Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bone morphogenetic protein signaling molecules in facial morphogenesis and dysmorphogenesis Ashique, M. Amir 2001

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

Item Metadata


831-ubc_2001-714284.pdf [ 14.03MB ]
JSON: 831-1.0090823.json
JSON-LD: 831-1.0090823-ld.json
RDF/XML (Pretty): 831-1.0090823-rdf.xml
RDF/JSON: 831-1.0090823-rdf.json
Turtle: 831-1.0090823-turtle.txt
N-Triples: 831-1.0090823-rdf-ntriples.txt
Original Record: 831-1.0090823-source.json
Full Text

Full Text

BONE MORPHOGENETIC PROTEIN SIGNALING MOLECULES IN FACIAL MORPHOGENESIS AND DYSMORPHOGENESIS  by  M . AMIR ASHIQUE B. Sc., The University of Manitoba, 1990 M. Sc., The University of Manitoba, 1994  A THESIS SUBMITTED IN PARTIAL FULFILLMENT 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 August 2001 © M. Amir Ashique, 2001  In presenting this thesis in partial fulfilment  of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or  by his or  her  representatives.  It  is understood that  copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  DE-6 (2/88)  ABSTRACT Facial morphogenesis is a dynamic multi-step process involving the formation of the neural crest cell derived facial prominences, their coordinated outgrowth, fusion and differentiation to generate the final facial form. Defects in each of these processes are associated with facial dysmorphogenesis. The identification of genes that are involved in cleft lip is the first aim of this thesis. PCR based differential display was used to compare gene expression profiles in cleft and non-cleft chicken embryos. The second manner to identify candidate genes involved in orofacial clefting was by examining expression patterns of previously cloned genes at the specific time and place where lip fusion occurs. This approach identified Bone morphogenetic proteins (BMPs) as potential genes that are involved with the outgrowth and fusion of facial prominences. Little was known about where BMP receptors and B M P antagonists are expressed in the face or the function of endogenous BMPs in the face. The expression of B M P ligands {Bmp-2, Bmp-4 and Bmp7) was correlated with the expression of BMP receptors (BmpR-IA, BmpR-IB and BmpRII) and noggin, a BMP antagonist, in order to predict where endogenous BMP signaling occurs in the face. Bmp-2, Bmp-4 and Bmp-7 are expressed in the zone of fusion between the frontonasal mass and maxillary prominences. Noggin expression is restricted to the frontonasal mass epithelium and is downregulated at the corners of the frontonasal mass just prior to fusion. Later in development, Bmps are expressed in the perichondrium of cartilage whereas noggin is expressed within differentiating cartilage. The BmpR-IA and BmpR-II are expressed ubiquitously whereas BmpR-IB is expressed in a subset of tissues. This expression data was used to design studies that examined BMP function during primary palate closure and facial cartilage morphogenesis.  n  Noggin protein was applied to regions with high BMP expression in the zone of fusion and subsequently induced clefts. These experiments show for the first time that endogenous BMPs regulate two aspects of lip closure; outgrowth of facial prominences and thinning of the frontonasal epithelium prior to fusion. We also examined the well-established retinoic acid model for cleft lip to see whether BMPs were mediating this defect. Ectopic retinoic acid increases Bmp-2 and Bmp-4 expression prior to the induction of cleft lip in chickens. Mimicking this result by applying BMP-2 to the face also induced clefts. These results are related to non-syndromic human orofacial clefting and demonstrate the utility of studying cleft lip in non-mammalian model systems.  Finally, experiments were carried out that would directly test the function of the B M P receptors in craniofacial development. Whereas expression of dominant-negative (dn) and constitutively active (ca) B M P type I receptors seldom affected early facial morphology, there were significant effects on chondrogenesis. These mutant forms of the type IA and IB receptors regulated the size and shape of cartilage elements. The dn form of the BMPRIB inhibited chondrogenesis, whereas ca-BMPRIA and ca-BMPRIB increased cartilage formation.  The embryos injected with active viruses also lacked  feather germs over much of the head and about 50% of specimens did not form an egg tooth.  It appears that type IA and type IB receptors play similar roles in regulating  cartilage and feather formation in the skull. signalling  molecules  as  critical  Overall, these findings identify B M P  regulators  of  facial  morphogenesis  and  dysmorphogenesis.  iii  T A B L E OF CONTENTS ABSTRACT LIST OF FIGURES LIST OF TABLES LIST OF ABBREVIATIONS ACKNOWLEDGEMENTS  :  CHAPTER I REVIEW OF T H E LITERATURE Normal Face development Origin of Tissues  ii vii vii ix xii  1 1 1  Neural crest formation and migration Somitomeric mesoderm  1 3  Outgrowth Fusion and merging of facial prominences Skeletal derivatives of the facial prominences Differentiation  3 5 7 8  Abnormal facial development Orofacial clefts in humans Phenocopying cleft-lip in chickens  Chicken model system Hydrocortisone Excision of epithelium SHH antibody Retinoic acid Retinoic acid physiology RA Receptors Biological functions of endogenous retinoic acid Genetic studies of endogenous R A functions Genes downstream of R A teratogenicity  Screening strategies to isolate novel, differentially expressed genes Subtractive hybridization Differential display R N A arbitrarily primed (RAP-) PCR Microarrays  Candidate genes involved in facial development Transforming growth factor-beta (TGFP) superfamily Bone morphogenetic proteins (BMPs).. BMP Receptors and signaling molecules BMPs in embryonic development  12 12 14  14 15 16 16 17 18 19 20 21 22  22 23 24 24 25  26 27 28 29 30  Aims of the present study  32  CHAPTER II Isolation of genes associated with retinoic acid-induced cleft-lip in chickens  34  Introduction  34  iv  Materials and Methods  36  Treatment of embryos with retinoic acid 36 Dissection of tissue 36 RNA extraction 36 Northern blot analysis 37 Reverse transcription-polymerase chain reaction (RT-PCR) differential display (DD) 37 Amplification and cloning of differential display gel bands : 39 Results Discussion  41 44  CHAPTER III Bone Morphogenetic proteins, receptors and antagonist: Expression during early chicken face morphogenesis 48 Summary Introduction Materials and Methods  Chicken embryos In situ hybridization probes Autoradiographic in situs Results Discussion  CHAPTER IV Bone Morphogenetic Proteins and the Mechanisms of Facial Clefting Summary Introduction Materials and Methods  In situ hybridization Bead preparation Skeletal staining TUNEL assay BrdU labeling Acridine Orange staining Epithelial stripping Results Discussion  48 49 56  56 56 56 58 77  88 88 89 93  93 .....93 93 93 93 94 94 95 107  CHAPTER V The type IA and type IB Bone morphogenetic protein receptors (BMPR) have similar functions in regulating chondrogenesis and feather formation in the chicken embryo 113 Summary Introduction Materials and Methods  113 114 118  Chicken Embryos In situ hybridization probes Virus Infection Skeletal staining Results Discussion  118 118 118 119 120 137  CHAPTER VI General Discussion and Future Directions  148  Literature Cited  157  vi  LIST OF FIGURES Figure 1.1 Schematic diagram of the embryonic chicken faces during outgrowth and differentiation of the facial prominences  4a  Figure 1.2 Schematic diagram of the skeletal elements in the chicken head  9a  Figure 2.1 Autoradiography of patterns of gene expression follow DD-PCR  42a  Figure 3.1 Expression in stage 10 and 15 chicken embryos  59a  Figure 3.2 Expression in stage 20 embryos  63a  Figure 3.3 Expression in stage 24 embryos  67a  Figure 3.4 Expression in stage 28 embryos  69a  Figure 3.5 Expression in stage 30 embryos  73a  Figure 3.6 Expression in the egg tooth of stage 32 embryos  76a  Figure 4.1 Expression of BMP-2 and noggin at the time of fusion of the primary palate and induction of Bmps with retinoic acid  96a  Figure 4.2 Effects of Noggin on facial morphogenesis  98a  Figure 4.3 Effects of BMP-2 on facial morphogenesis  102a  Figure 4.4 Model of B M P function and pathways in the frontonasal mass  109a  Figure 5.1 Targeting virus to the developing facial promiences Figure 5.2 Histochemical detection of alkaline phosphatase activity in chicken embryos infected with the RCASBP/alkaline phosphatase retrovirus  121  122a  Figure 5.3 Surface morphology of heads following injection of retroviruses containing B M P receptor constructs  127a  Figure 5.4 Morphology of the chondrocranium in cleared specimens  128a  Figure 5.5 Effects of B M P R viruses on cartilage formation and differentiation  129a  Figure 5.6 Expression of genes in adjacent frontal sections of a stage 29 embryo injected with ca-BMPR-IB at stage 14  134a  vii  LIST O F T A B L E S  Table 1.1 Neural crest derived skeletal components of the face  10  Table 4.1 Number of T U N E L positive cells in BMP-2 treated embryos  104  Table 5.1 Alkaline phosphotase activity showing viral spread following injection at multiple sites and stages with RCASBP/E retrovirus containing an alkaline phosphatase insert  