontonasal mass, which overlies the mesenchyme visualized in shallow sections but not mesenchyme of middle or deep sections (Fig. 7). Proliferation in the cpp mutant embryos, with reference to the role of the epithelium, will be discussed in the final chapter of this thesis. 4 3 Comparison of the cpp phenotype with other models for abnormal embryonic outgrowth Exposure of avian embryos to excess retinoids during facial development consistently results in truncation of the upper beak (Richman and Delgado, 1995; Tamarin et al., 1984; Wedden and Tickle, 1986). Similar to the cpp mutant embryos, the frontonasal mass is the only facial prominence affected in retinoic acid (RA)-treated embryos. Unlike the cpp mutation, however, RA-treated mesenchyme fails to grow out under the influence of normal epithelium, showing that mesenchyme is the target of RA (Wedden, 1987). As well, RA-treated mesenchyme does not show changes in cell proliferation or programmed cell death (McGonnell et al., 1998; Shen et al., 1997). Excess retinoids may, therefore, act to prevent mesenchyme from receiving its usual epithelial signals, whereas the situation with the cpp mutation seems to be the reverse. The cleft primary palate mutation provides an avian model for outgrowth in the face. Outgrowth is important in many other areas of the developing embryo, and parallels are often drawn between outgrowth in the face and in the limb buds. There are many examples of mutations affecting mesenchymal patterning in the limb bud, but one in particular affects the epithelium. The chicken eudiplopodia mutation causes Polydactyly, and the mechanism is formation of ectopic apical ectodermal ridges in the limb buds (Fraser and Abbott, 1971a; Fraser and Abbott, 1971b). Recombination experiments similar to those in this study showed that the eudiplopodia mutation affects the epithelium and not the mesenchyme at the stages examined (Carrington and Fallon, 1986; Fraser and Abbott, 1971b; Robert et al., 1991). Thus, the cpp and eudiplopodia mutations have in common the appearance of an ectopic ridge of ectoderm which promotes outgrowth along a different axis than in the normal embryo, essentially repatterning the underlying normal mesenchyme. One important difference is that the normal axis of growth persists in the eudiplopodia embryos, resulting in supernumary digits, while growth in a cranial direction is correlated with a loss of proximodistal outgrowth and distal structures in the cpp mutation. This may reflect the importance of limiting cell proliferation to certain regions of the developing facial prominences. The fate of cells produced by excessive proliferation was not followed beyond stage 28 in this study; so although it is clear that at this stage cell death does not contribute to the 44 morphology of the frontonasal mass, this process may play a later role in determination of the phenotype. An epithelially-mediated increase in cell proliferation in the frontonasal mass has been identified, leading to a disruption of proximodistal outgrowth and a complete loss of distal upper beak structures. Cleft primary palate mesenchyme remains capable of skeletogenesis, and responds normally to patterning cues from normal epithelium. This demonstrates not only that individual facial prominences are patterned separately, but also that mesenchymal cell proliferation alone is not sufficient for upper beak formation; regulation by the overlying epithelium is required to properly direct outgrowth. 45 Chapter 3: Length, morphology and egg tooth formation in grafts of frontonasal mass tissue INTRODUCTION The frontonasal mass forms distal midline structures of the chick upper beak; the prenasal cartilage, the premaxillary bone and an egg tooth at the distal tip, which is not a true tooth but is formed of keratinized epithelium. The egg tooth develops at stage 32 (approximately 7 days after fertilization), persists until hatching, when it cracks the shell, and is sloughed off shortly after (Bellairs and Osmond, 1998; Kingsbury et al., 1953). When grafted into a host limb bud, pieces of the frontonasal mass typically form a long cartilage rod (Wedden, 1987; Richman and Tickle, 1989). In most cases, an egg tooth also develops at the distal tip of the graft. An induction by frontonasal mass mesenchyme is necessary for egg tooth formation as isolated frontonasal mass epithelium will not form an egg tooth (Tonegawa, 1973). Quail-chick chimera experiments demonstrated that the left and right halves of the frontonasal mass epithelium, derived from ectoderm of the anterior neural ridge, contribute approximately equally to the egg tooth (Couly and Le Douarin, 1985). Frontonasal mass epithelium is not required for egg tooth formation, however, as recombination of frontonasal mass mesenchyme with various types of non-frontoriasal mass epithelium resulted in egg tooth formation in some grafts of each type (dorsal wing and flank epithelia, Richman and Tickle, 1992; cephalic and back epithelia, Tonegawa, 1973). Not all epithelium, however, is capable of supporting egg tooth formation; chorionic epithelium differentiates into many epithelial specializations but does not support egg tooth formation when recombined with upper beak (frontonasal mass) mesenchyme (Kato and.Hayashi, 1963). Egg tooth induction has therefore been considered a property of frontonasal mass mesenchyme, and the egg tooth is sometimes used to indicate the midline of the frontonasal mass in experiments using graft systems (Wedden, 1987; Richman and Tickle, 1989) or in ovo manipulations (Helms et al., 1997; Hu and Helms, 1999). Some previous studies of grafted frontonasal mass tissue reported an incidence of egg tooth formation of less than 100% even when the epithelium and mesenchyme were grafted intact (Tonegawa, 1973; Wedden, 1987), indicating that dissection or grafting may somehow disrupt egg tooth formation. The significance of 46 size or position (central or lateral) within the frontonasal mass for the ability of grafts to support or induce egg tooth formation has not yet been addressed. Tissue was dissected from all regions of the frontonasal mass for the experiments in this project, so it was important to determine whether egg tooth formation or any other patterning was position-dependent. Here, length; morphology and egg tooth formation were analysed with respect to the stage of the donor embryo, position of the graft tissue within the frontonasal mass, and treatment, whether pieces dissected from the frontonasal mass were grafted intact, stripped of epithelium or recombined. No differences were noted among grafts of different sizes (thirds or sixths) in the previous analysis (Chapter 2), so results were combined and analysed for length, morphology and egg tooth formation by stage and treatment but not size. Egg tooth formation is shown to be a property of all frontonasal mass tissue rather than a characteristic of the midline, and is not necessarily correlated with the amount of outgrowth in the normal chicken embryo. For this reason, egg tooth formation was not used to determine phenotype of embryos from the cpp strain in experiments described in the previous chapter. 47 MATERIALS AND METHODS Embryos Preparation of embryos, dissection of facial prominences, tissue separation, recombination, grafting, clearing, staining and camera lucida tracing of the grafts were as described in Chapter 2. Presence or absence of an egg tooth as well as morphology was noted as the graft and host limb were recovered, and specimens were photographed in PBS prior to fixation. Morphology was confirmed and selected grafts were traced and measured after the specimens were cleared and stained with Alcian blue. Statistical analysis Analysis was done on the normal specimens and recombinations of normal with cpp mutant tissue described in the previous chapter. Grafts of intact pieces or homotypic recombinations of frontonasal mass tissue from cpp mutant embryos are not considered in this analysis, as all failed to grow out or to form an egg tooth. Chi-square analyses were used to test for significant differences in morphology and in egg tooth formation with respect to stage and phenotype of donor, origin of tissue within the frontonasal mass (medial versus lateral) and treatment (intact grafts, isolated mesenchyme or recombinations), and ANOVA was used to analyse length. For all tests P<0.05 was considered significant. An estimation of power of the binomial test was done on each set used in Chi-square analysis to determine the power of the calculation, as in Chapter 2 (Zar, 1984). Photography and preparation of figures Images were captured with a Minolta RD175 digital camera and were processed using Adobe Photoshop version 4.0 (California). Grafts.were photographed using a Leica Wild M3C dissecting microscope. 48 RESULTS Length Each category of grafts of intact pieces of frontonasal mass tissue contained a few grafts which failed to grow out into rods (approximately 6%) despite attachment to the host limb and chondrogenesis (as detailed in the previous chapter). These grafts were excluded from the analysis of length and morphology. For the analysis of length, several (a minimum of 10 in each category, except as noted on Table 4) randomly-chosen grafts were traced and measured from each stage and treatment category, and the mean length and standard deviation calculated (Table 4). Measurements of stage 20 and 24 intact lateral pieces showed no significant difference with respect to stage. There were too few intact medial grafts to examine the effect of position in stage 20 and 24 intact grafts, however there were no significant differences among intact, lateral pieces and recombined, medial grafts at either stage 20 or 24. Among different types of graft (intact pieces, isolated mesenchyme and recombinations), the only significant difference found was that the few rods formed from grafts of isolated mesenchyme (4/19 including both medial and lateral, Table 1) were significantly shorter than rods from intact grafts and recombinations. This was the case even although nodules, which did not exhibit any proximodistal outgrowth at all and were the most frequent result of isolated mesenchyme grafts (15/19 nodules, Table 1), were excluded from the present analysis. There were no significant differences among kinds of recombinations, that is homotypic recombinations of normal tissue or heterotypic recombinations between normal and cpp mutant tissue. Morphology of rods In order to quantify outgrowth, two classes of graft outcomes were used (as outlined in the previous chapter); rods, which grew out along a proximodistal axis, and nodules, which did not grow out. The present analysis of morphology deals only with rods. Within the category of rods there were two distinct morphologies. Tapered rods had a rounded or tapered distal tip closely resemble the prenasal cartilage, and hammerhead rods consisted of a long rod with bulbous cartilage at the tip (Fig. 8). Both tapered and hammerhead rods grew out to similar lengths, and could support the formation of an egg tooth. The morphology of the rods and presence or absence of an 49 egg tooth was analysed within grafts of normal frontonasal mass tissue in order to examine the specification of medial versus lateral frontonasal mass pieces at different stages. The proportion of hammerheads was analysed with respect to graft type, position in the frontonasal mass and stage for all grafts from normal embryos. Among grafts of intact pieces, grafts of lateral pieces (31/101) formed hammerheads significantly more frequently than grafts of medial pieces (2/25; p=0.02). The incidence of hammerhead formation did not vary significantly with stage. Isolated lateral frontonasal mass mesenchyme from stage 28 embryos produced a hammerhead only once (1/19, Table 4), indicating that the absence of epithelium significantly affects the morphology as well as the frequency of outgrowth. Recombinations, which were always between an entire sheet of frontonasal mass epithelium and a medial piece of mesenchyme, produced a significantly higher proportion of hammerheads (46%, Table 4) than did grafts of intact frontonasal mass pieces (27%; p=0.002). Recombinations between frontonasal mass epithelium and lateral mesenchyme were not done, so analysis related to positional origin of mesenchyme could not be attempted. 50 Figure 8 : Morphology and egg tooth formation in grafts of normal and cpp mutant frontonasal mass tissue White arrow in A,B,D indicates egg tooth. Scale bar =1 mm A: An intact graft of the medial quarter of the frontonasal mass from a stage 20 normal donor embryo produced a long, straight rod with an egg tooth. This graft was classified as a tapered rod. B: An intact graft of the lateral quarter of the frontonasal mass from a stage 20 embryo produced a hammerhead rod with an egg tooth. C: Isolated medial mesenchyme from a stage 28 normal embryo grows out into a short rod. This specimen does not have an egg tooth. D: Recombination of medial mesenchyme from a stage 28 cpp mutant embryo with normal epithelium produced a hammerhead rod with an egg tooth. E: Recombination of medial normal frontonasal mass mesenchyme with cpp mutant frontonasal mass epithelium produced a large nodule without proximodistal growth or egg tooth. 5 1 Table 4: Length, morphology and egg tooth formation in frontonasal mass grafts from normal embryos and recombinations of normal arid cpp mutant frontonasal mass tissue. Treatment Graft Origin N Outcome Power stage/position rod (nodule) length (cm) a # of hheads egg tooth Estimate Intact stage 20 lateral 11(0) 0,38±0.13 2 11 (100%) 1.0 stage 20 medial 3(1). (not measured) 0 4 (100%) 1.0 stage 24 lateral 60 (6) 0.32+0.12 22 56 (85%) 0.97 stage 24 medial 5(0) (not measured) 1 4 (80%) 0.33 stage 28 lateral 30 (2) 0.28±0.07 7 23 (72%) 0.84 stage 28 medial 16(1) 0.35+0.11 1 15 (88%) 0.99 Isolated mesenchyme stage 28 lateral 2(8) (not measured) 0 l c 0.58 stage 28 medial 5(4) 0.14±0.01 b 1 6d 0.15 Recombination 20 mesenchyme 30 (0) 0.33±0.09 12 29 (97%) 1.0 wt FNM ep 24 mesenchyme 23 (0) 0.32±0.09 11 23 (100%) 1.0 (stages 20-28 28 mesenchyme 10(0) 0.21±0.09 6 10 (100%) 1.0 combined) c/?/?mesenchyme 38 (11) 0.34±0.11 21 31 (82%) 0.93 cpp FNM ep 28 mesenchyme 10(3) 0.25±0.10 4 8 (62%) 0.21 a: length was measured only for grafts which grew out into rods and were successfully cleared, stained and traced. b: lateral and medial combined c: egg tooth was formed on a nodule, not.a rod d: reflects the egg teeth formed on nodules, which were not included in analysis of length or morphology 53 Egg tooth formation The incidence of egg tooth formation was high but not completely correlated with growth in grafts from normal embryos; rods did form without an egg tooth, and nodules with egg teeth were also observed. For this reason, egg tooth formation was not used as a means of phenotyping grafts from cpp strain embryos (Chapter 2). Egg tooth formation was then analysed to determine whether stage, position or phenotype of tissues in recombinations influenced egg tooth formation. The total incidence of egg tooth formation in grafts of intact normal frontonasal mass pieces was 84%. This is similar to the 90% incidence found by Wedden (1987) for grafts of stage 24 frontonasal mass thirds, but less frequent than observed by Richman and Tickle (100%; 1989). No significant differences were found among intact grafts of different stages or origins within the frontonasal mass (medial versus lateral). Recombined normal grafts (normal epithelium with medial mesenchyme) formed an egg tooth in 98% of the cases (62/63, Table 4), an incidence significantly higher than intact grafts (p=0.002). Interestingly, grafts of isolated frontonasal mass mesenchyme were capable of inducing an egg tooth in the host limb epithelium, and this was the only type of graft in which the incidence of egg tooth formation was significantly higher in medial (6/9 or 67%) than in lateral (1/10, 10%, p=0.005) frontonasal mass tissue. Isolated mesenchyme grafts were done only with tissue from stage 28 embryos in the present study, however isolated mesenchyme from stage 24 embryos has previously been shown to form egg teeth in 29% of grafts (2/7; Richman and Tickle, 1989). Recombinations of cpp mutant and normal tissue were also capable of egg tooth formation (76% combined, Table 4). This was the case for cpp mutant mesenchyme recombined with normal epithelium, as well as the reciprocal. As with the intact normal grafts, no significant differences were found with respect to stage or position of the recombined graft. 54 DISCUSSION Recombinations of frontonasal mass mesenchyme and epithelium from donor embryos of any stage produced the highest proportion both of hammerhead rods and of egg teeth. Intact pieces from the lateral frontonasal mass formed hammerheads significantly more frequently than intact medial pieces, but position did not have an effect on egg tooth formation. Isolated stage 28 medial mesenchyme produced significantly higher numbers of rods and of egg teeth than lateral mesenchyme, but with respect to morphology only a single hammerhead rod formed from isolated mesenchyme grafts. Stage of the donor embryo was not found to affect length, morphology or egg tooth formation. Length of cartilage rods is affected by presence of facial epithelium but not by stage, position in the frontonasal mass or treatment The only difference in length among categories was that isolated mesenchyme from stage 28 embryos produced shorter rods. This category of graft was also significantly less likely to produce a hammerhead morphology. These results are in accordance with previous reports demonstrating that isolated mesenchyme from younger embryos (stages 20 and 24) does not grow out (Richman and Tickle, 1989; 1992). Results from this analysis show that by stage 28, frontonasal mass mesenchyme has gained some capacity for outgrowth but the addition of frontonasal mass epithelium to the graft significantly increases the length as well as the proportion of grafts that grow out. Stage of the donor embryo and position within the frontonasal mass did not affect the length of the grafts, nor did intact grafts differ significantly in length from any of the recombination treatments, including recombinations with cpp mutant tissue. This indicates that seven days of growth in the host limb is sufficient to allow grafts to reach their potential for outgrowth. It should be noted that variability in length contributes to a lack of significance among groups, however a certain amount of variability is intrinsic to the host limb graft system (Richman and Tickle, 1989). 