123  Table 5.2 Analysis of viral spread within the times in sections using a riboprobe made to the pol region of the RCASBP virus  125  Table 5.3 Morphology of virus injected embryos  130  LIST O F ABBREVIATIONS  BMP  Bone morphogenetic protein  BMPR  BMP receptor  °C  degrees centigrade  ca  constitutive active  CDMP  cartilage-derived morphogenetic protein  cDNA  complimentary D N A  CL/P  cleft lip with or without cleft palate  CO2  carbon dioxide  CRBP  cellular RBP  D  aspartic acid  dNTP  deoxynucleotide triphosphate  DEPC  diethyl pyrocarbonate  DD  differential display  DMSO  dimethyl sulfoxide  dn  dominant negative  DNA  deoxyribonucleic acid  DPP  decapentaplegic  DTT  dithiothreitol  EDTA  ethylenediaminetetraacetic acid  FGF  Fibroblast growth factor  FGFR2  FGF receptor 2  FHF  Fibroblast growth factor homologous factor  GDF  growth and differentiation factors  GITC  guanidium isothiocyanate  GS  glycine-serine  h/hr  hour  Hox  homeobox  L/ml/ul  liter/milliliter/microliter  mRNA  messenger R N A  Msx  Muscle-specific homeobox  Na  sodium  N-CAM  neural-cell adhesion molecule  OP  osteogenic protein  ptc  patched  PBS  phosphate buffered saline  PCR  polymerase chain reaction  PFA  paraformaldehyde  PNA  peanut agglutinin  Q  glutamine  R  rhombomere  RA  retinoic acid  RALDH  retinal dehydrogenase  RAP  RNA arbitrarily primed  RAR  retinoic acid receptor  RARE  R A response elements  PvBP  retinol binding protein  RBPr  RBP receptor  RDH  retinol dehydrogenase  RT  reverse transcription  RNA  ribonucleic acid  Rpm  revolutions per minute  RXR  retinoid X receptors  Sax  Saxophone  SHH  sonic hedgehog  H-TdR  labeled tissue Noden, 1975  TGF-P  transforming growth factor-beta  Tkv  Thick veins  TEM  transmission electron microscopy  TUNEL  terminal deoxynucleotide transferase (TdT)  3  mediated dUTP-digoxygenin nick end labelling VAD  vitamin A deficiency  %  percentage  ACKNOWLEDGEMENTS  First and foremost I have to thank my supervisor, Dr. Joy Richman, for the opportunity to experience developmental research and for her guidance. I would also like to thank the members of my advisory committee, Dr. Virginia Diewert, Dr. Gerald Weeks, Dr. Doug Waterfield and Dr. Chris Overall for directing me through my degree. Special thanks for letting me explore my individual ideas and guiding me along the way. Thank you also to Dr. Joe Tonzetich for his generous fellowship, which I was honored to receive. Over the course of my Ph.D. program I have had the pleasure of being surrounded by many great individuals. Thank you to all the lab technicians for their assistance and all members of the Richman Lab for the pleasure of their company. Special thanks to Mary MacDonald for patiently proofreading my thesis and being a wonderful office-mate. Good luck with Medical school, Mary. I would like to send special thanks to Dr. Waterfield for his tremendous assistance, encouragement and support during difficult times and for understanding that all I ever wanted was to do the best Ph.D. I was capable of doing. You are a wonderful person! Doing the right thing is always difficult, which is why its rewards are so great. My deepest appreciation goes to my family. In particular, my father, who dedicated his life for his children's education. I can not even begin to repay my kids for their understanding that their Ubbu, "has lots of work to do". If I could mold this Ph.D. in the form of a toy and give it to you to play with I would in a heart beat. You deserve my Ph.D. more than I do. To my wife, Sabreena, I can only fully repay with co-authorship. Sweetie, we did it! For giving me the patients to persevere I thank God, who is, incidentally, the only teacher who is always right. He revealed to us over 1400 years ago what we learn with amazement today: "We created humans from a quintessence of mud. Thereafter we cause him to remain as a drop of sperm in a place of rest (ie. the womb). Thereafter we fashioned the sperm into something that clings (to the endometrium), which We fashioned into a chewed lump (somite stage). The chewed lump is fashioned into bones, which are then covered with flesh. We develop out of it another creation. Blessed is God, the best of artisans. After that, at length ye will die. Again ye, on the Day of Judgement, will be raised up (and be held accountable for your actions)." Al-Quran 23:12-16  xii  CHAPTER I  REVIEW OFTHE LITERATURE  N o r m a l face development  In vertebrates, the face is formed during several sequential phases of development. The first phase involves the formation of cranial neural crest cells in the dorsal neural tube. Second, the neural crest cells migrate ventrally, populating the presumptive facial prominences. These prominences are buds of mesenchyme encased in epithelium that surround the primitive oral cavity.  Third, the neural crest-derived mesenchyme  undergoes a period of rapid proliferation and the facial prominences quickly enlarge in size.  Fourth, the facial prominences fuse together at the base of the nasal pit, which  establishes the basic plan of the face.  In the fifth phase, the facial mesenchyme  differentiates to form muscle, cartilage and bone (Sperber, 2001).  Origin of Tissues The origins of mesenchyme in the face are unique compared to other areas of the body. The majority of trunk mesenchyme is derived from the middle germ layer or mesoderm. In the face there is an additional contribution from cranial neural crest cells.  Neural crest formation and migration In early brain development, three bulges form at the cephalic end of the neural tube: forebrain  (prosencephalon),  midbrain  (mesencephalon)  and  hindbrain  1  (rhombencephalon). The rhombencephalon is divided into eight periodic swellings called rhombomeres (rl-8; Lumsden et al., 1991). As the neural tube closes along the dorsal midline, the cells occupying the crest of the neural folds, known as neural crest cells, migrate to different locations and have a wide variety of fates. Cranial neural crest cells migrate  ventrally beneath  the  head  ectoderm,  through the  paraxial mesoderm  (somitomeres) and into the face between the 6 and 9 somite stages (24-30 h incubation; Tosney, 1982; Noden, 1975). By grafting H-TdR labeled tissue (Johnson, 1966; Noden, 3  1975) and, more recently, using D i l labeling (Lumsden et al., 1991) the pathways of migration of cranial neural crest into the developing facial prominences have been elucidated. Neural crest cells from regions anterior to rhombomere 3 contribute to the face. The frontonasal mass lies between the nasal placodes and is populated by neural crest cells from the diencephalon (Le Lievre, 1978; Couly et al., 1993). The lateral nasal prominences are lateral to the nasal placodes and are populated by diencephalic and mesencephalic crest cells (Le Lievre, 1978). The maxillary prominences are inferior to the lateral nasal prominences and are populated by posterior diencephalic and mesencephalic crest (Le Lievre, 1978). The mandibular prominences form the inferior border of the stomodeum and have a dual contribution. The distal portion of the mandible is populated by mesencephalic (Couly et al., 1996; Noden, 1978; Le Lievre, 1974) and the proximal portion is derived from r l and r2 neural crest cells (Couly et al., 1996; Kontges and Lumsden, 1996).  2  Somitomeric mesoderm The paraxial mesoderm becomes segmented into whorls of cells called somitomeres, which form along the length of the embryo during gastrulation (Meier, 1979). In the trunk, the somitomeres become compacted and bound together by epithelium to form individual somites (Keynes and Stern, 1988).  Cranial somitomeres do not become  somites, but rather remain adjacent to the neural tube or disperse to the central regions of the branchial arches (Hacker and Guthrie, 1998; Trainor and Tarn, 1995). In chickens, the first pair of somitomeres form next to the prosencephalon while the remaining somitomeres extend caudally and are adjacent to more posterior brain regions (Meier, 1982;Noden, 1983a).  Outgrowth The difference in embryonic origin may be important in the behaviour of the facial mesenchyme as it grows and differentiates. Following neural-crest cell migration to the face, the facial prominences quickly enlarge (stage 20-28; Hamburger and Hamilton, 1951; equivalent to 3.5 to 5.5 days post-fertilization).  The prominences proliferate at  different levels and proliferation within the facial mesenchyme of each individual prominence is not uniform (McGonnell et al., 1998; Peterka and Jelinek, 1983; Patterson and Minkoff, 1985; Minkoff and Kuntz, 1977; 1978).  These regional differences in  proliferation as well as the direction of expansion of the facial mesenchyme (McGonnell et al., 1998) contribute to the changes in shape of the embryonic face as development proceeds (Figure 1.1). Growth of the facial prominences is dependent on epithelial-  3  F i g u r e 1.1 Schematic d i a g r a m of the embryonic chicken faces d u r i n g outgrowth a n d  differentiation of the facial prominences. Sketches are of the frontal view at stages 20 (A), 24 (B) and 28 (C). Depicted in A is the superior (or cranial)-inferior (or caudal) and medial-lateral axis. Depicted in B is the dorsal-ventral axis, and positions (•) of protein soaked bead placement in the corner of the frontonasal mass and cranial maxillary prominence for studies described in Chapter 4. Arrowheads in C identify regions where fusion occurs between the frontonasal mass and maxillary prominences and * indicates areas where merging occurs. Take note of the changes in shape of the facial prominences as development proceeds.  This is a modified version of Fig. 1 in Francis-West et al.,  1994. Key: ca, caudal; cr, cranial; d, dorsal; e, eye; f, frontonasal mass; 1, lateral nasal prominences; la, lateral; m, mandibular prominences; me, medial; n, nasal slit; v, ventral; x, maxillary prominences. Scale bars = 500 um.  4  mesenchymal interactions, as the epithelium regulates outgrowth (Wedden, 1987; Richman and Tickle, 1989; Hu and Helms, 1999).  Fusion and merging of facial prominences Two different processes consolidate the facial prominences that form the upper lip, merging and fusion.  Merging prominences have a groove or furrow between them,  which is filled in by the rapid proliferation of mesenchymal cells. Fusion, however, occurs between regions that are not connected, and therefore requires removal of the double layer of epithelium that forms when the prominences come into contact. Following removal of the epithelial seam there is ingrowth of mesenchyme to form a mesenchymal bridge.  In the face, a combination of fusion and merging takes place. Merging occurs between the paired mandibular prominences, and between the maxillary and lateral nasal prominences. In chickens, fusion takes place between the premaxilla and the distal end of the palatal shelves. The remainder of the secondary palate remains open in birds, unlike in mammals. Fusion also takes place between the lateral corners of the frontonasal mass and the maxillary prominences. The lateral edges of the frontonasal mass (the globular processes) in birds are equivalent to the medial nasal prominences in mammals. The frontonasal mass and maxillary prominences are already joined in proximal regions of the head but merge distally as cell proliferation and outgrowth proceeds. Fusion only occurs at the distal tips of the maxillary and lateral frontonasal regions. The frontonasal mass, lateral nasal and maxillary prominences must all come together in order to close  5  the primary palate and establish a smooth upper lip. Failure of the frontonasal mass and the maxillary prominences to fuse results in cleft lip.  Complete fusion involves removal of the apposed epithelia and the basement membrane components before mixing of mesenchymal cell populations can proceed (Iamaroon and Diewert, 1996). In the chicken embryo the distal tips of the facial prominences have not yet fused at stage 28.  At stage 29, however, fusion is complete and mesenchymal  breakthrough has taken place. Three theories have been proposed to explain the disappearance of the double layer of epithelium: 1) epithelial cells migrate to adjacent areas (Carette and Ferguson, 1992); 2) epithelial cells transform to mesenchymal cells (Fitchett and Hay, 1989; Martinez-Alvarez et al., 2000; Sun et al., 2000); and 3) epithelial cells undergo apoptosis (Clarke, 1990; for review see Shuler, 1995). Epithelial apoptosis can be detected in the anterior maxillary prominences, the groove between the lateral nasal and maxillary prominences and the coiners of the frontonasal mass in stage 24 and 28 chicken embryos (McGonnell et al., 1998; Shen et al., 1997). Recently, using T U N E L assay and transmission electron microscopy (TEM), Sun et al. (2000) showed that apoptosis in the primary palate is restricted only to the outer epithelial layer, the periderm, and functions to thin the epithelium during fusion. The basal epithelial layer makes up the epithelial seam and is removed soon after by epithelial-mesenchymal transformation. Therefore, in the chicken primary palate, the epithelium between the fusing prominences is removed by a combination of processes. Their T U N E L procedure may not be very sensitive, however, since others have reported that apoptosis is not only in the epithelium but also in the mesenchyme near regions where fusion occurs  6  (McGonnell et al., 1998; Shen et al., 1997). The molecular regulators of endogenous programmed cell death in the fusing facial prominences have not been identified.  Evidence of bidirectional cell mixing was reported in proximal face regions between the anterior maxillary prominences and the distal frontonasal mass while unidirectional migration of the cells occurred from the frontonasal mass into the lateral nasal process (McGonnell et al., 1998). These studies were performed between stages 20 and 28, prior to fusion of the distal tips of the prominences. In addition, their labeling was done near the base of the prominences, in regions that undergo merging as opposed to fusion. Little is known about how the mesenchymal bridge forms (see Sun et al., 2000).  What is  known is that there needs to be a sufficient volume of mesenchyme or the resulting bridge may break down (shown in A/WySn mouse embryos; Wang and Diewert, 1992; Diewert and Wang, 1992).  Skeletal derivatives of the facial prominences The frontonasal mass forms most of the upper beak and gives rise to the interorbital septum, the nasal septum, the prenasal cartilage and the premaxilla. The lateral nasal processes form the alae of the nose and give rise to the nasal conchae, nasal and lachrimal bones.  Externally, the maxillary prominences form the sides of the upper beak and  internally, the palatal shelves.  The maxillae form exclusively membranous bones,  including the pterygoid, jugal, quadratojugal, palatine and maxillary bones. The paired mandibular prominences give rise to the lower beak. The skeletal derivatives of the mandibular prominences include, Meckel's cartilage, the retroarticular process, the  7  mentomandibular bone, the dentary bone, the supraangular bone and the angular bone. Each facial prominence therefore contributes to a distinct region of the face (Figure 1.2). The prominences retain their distinct derivatives even when transplanted to ectopic sites (Wedden et al., 1987; Richman et al., 1997; Richman and Tickle, 1989, 1992). By stage 20, pattern is completely specified in the facial mesenchyme.  Differentiation Chick/quail chimeras have been extensively used to decipher the extent to which the cranial neural crest and mesodermal populations contribute to the tissues of the face. Derivatives of cranial neural crest cells include cartilage and bone, which are unique to the cranial region, in addition to derivatives also made by trunk neural crest cells such as pigment cells (melanocytes), sensory ganglia, parasympathetic ganglia and endocrine cells (Couly et al., 1996; Kontges and Lumsden, 1996; Couly et al., 1993; Noden 1983b; reviewed in Le Douarin et al., 1994; Hall and Hbrstadius, 1988).  Neural crest cells contribute to ventral head skeletal structures, including all cartilage and bone in the face, while the mesoderm differentiates to form bones of the chordal skull and cranial vault (Noden, 1983a, Noden, 1988; Table 1.1). The avian head is unique because it forms bone via both endochondral and intramembranous formation and has cartilage that does not ossify (Fig. 1.2; Table 1.1). Derivatives of the somitomeric mesoderm also include the muscles (myoblasts) and angioblasts, the posterior cranial base, otic capsule and calvaria (Noden, 1983a, Noden, 1988; Couly et al., 1995; Hacker and Guthie; 1998).  8  Figure  1.2  Schematic  diagram  of  the  skeletal  elements  in  the  chicken  head.  Cartilagenous elements are in light grey while membrane bones are in dark grey. Picture is from Francis-West et al., 1998.  9  9a  Table 1.1 Origin of the chicken head skeleton (from Couly et al., 1993; Hall, personal communication) A. Neural crest derived skeletal components Membrane bones Nasal Maxillary Palatine Quadratojugal Jugal Pterygoid Scleral ossicles Retroarticular process Mentomandibular bone Dentary bone Supraangulare bone Angulare bone Frontal Vomer Premaxillary bone Lachrimal bones  Cartilaginous bones Basipresphenoid Proximal Meckel's Quadrate Hyoid Pars cochlearis of otic capsule Columella  Cartilage Scleral cartilage Meckel's cartilage Interorbital septum Nasal conchae Nasal septum Prenasal cartilage  B. Paraxial mesodermal derived skeletal components Membrane bones Postorbital Supraoccipital Squamosal Parietal  Cartilaginous bones Temporal Basi postsphenoid Pars canalicularis Cochlearis of otic capsule Occipital  Cartilage  10  Skeletal growth via intramembranous ossification involves bone formed directly from mesenchymal cells (reviewed in Fang and Hall, 1997). This process is initiated when mesenchymal cells condense into compact regions.  Some of these cells develop into  capillaries while most differentiate into osteoblasts, bone precursor cells which deposit an osteoid matrix that is able to bind calcium and become calcified (Hall, 1997).  The  calcified matrix grows and osteoblasts that get trapped within the bone matrix become osteocytes (bone cells). Eventually, compact mesenchymal cells form a membrane (periosteum) that surrounds the bone. The inner surface of periostial cells that line the bone can also become osteoblasts and deposit additional matrix, resulting in many layers of bone.  