55 Neither morphology nor egg tooth formation are reliable indicators of the frontonasal mass midline The bulbous-tipped cartilage rod, termed here a hammerhead, was observed in earlier studies (Richman and Tickle, 1989) and noted to occur in grafts of frontonasal mass but not mandibular tissue (grafts from the maxillary prominence were also examined and exhibited no outgrowth). In another study, double cartilage rods with a wide or doubled egg tooth were observed in grafts from stage 20 (but not stage 24) embryos, and it was thought that this 'wide' morphology might indicate absence of a midline, i.e. that the frontonasal mass midline was not yet established at stage 20 (Wedden et al., 1988). Similarly, hammerhead grafts may indicate a loss of some positional identity, as the prenasal cartilage is normally tapered in vivo (Fig. 2C). Grafts from the lateral frontonasal mass form hammerheads more frequently than grafts from the medial region in the present study. One possible explanation for the prevalence of hammerhead formation in lateral grafts relates to the proliferation data outlined in the previous chapter, which indicates that cell proliferation is higher in the lateral region; this may account for the formation of hammerheads, which appear larger overall than tapered rods. However recombinations, which used medial (not lateral) mesenchyme combined with the entire frontonasal mass epithelium, had a higher proportion of hammerhead rods than either medial or lateral intact pieces. Amount of cell proliferation within the frontonasal mass of the donor embryo therefore does not account for the high proportion of hammerheads formed from recombinations. The amount of epithelium is another factor that varies among grafts of different treatments. Grafts of intact lateral pieces contain more epithelium than medial pieces, because the nasal pit epithelium is left in place. Recombinations contain more epithelium than either type of intact piece. It may therefore be that formation of the hammerhead is correlated with the amount of epithelium contained in the graft. Although the egg tooth is an epithelial specialization, its formation requires an induction from the mesenchyme (Tonegawa, 1973; Richman and Tickle, 1992). These results indicate that frontonasal mass mesenchyme can induce an egg tooth regardless of stage, phenotype or origin within the frontonasal mass. Egg tooth formation is not necessarily a characteristic of the frontonasal mass midline in a limb graft system, nor is 56 it necessarily correlated with outgrowth. Further, in the stage 28 cpp mutant embryos both the mesenchyme and the epithelium are capable of supporting egg tooth formation when combined with normal tissue in a rescue experiment. Dorsal wing ectoderm, which like cpp mutant frontonasal mass epithelium does not support outgrowth, is also capable of forming an egg tooth under the influence of isolated facial mesenchyme (Richman and Tickle, 1992 and results from the present study). These results not only indicate separate signalling pathways for outgrowth and epithelial specializations, but support previous findings of conservation of the epithelial-mesenchymal interactions required for epithelial specialization between the face and the limb (Richman and Tickle, 1992). 57 Chapter 4: The cleft primary palate mutation affects Fgf8 expression and alters epithelial-mesenchymal signalling in the frontonasal mass INTRODUCTION Patterning and growth of the vertebrate face proceeds via an intricate network of signalling between and within the epithelium and mesenchyme. Although the identity of the facial prominences is already laid down in the mesenchyme by stage 20 (Wedden, 1987; Richman and Tickle, 1989), grafting experiments have shown that an epithelial signal is necessary to evoke mesenchymal potential, and support outgrowth and formation of specific skeletal elements. While this signal may be provided by any facial epithelia (Richman and Tickle, 1989), the molecular nature of both the epithelial signal and the mesenchymal response is unknown. As outlined in the general introduction, genes necessary for epithelial-mesenchymal signalling and pattern formation in the limb have been localized to various regions of the face, but their functions and interactions are not necessarily the same among each of the facial prominences (Capdevila and Izpisua Belmonte, 2001). For example, placement or 'pre-pattern' of the limb is under complex regulation by the Hox code (Cohn et al., 1997; Davidson et al., 1991; Duboule, 1998) which is not active in the face. Comparison of gene expression in normal and mutant embryos will help establish which of the many genes present in the face, and specifically in the frontonasal mass, have a role in prominence outgrowth. Candidate genes examined in normal and cpp mutant embryos were chosen based on expression in the epithelium or mesenchyme of the frontonasal mass. The signalling molecules Fgf8, Shh and Bmp4 are all expressed in the facial epithelia at precisely the time of patterning and outgrowth. Fgf8 is expressed in a restricted region of facial epithelium (Richman et al., 1997; Vogel et al., 1996; Helms et al., 1997). Shh is expressed in stomodeal epithelium as well as in the ventral forebrain and is thought to have a role in patterning the frontonasal mass (Helms et al., 1997; Hu and Helms, 1999; Schneider et al., 2001). The Shh receptor Ptcl (Chen and Struhl, 1996) is expressed along the caudal edge of the frontonasal mass in a similar region as Shh during face development (Helms et al., 1997; Hu and Helms, 1999; Schneider et al., 2001); its expression in normal and cpp mutant embryos was examined. Bmp4 is one of 58 the Bmps (with 2 and 7) expressed in the frontonasal mass epithelium (Francis-West et al., 1994) as is the Bmp antagonist noggin (Ashique et al., 2002) also examined here. Bmp4 is upstream of Msxl and -2 in development of the tooth germ (Vainio et al., 1993), mandibular arch and maxillary process (Barlow and Francis-West, 1997), and regulates one of the Fgfs (Fgf4) in the limb AER (Pizette and Niswander, 1999). Expression of the transcription factors Msxl, Msx2 and AP2 was also examined. Msxl and Msx2 are expressed in the facial mesenchyme (Brown et al., 1997; Brown et al., 1993), Msxl in regions associated with high expansion during fusion of the prominences (McGonnell et al., 1998). AP2 is expressed in migrating cranial neural crest cells and also in the mesenchyme of facial prominences (Shen et al., 1997). Type II collagen, which is necessary for chondrogenesis (Kosher and Solursh, 1989; Thorogood et al., 1986) and is expressed throughout.the frontonasal mass mesenchyme (Matovinovic and Richman, 1997) was also examined. The possibility that frontonasal mass epithelium is not required to maintain mesenchymal gene expression at stage 28, just as epithelium is no longer required to support mesenchymal outgrowth (Chapter 2), was addressed in an organ culture system. Finally, a reported link between Shh and Fgf8 signalling in the face (Schneider et al., 2001) was examined in normal and cpp mutant embryos. FGF8 and SHH have been shown to interact with one another in the limb bud and are normally expressed in adjacent domains in the developing face (Hu and Helms, 1999; Richman et al., 1997 and these results). Shh and Fgf8 act in a feedback loop in the limb bud, with the early epithelial expression of Fgf8 inducing Shh in the underlying mesoderm (Crossley et al., 1996b; Lewandoski et al., 2000; Moon and Capecchi, 2000; Vogel et al., 1996). Shh then helps-to maintain Fgf8 in the AER, which in Shh-/- mice is initially normal but becomes disrupted and hypoplastic concomitant with loss of Fgf8 expression (Chiang et al., 2001). Both Fgf8 and Shh transcription are autoregulated in some interactions during normal development: Fgf8 in the limb (Crossley and Martin, 1995; Niswander et al., 1994) and Shh in the notochord and ventral floorplate (Ericson et al., 1996; Marti et al., 1995). Whether exogenous protein (FGF8 and SHH) or Shh antibody (5E1: Ericson et al., 1996) would alter the expression pattern of either Fgf8 or Shh in face cultures of normal embryos was therefore tested. 59 A change in the expression of one gene was found in the epithelium of cpp mutant embryos, both before and after the mutant phenotype becomes evident. No differences in expression of mesenchymal genes was found between normal and cpp mutant embryos. 60 MATERIALS AND METHODS Embryos Fertilized eggs from the Cleft primary palate (cpp) line were purchased from the Avian Sciences Center, University of California at Davis, California. White leghorn, and were purchased from Coastline Chicks (Abbotsford, B.C.) and from the University of Alberta (Edmonton, AB). Eggs were incubated at 38°C, windowed with Scotch tape and staged according to the Hamilton-Hamburger chicken staging table (Hamburger and Hamilton, 1951). The term normal is used to describe embryos from the white leghorn strain or from the cpp strain but with a normal phenotype. Embryos with the cpp phenotype are referred to as cpp mutants. Identification of cpp mutant embryos Mutant embryos were identified at stage 28 by the ectopic ridge that forms along the cranial margin of the frontonasal mass (Fig. 1). At earlier stages mutant embryos cannot be visually identified, so pieces of the frontonasal mass were grafted into a limb graft system and incubated for seven days. Failure of the graft to grow along a proximal-distal axis established the phenotype of the donor embryo as mutant (these grafts formed a healthy cartilaginous nodule, while grafts of wild-type frontonasal mass grew into a long rod). Cpp mutant donor embryos missing the grafted piece of the frontonasal mass were then processed for in situ hybridization. Organ culture Embryos were recovered at stage 20, 24 or 28 and the heads dissected into Hanks with 10% FCS. Either the frontonasal mass or the entire face was dissected away from the cranium and rinsed in culture media (BGJb with 10% FCS, 2mmol/L L-glutamine, 100 U/ml antibiotic-antimycotic (Gibco-BRL, Gaithersburg MD), 10 ug/ml ascorbic acid; (Tucker et al., 1999; Wang, 1999). The tissue was pipetted onto a square of filter paper (Nucleopore 0.1pm, Costar, Toronto ON) with the epithelial surface up (with or without epithelium) and cultured at 37°C, 5% CO2 after the method of Trowel (Saxen et al., 1962; Wang, 1999) supported at the air/medium interface on a stainless steel mesh grid. After 61 24 or 48 hours cultures were recovered by rinsing for 1 min in cold 100% methanol to fix them to the filter, fixed in 4% paraformaldehyde in PBS at 4°C overnight and processed to 70% ethanol for storage and subsequent whole mount in situ hybridization. For cultures of frontonasal mass mesenchyme, the frontonasal mass was treated with 2% trypsin in Hanks solution as described in Richman and Tickle (1989), the epithelium was stripped off with forceps and discarded, and the mesenchyme cultured as above. Application of exogenous FGF8, SHH protein or SHH monoclonal antibodies Heparin acrylic beads (Sigma, St. Louis MO) were soaked in FGF8 protein (R and D systems (Minneapolis, MN) at a concentration of 0.5 mg/ml or 1 mg/ml at room temperature for 2 hours and used immediately or stored overnight at 4°C and used the next day. Affigel blue beads were serially soaked 4 times in SHH protein (1 mg/ml, R and D Systems) as described (Lee et al., 2001) and implanted immediately or stored at 4°C and used the next day. Bioactivity of the proteins was confirmed by implanting FGF8 beads into the forebrain-midbrain isthmus (Crossley et al., 1996a) and by implanting SHH beads under the anterior AER (Drossopoulou et al., 2000; Yang et al., 1997). Duplication of the MH brain isthmus with FGF8 beads or duplication of digits with SHH beads indicated that proteins were bioactive. These control experiments were done with beads from the same soaking as used in the organ cultures. Although not every control experiment produced duplications, several specimens in each round did exhibit the expected duplications, confirming bioactivity of the protein'. For cultures, protein-soaked beads were placed either directly on top of the the frontonasal mass mesenchyme, or if the epithelium was present, a slit was made and the bead was tapped down slightly so that it contacted both the epithelium and the mesenchyme. Control beads (heparin or Affigel blue) were soaked in PBS and implanted as described. Shh 5E1 antibody developed by T.M. Jessell (Ericson et al., 1996) was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA 52242. Affigel blue beads were soaked in undiluted supernatant for several days at 4°C and implanted. Bioactivity of Shh Ab was assayed by implanting beads under the AER in the same manner as SHH protein beads and assaying for loss of digits (as in Hu and Helms, 1999). 62 In situ hybridization Whole mount in situ hybridization was done as described in Shen et al. (1997) and section in situs with radiolabeled probe as described (Rowe et al., 1991). Probes were obtained from the following individuals: Fgf8 from J.C. Izpisua Belmonte (Vogel et al., 1996), Msxl and Msx2 from S. Wedden (Brown et al., 1993, 1997), Bmp4 from P. Brickell (Francis-West et al., 1994), Noggin from R. Johnson (Capdevila and Johnson, 1998), Shh and Ptcl from C. Tabin (Shh: Riddle et al., 1993; Ptcl: Marigo et al., 1996). AP2 probe was as described in Shen et al. (1997). Photography and preparation of figures Images were captured with a Minolta RD175 digital camera and were processed using Adobe Photoshop version 4.0 (California). Stage 38 heads were photographed with a macrolens and substage illumination. 63 RESULTS Expression of Fgf8 but not mesenchymal patterning genes is altered in the cpp mutant frontonasal mass The mRNA expression patterns of several genes was examined in normal and cpp mutant embryos at three different stages (20, 24 and 28) during craniofacial development. Candidate genes known to be involved in patterning and outgrowth in the embryo nic face were chosen. Several genes expressed primarily in the epithelium were examined: Bmp4 and its antagonist Noggin, Fgf8 and Shh, as well as Ptcl, a Shh receptor expressed in both the epithelium and the mesenchyme. The mesenchymally-expressed genes AP2, collagen type II, Msxl and -2 were also examined (Table 5). Differences in expression between normal and cpp mutant embryos at stages 20 or 24, before the mutation is visible, were investigated since they could indicate primary effects of the cpp mutation. Expression at stage 28 was also examined. At this stage, cpp embryos are morphologically distinct so differences in expression between normal and cpp mutant embryos at this stage could be interpreted as signalling events downstream of the mutation. Mesenchymal genes show normal patterns of expression in the stage 28 cpp mutant frontonasal mass Whole mount in situ hybridization of stage 28 cpp mutant embryos (see Table 5 for N of all experiments) showed that expression of AP2 is confined to the corners of the frontonasal mass mesenchyme (globular processes), and to the mesenchyme of the lateral nasal and maxillary processes, as in normal embryos at the same stage (Shen et al., 1997 and Table 5, data not shown). In situ hybridization on sectioned embryos was used to examine another mesenchymally-expressed gene, collagen type II. Expression in normal and cpp mutant embryos was confined to the centre of the frontonasal mass, demarking the chondrogenic region (Fig. 9, compare A,B). In normal embryos, Msxl is expressed in the lateral third of the frontonasal mass at stages 20, 24 and 28. In addition there is expression in the lateral nasal prominences, the anterior two thirds of the maxillary prominence and the distal tips of the mandibular prominences 64 Table 5: In situ hybridization of face signalling genes in normal and cleft primary palate mutant embryos at stages 20, 24 and 28 Gene Phenotype Stage N and Type of ISH Epithelial B m p 4 N 28 4 frontal, 1 sagittal M 28 3 frontal, 1 sagittal N and M 20,24 1 each frontal, 1 each sagittal Fgf8 N 28 4 frontal, 2 sagittal, 2 whole mount M 28 3 frontal, 2 sagittal, 1 whole mount N 24 15 frontal, 3 sagittal, 1 whole mount M 24 7 frontal, 1 sagittal known N * 20 1 frontal, 1 sagittal known M * 20 1 frontal, 2 sagittal unknown phenotypef 20 20 frontal Noggin N and M 20, 24, 28 1 each frontal, 1 each sagittal Shh N 28 1 frontal, 1 sagittal, 2 whole mount M 28 1 frontal, 1 sagittal N 24 3 frontal, 1 sagittal, 1 whole mount M 24 4 frontal, 1 sagittal known N * 20 1 frontal, 1 sagittal known M * 20 1 frontal, 1 sagittal unknown phenotypef 20 20 frontal Mesenchymal AP2 N and M 28 2 each whole mount Collagen type II N 28 3 frontal M 28 2 frontal Msxl N 28 3 frontal, 3 whole mount M 28 2 frontal, 2 whole mount Msx2 N 28 4 frontal M 28 3 frontal N 24 1 frontal, 2 whole mount M 24 1 frontal, 1 whole mount N and M 20 1 each frontal Ptcl N and M 28 1 each frontal, 1 each sagittal N 24 3 frontal, 1 sagittal M 24 4 frontal, 1 sagittal known N * 20 1 frontal, 1 sagittal known M * 20 1 frontal, 1 sagittal unknown phenotypef 20 20 frontal | * Phenotype identified by grafting an intact piece f Phenotype was not identified and no differences in gene expression were found 65 Figure 9: Expression of Collagen type II, Msxl and -2, Bmp4 and its antagonist noggin is unaffected in cpp mutant embryos In situ hybridization of several candidate genes in frontal (A-N) and sagittal ( O - R ) sections of stage 28, 24 and 20 embryos showed no differences between normal and cpp mutant embryos. Expression in epithelium is indicated by white arrows. Asterisks indicate tissue removed for phenotyping. Scale bar in A=l mm and applies to all. A, B: Collagen type II is expressed in frontonasal mass mesenchyme at stage 28, cranially above and between the nasal pits, and in the central region but not in the globular processes. This pattern was observed in both normal (A) and cpp mutant ( B ) stage 28 embryos. C-H: Msxl and -2 are expressed in the mesenchyme of the globular processes (caudal lateral corners) of the frontonasal mass, as well as in the lateral nasal and maxillary processes. Expression of Msxl (C, D) and Msx2 (E, F) is similar in normal (C,E) and cpp mutant (D,F) embryos. Msx2 was also examined in stage 24 embryos, when it is usually expressed in the epithelium of the frontonasal mass and lateral nasal process as well as in the mesenchyme of the lateral corners of the frontonasal mass. Expression at this stage was the same in normal (G) and cpp mutant (H) embryos. I - R : Bmp4 is expressed at stage 28 in the frontonasal mass epithelium and in the underlying mesenchyme along the stomodeal margin, as well as in the mesenchyme of the lateral nasal process (Francis-West et al., 1994). This is the case for both normal (I) and cpp mutant embryos (J). Bmp4 is expressed similarly in normal (K,M,0) and cpp mutant (L,N,P) embryos at stages 24 (K,L frontal and 0,P sagittal) and 20 (M,N) as well, in the frontonasal mass and lateral nasal process epithelium, and in the distal but not the proximal invaginating nasal pit epithelium. Noggin is confined to a narrow band along the stomodeal margin of the frontonasal mass epithelium at stage 24 (Ashique et al., 2002 and Q ) . Noggin was similarly expressed in normal ( Q ) and cpp mutant ( R ) embryos at this stage. Key: fnm, frontonasal mass; lnp, lateral nasal process; md, mandibular arch; mx, maxillary process. 