Thus, the periosteum functions to regulate the rate of bone formation (Hall,  1997).  Cartilage formation begins with the condensation of mesenchymal cells (Hall and Miyake, 1992; 1995; 2000), which aggregate together into compact nodules, enlarge in size and produce several characteristic condensation specific markers such as peanut agglutinin (PNA; Stringa and Tuan, 1996), N-cadherin (Chimal-Monroy and Diaz de Leon, 1999), neural-cell adhesion molecule (N-CAM; Tavella et al., 1994), SOX9 (Ng et al., 1997; Zhao et a l , 1997), type II collagen mRNA (Kosher et al., 1986) and FGF receptor 2 mRNA (FGFR2 or Cek3; Matovinovic and Richman, 1997; Wilke et al., 1997; Noji et al., 1993; Patstone et al., 1993). Precartilagenous condensations represent a rough outline of the future skeletal elements (Griineberg, 1963).  These prechondrocytes  differentiate into chondrocytes and switch from the production of mesenchymal matrix (type I and type III collagens) to cartilagenous matrix (types II, IX and XI collagen).  11  Cartilage may remain as such, in which case it does not undergo hypertrophy or invasion by osteoblasts. In ossifying cartilage, the chondrocytes proliferate and continue to elaborate cartilage-specific extracellular matrix. The chondrocytes then stop dividing and increase in volume dramatically, becoming hypertrophic chondrocytes. Collagen X and fibronectin are added to the matrix, which enables the cartilagenous template to become mineralized by calcium carbonate.  Soon after, the hypertrophic chondrocytes die by  apoptosis. The vacated space eventually becomes bone marrow and allows for invasion of blood vessels, which bring in bone depositing osteoblasts.  In endochondral  ossification, all the cartilage is removed and replaced by bone and bone marrow (for review see Karsenty, 1998).  A b n o r m a l facial development  Orofacial clefts in humans Cleft lip with or without cleft palate (CL/P) occurs approximately 1 in 1300 live births and thus is one of the most commonly occurring birth defects in humans (Tolarova, and Cervenka, 1998). Clefts may be classified as syndromic or non-syndromic. Syndromes associated with clefts include, Van der Woude syndrome (lq32; Murray et al., 1990), Treacher Collins syndrome (5q; Dixon et al., 1992) and Waardenburg syndrome (PAX3; Tassabehji et al., 1992). Deletions or duplications in all chromosomes have also been associated with clefts, suggesting that many genes are involved in facial development and that cleft lip and palate can represent a common end-point in the disruption of processes involved in facial morphogenesis (Brewer et al., 1998; 1999). We are interested in the  12  non-syndromic forms of clefting, which are much more prevalent (~70%; Jones, 1988). Individuals affected by non-syndromic clefting have no other physical or developmental anomalies, no recognized maternal teratogen exposure or identifiable environmental characteristics (Murray, 1995).  The etiology of non-syndromic orofacial clefts is complex and is thought to include multiple environmental and genetic factors. Environmental factors that correlate with cleft lip and palate include insufficient vitamin intake and teratogens such as cigarette smoking, alcohol use, and prescription drugs (Wyszynski and Beaty, 1996; Romitti et al., 1999). Genetic analysis of cleft lip in a congenic mouse strain indicates that nonsyndromic CL/P is caused by two loci with epistatic interactions (Juriloff, 1995). Similar conclusions are reached from genetic research on humans, which include segregation analysis (Marazita et al., 1986; Mitchell, 1997) and twin studies (Mitchell and Risch, 1992). A few candidate genes associated with cleft lip or cleft palate have been identified (for review see Schutte and Murray, 1999) using genetic linkage and association studies in humans (TGFP3 and MSX1, Lidral et al., 1998; TGFoc, Ardinger, 1989; Mitchell, 1997; R A R a , Chenevix-Trench et al., 1992) and transgenic mouse models (TGFP3, Proetzel et al., 1995; Kaartinen et al., 1995; MSX1, Satokata and Maas, -1994). In the future, it might be possible to identify- and reverse such defects early in development  and thereby avoid costly  surgical care and the social  difficulties  encountered by affected individuals.  13  The first step in preventing such birth abnormalities is understanding the processes involved in normal facial morphogenesis and the genes that regulate them. Errors at any step in facial morphogenesis can potentially result in clefting. Adding further complexity to these processes is that attaining a critical size of the prominences, removal of the epithelia, fusion and sufficient merging of the mesenchyme all need to occur within a narrow time window for normal development to occur.  Craniofacial developmental  biology therefore has direct relevance to the clinical treatment of cleft lip and other facial syndromes.  Unfortunately, the majority of orofacial clefts in mutant mouse embryos  involve the secondary palate or are midline clefts between the medial nasal prominences, and therefore are not particularly informative for understanding the etiology of cleft lip (see Francis-West et al., 1998). Therefore, a reproducible cleft lip that resembles the human defect would be an extremely valuable model system.  Phenocopying cleft lip in chickens  Chicken model system The five phases of face development are very similar in all vertebrates and are highly conserved through evolution.  Similarities in the origin of embryonic tissues, general  morphology (Hall and Horstadius, 1988), patterns of gene expression (Francis-West et al., 1998), and mechanisms of development of the face have allowed other vertebrates, notably rodents and chickens, to be used as valuable model systems to understand normal human craniofacial development and related disorders. The chicken model is particularly advantageous because of the ease with which the embryo can be directly manipulated in  14  ovo at all critical phases of face development. Several methods, including manipulations and/or exposure to teratogens have been used to generate unilateral or bilateral cleft lip in chickens (reviewed in Young et al., 2000). Similarly, exposure to teratogens in rodents prevents fusion and result in clefts of the lip, midface and palate (reviewed in Diewert and Wang, 1992).  Stage and position-specific manipulations can act as 'conditional  knockouts' or 'knockdowns'. In general these are much easier to perform in chickens than in mice. Traditional knockout studies potentially disrupt several processes in facial morphogenesis or facial defects may even be secondary to abnormal brain development (Opitz and Gilbert, 1982; Couly and Le Douarin, 1985; 1987). The chicken embryo allows us to target specific processes in primary palate formation in a spatial and temporal specific manner. By studying the molecular and cellular basis by which teratogens exert their effects, we can gain a better understanding of both normal and abnormal development.  Below are described several experimental approaches used to  study cleft lip in chicken embryos.  Hydrocortisone treatment An example of a teratogen that causes cleft lip in chickens is hydrocortisone (Peterka and Jelinek, 1983). A single (3ul) intraamniotic injection of hydrocortisone (O.lp.g/p.1) on day 4 (stage 21-24) resulted in a full bilateral cleft. Clefts were formed due to hypoplasia of the frontonasal mass prior to fusion. The reduced size of the frontonasal mass prevents contact with the maxillary prominence and fusion fails to occur. The suppressed growth must be temporary and/or restricted to the lateral regions of the frontonasal mass since, in every case, a full-length upper beak and egg tooth form.  15  Excision of epithelium Tissue (epithelial-mesenchymal)  recombination studies have shown that the facial  epithelium is required for outgrowth of the facial prominences (Wedden, 1987; Richman and Tickle, 1989). When the frontonasal mass or maxillary prominence was stripped of Shh expressing ectoderm in ovo, at stage 25-26, outgrowth of the corresponding prominence was interrupted and clefts resulted (Hu and Helms, 1999).  Extirpation at  earlier stages, or of adjacent S7z/2-negative frontonasal ectoderm, had no effect on subsequent facial morphology.  These data show that insufficient signaling from the  epithelium will prevent fusion and at later stages not all-epithelial regions are equal in supporting facial outgrowth.  SHH antibody By targeting genes specifically expressed  in the ectodermal region that supports  outgrowth of the frontonasal mass, we are able to understand the genetic requirements for outgrowth and fusion.  From the extirpation studies above, Hu and Helms (1999)  predicted that S H H might be supporting outgrowth.  To test this they implanted S H H  antibody-soaked beads into the frontonasal mesenchyme. Embryos treated in this manner mimicked the epithelial stripping defect; cleft lip occurred due to reduced outgrowth of the frontonasal mass.  16  Retinoic acid The discovery that retinoid excess causes dysmorphogenesis implicates retinoids as key patterning factors.  