66 (Mina et al., 1995; Nishikawa et al., 1994). No differences were noted in Msxl or Msxl expression between normal and cpp mutant embryos at either stage 24 or 28, in the frontonasal mass epithelium as well as in the mesenchyme (Fig. 9C,D and compare E,F and G,H). Expression of Bmp4 and its antagonist noggin are not affected by the cpp mutation Bmp4 is normally expressed in the epithelium of the frontonasal mass, medial maxillary prominences and the midline of the mandibular promiences (Francis-West et al., 1994). Bmp4 was examined at stages 20, 24 and 28 (n=9 normal, 8 cpp). The Bmp antagonist Noggin is expressed in a limited region of the frontonasal mass epithelium (Ashique et al., 2002). Expression in normal and cpp mutants was similar for both genes (Fig. 9,1-R). Shh and its receptor Ptcl are expressed similarly in normal and cpp mutant embryos Shh is expressed in the ventral forebrain, the stomodeal epithelium and the intraoral side of the maxillary prominence epithelium between stages 20 and 29 (Fig. 10; Helms et al., 1997; Hu and Helms, 1999; Schneider et al., 2001), and its receptor Ptcl (Chen and Struhl, 1996) is expressed in the same epithelium as Shh transcripts, as well as in the adjacent mesenchyme (Fig. 10; Hu and Helms, 1999). Shh was expressed strongly in the ventral forebrain but was faint in the frontonasal mass epithelium in both normal and cpp mutant embryos at stage 20 (Fig. 10A,D). Epithelial expression was strong along the caudal margin of the frontonasal mass where it meets the stomodeal epithelium by stage 24 in both normal and cpp mutant embryos (compare Fig. 10B,E). At stage 28, Shh was expressed in the deep (stomodeal; see Fig. 10F) but not rostral frontonasal mass epithelium (Fig. 10C) in normal and cpp mutant embryos, and was not expressed in the epithelium at the lateral corners of the frontonasal mass. Ptcl was evident in the ventral fore-, mid- and hindbrain, and in the mesenchyme of the mandibular arches, more intense along the caudal margin, and in the medial region of the frontonasal mass, in both the epithelium and mesenchyme. The signal was faint in shallow sections and more intense in deeper sections, especially in medial mesenchyme 68 Figure 10: Shh and its receptor Ptcl are expressed similarly in normal and cpp mutant embryos In situ hybridization of stage 20, 24 and 28 normal and cpp mutant embryos. Scale bar in A=lmm and applies to all. Arrows indicate limits of expression. Asterisk indicates tissue removed in phenotype assay. Shh A-F: At stage 20 the pattern of expression is similar in normal (A) and cpp mutant (D) embryos. Shh is expressed strongly in the ventral forebrain, caudal mandibular epithelium and notochord, and is present but less intense in the frontonasal mass epithelium. At stage 24, expression is strong in the medial frontonasal mass epithelium of both normal (B) and cpp mutant (E) embryos. No transcripts are seen in the corners of the frontonasal mass (globular processes) or in the nasal pit epithelium. Expression in the brain and mandibular arch is similar to expression at stage 20. At stage 28, Shh is expressed in the stomodeal epithelium and in the adjacent deep frontonasal mass epithelium of both normal (C) and cpp mutant (F) embryos. Note that F is a deeper section than C, so expression is stronger and extends further laterally. Ptcl G - L : Ptcl is expressed with equal intensity in the ventral forebrain and the frontonasal mass epithelium, and more weakly in the frontonasal mass mesenchyme at stage 20 (G). Expression is very similar in the cpp mutant (J). At stage 24, expression is strong in the caudal frontonasal mass epithelium and in the medial frontonasal mass mesenchyme in both the normal (H) and cpp mutant (K). Note that expression does not extend laterally as far as the nasal pits (arrows). At stage 28, Ptcl expression is stronger in the epithelium than in the mesenchyme and extends laterally as far as, but not into, the nasal pit epithelium (I, arrows). Expression is still visible in the ventral telencephalon (L). Key: fnm, frontonasal mass; lnp, lateral nasal process; md, mandibular arch; mx, maxillary process. 69 70 Figure 11: Fgf8 expression is extended throughout the cpp mutant frontonasal mass epithelium Expression of Fgf8 in normal (A-C, G-I) and cpp mutant (D-F, J-L) embryos at stages 20-28. Scale bar in A=lmm and applies to A-F, G and J. Scale bar in L=lmm and applies to H,I,K,L. At stage 20, Fgf8 is normally expressed in the ventral forebrain (A) and throughout the frontonasal mass epithelium (G). Expression in normal stage 20 embryos (D) is similar to cpp mutant embryos whose phenotype had been identifed with the grafting assay (J). At all stages, expression of Fgf8 elsewhere in the craniofacial region (the mid-hindbrain isthmus and the epithelium at the junction of the maxillary and mandibular processes) was normal in normal and cpp mutant embryos. At stages 24 and 28, Fgf8 mRNA is expressed only in the epithelium bordering the nasal pits and at the junction of the maxillary process and the mandibular arch (B,C). In cpp embryos, ectopic Fgf8 expression extends throughout the frontonasal mass epithelium at both stages (E,F). Expression in the caudal frontonasal mass extends medially from the nasal pits along the stomodeal margin, and laterally to include the epithelium of the lateral nasal processes (F). This is the case throughout the frontonasal mass, seen in sections along the midsagittal plane (L). Ectopic expression is of similar intensity to normal expression in the nasal pit epithelium (F). The extended pattern and intensity are consistent among all of the stage 28 cpp mutant embryos examined (n=6). Key: Arrows indicate ectopic Fgf8 expression; triangles, expression in the mid-hindbrain isthmus; asterisks, tissue removed for the phenotype assay, e, eye; fnm, frontonasal mass; md, mandibular arch; mx, maxillary process, r, rhombencephalon. 71 stage 20 stage 24 stage 28 72 underlying the ventral forebrain (Fig. 10G,K,L). Expression was stronger and spread slightly more laterally at stage 20 than at stages 24 and 28, although like Shh, Ptcl transcripts were not seen in nasal pit epithelium at any stage. No differences were found between normal and cpp mutant patterns of expression at any of the three stages studied. Fgf8 expression is maintained throughout the frontonasal mass epithelium until stage 28 in cpp mutant embryos In the normal embryo, Fgf8 is expressed in the anterior commisural plate, the mid-hindbrain isthmus and throughout the frontonasal mass epithelium of the chicken until approximately stage 20 (Schneider et al., 2001; Vogel et al., 1996 and Fig. 11A,G). This pattern is conserved in mammals, with Fgf8 expression throughout the medial nasal process epithelium in the mouse (Bachler and Neubuser, 2001). After stage 20, Fgf8 transcripts are restricted to the epithelium surrounding the nasal pits, the corners of the frontonasal mass and the junction of the maxillary and mandibular processes (Richman et al., 1997 and Fig. 11B,C). In the cpp mutation there is no restriction of Fgf8 expression after stage 20; it is detected throughout the frontonasal mass epithelium at stage 24 and at stage 28 (Fig. 11E,F,K,L).. An in situ hybridization done on twenty non-phenotyped stage 20 cpp strain embryos (i.e., not phenotyped by grafting) showed no differences among them in Fgf8 expression. A similar experiment, done at stage 24, identified 6 of 20 embryos with Fgf8 expression throughout the frontonasal mass epithelium. This is consistent with the 25% frequency of the mutant phenotype observed during breeding of cpp strain birds (Yee and Abbott, 1978). Normal and cpp mutant embryos can therefore be distinguished on the basis of Fgf8 expression at stages 24 and later, although not at stage 20. Gene expression within the stage 28 frontonasal mass is independent of other facial prominences and is maintained in organ culture Identification of numerous face signalling genes that are unaffected by the cpp mutation raised the question of whether epithelium is normally required for the maintenance of mesenchymal genes in the stage 28 frontonasal mass. This stage is significant because grafting experiments in these and other studies identified a capacity 73 for outgrowth inherent in frontonasal mass mesenchyme at stage 28 which was not present at stage 20 or 24 (Richman and Tickle, 1989, 1992). Independence of some or all mesenchymal gene expression from epithelial signalling would explain both why expression of so many important signalling genes appears unaffected by the cpp mutation, and why the cpp mesenchyme is capable of responding to patterning by normal epithelium even after the phenotype is morphologically evident. Whether the epithelium is required to maintain expression of mesenchymal genes in the frontonasal mass at stage 28 was therefore addressed in an organ culture system. Maintenance of gene expression in whole face and frontonasal mass organ cultures was first tested. Mouse mandibular arches isolated in organ culture have been shown to undergo normal signalling interactions, up to and including those required to induce tooth formation (Tucker et al., 1998b; Wang et al., 1998). Whole face chicken cultures have also been shown to develop and undergo relatively normal skeletogenesis when grown in organ culture for up to 7 days from stage 20 (Hu and Helms, 2001). Whole mount in situ hybridization of all genes studied (AP2, Fgf8, Msxl and Shh) showed the same expression in stage 28 heads and in stage 28 frontonasal mass organ cultures (Table 6, Fig. 12), demonstrating that when isolated in organ culture the frontonasal mass, like the mandibular arch, maintains normal gene expression for at least 48 hours. AP2 and Msxl were expressed in the mesenchyme of the globular processes (caudal lateral corners of the frontonasal mass) in all embryos and cultures, and there was no difference in expression patterns between normal and cpp mutant embryos or cultures (Table 6, Fig. 12E,F,I,J). Shh was expressed in the epithelium of the caudal frontonasal mass and stomodeal cavity, and again no difference in expression between normal and cpp mutant cultures, was noted. Fgf8 is usually expressed in the epithelium of the nasal pits and between the maxillary process and mandibular arch, as well as in the forebrain and mid-hindbrain isthmus (Richman et al., 1997; Fig 12C). This expression was maintained in culture (Table 6, Fig. 12G). In cpp mutant embryos, expression extends throughout the frontonasal mass epithelium and is most intense in the epithelium along the cranial margin, overlying the ectopic ridge. This remained the case in cultures of cpp mutant whole face or frontonasal mass (Table 6, Fig 12P). 74 o a. W w • oo o to s o a cu W) o cu B o tj "a. o. s "3 u CU B 4) Ml Ml S • mm 13 B M) es C f l C5 B O *J S o o E o cn cn CU U CL. X u o Z o o Z Z CN ' CN ^5 3 ca g o z O CN CN (=1 O 9-x CJ O CN CN o x a> o Z g g 6 ca ca <u Cl o T3 X ) O O g 6 s 6 o O o C/3 oo T3 T3 •a a 3 o o o z o c p 00 • CA <G ^ 0) WD 1- rv. c H H TO w o z cn cn 53 O ID O CN CN O Z CJ CJ 6 s o o o CO + & + + Z 2 o O O z Z CN CN •o CN CN o cj o Z Tt T t Hours Hours Tt CN Tt CN Tt CN Tl-CN Tt CN Tt CN Tt CN Tt CN Tt CN 00 T t T t CN Tt CN 00 Tt T t CN tage 00 CN 00 CN OO CN 00 CN oo CN 00 CN 00 CN O CN O CN o CN Tt CN O CN o CN Tt CN GO o X> B « X i X i GO i~ o cu S 01 u s o ca fe cj o ca CJ CJ ca u o ca fe cu c. >-> o c cu 0. o 13 -3 z z 13 2 o z o z o z o z •a «s cu pa § o O p. s o o o o o p a u CJ o I o O 00 oo 00 fe a o O fe fe . fe 5T CJ CU CJ O CJ o ca ca ca fe fe fe o Z 73 o o Z Z 3 33 CO GO 7 5 Epithelium is required for the maintenance and regulation of some but not all mesenchymally-expressed genes in the frontonasal mass An epithelial requirement for maintenance of mesenchymal gene expression was tested by stripping the frontonasal mass of epithelium and culturing the isolated mesenchyme. Cultures of frontonasal mass mesenchyme from which the epithelium had been stripped continued to express AP2 and Msxl in their characteristic bilateral domains. This was the case for cultures from both normal and cpp mutant embryos (Table 6, Fig. 12I,J). Interestingly, there is a requirement for the epithelium to maintain expression of AP2 in frontonasal mass mesenchyme isolated at stage 24 (Plut and Richman, unpublished results), but by stage 28 this is no longer the case. Epithelium is not required to direct bilateral AP2 expression within frontonasal mass mesenchyme In order to maximize the number of cultures obtained from cpp mutants, the frontonasal mass of some embryos was bisected (so that each half contained one nasal pit but not the lateral nasal prominence) and the halves cultured separately. A frontonasal mass half cultured with the epithelium intact showed a new expression domain adjacent to the cut edge. The bilateral expression of AP2 indicated that the pattern observed in a full frontonasal mass was re-established in a half (i.e. the midline was re-set). Next, an epithelial requirement for this repatterning was tested in frontonasal mass halves cultured without epithelium. Again, after 24 hours in culture bilateral expression patterns of AP2 and Msxl were observed (Fig. 121; Table 6). The maintenance and regulation of bilateral expression for these genes is therefore independent of frontonasal mass epithelium by stage 28. Cultures of smaller fractions (lateral quarters) of frontonasal mass mesenchyme did not express either AP2 or Msxl (data not shown), so for all subsequent experiments the frontonasal mass was cultured either whole or halved. 76 Figure 12: Culture of stage 28 normal and cpp mutant frontonasal mass shows that epithelium is not required to maintain AP2 or Msxl, nor does exogenous SHH protein affect their expression Expression of AP2, Msxl, Fgf8 and Shh mRNA visualized with a whole mount in situ hybridization. Note that beads in E and H fell out during the in situ hybridization. Scale bar in A=l mm and applies to A-D. Scale bar in E=500pm and applies to E-P. Expression indicated with arrows. A-D: normal expression in the normal stage 28 chicken embryo. A, B: AP2 and Msxl are expressed in the mesenchyme of the globular processes. Msxl is also evident in the lateral nasal and maxillary processes. C: Fgf8 is expressed in the epithelium lining the nasal pits. D: Shh is expressed in the epithelium of the caudal margin of the frontonasal mass and of the stomodeal cavity. E-H: Expression of all genes in the intact frontonasal mass is maintained after 24 hours in culture. E: AP2 expression is maintained in the lateral regions of the frontonasal mass. Note that the mesenchyme is growing out from under the caudal edge of the frontonasal mass epithelium caudally. F: Msxl is visible in the caudal lateral corners of the culture, although the morphology of the globular processes is not maintained. G: Fgf8 expression is visible in the epithelium of the left nasal pit; the right nasal pit epithelium of this specimen also expresses Fgf8 but can only be seen by turning the culture on its side. H: Shh expression is visible in the caudal frontonasal mass. Note that expression is confined to the medial epithelium, as in the intact face. I, J: Cultures of isolated frontonasal mass mesenchyme show normal expression of AP2 (I) and Msxl (J). Note that I is a frontonasal mass half, and expression of AP2 is bilateral as in the whole frontonasal mass. Beads are control beads soaked in PBS with 1% BSA. K-M: Normal frontonasal masses cultured for 24 hours with two SHH beads show no difference in expression of AP2 (K), Msxl (L) or Fgf8 (M). Two beads were placed over a slit in the epithelium of each culture. One bead in K has sunk into the mesenchyme. Note that L is a frontonasal mass half, which expresses Msxl bilaterally as in a whole frontonasal mass. 7 7 N-P: Cpp mutant frontonasal masses cultured for 24 hours with two SHH beads. No differences in expression from the normal were observed. N: Frontonasal mass morphology is evident in this culture, as in O. The nasal pits can be distinguished in O, as can Msxl expression in parts of the lateral nasal processes included in the culture. P: Note Fgf8 expressed throughout the frontonasal mass epithelium (compare with M). Expression is unchanged near the SHH beads. Key: fnm, frontonasal mass; lnp, lateral nasal process; mx, maxillary process; s, stomodeum. 