Excess retinoic acid (RA, all-trans, 13-cis, or 9-cis) is a potent  teratogen that causes an assortment of birth defects in many animals, including vertebral and craniofacial malformations (Cohlan, 1953; Kalter, I960; Hassell et al., 1977; Yip et a l , 1980; Jelinek and Kistler, 1981; Wiley et al., 1983; Fernhoff and Lammer, 1984). In humans, defects associated with excess retinoic acid include micrognathia, microtia and cleft palate but not cleft lip (Lammer et al., 1985).  Early studies in chickens examining  the role of excess retinoic acid on facial clefting were relatively non-specific. A retinoic acid solution was added systemically, leading to global effects on development (Hassell et al., 1977; Jelinek and Kistler, 1981). While some specificity is introduced by varying the timing of administration, it is difficult to identify primary versus secondary effects. Defects were much more consistent and easier to interpret once a carrier, formate-ion exchange beads, was used to locally apply retinoic acid (Eichele et al., 1984). At high concentrations, retinoic acid-bead implantations into the limb resulted in limb truncations and reproducible, full bilateral cleft lip (Tamarin, 1984; Wedden and Tickle, 1986). In chickens, retinoic acid inhibits outgrowth of the frontonasal mass (Wedden, 1987), maxillary (Helms et al., 1997) and lateral nasal prominences (Richman, unpublished data).  The mechanisms by which outgrowth is restricted in response to R A treatment  include changes in gene expression (Shen et al., 1997; Helms et al., 1997; Brown et al., 1997; Rowe et al., 1991), reduced proliferation (McGonnell et al., 1998) and the induction of apoptosis (Shen et al., 1997).  By placing beads directly into the face  (Richman and Leon Delgado, 1995; Shen et al., 1997) instead of under the apical  17  ectodermal ridge (Wedden and Tickle, 1986; Tamarin, 1984; Helms et al., 1997; Brown et al., 1997) a unilateral or bilateral cleft lip can be created in a dose-dependent manner (Richman and Delgado, 1995). To study phenotypes that more closely resemble human cleft lip, low doses of retinoic acid in the nasal pit can be used, which does not affect outgrowth of the nose. The disadvantage of the lower dose is that it is difficult to control the quantity of R A that is applied and as a result the magnitude of the cleft varies considerably (Richman and Delgado, 1995). Therefore, in order to generate consistant clefts we used higher dosages for our experiments.  Retinoic acid physiology Retinoic acid (RA), retinol and retinaldehyde, collectively known as retinoids, are biological derivatives of vitamin A and serve important roles in embryogenesis.  The  major form of vitamin A is retinol, which is stored in the liver or bound to retinol binding protein (RBP) in the circulation. Retinol uptake is mediated by a cell surface receptor for RBP (RBPr; Bavik et al., 1997; Ward et al., 1997).  Accumulation of retinol within  embryonic cells occurs in cells that express a high-affinity intracellular RBP, cellular CRBP I (Ong et al., 1994; Gustafson et al., 1993).  Target cells oxidize retinol to  retinaldehyde and subsequently to R A , which are enzymatic reactions carried out by retinol dehydrogenase (RDH) and retinal dehydrogenase (RALDH) respectively (Bhat et al., 1995; Penzes etal., 1997). There are two isoforms of retinoic acid, all-trans-RA and 9-cis-RA.  The physiological effects of RA, the active form of vitamin A , are transduced  by binding to its intracellular receptors.  18  R A Receptors The discovery of retinoic acid receptors (RARs) unraveled the mechanism by which R A induces its biological effects (Giguere et al., 1987; Petkovitch et al., 1987). RARs are ligand-inducible transcriptional regulators classified structurally and functionally as members of the steroid/thyroid hormone receptor superfamily of transcription factors. A second family of retinoic acid receptors, retinoid X receptors (RXRs), was isolated soon after (Mangelsdorf et al., 1990). Numerous isoforms of the receptors are known today, which are derived by differential promoter usage and alternative splicing. (RARoc, RAR(3 and RARy) can be activated by a\Urans-RA  The RARs  or 9-cis-RA but RXRs  (RXRct, RXRp, RXRy) are activated only by 9-cis-RA. The RXRs are unique in that they function as heterodimeric partners for not only RARs but also a number of other nuclear receptors. RARs therefore bind cooperatively with RXRs in the promoter region of target genes that contain R A response elements (RARE).  R A R homodimers bind weakly to  RAREs or not at all (Gudas et a l , 1994), whereas R X R homodimers can bind to certain promoter sequences containing R A R E and stimulate transcriptional activation (Lee et al, 1993; Wilson et al., 1992).  R A has been shown to regulate Hox (homeobox) gene  expression, genes that specify anterior-posterior polarity in vertebrates, through receptor binding to RAREs (Hoxal; Langston and Gudas, 1992; Hoxbl, Marshall et al., 1996; Hoxd4, Popped and Featherstone, 1993).  RAR and R X R are expressed in a wide range of organs, with much overlap in receptor expression during development, suggesting a role for R A in the morphogenesis of such tissues. RARct is almost ubiquitously expressed in all tissues (Ruberte et al., 1990; 1991)  19  however more specific localization is observed for RARp (head and face mesenchyme, nasal epithelium, eyes, ears, limb mesenchyme, kidney; Dolle et al., 1989; 1990; Ruberte et al., 1991; Smith and Eichele, 1991; Rowe et al., 1991) and RARy (in cartilage, sclerotome and keritinizing epithelia; Ruberte et al., 1990). R X R expression patterns show less tissue specificity than RARs (Dolle et al., 1994). In the chick face, RAR{3 is expressed in all the facial prominences (Rowe et al., 1991; Rowe et al., 1992; Smith et al., 1991). The localized function of R A is not only dependent on the presence of the R A receptors, but also on the availability of retinoids, ie. RA. RALDH-2 transcription in the chicken face is localized to the nasal placodes and the developing branchial arches of stage 17 embryos (later stages were not reported; Berggren et al., 1999). The localization of RALDH-2 to the face provides further evidence to support a role for endogenous R A in face morphogenesis.  Biological functions of endogenous retinoic acid Retinoids play essential roles in vertebrate development. Evidence to support this comes from a spectrum of congenital malformations exhibited by mouse fetuses from females with inadequate dietary intake of vitamin A.  The features of vitamin A deficiency  (VAD) syndrome include defects in heart and aortic arches, urogenital tract, respiratory tract, diaphragm, eye and craniofaciakmorphogenesis (cleft face, palate and lip; MorrissKay and Sokolova, 1996). Vitamin A deprivation in quails causes development to arrest at about stage 20/21, prior to face formation. Embryos exhibit defects in limb (Stratford et al., 1999), hindbrain (Maden et al., 1996), heart (Heine et al., 1985) and somite formation (Maden et al., 2000).  20  Genetic studies of R A endogenous functions Mouse knockouts of various RARs and RXRs have been used to evaluate the function of R A and, more precisely, the RARs and RXRs in vivo. Mice lacking one R A receptor are normal and fertile (Li et al., 1993; Lohnes et al., 1993; Lufkin et al., 1993; Mendelsohn et al., 1995).  Either the phenotype is too subtle and is not manifested in the detection  procedures or some of the functions of the mutated receptor are taken over by another. Disruption of all isoforms of the R A R receptors and double receptor knockout mice are more informative. R A R compound mutants exhibit all malformations typical of fetal VAD,  including in limb, axial and craniofacial morphogenesis (in embryos with  disruption of all isoforms of R A R a ; Lufkin et al., 1993, or RARy; Lohnes et al., 1993). In the face, RARa/y double knockout mice exhibit median clefts of the primary palate (Lohnes et al., 1994). This defect was duplicated in embryos exposed to a R A R panantagonist, confirming that endogenous  R A is required for facial morphogenesis  (Kochhar et al., 1998).  The above studies identify a developmental requirement for retinoids and suggest that a mechanism to maintain a critical balance of retinoid activity must be present during normal face morphogenesis. JSeyere. vitamin A deficiency is uncommon in humans, Jjut some investigators feel that mild deficiencies combined with genetic defects may be an underestimated cause of human abnormalities (Morriss-Kay and Ward, 1999). This may be especially true for those with a genetic disposition to cleft lip.  21  Genes downstream of R A teratogenicity The mechanism by which R A induces craniofacial defects at a molecular level is starting to be understood. Retinoic acid decreases the expression of shh (Helms et al., 1997) and fibroblast growth factor homologous factor-4 (FHF-4; Mufioz-Sanjuan et al., 2001). Expression of fg/8 (Helms et al., 1997), fgf receptors (Fgfr2 and Fgfr3; Richman and Delgado, 1995) and Bmp-2 (Munoz-Sanjuan et al., 2001) are unaffected. However, in my opinion Bmp-2 appears to be upregulated in Figure 3L of Munoz-Sanjuan et al. (2001). Retinoic acid also alters the expression of numerous transcription factors. For example, RAR/3 is ectopically expressed (Rowe et al., 1991) while activating protein-2 (AP-2; Shen et al., 1997), Msx-1 and Msx-2 (Brown et al., 1997) are all downregulated following R A application. Several genes shown to be downregulated by R A treatment have later been shown to be associated with cleft lip in chickens (Shh, Hu and Helms, 1999), cleft lip and/or palate in humans (MSX1; van den BooGaard et al, 2000), cleft palate (Msx-1, Ichiro and Maas, 1994) or cleft lip and palate in mice (Msx-1 and Msx-2, Maas, personal communication). Thus, understanding the molecular mechanism by which R A induces cleft lip has proven to be an excellent model system to identify genes involved in facial morphogenesis/dysmorphogenesis.  This approach also provides another method to  identify candidate genes potentially involved in orofacial clefting.  S c r e e n i n g s t r a t e g i e s to i s o l a t e n o v e l , d i f f e r e n t i a l l y e x p r e s s e d genes  Control of gene expression lies at the heart of embryology, cell biology and neoplasia. The examination of expression changes in embryos treated with teratogens or other manipulations is an excellent way to identify genes that may mediate normal and  22  abnormal morphogenesis. Genes whose expression is altered in the position of the future defect are studied. The advantage is that genes selected in this manner are likely to be causally related to the defect. The disadvantage is that these techniques examine genes that have already been identified and characterized, thus missing many other potentially important genes. A number of techniques are available for the detection and cloning of novel transcripts that are differentially expressed.  Subtractive hybridization Subtractive hybridization enriches those sequences of R N A that are expressed at different levels between two cell populations (Sargent, 1987).  Complementary D N A (cDNA)  prepared from the tissue of interest (for example, the frontonasal mass) is hybridized to an excess of (complementary) poly (A)+ R N A from a second, similar tissue type, for use as a subtractive driver. The unique cDNAs that do not anneal to the driver poly A+ R N A are isolated (for example, by column chromatography). The enriched cDNAs are then used to generate subtractive cDNA libraries for screening, or to make radiolabeled probes to screen libraries differentially.  Subtractive hybridization is a well-established  technique that has the potential to clone previously unidentified transcripts. Limitations in this technique include that it is labor intensive, requires use of unstable poly A+ RNA, is prone to picking up abundant sequences, and either up-regulated or down-regulated genes can be identified, but not both. The advantage of this procedure is that, even if it doesn't identify significant genes, the tissue-specific cDNA library can be used for other experiments.  23  Differential display Differential display (Liang and Pardee, 1992) compares the mRNA content of different tissue or cell culture extracts. Following reverse transcription (using a primer anchored in the poly-A region) and polymerase chain reaction (PCR; with degenerate primers) a population of radioactive cDNAs are generated that are representative of the m R N A species present in each extract. Different primer pairs are used in independent reactions in order to amplify a distinct subset of mRNA transcripts. The PCR is performed with a radiolabeled nucleotide so that the products of the reactions can be resolved by electrophoresis  and then  visualized  using  autoradiographic film.  The  genes  corresponding to differential band intensities can then be purified, reamplified and either cloned or used as probes to screen cDNA libraries. Differential display uses small tissue amounts, total R N A and allows the simultaneous analysis of multiple tissues. technique, however,  can also be labor intensive  (especially  This  when many primer  combinations are used for adequate representation of mRNAs), often generate false positive clones, and true positive clones are sometimes identified as rRNA or housekeeping genes.  This strategy also preferentially amplifies 3' non-coding parts of the  message, which restricts direct database sequence comparisons.  At the end of the  procedure, a library-screening step is required to clone the full length coding sequence.  R N A arbitrarily primed (RAP-) PCR RAP-PCR is another R N A fingerprinting strategy similar to differential display (Welsh et al., 1992).  RAP-PCR, however, uses an arbitrary primer in the reverse transcription  reaction instead of a primer anchored in the poly A region. The PCR reaction is then  24  performed with this same primer along with many possible second arbitrary primers. The advantages and limitations of using the differential display technique also apply to RAPPCR.  RAP-PCR, however, eliminates the problem of preferentially amplifying the 3'-  untranslated region of transcripts.  Microarrays Microarray technology allows the simultaneous monitoring of gene expression for tens of thousands of genes and has been described as a recent breakthrough in experimental molecular biology (see Duggan et al., 1999). A microarray is a glass slide, onto which single-stranded D N A molecules are attached at fixed locations. There may be tens of thousands of spots on an array, each representing a single expressed sequence from a particular tissue. Microarrays exploit the preferential binding of complementary singlestranded nucleic acid sequences and thereby allow the comparison of mRNA abundance from two different samples. Briefly, isolated R N A from a sample population and control or reference populations are labeled with a red and a green dye, respectively. Both R N A extracts are washed over the array, allowing complementary gene sequences to hybridize with sequences on the spots.  Excitation from laser emissions causes the dye labels to  flouresence, allowing the estimation of the relative expression levels of the hybridized genes in the sample compared to the reference population. For example, if the mRNA complementary to a spot on the microarray is most abundant in the sample the spot will be red but if the mRNA is greater in control populations it will be green. If R N A from both populations bind equally well the spot will be yellow and if there is no binding it will be black.  By obtaining the 'gene expression profile' from nearly all-possible  25  conditions (that is, different developmental stages, assorted tissues or dysplasia) and combining all of this information in a database, the dynamic expression of each gene in the genome can be characterized.  There are several advantages of using microarrays over other differential hybridization techniques: 1) in a single run, microarrays allows the analysis of up to 40,000 genes (depending on the size of the array); 2) each of the cDNAs has already been cloned so the identity of the differentially-expressed gene is clear ie. housekeeping genes can be eliminated from consideration; 3) there are built in positive controls with genes that are known to be differentially expressed; 4) the cDNA on the slide has been 'normalized' so that only rare, interesting sequences are represented. The disadvantages are that a species and preferably stage specific array is preferred because the stringency for hybridization is very high. Thus, for our work we need a chicken face microarray, which we would have to make ourselves.  Candidate genes involved in facial development The expression patterns of many genes have been described in the face, several of which may have roles in patterning, epithelial-mesenchymal interactions, proliferation and differentiation (Francis-West et al., 1998). Examples of potential signaling molecules that are localized to specific regions and tissues within the -face include FGFs (fibroblast growth factors; Wall and Hogan, 1995; Barlow and Francis-West, 1997; Richman et al., 1997;  Munoz-Sanjuan et al., 2001;  Bachler and Neubuser, 2001), B M P (bone  morphogenetic protein; Francis West et al., 1994; Wall and Hogan, 1995; Wang et al.,  26  1999; Shigetani et al., 2000) SHH (Sonic hedgehog) families (Wall and Hogan, 1995; Helms et al., 1997; Ahlgren and Bronner-Fraser, 1999), Wnt5a (Wingless) family (Dealy et al., 1993; Yamaguchi et al., 1999; Ladher et al., 2000). A l l of the aforementioned genes are secreted proteins that signal through cell surface receptors. In addition, there are many transcription factors (Msx-1, Msx-2, AP-2, Slug, Tbx-2, Tbx-3, Dlx-1,2,3,5, Barx-1, R A R a , RARp, RARy, Gli-2,3, Gsc) expressed iri the face (Francis-West et al., 1998; Davideau et al., 1999; Barlow et al., 1999; Shen et al., 1997), some of which lead to facial defects when they are knocked out (for review see Francis West et al., 1998). Subsets of these genes have been shown to work within the same signalling pathway (eg. BMPs and Msx genes) in the face (Barlow and Francis-West, 1997; Wang et al., 1999). In choosing genes to study in detail in this thesis, we selected a growth factor family, BMPs, which have known expression in the primary palate (Francis-West et al., 1996) and for whom reagents exist to study function in facial morphogenesis.  