7 8 Neither Fgf8 nor Shh are autoregulated in the stage 28 frontonasal mass The effects of FGF8 have previously been examined both in vivo and in culture. One of the main uses for the culture technique has been to allow 'add-back' experiments in which the epithelium is removed and the effects of proteins on gene expression in the mesenchyme may be examined one at a time. The experiments described here, however, were designed to test the effect of exogenous proteins and Shh antibody on Shh and Fgf8, which in the frontonasal mass are expressed in the epithelium. These experiments were therefore done on face cultures with the epithelium intact. The presence of the epithelium was not expected to confound the results, as in vivo studies in which the epithelium is not stripped have also examined the effect of several FGFs on transcription of downstream genes, and have demonstrated that exogenous FGF8 can induce, for example, Lhx6 in mandibular mesenchyme and Clim2 in both mandibular and limb bud mesenchyme (Tucker et al., 1999). Whole mount in situ hybridization demonstrated that neither Fgf8 nor Shh expression was changed by addition of SHH protein (Fig. 12 M-P, Table 6), FGF8 protein or Shh Ab (Fig. 13, Table 6). The culture system did, however, support normal changes in transcription; tissue from stage 20 embryos cultured for 24 hours expressed Fgf8 throughout the frontonasal mass epithelium, similar to the pattern seen at stage 20 in vivo (n=4; Fig. 13 C,G). After 48 hours of incubation, Fgf8 expression in cultures becomes restricted to the epithelium surrounding the nasal pits, as in stage 24 embryos (Table 6, Fig. 13 A). Tissue from stage 24 or 28 normal embryos expressed Fgf8 only in the nasal pit epithelium (Fig. 13E). This demonstrates that normal downregulation of Fgf8 occurs within 24 to 48 hours in culture, slightly later than would be expected in vivo. Any effect of the exogenous proteins or antibody on Fgf8 or Shh transcription would therefore have been observed. Exogenous application of SHH to the nasal pit truncates the frontonasal mass-derived premaxillary bone SM-expressing cell pellets have been shown to induce an ectopic upper beak when injected into the chicken frontonasal mass (Hu and Helms, 1999). The possibility that SHH might induce normal outgrowth and rescue the cpp defect was therefore tested 80 by implanting SHH-soaked beads into the stage 28 frontonasal mass in ovo and examining the facial skeleton at stage 38. SHH-soaked beads implanted into the midline of the frontonasal mass (n=4) or the right nasal pit (n=l) had no effect on stage 28 cpp mutant embryos. SHH beads implanted into the midline of the frontonasal mass of normal stage 28 embryos also had no effect (n=10), but beads implanted into the right nasal pit resulted in a truncation of the premaxillary bone on the right side and a deviation of the upper beak toward the left (4/9, Fig. 14). This phenotype varied from a slight truncation of the premaxillary bone visible only after clearing the specimen (2/9) to a visible cleft on the right and substantial deviation of the upper beak (2/9, Fig. 14). The prenasal cartilage, though deviated in the two most severely affected specimens, was of normal length and morphology (Fig. 14), and no other skeletal elements were affected. The lack of phenotype after exogenous SHH was implanted into the midline implies that the absolute level of SHH protein is less significant than the ectopic presence of SHH in the right nasal pit, where it is not usually present at any stage (Fig. 10A-C). 81 Figure 13: Absence of an FGF8-SHH feedback loop in the frontonasal mass Exogenous FGF8 and Shh antibody were added to face cultures from stage 20 and 24 normal chicken embryos. Scale bar in A=l mm and applies to all. A, B: Control cultures from stage 20 embryos incubated for 24 hours with beads soaked in 1% BSA in PBS. A: Fgf8 expression is highest in the epithelium surrounding the nasal pits but is also present in the frontonasal mass epithelium. Strong expression in the forebrain may be seen through the surface epithelium. B: Shh is expressed strongly in the epithelium of the stomodeal cavity and is also present in the caudal frontonasal mass epithelium. C, D: Two FGF8 protein-soaked beads were added to stage 20 faces cultured for 24 hours. One bead was placed at the cranial margin and one at the stomodeal margin of the frontonasal mass. Expression of both Fgf8 and Shh are normal. Pins are used for photos only as cultures curl during in situ hybridization. E,F: FGF8 protein beads were added as above to stage 24 faces cultured for 24 hours. E: Nasal pits are narrowed to slits and prominences are closely apposed, indicating that the face is close to stage 28. Expression of Fgf8 is normal and can be seen in the epithelium at the junction of the maxillary process and mandibular arch. F: This culture also appears to be close to stage 28, and Shh is restricted to the stomodeal epithelium, as in vivo. G,H: Shh Ab-soaked Affigel blue beads were added to cultures as above. The cranialmost bead in G and H has fallen off during in situ hybridization. G: Fgf8 expression is normal for a stage 20 culture and is unchanged near the bead. H: Shh is expressed normally, including around the bead. I,J: Shh-Ab soaked beads were applied to stage 24 face cultures as above. I: No effect is seen on Fgf8 expression. In this specimen the beads have rolled together during culture. J : No effect is seen on Shh expression in the stomodeal epithelium. The cranialmost bead has fallen off the culture during in situ hybridization. Key: e, eye; fb, forebrain; fnm, frontonasal mass; lnp, lateral nasal process; md, mandibular arch; mx maxillary process. Asterisk indicates right nasal pit in all cultures. 82 Figure 14: Exogenous SHH causes truncation of the premaxillary bone SHH-soaked bead applied to the midline (A-C) or right nasal pit (D-F) of normal embryos or cpp mutant embryos (G-I) at stage 28. A,D,G are lateral views of the right side. B,E,H are superior views of the upper beak. C,F,I are views of the palatal surface of the upper beak with the lower beak removed. Asterisk indicates missing bone (cleft). Scale bar in A= 1cm and applies to all. A, B, C: SHH bead implanted into the midline of the frontonasal mass has no effect on skeletal derivatives. Note that the left jugal was pulled away from the quadratojugal during clearing and staining. D,E,F: SHH bead implanted into the right nasal pit causes a truncation of the premaxillary and maxillary bones on the right (implanted) side. Note that although the prenasal cartilage is deviated to the left, it is of normal length. The nasal conchae are also similar size and symmetrical. G,H,I: SHH bead implanted into the midline of a cpp mutant embryo at stage 28 has no effect. The premaxillary bone and prenasal cartilage are absent and the nasal conchae are widened symmetrically as in untreated cpp mutant embryos (compare to Fig. 2D) Key: f, frontal; ios, interorbital septum; j, jugal; mx, maxillary bone; n, nasal bone; nc, nasal conchae; p, palatine bone; pmx, premaxillary bone; pnc, prenasal cartilage; qj, quadratojugal. 84 DISCUSSION Examination of candidate signalling genes in cpp mutant embryos showed that only expression of Fgf8, an epithelially secreted growth factor, was affected by the mutation. This is in accordance with grafting studies (Chapter 2), which demonstrated that the frontonasal mass epithelium is the only tissue affected by the cpp mutation. Fgf8 is normally expressed throughout the frontonasal mass until stage 20 (similar to previous reports of expression at stage 18, Vogel et al., 1996; Schneider et al., 2001). At stages 24 and 28, Fgf8 is restricted to the epithelium bordering the nasal pits in normal but not in cpp embryos. Expression of members of the Msx, Bmp, and Shh signalling families, as well as other mesenchymal genes, is unchanged in mutant embryos. Culture experiments demonstrated that expression of AP2 and Msxl in the stage 28 frontonasal mass mesenchyme is independent of the epithelium. Application of SHH and FGF8 protein and Shh Ab in culture showed that at the stages examined, Shh and Fgf8 transcription are independent of one another. Finally, implantation of SHH beads into stage 28 embryos in ovo revealed a position-dependent effect of Shh signalling on morphogenesis but not outgrowth of the frontonasal mass-derived premaxillary bone. Fgf8 must be suppressed for normal outgrowth of the upper beak Fgf8 is a growth factor signalling molecule expressed in the epithelium in many regions where epithelial-mesenchymal interactions are taking place, and is implicated in cell proliferation or survival and outgrowth (limb, Crossley et al., 1996; Lewandoski et al., 2000; Moon and Capecchi, 2000; Vogel et al., 1996; pituitary, Trier et al., 1998; tooth germ, Kettunen et al., 1998). In the head, Fgf8 has roles in patterning the mid-hindbrain isthmus of the brain (Crossley et al., 1996a; Martinez et al., 1999), in early specification of the maxillomandibular region (Shigetani et al., 2000), and in later patterning of the mouse tooth (Kettunen and Thesleff, 1998). The requirement for the Fg/S-expressing apical ectodermal ridge (AER) to pattern the proximodistal axis of the limb (Crossley et al., 1996; Vogel et al., 1996) led to searches for a similar patterning region or gene in facial primordia (e.g., Helms et al., 1997; Hu and Helms, 1999), but the limited domain of Fgf8 in the face would seem to preclude a role similar to the one it plays in the limb. In fact, there have been no morphologically or molecularly distinct patterning centres 86 (like the limb AER) identified in the face to date (Wedden et al., 1988; Francis-West et al., 1998; Richman and Tickle, 1992). Rather, facial epithelium from different prominences interchangeably supports outgrowth in facial mesenchyme (Richman and Tickle, 1989), and even limb mesenchyme (Richman and Tickle, 1992). The forebrain has been advanced as a possible source of molecular signals that act on the frontonasal mass (Nasrallah and Golden, 2001). Earlier studies have shown that the brain has a mechanical influence on craniofacial morphology, and many craniofacial anomalies are associated with abnormalities in brain development (Diewert et al., 1993; Johnston and Bronsky, 1991; Patterson and Minkoff, 1985). The brain does not, however, influence proximodistal outgrowth. Substantial growth of facial prominences can occur in the limb bud system without accompanying brain tissue, as shown in this and other grafting studies; thus at these later stages of development the role of supporting outgrowth falls to the epithelium of each prominence (Wedden, 1987; Richman and Tickle, 1989, 1992). The ectopic Fgf8 expression observed in the cpp mutant does not necessarily prove that FGF8 protein is synthesized and is functioning in the same region of the face, but it is an essential step in identifying the downstream events in the'mutation. As the only molecular effect of the cpp phenotype identified is persistence of Fgf8 throughout the entire frontonasal mass epithelium, the cpp mutation may be in a regulatory factor that restricts Fgf8 to the nasal pit epithelium after stage 20. Fgf8 is expressed normally in cpp mutant embryos outside of the frontonasal mass, e.g. in the epithelium at the juncture of the maxillary process and mandibular arch, in the limb and in the brain. Normal expression of Fgf8 in these regions and normal function of the FGF8 protein elsewhere in the embryo suggest that the mutation is not in the Fgf8 coding region. A possibility that would allow for the abnormal frontonasal mass expression, but normal expression and function elsewhere, is a mutation in a promoter or enhancer element in the Fgf8 gene. One promoter region, identified in the murine Fgf8 gene in an ES (embryonic stem) cell line , contains DNA-binding sites for several transcription factors including AP2, Msx2 and Engrailed (Gemel et al., 1999). Significantly, Engrailed is expressed prior to Fgf8 in the mid-hindbrain isthmus during brain development (McMahon et al., 1992) and in the limb (Loomis et al., 1996; 2001), regions where it is postulated to restrict the expression 87 of Fgf8 (Lee et al., 1997; Loomis et al., 1996). A mutation in either the Fgf8 promoter region or in a transcription factor that binds to the promoter could alter spatial or temporal aspects of Fgf8 expression without affecting function of the protein, and would explain why the cpp phenotype is confined to the frontonasal mass. Aside from transcription factors, FGF8 function is modulated through many mechanisms, one of which is a positive feedback loop autoregulating transcription (Crossley et al., 1996). Any local regulation of FGF8 may therefore indirectly affect transcription. For example, there are eight different isoforms of the murine FGF8 protein, which have considerable although not complete redundancy of function (Blunt et al., 1997). A requirement for heparin/heparan sulfate in binding of FGF to FGF receptors (FGFRs) has been shown for FGF1, FGF2, and FGF4 (Guimond et al., 1993), suggesting that local patterns of heparin/heparan sulfate-containing cell surface proteoglycans could play a role in Fgf signalling (Dealy et al., 1997; Ornitz, 2000; Sherman et al., 1998; Tucker et al., 1999). Ectopic expression of Fgf8 could be confined to the frontonasal mass epithelium through any one of these mechanisms. Cpp was first isolated against the background of a mutation, ectrodactyly, that produced birds with limb truncations (loss of proximal bones and digits in the leg but not the wing), as well as the upper beak truncation seen in the cleft primary palate phenotype (Abbott and MacCabe, 1966). It was subsequently found that the limb defect {ectrodactyly) is only present when birds from the cpp line are crossed with another mutation, scaleless. It should be noted that scaleless birds lack scales, foot pads, spurs and most feather follicles, which are all epithelial specializations, but limb and face development is normal (Abbott and MacCabe, 1966; Goetinck and Abbott, 1963). Scaleless was studied using tissue recombinations, and the defect was found to be in the ectoderm (Goetinck and Abbott, 1963). Ectrodactylous embryos therefore have defects in face, hindlimbs and epithelial specializations, and while the face and epithelial defects can occur as single mutations {cpp and scaleless, respectively), the limb defect is only seen against a scaleless background. Although molecular signalling in ectrodactyly has not yet been examined, it is tempting to speculate that Fgf8 signalling in the limb AER is affected; consider that the ectopic Fgf8 expression in frontonasal mass epithelium shown in this thesis for cpp should be apparent in the frontonasal mass of ectrodactyly embryos 88 as well, since the ectrodactyly genotype consists simply of co-occurrence of the cpp and the scaleless mutations (Abbott and MacCabe, 1966; Yee and Abbott, 1978). Existence of cpp as a mutation separate from ectrodactyly may indicate that there are separate regulatory elements controlling expression of Fgf8 in the AER and in the face. Neither Fgf8 nor the cleft primary palate mutation is upstream of Bmp or Msx signalling in the stage 20-28 face To discuss the finding that the cpp mutation affects Fgf8 alone of all the genes examined, I first address a possible limitation of the technique. In situ hybridization identifies areas of mRNA transcription, and as such permits only an indirect observation of molecular signalling. Expression in a given region of the embryo (as detected by in situ hybridization) does not necessarily imply that the protein products will be synthesized, or that if synthesized they will function in the same area, as secreted signalling molecules such as the Bmps, Fgfs and Shh may participate in long-range signalling interactions. In situ hybridization is, however, an indication that a signalling event is underway. Cells expressing specific mRNA transcripts are able to synthesize those proteins without further inductive interactions, which is to say the cells are determined. Conversely, because transcription is a necessary first step in protein synthesis, lack of expression in a region where it might be expected can predict the later absence of that protein (molecular signal). Similarly, failure to detect the mRNA of intracellular or nuclear products such as the transcription factors AP2, Msxl and Msx2 implies that these peptides are not active, since they function at their sites of transcription. Changes in mRNA expression are therefore useful indications of later changes in molecular signalling events, especially when comparing groups of embryos with known differences (such as stage or phenotype). Bmp4 is a secreted signalling molecule, so does not necessarily function in the same region in which it is synthesized. In the face, it is produced in the epithelium but acts primarily on the underlying mesenchyme; this interaction has been studied extensively with respect to tooth induction (Thesleff and Sharpe, 1997; Tucker et al., 1998a; Tucker et al., 1998b; Vainio et al., 1993; Wang, 1999). BMP4 antagonizes FGF8 in a feedback loop that establishes boundaries in several phases of face development, 89 such as early patterning of the maxillomandibular region (Shigetani et al., 2000), mandibular arch (Stottmann et al., 2001), and tooth formation (Neubuser et al., 1997; Tucker et al., 1998a). In the mandible, the two genes are expressed in mutually exclusive regions; Fgf8 in the lateral and Bmp4 in the medial epithelium, where it is thought to antagonize Fg/8-induced Pax9 expression in specifying sites of tooth induction (Neubuser et al., 1997). In the frontonasal mass epithelium, unlike the mandibular arch, there is overlap between the expression domains of Fgf8 and Bmp4 (Fgf8, Richman et al., 1997; Bmp4, Francis-West et al., 1994; Ashique et al., 2002 and compare Figs 9K, 11B). At stages 24 and 28, Fgf8 and Bmp4 expression overlaps in the epithelium at the lateral corners of the frontonasal mass, corresponding to the globular processes. Fgf8 extends laterally from this area of co-expression into the epithelium lining the nasal pits on both the frontonasal mass and lateral nasal processes. Bmp4 extends medially throughout the caudal edge of the frontonasal mass epithelium. The expression of Bmp4 and Fgf8 in the frontonasal mass epithelium therefore approximately delineates lateral (Fgf8) and medial (Bmp4) domains along the caudal frontonasal mass. That co-expression is seen in the epithelium of the globular processes is significant because strict control of mesenchymal proliferation and epithelial cell death appears to be required in this area for fusion of the frontonasal mass and lateral nasal prominences to occur (McGonnell et al., 1998; Minkoff and Kuntz, 1977). Overexpression of the Bmp antagonist noggin in the limb bud led to persistence of the AER, including spatial and temporal extension of Fgf8 expression (Pizette and Niswander, 1999). Downregulation or antagonism of Bmp4 might therefore lead to ectopic Fgf8 expression in the cpp mutation. However, expression of Bmp4 and noggin were unchanged in cpp mutant embryos, so that Bmp4, noggin and Fgf8 were coexpressed along the caudal margin of the frontonasal mass epithelium at stage 24. Thus, failure to downregulate Fgf8 in the cpp frontonasal mass is not due to a decrease in BMP signalling. Aside from its role in boundary specification, BMP4 also induces chondrogenesis (Buckland et al., 1998; Semba et al., 2000). Chondrogenesis of the mesenchyme is unimpeded in the cpp mutation, as shown first by normal expression of collagen type II in cpp mutant embryos, later by outgrowth of cartilage rods in recombinations of cpp 90 mesenchyme and normal epithelium, and finally by the cartilage and bone seen in cleared preparations of stage 38 cpp mutant embryo heads. As mRNA expression of Bmp4, craniofacial development until stage 28 (including maxillomandibular specification), and chondrogenesis are all normal, BMP signalling appears to be unaffected by the cpp mutation. The normal patterns of Msxl and Msx2 expression observed in the cpp mutant embryos for each of these genes show that they function in the same regions of the cpp frontonasal mass as in the normal. Msxl is consistently expressed in lateral mesenchyme of the frontonasal mass, roughly correlated with Fgf8 in the overlying nasal pit epithelium, and this expression pattern has been used to implicate Msxl in mesenchymal cell proliferation in this region (McGonnell et al., 1998). Both genes, however, are absent in the highly-proliferating ectopic ridge mesenchyme of cpp mutants, and are present in their normal patterns in lateral frontonasal mass mesenchyme. The normal mesenchymal expression of Msxl and Msxl, and the observation that the mesenchyme of cpp mutant embryos responds normally to normal epithelium, suggests it is unlikely that Msxl and -2 are downstream of the cpp mutation. It has been proposed that the role of Msxl and -2 is to suppress cellular differentiation, thereby indirectly encouraging cell proliferation (Zhang et al., 1997). Msxl and —2 may therefore act to support proliferation in facial mesenchyme during normal development, without necessarily being required to induce ectopic proliferation in the cpp mutant embryos. Shh has a role in normal patterning of the upper face but is not implicated in the cpp mutation It has been hypothesized that both Fgf8 and Shh are downstream of a retinoid signalling pathway during a specific window in early face and forebrain development. This was demonstrated by the rescue of RA antagonist-induced defects with implantation of SHH+FGF8-soaked beads (Schneider et al., 2001). The retinoid antagonist was most effective prior to stage 14, suggesting that subsequent control of facial growth may be retinoid-independent (Schneider et al., 2001). The continued expression of Fgf8, Shh and Ptcl until stage 28 in normal embryos, however, implies that these genes continue to function whether within an endogenous retinoid pathway or not. Attempts to rescue 91 retinoid defects with either SHH or FGF8 have not been reported, and it is not known what interactions, if any, the Fgf8 and Shh signalling pathways undergo in the face. Fgf8 and Shh have been shown to act synergistically in the limb (Crossley et al., 1996; reviewed in Capdevila and Izpisua Belmonte, 2001) and in neurogenesis (Ye et al., 1998). They are induced separately, however, as shown by Fgf8 expression in the single nasal placode of the Shh-/- mouse face (Chiang et al., 1996) and in Shh -I- mouse limbs (Sun et al., 2000b), and by the presence of Shh in mouse limbs lacking Fgf8 (Fgf8 conditional knockouts: Lewandoski et al., 2000; Moon and Capecchi, 2000). While a specific role for Fgf8 in upper face patterning has not yet been established (Trumpp et al., 1999), Shh signalling determines the midline of the upper face (Chiang et al., 1996; Hu and Helms, 1999; Schneider et al., 2001). In the cpp mutation, however, the midline is intact. The width of the face is the same as in normal embryos, nor is there any effect of the mutation on the morphology of the cranial vault or eyes. Furthermore, expression of Shh and its receptor Ptcl is normal in cpp mutant embryos, in which Fgf8 is ectopically expressed until stage 28. Finally, expression of Shh was not affected by ectopic FGF8 in organ cultures of the face. These observations demonstrate that Shh is not downstream of Fgf8 at the stages examined. A separate possibility is that Fgf8 is downstream of Shh. If this were the case, the ectopic Fgf8 expression seen in cpp mutant embryos at stages 24-28 could result from an earlier alteration in Shh signalling. However, the normal expression of Fgf8 seen in normal faces cultured with SHH or with Shh antibody demonstrates that Fgf8 transcription is not regulated by Shh signalling. SHH is necessary for formation of the upper face (Chiang et al., 1996), and localized increases in SHH can induce ectopic outgrowths containing prenasal cartilage and an egg tooth in the frontonasal mass (Hu and Helms, 1999). The possibility that exogenous SHH could counteract the truncation of proximodistal outgrowth observed in the cpp mutation was therefore tested. Previous SHH gain-of-function experiments which led to ectopic outgrowth from the frontonasal mass were conducted at stages 20-21 and were correlated with increased cell proliferation followed by a widening of the frontonasal mass along the medial-lateral axis (Hu and Helms, 1999). Significantly, the formation of a SHH-induced ectopic outgrowth from the frontonasal mass did not 92 interfere with development of the normal upper beak skeleton, which appears to form and articulate properly with derivatives of the maxillary process (Fig. 5 in Hu and Helms, 1999). However, data from SHH beads implanted into the frontonasal mass of stage 28 cpp mutant embryos demonstrates that SHH delivered using this method is not sufficient to induce outgrowth of facial prominences in this system. Further, the clefting observed after implanting SHH-soaked beads into the right nasal pit of stage 28 normal embryos suggests that SHH does not simply induce growth. Clefting may result from the failure of facial prominences to grow sufficiently to meet, a mechanism that is postulated to explain Shh antibody-induced clefts, in which the premaxilla and maxillary bone are both truncated and fail to articulate (Hu and Helms, 1999). However, clefts also occur when prominences meet and fail to fuse. It is possible that the late SHH gain-of-function in the present study interfered with fusion of the prominences, perhaps by supporting persistence of the epithelium. Such a mechanism would explain why implants into the nasal pit but not the midline of the frontonasal mass resulted in truncations of the premaxilla and maxillary bone, clefts very similar to the Shh antibody (loss-of-function) phenotype observed by Hu and Helms (1999). The cpp mutation does not affect Shh or Ptcl expression, or the medial-lateral axis (width) of the frontonasal mass. Exogenous SHH does not rescue the truncated upper beak of cpp mutant embryos, which fails to grow out despite adequate width. These studies on the cpp mutation may therefore support a model of facial patterning in which Shh signalling patterns frontonasal mass growth along a medial-lateral axis, while Fgf8 directs outgrowth along the proximodistal axis (Fig. 15). This investigation of molecular signalling in the cpp mutation suggests a previously unsuspected requirement for Fgf8 downregulation after stage 20 in frontonasal mass outgrowth. The role of Fgf8 signalling in normal cell proliferation and outgrowth will be further addressed in Chapter 5, the general discussion. 9 3 Figure 15: Molecular signalling and cell proliferation in the stage 28 normal and cpp mutant embryo frontonasal mass Patterns of cell proliferation (Chapter 2) and gene expression (Chapter 4) are represented in schematic diagrams of the stage 28 normal (A,C) and cpp mutant (B,D) face. A and B are camera lucida tracings of a frontal view, while C and D are enlarged schematic diagrams representing a sagittal section through the frontonasal mass along the midline (at level of lines in A, B). Fgf8 is shown in green (small arrows), Shh by large dark arrows, and cell proliferation by orange asterisks. Circles correspond to the chondrogenic region of the frontonasal mass as visualized by type JJ collagen mRNA and by a relatively low level of cell proliferation. A: Fgf8 expression (small arrows) in the upper face of the normal embryo is confined to the epithelium of the nasal pits. Shh (large arrows) is expressed in the stomodeal epithelium and along the caudal margin of the frontonasal mass. Note that expression of Fgf8 in the frontonasal mass epithelium is continuous with. Shh, and that areas of relatively high cell proliferation in frontonasal mass mesenchyme (asterisks) correspond with epithelial expression of these two genes. B: In the stage 28 cpp mutant embryo, Shh is expressed in its normal domain (as is its receptor Ptcl, not shown). Note that the width of the frontonasal mass is similar to a normal embryo. Fgf8, however, is expressed throughout the frontonasal mass epithelium (stripes), and cell proliferation (asterisks) is high throughout the underlying mesenchyme. C, D: In the normal stage 28 embryo (C), cell proliferation (asterisks) is induced only in the lateral and caudal edges of the frontonasal mass, while the centre of the frontonasal mass undergoes chondrogenic condensation (large circle) and outgrowth. Cells may proliferate and move into the central region of the frontonasal mass from the lateral (short green arrows in A and C) and caudal (long arrows in A and C) margins. D: Ectopic Fgf8 throughout the frontonasal mass epithelium causes widespread proliferation in the underlying mesenchyme (short arrows). Condensing cells in the centre of the frontonasal mass continue to proliferate at low levels (circle), but may be blocked from growing along a proximodistal axis by the surrounding proliferative, undifferentiated cells which do not contribute to the chondrogenic condensation (see Chapter 5, general discussion). 94 95 Chapter 5: General discussion The cpp mutation affects the epithelium but not the mesenchyme of the chicken frontonasal mass through a failure to downregulate Fgf8, a secreted growth factor that has roles in epithelial-mesenchymal interactions directing patterning and outgrowth throughout development. Persistence of Fgf8 throughout the frontonasal mass epithelium is correlated with an increase in cell proliferation throughout the underlying mesenchyme, which appears to disrupt normal patterning of the frontonasal mass to such a severe extent that all skeletal derivatives of this prominence are lost. The cpp mutation demonstrates that substantial molecular control is required for proper morphogenesis of the facial prominences, and provides a system within, which to dissect the signalling interactions that direct growth and patterning in the face. Fgf8 induces cell proliferation during limb bud development The best evidence for the role of Fgf8 in inducing cell proliferation comes from studies of the limb bud. FGFs produced in the AER of the limb maintain the underlying progress zone mesoderm in a proliferative, undifferentiated state (Cohn et al., 1995; Reiter and Solursh, 1982; Summerbell et al., 1973). An increase in the range of Fgf8 expression along the flank, accomplished with a retroviral vector, leads to subsequent formation of ectopic limb buds (Crossley et al., 1996; Vogel et al., 1996) or excess soft tissue at the distal tip of the limb (Pizette and Niswander, 1999). Formation of ectopic tissue is associated with maintenance of proliferation over a greater area of the flank in at least one of these systems (Pizette and Niswander, 1999), although migration of flank cells into the presumptive limb bud area is also a factor (Tanaka et al., 2000). The study of limb patterning that most closely examined proliferation found a positive correlation between Fgf8 expression in the AER and proliferation in the underlying mesoderm, labelled with BrdU. This correlation was observed in both normal and ectopically Fgf8-expressing limb buds (Pizette and Niswander, 1999). Formation of an ectopic AER on the dorsal or ventral surface of the limb bud results in a secondary axis, with duplication of skeletal elements. Ectopic AERs have been described in mutations such as eudiplopodia (Goetinck, 1964) as well as in surgical 96 manipulations (Carrington and Fallon, 1986). In these systems, it is assumed that the ectopic AER secretes Fgf8\ although this was not addressed directly with in situ hybridization experiments, a later study in which ectopic AERs were induced with FGF8 does show Fgf8 in the ectopic AER (Crossley et al., 1996). During normal development the AER maintains the underlying mesenchyme in a proliferative, undifferentiated state (Reiter and Solursh, 1982; Solursh et al., 1981; Summerbell et al., 1973), while axial (AP, DV and PD) patterning and differentiation events are induced by other pathways such as the retinoid-Shh cascade (reviewed in Capdevila and Izpisua-Belmonte, 2001). Fgf8 therefore supports proliferation (as well as cell survival, Trumpp et al., 1999) both in normal development and in cases where an ectopic AER is formed. Ectopic Fgf8 expression in cpp mutant embryos is correlated with increased cell proliferation The importance of cell proliferation to normal face morphogenesis was well established long before any molecular signals inducing proliferation were characterized (Minkoff, 1984; Minkoff, 1991; Minkoff and Kuntz, 1978; Minkoff and Martin, 1984). Differential rates of proliferation were shown to be the mechanism for the initiation of outgrowth in the limb bud. In that system, cell proliferation in flank mesoderm decreases sharply at stages 17-18, while proliferation in the presumptive limb region is maintained (Searls and Janners, 1971). This mechanism was later observed in the face, where non-bud mesenchyme undergoes a similar decline in proliferation, accompanied by invagination of placodal ectoderm to form grooves between the bud-like prominences (Minkoff and Kuntz, 1977; 1978). Maintenance of proliferation in the presumptive limb area of the flank is due to FGFs secreted from the AER (Cohn et al., 1995; Crossley et al., 1996b; Niswander et al., 1993). Proliferation of facial mesenchyme is dependent upon facial epithelium, a dependence that lessens as development proceeds (Saber et al., 1989), but maintenance of proliferation in the face has not yet been examined specifically with respect to Fgf8 signalling. In normal embryos, all frontonasal mass epithelium expresses Fgf8 until approximately stage 20 (facial prominences at this stage are well-established buds) after which expression is restricted to the epithelium adjacent to the nasal pits (Richman et al., 97 1997; Schneider et al., 2001; Vogel et al., 1996). This Fgf8 expression pattern overlies areas of high proliferation in the mesenchyme of both the frontonasal mass and the lateral nasal processes in the face (McGonnell et al, 1998 and the present study; see Figs. 6, 7, 10, 15). In cpp mutant embryos, restriction of Fgf8 to the nasal pit epithelium does not take place, and the underlying mesenchyme then proliferates at a higher rate than in similar regions of normal embryos. As FGF8 acts as a mitogen in many developing systems (Cristen and Slack, 1997; Kettunen et al., 1998; Mina et al., 2002; Pizette and Niswander, 1999; Vogel et al., 1996), a conclusion of the present study is that the extension of the Fgf8 domain observed in the cpp mutant is responsible for the increase in cell proliferation in the underlying mesenchyme. Increased cell proliferation led to formation of the ectopic ridge observed at stage 28, however the fate of cells in this region was not followed. Although programmed cell death does not appear to initially influence the morphology of the mutant frontonasal mass, it may play a later role. Selective restriction of proliferation may be required to direct outgrowth of the facial prominences Compared with the unidirectional outgrowth and divergence of a limb bud, which articulates with other structures only at the base, outgrowth in the face is complicated by the requirement for several buds (prominences) to converge onto a central focal point (e.g. the distal tips of the upper and lower beak in birds) and eventually articulate with one another at both ends. Regional restriction of proliferation between and within the prominences may therefore be a major mechanism of patterning in the face, as opposed to the widespread induction of proliferation seen along the entire AP axis of the limb bud. Proliferation in the mesenchyme at the bases of the prominences, pushing outgrowth proximal to distal, might therefore contribute more effectively to face outgrowth than proliferation at the distal tip, the major mechanism of growth in the limb bud (Reiter and Solursh, 1982). Previous observations that normal facial epithelium of any facial prominence is interchangeable for the purpose of promoting outgrowth (Richman and Tickle, 1989; 1992), are reaffirmed by the failure of cpp mutant epithelium to support proximodistal outgrowth in both frontonasal mass and mandibular mesenchyme. The facial epithelia 98 homogeneously support outgrowth, thus the persistence of individual prominences throughout a substantial period of development (stages 18-28; approximately three days) appears to come about through a restriction of growth in the junctions between each prominence. In one study of facial prominence outgrowth and fusion, a negative correlation was described between a gap junction protein, connexin 43, and areas of outgrowth; conversely, high expression of connexin 43 was seen during fusion of the prominences (Minkoff et al., 1997). This is in contrast to the relatively unrestricted growth seen in a single-bud system (the limb), and suggests that complex control is required to suppress outgrowth between the facial prominences, and/or to direct outgrowth along the proximodistal axis of each prominence. Proliferation concentrated at the base or sides of the prominences rather than the distal tips has been observed previously (McGonnell et al., 1998; Peterka and Jelinek, 1983; Peterka et al., 1997) and confirmed for the frontonasal mass by proliferation studies in this thesis (Chapter 2). It is possible, then, that the Fgf8 normally produced only in the base and lateral epithelium of the prominences contains proliferation to these regions of the face, just as it does in its domain at the distal tip of the limb buds. The widespread cell proliferation observed in the cpp mutant frontonasal mass may preempt normal patterning, as large numbers of cells in the mutant frontonasal mass are maintained in a proliferative state. The centralmost frontonasal mass mesenchyme of the cpp embryo, which is not in contact with epithelium, has a lower proliferation index than adjacent regions. This area of low proliferation expresses type II collagen, which is characteristic of cartilagenous condensations (Hall and Miyake, 1995). In the cpp mutation the chondrogenic core appears normal, but is surrounded by the highly proliferating ectopic ridge and superficial mesenchyme. Cells in these regions of the frontonasal mass may not be capable of differentiation as long as high rates of proliferation are maintained (Stein and Lian, 1993; Zhu and Skoultchi, 2001). The fate of these ectopic proliferating cells was not followed in this study, however given the absence of frontonasal mass derivatives it is possible that they die off at a later stage. Restriction of Fgf8 to the nasal pit epithelium in the normal embryo may therefore be required to permit cells in the central frontonasal mass mesenchyme to decrease in proliferation and undergo chondrogenic condensation and differentiation (Fig. 15). 