TGF-P superfamily Over the last 20 years, a large family of secreted signaling molecules, the transforming growth factor-P family, has been isolated that regulates a plethora of biological activities during normal growth and development. An invariant feature of these growth factors is the presence of seven cysteine residues, six of which are closely grouped to make-arigidstructure called a cysteine knot (Daopin et af, 1992; Schlunegger and Grutter, 1992) within the mature peptide region. Based on sequence comparisons between their bioactive domains, members of this family are divided into subfamilies (Massague, 1998). A major subdivision is the BMP family.  27  Bone morphogenetic proteins Pioneering work by Urist and his colleagues demonstrated that demineralized bone or dentin extracts could induce ectopic bone and cartilage when implanted either subcutaneously or intramuscularly in rats (Urist 1965: Urist et al., 1973, 1979). The search for factors responsible for this effect resulted in the cloning of the first BMPs more than 20 years later (Wozney et al., 1988). To date, more than 30 members of this family have been identified in organisms ranging from sea urchin to mammals, and are termed BMPs, osteogenic proteins (OPs), cartilage-derived morphogenetic proteins (CDMPs), growth and differentiation factors (GDFs) or inhibin. These subclassification are segregated according to structural similarities (for review see Ducy and Karsenty, 2000).  Biologically active BMPs are glycosylated dimeric molecules (Griffith et al., 1996). Whether BMPs normally exist solely as homodimers is not known but BMP-4/7 recombinant heterodimers have greater bioactivity than homodimers (Hazama et al., 1995; Suzuki et al., 1997). The homologs of many mammalian BMPs have been identified in invertebrates such as Drosophila (decapentaplegic, dpp) and Caenorhabditis elegans (Daf-7; Ren et al., 1996), and in vertebrates such as zebrafish and Xenopus (Vgl; Weeks and Melton, 1987). BMP-2 and BMP-4 are so similar to dpp, sharing 84-88% sequence homology, that they can substitute for one another in biological assays. BMP-4 can rescue embryonic defects in dpp mutant flies (Padgett et al., 1993) and recombinant dpp can induce endochondral bone formation in mammals (Sampath et al., 1993).  28  The name, bone morphogenetic protein, is misleading because there is tremendous genetic and experimental evidence that these molecules regulate numerous biological processes.  Expression and functional studies have identified B M P to be involved in  patterning, morphogenesis and differentiation of embryonic organs and tissues where reciprocal interactions between epithelial and mesenchymal cells are important (for review see Hogan, 1996; Kingsley, 1996; Wozney, 1998).  B M P Receptors and BMP signaling molecules The biological effects of BMPs are mediated by transmembrane serine/ threonine kinases cell surface receptors, classified structurally and functionally as type I and type II. The receptor types are approximately 55 kDa (-500 amino acids) and 70 kDa (~570 amino acids) glycoproteins, respectively. Two type I receptors have been identified, BMPR-IA (ALK-3, BRK-1) and BMPR-IB (ALK-6, BRK-2; for review, see Massague, 1998). Heterotetramers of type IA, type IB and type II receptors form when ligand binds (Yamashita et al., 1994). BMPRs are evolutionarily conserved and homologs of the type I and type II receptors have been isolated in non-vertebrates like Drosophila (Thick veins (Tkv), Saxophone (Sax), and Punt; Nellen et al., 1994; Brummel et al., 1994; Xie et al., 1994) and C. elegans (Daf-1 and Daf-4; Georgi et al., 1990;_Estevez et al., 1993).  The extracellular and intracellular domains of the BMP receptors are approximately 150 and 475 amino acids, respectively.  Immediately preceding the type I receptor kinase  domain is a highly conserved 30 amino acid region call the glycine/serine-rich (GS)  29  domain because of its characteristic SGSGSG sequence (Wrana et al., 1994). The GS domain regulates the receptors' kinase activity. Mutation of a glutamine (Q) residue to aspartic acid (D) in the GS domain activates the kinase domain and constitutive signaling activity in the cell (Zou et al., 1997). Downstream of the receptors are receptor-activated Smads (1,5,8) that can activate gene transcription (Heldin et al., 1997; Massague, 1998; Hoodless and Wrana et al., 1998).  BMPs in embryonic development Studies involving the localization of BMP signaling molecules, ectopic gain-of-function and loss-of-function studies demonstrate a broad range of biological functions for BMPs in addition to post-fetal cartilage and bone induction.  These functions include cell  proliferation, differentiation, apoptosis, morphogenesis, and organogenesis.  Early in development, BMPs regulate dorsal-ventral axis formation. In Drosophila, dpp acts as a dorsalizing factor (Ferguson and Anderson, 1992; Wharton et al., 1993) while in Xenopus, BMP-4 is a ventralizing factor during mesoderm formation (Jones et al., 1996). BMPs therefore act as morphogens, by providing positional information necessary for pattern formation (Kay and Smith, 1989; Wolpert, 1989). In addition, BMPs are implicated in the patterning of other organs such as the somites, teeth, feathers, limbs, forebrain and head (Drossopoulou et al., 2000; Golden et al., 1999; Tonegawa and Takahashi, 1998; McMahon ef al., 1998; Capdevila and Johnson, 1998; Noramly and Morgan, 1998; Marcelle et al., 1997; Vainio et al., 1993).  30  The role of endogenous BMPs can be studied with knockout mice or with the application of antagonists.  Knockout mice have not been particularly informative for facial  morphogenesis since they either die very young or have no facial phenotype (BMP-2, Zhang and Bradley, 1996; BMP-4, Winnier et al., 1995; BMP-7, Dudley et al., 1995; Luo et al., 1995; BMPR-IA, Mishina et al., 1995; BMPR-IB, Baur et al., 2000; L i et al., 2000; BMPRII, Beppu et al., 2000).  The list of BMP antagonists is growing and includes  Noggin, Follistatin, Chordin, Gremlin/Drm, Caronte/Cerberus and Dan (Smith, 1999). The antagonists bind to BMPs and prevent them from interacting with and activating their receptors (Zimmerman et al., 1996; Piccolo et al., 1996, 1999; Hsu et al., 1998; Yokouchi et al., 1999; Esteban et al., 1999). Developmental processes regulated by B M P antagonists include, joint formation (Noggin, Brunet et al., 1998), patterning the neural tube and somites (Noggin, McMahon et al, 1998), hair-follicle induction (Noggin, Botchkarev et al., 1999), limb patterning and outgrowth (Gremlin, Zuniga et al., 1999; Capdevila et al., 1999; Merino et al., 1999), left-right asymmetry (Caronte, Esteban et al., 1999; Yokouchi et al., 1999; Cerberus, Zhu et al., 1999), overall growth, secondary palate formation, tooth and skeletal development  (follistatin; Matzuk et al., 1995), ear  development (Chordin; Bachiller et al., 2000) and primitive streak formation (Chordin; Streit et al., 1998). In Xenopus, dorsalizing factors from the dorsal lip of the blastopore, Spemann's organizer, were identified as B M P antagonists (Noggin, Chordin and Follistatin) and later the level of B M P activity was shown to be a key regulator of dorsalventral axis formation (De Robertis and Sasai, 1996).  B M P activity, balanced by its  antagonists, also governs whether ectoderm will form epidermis (in the presence of B M P activity), or neural tissue (in its absence; Lamb et al., 1993; Hemmati-Brivanlou et al.,  31  1994; Sasai et al., 1994, 1995; Wilson and Hemmati-Brivanlou, 1995). Only noggin and gremlin/drm have been localized to the chicken face. Noggin is expressed in frontonasal mass epithelium (see Chapter 2). At stages 22-26, gremlin is expressed in the maxilla, mandible, second and third branchial arches but not the prominences giving rise to the upper beak (Bardot et al., 2001). In the mouse face, chordin is ubiquitously expressed (Scott et al., 2000). Follistatin is expressed in the mesenchyme surrounding the nasal epithelium and in the tongue (Feijen et al., 1994). Noggin is expressed in the face of older mouse embryos, after cartilage differentiation (Brunet et al., 1998). Cerberus-like does not appear to be expressed in the mouse face (Stanley et al., 2000), although D A N is (Pearce et al., 1999).  The complexity of the BMP signaling system indicates that their activity is very closely regulated. The various defects following mutations in BMP signaling molecules attest to critical functional roles for BMPs in a wide range of organs and phases of development. Their intricate functioning makes them good candidates for the complex control of facial morphogenesis.  AIMS OF THE PRESENT STUDY  In the present study we explore the functions of BMPs during chicken facial morphogenesis.  The general aims of this project were:  32  1. To identify candidate genes involved in cleft lip; 2. To test the function of the candidate genes in normal and abnormal facial development;  3. To identify genes that are important in craniofacial skeletal development.  The specific aims were: 1. Use differential display to identify genes associated with RA-induced cleft lip in chickens; 2. To localize the expression of BMP ligands, BMP receptors and antagonists during the different phases of craniofacial morphogenesis; 3. To study the function of B M P signaling in primary palate morphogenesis and dysmorphogenesis; 4. To determine the roles of BMP receptor signaling in craniofacial skeletogenesis.  