99 Consequences of Fgf8 overexpression: truncations in the limb and in the face Many of the studies demonstrating the mitogenic or inductive effects of FGF8 in the face involved add-back experiments: stripping of Fg/S-expressing epithelium and replacement with FGF8 protein-soaked beads (Kettunen et al., 1998; Mina et al., 2002; Neubiiser et al., 1997; Tucker et al., 1999b). There has been no attempt to introduce a secondary axis into a face prominence using FGF8 or any other FGF in the presence of epithelium, as there has been in the limb (Crossley et al., 1996; Vogel et al., 1996). Addition of FGF8 to the presumptive face region early in development, during neural tube closure (stage 10), however, resulted in gain of posterior (mandibular-type) structures at the expense of anterior (premaxillary) structures; primarily nerve morphology was examined, but the entire premaxillary region including the eye was reported as being reduced (Shigetani et al., 2000). Similarly, the phenotype of the cpp mutation predicts that rather than promoting outgrowth, the addition of FGF8 to the frontonasal mass of a normal embryo (i.e., at a later stage than the study by Shigetani and colleagues) would lead to a loss of frontonasal mass outgrowth along the proximodistal axis. Loss of proximodistal outgrowth (as seen in the cpp mutation) is not due to an inability of the frontonasal mass to grow along a secondary axis, as supernumary upper beaks have been induced with cell pellets expressing ectopic SHH (Hu and Helms, 1999). Rather, loss of proximodistal growth in cpp embryos may be due to proliferation superceding differentiation into cartilage. Support for the proposal that excess FGF8 leads to the truncated upper beak in the cpp mutation is provided by some of the same limb studies that showed ectopic structures can result from Fgf8 overexpression. In one of the first studies to demonstrate that Fgf8 is the endogenous signal that initiates limb development, Vogel et al. (1996) applied FGF8 protein beads to the flank and induced ectopic limb buds. Also in that study, however, was the observation that if the beads were placed too close to an existing presumptive limb field, those limbs (in this case, wings) were truncated at various points along the proximodistal axis. The investigators then grafted Fg/8-transfected cells, which provide a localized and continuous source of FGF8, into the same region. Limb truncations were again observed, this time more severe and consistent, often affecting the 100 leg as well as the wing and extending even to the limbs on the contralateral side. Finally, injections of Fgf8 virus directly into the stage 17 forelimb bud resulted in shortening of the humerus, radius and ulna. One of the conclusions from this study was that a continuous or widespread source of FGF8 (such as the transfected cells or virus) was correlated with truncations and deletions of existing limbs (Vogel et al., 1996). Although they did not examine proliferation either in the ectopic limb buds produced by the FGF8 beads, or in truncated limbs, Vogel and colleagues speculate, as I do regarding the cpp mutation, that increases in proliferation may occur at the expense of differentiation, leading to loss of skeletal elements. Exogenous FGFs have not been applied to facial prominences in vivo (the experiments of Shigetani and colleagues were done at a stage prior to formation of facial prominences), but facial prominence mesenchyme has been recombined with limb ectoderm containing the Fg/8-expressing AER (Richman and Tickle, 1992). Recombinations of mandibular mesenchyme with limb ectoderm failed to grow out, forming nodules despite the persistence of the AER for up to 48 hours (Richman and Tickle, 1992); this failure of outgrowth is similar to the result from recombinations of mandibular mesenchyme with cpp mutant epithelium (in the present study). These studies, including the present work on the cpp mutation, indicate that persistent, exogenous Fgf8 signalling leads to excess proliferation, and that if proliferation persists in a region that would ordinarily be undergoing differentiation (e.g. skeletogenesis), the differentiation does not occur. From this model, excess Fgf8 signalling might be predicted to result in truncations of outgrowth. In an examination of the role of BMPs in regression of the AER, misexpression of the Bmp antagonist noggin led to persistence of the AER, identified morphologically and from Fgf8 expression (Pizette and Niswander, 1999). Examination of proliferation in noggin-injected limb buds showed that the region of Fgf8 expression in the epithelium always corresponded with a significantly greater number of BrdU-labelled cells in the underlying mesoderm. A clear correlation was therefore shown between epithelial Fgf8 expression, both normal and ectopic, and proliferation in adjacent mesenchyme. Noggin-infected limbs exhibited digit truncations. These, however, were attributed to antagonism of Bmp-mediated chondrogenesis rather than to any effect of Fgf8 overexpression. 101 Overgrowth of soft tissue at the distal tip of the limb, which was observed in limbs with and without digit truncations, was attributed to persistence of Fgf8 signalling. The noggin-mediated repression of Bmp signalling and subsequent chondrogenesis makes it difficult to determine what contribution, if any, was made to this limb truncation phenotype by persistence of Fgf8. Proliferation data from this study, however, correlate well with data from the face (Chapter 2). Furthermore, the lack of any effect of the cpp mutation on Bmp4 and noggin expression, or on the initiation of chondrogenesis, support the explanation that in cpp embryos the primary effect of the mutation is on Fgf8 signalling. Downregulation of Fgf8 in the frontonasal mass epithelium during normal development may be due to a signal from underlying mesenchyme Epithelial-mesenchymal interactions have been discussed in terms of epithelial signals acting on mesenchyme. Mesenchyme may, however, direct changes in the overlying epithelium, and many developmental systems rely on reciprocal epithelial-mesenchymal inductions to direct differentiation of increasingly complex structures or regions (Jacobson and Sater, 1988; Thesleff and Sharpe, 1997; Vaahtokari et al., 1996). The source of a putative Fgf8 downregulator, inactive in the cpp mutation, may normally be the frontonasal mass mesenchyme. The requirement for a signal from the mesenchyme to downregulate Fgf8 in the overlying epithelium would explain why failure of outgrowth persists in frontonasal mass pieces from cpp embryos grafted intact (or as a homotypic recombination) to the host limb, while recombinations of cpp mutant mesenchyme with normal epithelium successfully grow out. Cpp mesenchyme would not produce a signal downregulating Fgf8, but normal epithelium would already have received this signal by the stage at which the tissues were recombined. Recombinations of cpp mutant epithelium with normal mesenchyme did grow out more often than isolated mesenchyme, although this difference was not found to be statistically significant. This intermediate result could be explained by a partial or late downregulation of Fgf8 in cpp mutant epithelium by normal mesenchyme. Mesenchyme from the medial region of the mandibular arches contributes significantly to overall growth and is dependent on signals from overlying epithelium to 102 support proliferation and outgrowth (Mina et al., 2002). FGFs are implicated in the epithelial induction, as medial mandibular mesenchyme grows out in response to FGF2 and FGF 4 beads in a limb graft system (Richman et al., 1997) and in culture (Mina et al., 2002). Fgf8, however, is not expressed in the epithelium of this region (Mina et al., 2002; Richman et al., 1997) nor is Fgf8 required for formation of distal mandibular structures (Trumpp et al., 1999). Medial mandibular mesenchyme is capable of responding to the proper signal for outgrowth, however that signal is not Fgf8. It is possible therefore that the failure of both normal mandibular and cpp mutant frontonasal mass mesenchyme to grow out in response to cpp mutant epithelium is due to the same mechanism, overexpression of Fgf8 (not normally present in either the central frontonasal mass or mandibular epithelium) followed by excessive proliferation instead of outgrowth. Stage 24 frontonasal mass mesenchyme recombined with limb AER grew out into prenasal cartilage-like rods (Richman and Tickle, 1992). The existence of a Fgf8-downregulating factor in the frontonasal mass mesenchyme, but not in mandibular mesenchyme, is compatible with the results of the present study and that of Richman and Tickle (1992). Comparison of the cpp mutation with other models of abnormal outgrowth Cleft lip, although it can occur as a component of genetic syndromes or as a result of exposure to teratogens, is most often nonsyndromic and multifactorial in etiology (Schutte and Murray, 1999). Animal models for these malformations include loss of function mutations (often identified as human clinical conditions); transgenic mouse models and mice with conditional deletions or single gene knockouts; teratogen-induced defects; and naturally occurring mutations, often with an unknown genotype. Through these approaches a large body of literature has been accumulated implicating failure of facial prominence outgrowth in many cleft disorders. I now consider these other models for abnormal outgrowth to highlight the unique aspects of the cpp mutation. Loss of function models Loss of function mutations in mouse and human populations have proven to be a valuable resource for teasing apart the intricacies of molecular signalling networks. 103 Some gene knockouts that produce craniofacial defects were used to choose candidate genes for this study, as outlined in the introduction to Chapter 4 (Jabs et al., 1993; Satokata and Maas, 1994; Schorle et al., 1996; van den Boogaard et al., 2000). Sonic hedgehog, a secreted signalling protein identifed as the molecular signal encoding polarizing activity of the limb bud (Riddle et al., 1993), was implicated in face patterning by both a mouse knockout (Chiang et al., 1996) and a human condition, holoprosencephaly (Belloni et al., 1996; Roessler et al., 1996). Knockouts of Fgf8 and Bmp4 cause early lethality (Sun et al., 2000b; Winnier et al., 1995) but expression in the face combined with a role in limb outgrowth indicates that they may act as face patterning genes later in development (Barlow and Francis-West, 1997; Crossley et al., 1996b; Francis-West et al., 1994; Richman et al., 1997; Vogel et al., 1996). These studies demonstrate that while human conditions and mouse knockouts can help to identify genes involved in face patterning, experimental systems that allow finer examination of molecular signalling are often required to determine gene functions and interactions. The cpp mutation, which affects expression of only a single gene in the complex network of signalling that occurs throughout facial development, could prove to be one such system. Terato gen-induced clefts Retinoid derivatives in deficiency or excess induce various craniofacial deformities in humans and mice, and have been for many years a model for teratogenic clefting (Richman, 1992; Sulik et al., 1988). Exogenous retinoic acid implanted into chicken embryos results in a truncation of the upper beak (Tamarin et al., 1984) superficially similar to the phenotype of the cpp mutation. However, as discussed in Chapter 2, the target of retinoid action in the developing upper face (stages 20-24) is the frontonasal mass mesenchyme (Wedden, 1987), while the cpp mutation affects the epithelium by stage 24. Hydrocortisone, another teratogen that can produce bilateral cleft beak in chicken embryos, produces hypoplasia of the globular processes of the frontonasal mass, which do not meet, and so fail to fuse with, the maxillary processes (Peterka and Jelinek, 1983). Hydrocortisone does not affect proximodistal outgrowth, as the prenasal cartilage in treated embryos forms normally, however both the frontonasal 104 mass-derived premaxilla and maxillary process-derived maxillary bones are truncated (Fig. 7 in Peterka and Jelinek, 1983). A similarly hypoplastic effect of corticoids, i.e. a reduction in the size of palatal shelves, was seen in mouse embryos (Diewert and Pratt, 1981). In both of these studies of corticoid teratogenicity, clefts were caused by hypoplasia but proximodistal outgrowth was unaffected. In contrast, the clefting effect of the cpp mutation appears to be secondary to loss of proximodistal outgrowth. Chicken mutations affecting outgrowth Naturally occuring chicken mutations that cause truncated limbs often also have effects in the face, suggesting some conservation of signals involved in bud outgrowth. Many appear to be single gene mutations, however as avian embryos are less amenable than mammalian (i.e., mouse) embryos to analysis of genotype, the genes affected have for the most part not yet been identified. There has been considerable examination of tissue-level interactions in mutations affecting the limb, but surprisingly little attention has been paid to the face defects. Diplopodia mutant embryos, which have extra preaxial digits on all limbs, also have shortened upper beaks (Abbott, 1983), as do some limbless and wingless embryos (Abbott and Pisenti, 1993). Similarly the talpid2 (Goetinck and Abbott, 1964; MacCabe and Abbott, 1974) and talpid3 (Ede and Kelly, 1964; Hinchliffe and Ede, 1967) chicken mutations have bilaterally cleft upper beaks (which are also shortened to varying degrees) as well as Polydactyly. Talpid2 mutants overexpress Shh and Fgf8 but not Msxl in the frontonasal mass (Schneider et al., 1999), and the beak defect appears similar to the clefts produced by exogenous SHH applied to the right nasal pit in the present study. Examination of the molecular basis for these naturally occuring chicken mutations is emerging as a extensive area of interest for developmental biologists. Separation of the cpp mutant strain from ectrodactyly, which exhibited both limb and upper beak truncations, provides an interesting system in which it may eventually be possible to closely examine and compare molecular control of Fgf8 expression and function in the face versus the limb. Perturbations such as those described above raise interesting questions about the nature of molecular control of development, questions that cannot always be addressed in 105 the system within which they arose. Teratogens, natural mutations and targeted gene deletions are all valuable tools in the study of embryonic development. It is clear from this study of a single mechanism, outgrowth, that a complete understanding of any facial defect relies on making use of all available avenues of research. Elucidating the mechanism of the cpp mutation contributes to the study of outgrowth both as a developmental process and in the genesis of facial clefting. Future directions Identification of the mechanism of the cleft primary palate mutation as an Fgf8-induced increase in cell proliferation raises questions about the control of proliferation and differentiation during development, in the face and elsewhere. The fate of cells in the ectopic frontonasal mass ridge in the cpp mutant embryos was not followed in this study, and it would be worth determining whether these cells do contribute substantially to the truncated structures of the upper beak, and if so, when. Any one of several methods of fate mapping could be attempted; for example, single-cell injection of Dil, a lipophilic label, or infection with a green fluorescent protein (GFP)-expressing retrovirus. Although the ectopic ridge of cpp mutant embryos appears to persist, the complete lack of skeletal elements derived from the frontonasal mass suggests that cell death may contribute to the phenotype at some point later than stage 28. Accordingly, cell death could be examined at a later stage. Fgf8 expression has not yet been examined in scaleless or ectrodactyly mutant chicken embryos, and this would be an essential first step in determining whether perturbation of Fgf8 signalling is also implicated in the limb truncations (ectrodactyly) and loss of epithelial specializations (scaleless). Another interesting avenue of investigation may be to examine intracellular control of proliferation, to demonstrate if possible a direct link between Fgf8 signalling and excess proliferation in the cpp frontonasal mass. This could be done in cultures of cells from the ectopic ridge of cpp mutant embryos, and/or normal frontonasal mass cells, to determine whether the same intracellular signalling pathways are activated by Fgf8 in normal and cpp frontonasal mass mesenchyme. 106 As more is discovered about transcriptional control of Fgf8 gene expression, it may become possible to identify the specific elements that control downregulation of Fgf8 in the normal frontonasal mass betweeen stage 20 and 24, and the failure of this event in the cpp mutant. Despite the lack of effect from the cpp mutation on Shh or Bmp signalling, further investigations of facial outgrowth should continue to examine possible roles for these and other molecular networks. Concluding remarks There exist many developmental processes that have been studied at the level of tissue interactions but are being reexamined in the light of molecular signalling events. It is now apparent that signalling networks are often recapitulated throughout development. Understanding spatial and temporal control of signalling cascades can be greatly facilitated by a system in which a key process, Fg/8-induced mitogenesis, is affected in a consistent and restricted manner. Ultimately, it would be appealing to identify the cpp gene, however much may still be learned in the meantime about patterning events unique to facial outgrowth from the cleft primary palate mutation. 107 Literature Cited Abbott, U. K. (1983). Genetic modification of limb morphogenesis. Prog Clin Biol Res 110 Pt A, 13-31. Abbott, U. K., and MacCabe, J. A. (1966). Ectrodactyly: a new embryonic lethal mutation in the chicken. J Hered 57, 207-11. Abbott, U. K., and Pisenti, J. M. (1993). Making the connection: exploring classical concepts in normal and abnormal limb development using contemporary approaches. Prog Clin Biol Res 383A, 99-112. Ashique, A. M., Fu, K., and Richman, J. M. (2002). Endogenous bone morphogenetic proteins regulate outgrowth and epithelial survival during avian lip fusion. Development 129, in press. Bachler, M., and Neubuser, A. (2001). Expression of members of the Fgf family and their receptors during midfacial development. Mech Dev 100, 313-6. Barlow, A. J., and Francis-West, P. H. (1997). Ectopic application of recombinant BMP-2 and BMP-4 can change patterning of developing chick facial primordia. Development 124, 391-8. Bee, J., and Thorogood, P. (1980). The role of tissue interactions in the skeletogenic differentiation of avian neural crest cells. Dev Biol 78, 47-62. Bellairs, R., and Osmond, M. (1998). "The Atlas of Chick Development." Academic Press, Toronto. Belloni, E., Muenke, M., Roessler, E., Traverso, G., Siegel-Bartelt, J., Frumkin, A., Mitchell, H. F., Donis-Keller, H., Helms, C , Hing, A. V., Heng, H. H., Koop, B., Martindale, D., Rommens, J. M., Tsui, L. C , and Scherer, S. W. (1996). Identification of Sonic hedgehog as a candidate gene responsible for holoprosencephaly. Nat Genet 14, 353-6. Blunt, A. G., Lawshe, A., Cunningham, M. L., Seto, M. L., Ornitz, D. M., and MacArthur, C. A. (1997). Overlapping expression and redundant activation of mesenchymal fibroblast growth factor (FGF) receptors by alternatively spliced FGF-8 ligands. JBiol Chem 272, 3733-8. Boshart, L., Vlot, E. A., and Vermeij-Keers, C. (2000). Epithelio-mesenchymal transformations in the embryonic face: implications for craniofacial malformations. Eur J Plast Surg 23,217-223. 108 Brown, J. M., Robertson, K. E., Wedden, S. E., and Tickle, C. (1997). Alterations in Msx 1 and Msx 2 expression correlate with inhibition of outgrowth of chick facial primordia induced by retinoic acid. Anat Embryol (Berl) 195, 203-7. Brown, J. M., Wedden, S. E., Millburn, G. H., Robson, L. G., Hill, R. E., Davidson, D. R., and Tickle, C. (1993). Experimental analysis of the control of expression of the homeobox-gene Msx-1 in the developing limb and face. Development 119, 41-8. Buckland, R. A., Collinson, J. M., Graham, E., Davidson, D. R., and Hill, R. E. (1998). Antagonistic effects of FGF4 on BMP induction of apoptosis and chondrogenesis in the chick limb bud. Mech Dev 71, 143-50. Capdevila, J., and Izpisua Belmonte, J. C. (2001). Patterning mechanisms controlling vertebrate limb development. Annu Rev Cell Dev Biol 17, 87-132. Capdevila, J., and Johnson, R. L. (1998). Endogenous and ectopic expression of noggin suggests a conserved mechanism for regulation of BMP function during limb and somite patterning. Dev Biol 197, 205-17. Carrington, J. L., and Fallon, J. F. (1986). Experimental manipulation leading to induction of dorsal ectodermal ridges on normal limb buds results in a phenocopy of the Eudiplopodia chick mutant. Dev Biol 116, 130-7. Chai, Y., Sasano, Y., Bringas, P., Jr., Mayo, M., Kaartinen, V., Heisterkamp, N., Groffen, J., Slavkin, H., and Shuler, C. (1997). Characterization of the fate of midline epithelial cells during the fusion of mandibular prominences in vivo. Dev Dyn 208, 526-35. Chen, Y., and Struhl, G. (1996). Dual roles for patched in sequestering and transducing Hedgehog. Cell 87, 553-63. Chiang, C , Litingtung, Y., Harris, M. P., Simandl, B. K., Li, Y., Beachy, P. A., and Fallon, J. F. (2001). Manifestation of the limb prepattern: limb development in the absence of sonic hedgehog function. Dev Biol 236,421-35. 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, 407-13. Christen, B., and Slack, J. M. (1997). FGF-8 is associated with anteroposterior patterning and limb regeneration in Xenopus. Dev Biol 192, 455-66. Cohn, M. J., Izpisua-Belmonte, J. C , Abud, H., Heath, J. K., and Tickle, C. (1995). Fibroblast growth factors induce additional limb development from the flank of chick embryos. Cell 80, 739-46. 109 Cohn, M. J., Patel, K., Krumlauf, R., Wilkinson, D. G., Clarke, J. D., and Tickle, C. (1997). Hox9 genes and vertebrate limb specification. Nature 387, 97-101. Couly, G., Creuzet, S., Bennaceur, S., Vincent, C , and Le Douarin, N. M. (2002). Interactions between Hox-negative cephalic neural crest cells and the foregut endoderm in patterning the facial skeleton in the vertebrate head. Development 129, 1061-73. Couly, G., Grapin-Botton, A., Coltey, P., Ruhin, B., and Le Douarin, N. M. (1998). Determination of the identity of the derivatives of the cephalic neural crest: incompatibility between Hox gene expression and lower jaw development. Development 125, 3445-59. Couly, G. F., and Le Douarin, N. M. (1985). Mapping of the early neural primordium in quail-chick chimeras. I. Developmental relationships between placodes, facial ectoderm, and prosencephalon. Dev Biol 110, 422-39. Crackower, M. A., Motoyama, J., and Tsui, L. C. (1998). Defect in the maintenance of the apical ectodermal ridge in the Dactylaplasia mouse. Dev Biol 201, 78-89. 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, 439-51. Crossley, P. H., Martinez, S., and Martin, G. R. (1996a). Midbrain development induced by FGF8 in the chick embryo. Nature 380, 66-8. Crossley, P. H., Minowada, G., MacArthur, C. A., and Martin, G. R. (1996b). Roles for FGF8 in the induction, initiation, and maintenance of chick limb development. Cell 84, 127-36. Davidson, D. R., Crawley, A., Hill, R. E., and Tickle, C. (1991). Position-dependent expression of two related homeobox genes in developing vertebrate limbs. Nature 352, 429-31. Dealy, C. N., Clarke, K., and Scranton, V. (1996). Ability of FGFs to promote the outgrowth and proliferation of limb mesoderm is dependent on IGF-I activity. Dev Dyn 206,463-9. Dealy, C. N., Seghatoleslami, M. R., Ferrari, D., and Kosher, R. A. (1997). FGF-stimulated outgrowth and proliferation of limb mesoderm is dependent on syndecan-3. Dev Biol 184, 343-50. DeBeer, G. R. (1985). "The Development of the Vertebrate Skull." The University of Chicago Press, Chicago. 110 Diewert, V. M. (1985). Development of human craniofacial morphology during the late embryonic and early fetal periods. Am J Orthod 88, 64-76. Diewert, V. M., Lozanoff, S., and Choy, V. (1993). Computer reconstructions of human embryonic craniofacial morphology showing changes in relations between the face and brain during primary palate formation. / Craniofac Genet Dev Biol 13, 193-201. Diewert, V. M., and Pratt, R. M. (1981). Cortisone-induced cleft palate in A/J mice: failure of palatal shelf contact. Teratology 24, 149-62. Drossopoulou, G., Lewis, K. E., Sanz-Ezquerro, J. J., Nikbakht, N., McMahon, A. P., Hofmann, C , and Tickle, C. (2000). A model for anteroposterior patterning of the vertebrate limb based on sequential long- and short-range Shh signalling and Bmp signalling. Development 127, 1337-48. Duboule, D. (1998). Vertebrate hox gene regulation: clustering and/or colinearity? Curr Opin Genet Dev 8, 514-8. Dunlop, L. L., and Hall, B. K. (1995). Relationships between cellular condensation, preosteoblast formation and epithelial-mesenchymal interactions in initiation of osteogenesis. Int J Dev Biol 39, 357-71. Ede, D. A., and Kelly, W. A. (1964). Developmental abnormalities in the head region of the talpid3 mutation of the fowl. J Embryol Exp Morphol 12, 161-182. Ericson, J., Morton, S., Kawakami, A., Roelink, H., and Jessell, T. M. (1996). Two critical periods of Sonic Hedgehog signaling required for the specification of motor neuron identity. Cell 87, 661-73. Fitch, N. (1957). An embryological analysis of two mutants in the mouse, both producing cleft palates. J Exp Zool 136, 329-357. Fitchett, J. E., and Hay, E. D. (1989). Medial edge epithelium transforms to mesenchyme after embryonic palatal shelves fuse. Dev Biol 131, 455-74. Francis, P. H., Richardson, M. K., Brickell, P. M., and Tickle, C. (1994). Bone morphogenetic proteins and a signalling pathway that controls patterning in the developing chick limb. Development 120, 209-18. Francis-West, P., Ladher, R., Barlow, A., and Graveson, A. (1998). Signalling interactions during facial development. Mech Dev 75, 3-28. Francis-West, P. H., Tatla, T., and Brickell, P. M. (1994). Expression patterns of the bone morphogenetic protein genes Bmp-4 and Bmp-2 in the developing chick face suggest a role in outgrowth of the primordia. Dev Dyn 201, 168-78. I l l Fraser, R. A., and Abbott, U. K. (1971a). Studies on limb morphogenesis. V. The expression of Eudiplopodia and its experimental modification. J Exp Zool 176, 219-36. Fraser, R. A., and Abbott, U. K. (1971b). Studies on limb morphogenesis. VI. Experiments with early stages of the polydactylous mutant Eudiplopodia. J Exp Zool 176, 237-48. Gemel, J., Jacobsen, C., and MacArthur, C. A. (1999). Fibroblast growth factor-8 expression is regulated by intronic engrailed and Pbxl-binding sites. J Biol Chem 274, 6020-6. Goetinck, P. F. (1964). Studies on limb morphogenesis II: Experiments with the polydactylous mutant eudiplopodia. Dev Biol 10, 71-91. Goetinck, P. F., and Abbott, U. K. (1963). Tissue interactions in the scaleless mutant and the use of scaleless as an ectodermal marker in studies of normal limb differentiation. J Exp Zool 154, 7-19. Goetinck, P. F., and Abbott, U. K. (1964). Studies on limb morphogenesis I: Experiments with the polydactylous mutant talpid2. J Exp Zool 155, 161-170. Guimond, S., Maccarana, M., Olwin, B. B., Lindahl, U., and Rapraeger, A. C. (1993). Activating and inhibitory heparin sequences for FGF-2 (basic FGF): distinct requirements for FGF-1, FGF-2, and FGF-4. JBiol Chem 268, 23906-14. Hall, B. K. (1978). Initiation of osteogenesis by mandibular mesenchyme. Arch Oral Biol 23, 1157-61. Hall, B. K. (1980). Tissue interactions and the initiation of osteogenesis and chondrogenesis in the neural crst-derived mandibluar skeleton of the embryonic mouse as seen in isolated murine tissues and in recombinations of murine and avian tissues. / Embryol Exp Morphol 58, 251-264. Hall, B. K. (1981). The induction of neural crest-derived cartilage and bone by embryonic epithelia: an analysis of the mode of action of an epithelial-mesenchymal interaction. J Embryol Exp Morphol 64, 305-20. Hall, B. K. (1988). "The Neural Crest." Oxford University Press, Toronto. Hall, B. K. (1999). "The Neural Crest in Development and Evolution." Springer-Verlag, New York. Hall, B. K., and Miyake, T. (1992). The membranous skeleton: the role of cell condensations in vertebrate skeletogenesis. Anat Embryol (Berl) 186, 107-24. 112 Hall, B. K., and Miyake, T. (1995). Divide, accumulate, differentiate: cell condensation in skeletal development revisited. Int J Dev Biol 39, 881-93. Hall, B. K., and Tremaine, R. (1979). Ability of neural crest cells from the embryonic chick to differentiate into cartilage before their migration away from the neural tube. Anat Rec 194,469-75. Hamburger, V., and Hamilton, H. (1951). A series of normal stages in the development of the chick embryo. J Morphol 88,49-92. Heikenheimo, M., Lawshe, A., Shackleford, G. M., Wilson, D. B., and MacArthur, C. A. (1994). Fgf-8 expression in the post-gastrulation mouse suggests roles in the development of the face, limbs and central nervous system. Mech Dev 48, 129-38. Helms, J. A., Kim, C. H., Hu, D., Minkoff, R., Thaller, C., and Eichele, G. (1997). Sonic hedgehog participates in craniofacial morphogenesis and is down-regulated by teratogenic doses of retinoic acid. Dev Biol 187, 25-35. Hinchliffe, J. R., and Ede, D. A. (1967). Limb development in the polydactylous talpicf mutant of the fowl. J Embryol Exp Morphol 17, 385-404. Hu, D., and Helms, J. (2001). Organ culture of craniofacial primordia. Methods 24, 49-54. Hu, D., and Helms, J. A. (1999). The role of sonic hedgehog in normal and abnormal craniofacial morphogenesis. Development 126, 4873-84. Jabs, E. W., Muller, U., Li, X., Ma, L., Luo, W., Haworth, I. S., Klisak, I., Sparkes, R., Warman, M. L., Mulliken, J. B., and et al. (1993). A mutation in the homeodomain of the human MSX2 gene in a family affected with autosomal dominant craniosynostosis. Cell 75, 443-50. Jacobson, A. G., and Sater, A. K. (1988). Features of embryonic induction. Development 104, 341-59. Johnson, R. L., and Tabin, C. J. (1997). Molecular models for vertebrate limb development. Cell 90, 979-90. Johnston, M. C. (1966). A radioautographic study of the migration and fate of cranial neural crest cells in the chick embryo. Anat Rec 156, 143-55. Johnston, M. C., and Bronsky, P. T. (1991). Animal models for human craniofacial malformations. J Craniofac Genet Dev Biol 11, 277-91. Kato, Y., and Hayashi, Y. (1963). The inductive transformation of the chorionic epithelium into skin derivatives. Exp Cell Res 31, 599-602. 113 Kettunen, P., Karavanova, I., and Thesleff, I. (1998). Responsiveness of developing dental tissues to fibroblast growth factors: expression of splicing alternatives of FGFR1, -2, -3, and of FGFR4; and stimulation of cell proliferation by FGF-2, -4, -8, and -9. Dev Genet 22, 374-85. Kettunen, P., and Thesleff, I. (1998). Expression and function of FGFs-4, -8, and -9 suggest functional redundancy and repetitive use as epithelial signals during tooth morphogenesis. Dev Dyn 211, 256-68. Kingsbury, J. W., Allen, V., and Rotheram, B. A. (1953). The histological structure of the beak in the chick. Anat Rec 116, 95-115. Kollar, E. J., and Baird, G. R. (1969). The influence of the dental papilla on the development of tooth shape in embryonic mouse tooth germs. J Embryol Exp Morphol 21, 131-48. Kollar, E. J., and Fisher, C. (1980). Tooth induction in chick epithelium: expression of quiescent genes for enamel synthesis. Science 207, 993-5. Kosher, R. A., and Solursh, M. (1989). Widespread distribution of type II collagen during embryonic chick development. Dev Biol 131, 558-66. Langille, R. (1993a). Formation of the vertebrate face: differentiation and development. Amer Zool 33, 462-471. Langille, R. M., Hall, B.K. (1993b). Pattern Formation and the Neural Crest. In "The Skull" (J. Hanken, Hall, B.K., Ed.), Vol. 1, pp. 77-111. University of Chicago Press, Chicago. Le Lievre, C. S., and Le Douarin, N. M. (1975). Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos. J Embryol Exp Morphol 34, 125-154. Lee, S. H., Fu, K. K., Hui, J. N., and Richman, J. M. (2001). Noggin and retinoic acid transform the identity of avian facial prominences. Nature 414, 909-12. Lee, S. M., Danielian, P. S., Fritzsch, B., and McMahon, A. P. (1997). Evidence that FGF8 signalling from the midbrain-hindbrain junction regulates growth and polarity in the developing midbrain. Development 124, 959-69. Lewandoski, M., Sun, X., and Martin, G. R. (2000). Fgf8 signalling from the AER is essential for normal limb development. Nat Genet 26,460-3. Logan, C , Hornbruch, A., Campbell, I., and Lumsden, A. (1997). The role of Engrailed in establishing the dorsoventral axis of the chick limb. Development 124, 2317-24. 114 Loomis, C. A., Harris, E., Michaud, J., Wurst, W., Hanks, M., and Joyner, A. L. (1996). The mouse Engrailed-1 gene and ventral limb patterning. Nature 382, 360-3. Loomis, C. A., Kimmel, R. A., Tong, C. X., Michaud, J., and Joyner, A. L. (1998). Analysis of the genetic pathway leading to formation of ectopic apical ectodermal ridges in mouse Engrailed-1 mutant limbs. Development 125, 1137-48. MacCabe, J. A., and Abbott, U. K. (1974). Polarizing and maintenance activities in two polydactylous mutants of the fowl: diplopodia and talpid. J Embryol Exp Morphol 31, 735-46. Marigo, V., Davey, R. A., Zuo, Y., Cunningham, J. M., and Tabin, C. J. (1996a). Biochemical evidence that patched is the Hedgehog receptor. Nature 384, 176-9. Marigo, V., Scott, M. P., Johnson, R. L., Goodrich, L. V., and Tabin, C. J. (1996b). Conservation in hedgehog signaling: induction of a chicken patched homolog by Sonic hedgehog in the developing limb. Development 122, 1225-33. Marti, E., Takada, R., Bumcrot, D. A., Sasaki, H., and McMahon, A. P. (1995). Distribution of Sonic hedgehog peptides in the developing chick and mouse embryo. Development 121, 2537-47. Martinez, S., Crossley, P. H., Cobos, I., Rubenstein, J. L., and Martin, G. R. (1999). FGF8 induces formation of an ectopic isthmic organizer and isthmocerebellar development via a repressive effect on Otx2 expression. Development 126, 1189-200. Matovinovic, E., and Richman, J. M. (1997). Epithelium is required for maintaining FGFR-2 expression levels in facial mesenchyme of the developing chick embryo. Dev Dyn 210, 407-16. McCann, J. P., Owens, P. D., and Wilson, D. J. (1991). Chick frontonasal process excision significantly affects mid-facial development. Anat Embryol (Bed) 184, 171-8. McGonnell, I. M., Clarke, J. D., and Tickle, C. (1998). Fate map of the developing chick face: analysis of expansion of facial primordia and establishment of the primary palate. Dev Dyn 212, 102-18. McMahon, A. P., Joyner, A. L., Bradley, A., and McMahon, J. A. (1992). The midbrain-hindbrain phenotype of Wnt-1-/Wnt-1- mice results from stepwise deletion of engrailed-expressing cells by 9.5 days postcoitum. Cell 69, 581-95. Mina, M., Gluhak, J., Upholt, W. B., Kollar, E. J., and Rogers, B. (1995). Experimental analysis of Msx-1 and Msx-2 gene expression during chick mandibular morphogenesis. Dev Dyn 202, 195-214. 115 Mina, M., and Kollar, E. J. (1987). The induction of odontogenesis in non-dental mesenchyme combined with early murine mandibular arch epithelium. Arch Oral Biol 32, 123-7. Mina, M., Wang, Y. H., Ivanisevic, A. M., Upholt, W. B., and Rodgers, B. (2002). Region- and stage-specific effects of FGFs and BMPs in chick mandibular morphogenesis. Dev Dyn 223, 333-52. Minkoff, R. (1980). Regional variation of cell proliferation within the facial processes of the chick embryo: a study of the role of 'merging' during development. J Embryol Exp Morphol 57, 37-49. Minkoff, R. (1984). Cell cycle analysis of facial mesenchyme in the chick embryo. I. Labelled mitoses and continuous labelling studies. J Embryol Exp Morphol 81, 45-59. Minkoff, R. (1991). Cell proliferation during formation of the embryonic facial primordia. J Craniofac Genet Dev Biol 11, 251-61. Minkoff, R., and Kuntz, A. J. (1977). Cell proliferation during morphogenetic change; analysis of frontonasal morphogenesis in the chick embryo employing DNA labeling indices. J Embryol Exp Morphol 40, 101-13. Minkoff, R., and Kuntz, A. J. (1978). Cell proliferation and cell density of mesenchyme in the maxillary process and adjacent regions during facial development in the chick embryo. J Embryol Exp Morphol 46, 65-74. Minkoff, R., and Martin, R. E. (1984). Cell cycle analysis of facial mesenchyme in the chick embryo. II. Lavel dilution studies and developmental fate of slow cycling cells. J Embryol Exp Morphol 81, 61-73. Minkoff, R., Parker, S. B., Rundus, V. R., and Hertzberg, E. L. (1997). Expression patterns of connexin43 protein during facial development in the chick embryo: associates with outgrowth, attachment, and closure of the midfacial primordia. Anat Rec 248, 279-90. Moon, A. M., and Capecchi, M. R. (2000). Fgf8 is required for outgrowth and patterning of the limbs. Nat Genet 26, 455-9. Munoz-Sanjuan, I., Cooper, M. K., Beachy, P. A., Fallon, J. F., and Nathans, J. (2001). Expression and regulation of chicken fibroblast growth factor homologous factor (FHF)-4 during craniofacial morphogenesis. Dev Dyn 220, 238-45. Nasrallah, I., and Golden, J. A. (2001). Brain, eye, and face defects as a result of ectopic localization of Sonic hedgehog protein in the developing rostral neural tube. Teratology 64, 107-13. 116 Neubiiser, A., Peters, H., Balling, R., and Martin, G. R. (1997). Antagonistic interactions between FGF and BMP signaling pathways: a mechanism for positioning the sites of tooth formation. Cell 90, 247-55. Nishikawa, K., Nakanishi, T., Aoki, C., Hattori, T., Takahashi, K., and Taniguchi, S. (1994). Differential expression of homeobox-containing genes Msx-1 and Msx-2 and homeoprotein Msx-2 expression during chick craniofacial development. Biochem Mol Biollnt 32, 763-71. Niswander, L., Jeffrey, S., Martin, G. R., and Tickle, C. (1994). A positive feedback loop coordinates growth and patterning in the vertebrate limb. Nature 371, 609-12. Niswander, L., and Martin, G. R. (1993). FGF-4 and BMP-2 have opposite effects on limb growth. Nature 361, 68-71. Niswander, L., Tickle, C., Vogel, A., Booth, I., and Martin, G. R. (1993). FGF-4 replaces the apical ectodermal ridge and directs outgrowth and patterning of the limb. Cell 75, 579-87. Noden, D. M. (1975). An analysis of migratory behavior of avian cephalic neural crest cells. Dev Biol 42, 106-30. Noden, D. M. (1978). The control of avian cephalic neural crest cytodifferentiation. II. Neural tissues. Dev Biol 67, 313-29. Noden, D. M. (1983a). The embryonic origins of avian cephalic and cervical muscles and associated connective tissues. Am J Anat, 257-276. Noden, D. M. (1983b). The role of the neural crest in patterning of avian cranial skeletal, connective, and muscle tissues. Dev Biol, 144-165. Ohuchi, H., Nakagawa, T., Yamamoto, A., Araga, A., Ohata, T., Ishimaru, Y., Yoshioka, H., Kuwana, T., Nohno, T., Yamasaki, M., Itoh, N., and Noji, S. (1997). The mesenchymal factor, FGF 10, initiates and maintains the outgrowth of the chick limb bud through interaction with FGF8, an apical ectodermal factor. Development 124, 2235-44. Ornitz, D. M. (2000). FGFs, heparan sulfate and FGFRs: complex interactions essential for development. Bioessays 22, 108-12. Parr, B. A., and McMahon, A. P. (1995). Dorsalizing signal Wnt-7a required for normal polarity of D-V and A-P axes of mouse limb. Nature 374, 350-3. Patterson, S. B., and Minkoff, R. (1985). Morphometric and autoradiographic analysis of frontonasal development in the chick embryo. Anat Rec 212, 90-9. 117 Peterka, M., and Jelinek, R. (1983). Origin of hydrocortisone induced orofacial clefts in the chick embryo. Cleft Palate J 20, 35-46. Peterka, M., Peterkoya, R., and Likovsky, Z. (1997). Cleft beak induced by hydrocortisone in the chick is prevented by increased cell division after experimental reduction of amniotic fluid. Anat Embryol (Berl) 195, 387-91. Pizette, S., Abate-Shen, C , and Niswander, L. (2001). BMP controls proximodistal outgrowth, via induction of the apical ectodermal ridge, and dorsoventral patterning in the vertebrate limb. Development 128, 4463-74. Pizette, S., and Niswander, L. (1999). BMPs negatively regulate structure and function of the limb apical ectodermal ridge. Development 126, 883-94. Plant, M. R., MacDonald, M. E., Grad, L. I., Ritchie, S. J., and Richman, J. M. (2000). Locally released retinoic acid repatterns the first branchial arch cartilages in vivo. Dev Biol 222, 12-26. Poswillo, D. (1975a). Causal mechanisms of craniofacial deformity. Br Med Bull 31, 101-6. Poswillo, D. (1975b). Hemorrhage in development of the face. Birth Defects Orig Artie Ser 11, 61-81. Reiter, R. S., and Solursh, M. (1982). Mitogenic property of the apical ectodermal ridge. Dev Biol 93, 28-35. Richman, J. M. (1992). The role of retinoids in normal and abnormal embryonic craniofacial morphogenesis. Crit Rev Oral Biol Med 4, 93-109. Richman, J. M., and Crosby, Z. (1990). Differential growth of facial primordia in chick embryos: responses of facial mesenchyme to basic fibroblast growth factor (bFGF) and serum in micromass culture. Development 109, 341-8. Richman, J. M., and Delgado, J. L. (1995). Locally released retinoic acid leads to facial clefts in the chick embryo but does not alter the expression of receptors for fibroblast growth factor. J Craniofac Genet Dev Biol 15, 190-204. Richman, J. M., Herbert, M., Matovinovic, E., and Walin, J. (1997). Effect of fibroblast growth factors on outgrowth of facial mesenchyme. Dev Biol 189, 135-47. Richman, J. M., and Tickle, C. (1989). Epithelia are interchangeable between facial primordia of chick embryos and morphogenesis is controlled by the mesenchyme. Dev Biol 136, 201-10. 118 Richman, J. M., and Tickle, C. (1992). Epithelial-mesenchymal interactions in the outgrowth of limb buds and facial primordia in chick embryos. Dev Biol 154, 299-308. Riddle, R. D., Johnson, R. L., Laufer, E., and Tabin, C. (1993). Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75, 1401-16. Robert, B., Lyons, G., Simandl, B. K., Kuroiwa, A., and Buckingham, M. (1991). The apical ectodermal ridge regulates Hox-7 and Hox-8 gene expression in developing chick limb buds. Genes Dev 5, 2363-74. Rodriguez-Esteban, C., Schwabe, J. W., Pena, J. D., Rincon-Limas, D. E., Magallon, J., Botas, J., and Belmonte, J. C. (1998). Lhx2, a vertebrate homologue of apterous, regulates vertebrate limb outgrowth. Development 125, 3925-34. Roessler, E., Belloni, E., Gaudenz, K., Jay, P., Berta, P., Scherer, S. W., Tsui, L. C., and Muenke, M. (1996). Mutations in the human Sonic Hedgehog gene cause holoprosencephaly. Nat Genet 14, 357-60. Rowe, A., Richman, J. M., and Brickell, P. M. (1991). Retinoic acid treatment alters the distribution of retinoic acid receptor-beta transcripts in the embryonic chick face. Development 111, 1007-16. Saber, G. M., Parker, S. B., and Minkoff, R. (1989). Influence of epithelial-mesenchymal interaction on the viability of facial mesenchyme in vitro. Anat Rec 225, 56-66. Satokata, I., and Maas, R. (1994). Msxl deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development. Nat Genet 6, 348-56. Saunders, J. W. (1948). The proximo-distal sequence of origin of the parts of the chick wing and the role of the ectoderm. J Exp Zool 108, 363-402. Saunders, J. W., and Gasseling, M. T. (1968). Ectoderm-mesoderm interaction in the origins of wing symmetry. In "Epithelial-Mesenchymal Interactions" (R. Fleischmajer and R. E. Billingham, Eds.), pp. 289-314. Williams & Wilkins, Baltimore. Saxen, L., Vainio, T., and Toivonen, S. (1962). Effect of polyoma virus on mouse kidney rudiments in vitro. J Natl Cancer Inst 29, 597-631. Schneider, R. A., Hu, D., and Helms, J. A. (1999). From head to toe: conservation of molecular signals regulating limb and craniofacial morphogenesis. Cell Tissue Res 296, 103-9. Schneider, R. A., Hu, D., Rubenstein, J. L., Maden, M., and Helms, J. A. (2001). Local retinoid signaling coordinates forebrain and facial morphogenesis by maintaining FGF8 and SHH. Development 128, 2755-67. 119 Schorle, H., Meier, P., Buchert, M., Jaenisch, R., and Mitchell, P. J. (1996). Transcription factor AP-2 essential for cranial closure and craniofacial development. Nature 381, 235-8. Schutte, B. C , and Murray, J. C. (1999). The many faces and factors of orofacial clefts. Hum Mol Genet 8, 1853-9. Searls, R. L., and Janners, M. Y. (1971). The initiation of limb bud outgrowth in the embryonic chick. Dev Biol 24, 198-213. Semba, I., Nonaka, K., Takahashi, I., Takahashi, K., Dashner, R., Shum, L., Nuckolls, G. H., and Slavkin, H. C. (2000). Positionally-dependent chondrogenesis induced by BMP4 is co-regulated by Sox9 and Msx2. Dev Dyn 217, 401-14. Shen, H., Wilke, T., Ashique, A. M., Narvey, M., Zerucha, T., Savino, E., Williams, T., and Richman, J. M. (1997). Chicken transcription factor AP-2: cloning, expression and its role in outgrowth of facial prominences and limb buds. Dev Biol 188, 248-66. Sherman, L., Wainwright, D., Ponta, H., and Herrlich, P. (1998). A splice variant of CD44 expressed in the apical ectodermal ridge presents fibroblast growth factors to limb mesenchyme and is required for limb outgrowth. Genes Dev 12, 1058-71. Shigetani, Y., Nobusada, Y. & Kuratani S. (2000). Ectodermally derived FGF8 defines the maxillomandibular region in the early chick embryo: epithelial-mesenchymal interactions in the specification of the craniofacial ectomesenchyme. Dev Biol, 73-85. Solursh, M., Singley, C. T., and Reiter, R. S. (1981). The influence of epithelia on cartilage and loose connective tissue formation by limb mesenchyme cultures. Dev Biol 86, 471-82. Stein, G. S., and Lian, J. B. (1993). Molecular mechanisms mediating proliferation/differentiation interrelationships during progressive development of the osteoblast phenotype. Endocr Rev 14, 424-42. Stottmann, R. W., Anderson, R. M., and Klingensmith, J. (2001). The BMP antagonists Chordin and Noggin have essential but redundant roles in mouse mandibular outgrowth. Dev Biol 240, 457-73. Sulik, K. K., Cook, C. S., and Webster, W, S. (1988). Teratogens and craniofacial malformations: relationships to cell death. Development 103 Suppl, 213-31. Summerbell, D.. (1974). A quantitative analysis of the effect of excision of the AER from the chick limb-bud. J Embryol Exp Morphol 32, 651-60. Summerbell, D., and Lewis, J. H. (1975). Time, place and positional value in the chick limb-bud. J Embryol Exp Morphol 33, 621-43. 120 Summerbell, D., Lewis, J. H., and Wolpert, L. (1973). Positional information in chick limb morphogenesis. Nature 244, 492-6. Sun, D., Baur, S., and Hay, E. D. (2000a). Epithelial-mesenchymal transformation is the mechanism for fusion of the craniofacial primordia involved in morphogenesis of the chicken lip. Dev Biol 228, 337-49. Sun, X., Lewandoski, M., Meyers, E. N., Liu, Y. H., Maxson, R. E., Jr., and Martin, G. R. (2000b). Conditional inactivation of Fgf4 reveals complexity of signalling during limb bud development. Nat Genet 25, 83-6. Tamarin, A., Crawley, A., Lee, J., and Tickle, C. (1984). Analysis of upper beak defects in chicken embryos following with retinoic acid. J Embryol Exp Morphol 84, 105-23. Tanaka, M., Cohn, M. J., Ashby, P., Davey, M., Martin, P., and Tickle, C. (2000). Distribution of polarizing activity and potential for limb formation in mouse and chick embryos and possible relationships to Polydactyly. Development 127, 4011-21. Thaller, C , and Eichele, G. (1987). Identification and spatial distribution of retinoids in the developing chick limb bud. Nature 327, 625-628. Thaller, C , and Eichele, G. (1988). Characterization of retinoid metabolism in the developing chick limb bud. Development 103, 473-83. Thesleff, I., Partanen, A. M., and Vainio, S. (1991). Epithelial-mesenchymal interactions in. tooth morphogenesis: the roles of extracellular matrix, growth factors, and cell surface receptors. / Craniofac'Genet Dev Biol 11, 229-37. ~ Thesleff, I., and Sharpe, P. (1997). Signalling networks regulating dental development. Mech Dev 67, 111-23. Thorogood, P., Bee, J., and von der Mark, K. (1986). Transient expression of collagen type II at epitheliomesenchymal interfaces during morphogenesis of the cartilaginous neurocranium. Dev Biol 116, 497-509. Tonegawa, Y. (1973). Inductive tissue interactions in the beak of a chick embryo. Dev Growth Differ 15, 57-71. Trainor, P. A., Ariza-McNaughton, L., and Krumlauf, R. (2002). Role of the isthmus and FGFs in resolving the paradox of neural crest plasticity and prepatterning. Science 295, 1288-91. Trumpp, A., Depew, M. J., Rubenstein, J. L., Bishop, J. M., and Martin, G. R. (1999). Cre-mediated gene inactivation demonstrates that FGF8 is required for cell survival and patterning of the first branchial arch. Genes Dev. 3136-48. 121 Tucker, A. S., Al Khamis, A., Ferguson; C. A., Bach, I., Rosenfeld, M. G., and Sharpe, P. T. (1999). Conserved regulation of mesenchymal gene expression by Fgf-8 in face and limb development. Development 126, 221-8. Tucker, A. S., Al Khamis, A., and Sharpe, P. T. (1998a). Interactions between Bmp-4 and Msx-1 act to restrict gene expression to odontogenic mesenchyme. Dev Dyn 212, 533-9. Tucker, A. S., Matthews, K. L., and Sharpe, P. T. (1998b). Transformation of tooth type induced by inhibition of BMP signaling. Science 282, 1136-8. Tyler, M. S. (1978). Epithelial influences on membrane bone formation in the maxilla of the embryonic chick. Anat Rec 192, 225-33. Tyler, M. S., and Hall, B. K. (1977). Epithelial influences on skeletogenesis in the mandible of the embryonic chick. Anat Rec 188, 229-39. Vaahtokari, A., Aberg, T., Jernvall, J., Keranen, S., and Thesleff, I. (1996). The enamel knot as a signaling center in the developing mouse tooth. Mech Dev 54, 39-43. Vainio, S., Karavanova, I., Jowett, A., and Thesleff, I. (1993). Identification of BMP-4 as a signal mediating secondary induction between epithelial and mesenchymal tissues during early tooth development. Cell 75, 45-58. van den Boogaard, M. J., Dorland, M., Beemer, F. A., and van Amstel, H. K. (2000). MSX1 mutation is associated with orofacial clefting and tooth agenesis in humans. Nat Genet 24, 342-3. Vogel, A., Rodriguez, C , and Izpisua-Belmonte, J. C. (1996). Involvement of FGF-8 in initiation, outgrowth and patterning of the vertebrate limb. Development 122, 1737-50. Vogel, A., Rodriguez, C , Warnken, W., and Izpisua Belmonte, J. C. (1995). Dorsal cell fate specified by chick Lmxl during vertebrate limb development. Nature 378, 716-20. Wanek, N., Gardiner, D. M., Muneoka, K., and Bryant, S. V. (1991). Conversion by retinoic acid of anterior cells into ZPA cells in the chick wing bud. Nature 350, 81-3. Wang, K. Y., and Diewert, V. M. (1992). A morphometric analysis of craniofacial growth in cleft lip and noncleft mice. J Craniofac Genet Dev Biol 12, 141-54. Wang, Y.-H., Rutherford, B, Upholt, WB, Mina M. (1999). Effects of BMP-7 on mouse tooth mesenchyme and chick mandibular mesenchyme. Dev Dyn, 320-35. Wang, Y.-H., Upholt, W. B., Sharpe, P. T., Kollar, E. J., and Mina, M. (1998). Odontogenic epithelium induces similar molecular responses in chick and mouse mandibular mesenchyme. Dev Dyn 213, 386-97. 122 Wedden, S. E. (1987). Epithelial-mesenchymal interactions in the development of chick facial primordia and the target of retinoid action. Development 99, 341-51. Wedden, S. E., Ralphs, J. R., and Tickle, C. (1988). Pattern formation in the facial primordia. Development 103 Suppl, 31-40. Wedden, S. E., and Tickle, C. (1986). Quantitative analysis of the effect of retinoids on facial morphogenesis. J Craniofac Genet Dev Biol Suppl 2, 169-78. Winnier, G., Blessing, M., Labosky, P. A., and Hogan, B. L. (1995). Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev 9, 2105-16. Yang, Y., Drossopoulou, G., Chuang, P. T., Duprez, D., Marti, E., Bumcrot, D., Vargesson, N., Clarke, J., Niswander, L., McMahon, A., and Tickle, C. (1997). Relationship between dose, distance and time in Sonic Hedgehog-mediated regulation of anteroposterior polarity in the chick limb. Development 124, 4393-404. Ye, W., Shimamura, K., Rubenstein, J. L., Hynes, M. A., and Rosenthal, A. (1998). FGF and Shh signals control dopaminergic and serotonergic cell fate in the anterior neural plate. Cell 93, 755-66. Yee, G. W., and Abbott, U. K. (1978). Facial development in normal and mutant chick embryos. I. Scanning electron microscopy of primary palate formation. J Exp Zool 206, 307-21. Zar, J. H. (1984). "Biostatistical Analysis." Prentice-Hall, Toronto. Zhang, H., Hu, G., Wang, H., Sciavolino, P., Iler, N., Shen, M. M., and Abate-Shen, C. (1997). Heterodimerization of Msx and Dlx homeoproteins results in functional antagonism. Mol Cell Biol 17, 2920-32. Zhu, L., and Skoultchi, A. I. (2001). Coordinating cell proliferation and differentiation. Curr Opin Genet Dev 11, 91-7. 123 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0090674/manifest

Comment

Related Items