33  CHAPTER II  I s o l a t i o n o f genes a s s o c i a t e d w i t h r e t i n o i c a c i d - i n d u c e d cleft l i p i n c h i c k e n s  Introduction  Cleft lip with or without cleft palate (CL/P) is a commonly occurring birth defect that occurs when the facial prominences surrounding the nasal pit fail to fuse together. Fusion takes place primarily between the frontonasal mass (or medial nasal prominences in mice and humans) and the maxillary prominences. Morphological defects early in face development, such as reduced size (or outgrowth) of the facial prominences, have been associated with orofacial clefts (Trasler, 1968; Diewert and Wang, 1992). The reduced size prevents contact between the prominences and fusion is unable to take place. Identifying the molecular regulators of facial outgrowth would help understand normal morphogenesis and disorders resulting from alterations in this developmental process.  To address this issue we used the RA-cleft lip-chicken model system, which has numerous advantages.  The chicken embryo was used because it resembles mammalian  embryonic face and is easily manipulated at nearly all stages of face development. When the chicken "embryo is exposed to excess R A a full bilateral cleft lip is generated in a consistent and reproducible manner (Tamarin et al., 1984, Richman and Delgado, 1995). The RA-induced clefts are the result of reduced outgrowth of the prominences giving rise to the upper beak (frontonasal mass (Wedden, 1997) and the paired-lateral nasal (McGonnell et al., 1998) and maxillary prominences (Helms et al., 1997; McGonnell et  34  al., 1998). The frontonasal mass forms most of the upper beak and is most affected by R A (Tamarin et al., 1984). In spite of being in close proximity the paired-mandibular prominences outgrow normally and the lower beak is unaffected even at very high R A concentrations.  Careful analysis, shortly after R A treatment, shows that outgrowth is  reduced due to elevated levels of apoptosis, reduced proliferation and changes in the directional expansion of the face (McGonnell et al., 1998). These changes also correlate with changes in expression of several transcription factors and secreted signaling molecules (see thesis introduction for details).  This method however only allows  analysis of genes that have already been isolated. Here we combined the RA-cleft lipchicken model system with differential display-polymerase chain reaction (DD-PCR) in order to identify new and novel genes downstream of R A that are potentially involved in outgrowth of the facial prominences.  35  Materials and Methods  Treatment of embryos with retinoic acid. Fertile, white leghorn chicken embryos are incubated at 3 8 ° C until they reach stage 20. AG1X-2 beads (Biorad, formate form) of 100pm diameter are soaked in 5mg/ml retinoic acid (dissolved in dimethyl sulfoxide, DMSO) for 20 minutes.  Beads are rinsed 3 times in culture medium for a total of 20  minutes and implanted into the right nasal pit of the host embryo (Eichele et al., 1984; Richman and Leon Delgado, 1995). Treated embryos are returned to the incubator for 24 hours.  In order to generate irreversible full bilateral clefts the embryos required a  minimum exposure of 16 hours (Wedden and Tickle, 1986; Richman and Leon Delgado, 1995).  Dissection of tissue. Once embryos reach stage 24 (24 hours post-bead implantation), they are removed from the egg and the frontonasal process is dissected in chilled phosphate buffered saline (PBS) solution using iridectomy scissors and fine forceps. Batches of explanted frontonasal tissue are rapidly frozen in liquid nitrogen and stored at -70°C.  RNA extraction. Two separate batches of frozen R A treated frontonasal prominences and two batches of non-treated frontonasal promiences were homogenized in guanidium isothiocyanate (GITC) and then R N A was extracted in acid phenol (Chomzynski and Saachi, 1987). Briefly, tissue was solubilized in denaturing solution (4M GITC, 25 m M NaCitrate pH 7.0, 0.5% N-Lauryl Sarcosine, 0.1M P-mercaptoethanol) with a hand-held glass homogenizer.  To this homogenized mix was added, 1 vol water saturated phenol,  0.2 vol chloroform:isoamyl alcohol (49:1), 0.1 vol 2 M NaAcetate pH 4. The combined  36  mix was vortexed, set on ice for a minimum of 15 minutes and then centrifuged in a microfuge at 10,000 rpm for 20 minutes (4°C). The upper aqueous layer was removed precipitated with 1 vol isopropanol (mix, let sit for 1 hour at - 2 0 ° C , spin at 10,000 at 4 ° C to pellet).  The pellet was then resuspended in denaturing solution, precipitated and  pelleted as described before.  At this time the pellet was washed with 75% ethanol to  remove salts and resuspend in DEPC-H2O.  Any contaminating genomic D N A was  removed from all samples by treating the total R N A for 30 minutes at 3 7 ° C with 10 U RNase-free DNase I followed by phenol-chloroform extraction and ethanol precipitation. In parallel, total R N A was extracted from normal stage 24 frontonasal prominences.  Northern blot analysis. Total R N A (lOpg) was resolved by electrophoresis on a 1.2% formaldehyde gel and transferred to a nitroplus 2000 membrane (Micon Separations Inc.). Blots were prehybridized with 50% formamide, 5x SSC, lOxDenhardt's solution, 0.1 mg/ml denatured salmon sperm DNA, 0.2% SDS, 0.05 M sodium phosphate buffer pH 6.8, and 0.001M E D T A for 2 hours at 4 2 ° C .  Hybridization was performed for 16  hours at 42 C in the above prehybridization mix plus 10% dextran sulphate with 2-5x10 cpm of P-labelled probes prepared according to methods described in the T7-random 32  priming kit (Pharmacia).  Filters were washed twice at 65°C with 2xSSC containing  0.5% SDS for 30 minutes, followed by 0.2xSSC and 0.5% SDS for 40 minutes before exposure to X-ray film at -70°C.  Reverse transcription-polymerase chain reaction (RT-PCR) differential display (DD). Differential display was performed essentially as described by Liang et al., 1993. Primers anchored to the poly A-region are used in the reverse transcription reactions. Three R T reactions were performed using the total DNA-free R N A (0.5 pg RNA/reaction) in lx  37  transcription buffer; 10 mmol/L dithiothreitol; 20  umol/L each deoxynucleotide  triphosphate (dNTP; dGTP, dATP, dCTP, dTTP); and 1 umol/L of either H-T11G, H T11C, H-T11A oligonucleotide primer (where H is a Hindlll restriction site). Samples were heated to 6 5 ° C for 5 minutes and cooled to 3 7 ° C for 10 minutes before addition of 100U Moloney Murine Leukemia Virus RTase. After incubation at 3 7 ° C for one hour, the reactions were heated to 9 5 ° C for 5 minutes before storage at - 2 0 ° C . A proportion of the reverse transcribed products was amplified by PCR using the original c D N A primer plus one of the five arbitrary primers. PCR was performed on a Perkin-Elmer Thermal Cycler (94°C for 30 seconds, 5 2 ° C for 2 minutes, 72 minutes for 30 seconds for 40 cycles, followed by a 5-minutes extension at 72°C).  The 20pl PCR reaction mix  contained 2 ul RT cDNA template, 50 mmol/L K C L , 10 mmol/L Tris-HCL (pH 8.3), 2.5 mmol/L MgC12, 0.5 umol/L H - T l IN primer, 0.2 umol/L arbitrary primer, 2 umol/L each dNTP, 10 deoxynucleotide triphosphate (dNTP; dGTP, dATP, dCTP, dTTPCi 35S-ATP (Dupont N E N , Albany, NY), and 1 U Taq D N A plymerase. Arbitrary primers used were as described in the GeneHunter Kit (E-JR-A, gcgaattctatccatg; E-JR-B, gcgaattctgattgcc; E-JR-C, gcgaattctgctctca; E-JR-D, gcgaattcttgcttga; E-JR-E, gcgaattctagtatgg; where E is an EcoRl restriction site). PCR products from both samples were electrophoresed together to enable identification of bands with differing intensities.  PCR amplification  products (4ul) were denatured by heating for 3 minutes to 8 0 ° C in sequencing geLstop buffer and electrophoresed at 55 W constant power on a 6% denaturing polyacrylimide gel. Gels were dried onto Whatman 3 M M filter paper (VWR, Seattle, WA) and exposed for 1 to 4 days to Kodak XAR-5 film (CMX, Seattle, WA) with fluorescent ink orientation markers. These markers were used to accurately position the radiograph over  38  the gel during excision of the band. Differential expression was verified by multiple D D PCR reactions, using different R N A preparations and numerous gel electrophoesis. A l l DD-PCR reactions were repeated at least twice using each reverse transcription reaction product. Two reverse transcription reaction products were generated from two separate R N A preparations. Each DD-PCR reaction was run on a sequencing gel twice.  Amplification and cloning of differential display gel bands. Bands that appeared to consistently have different band intensities from control samples were excised from the dried sequencing gels with a razor blade. Therefore, only bands clearly differential on at least 4 gels were only selected. The gel piece was soaked for 15 minutes in 100 ul H 0 , 2  then boiled for 15 minutes to elute the DNA. The eluate was transferred to a fresh tube. The D N A was precipitated by addition of 3mmol/L sodium acetate (pH 5.2) to a concentration of 0.3 mol/L and 2.2 volumes of 100% ethanol, followed by 30 minutes of incubation at - 8 0 ° C , and centrifugation at 13000 rpm for 15 minutes in a microfuge. The D N A pellet was washed in 70% ethanol and resuspended in water. Reamplification by PCR was performed in 40 pi using the appropriate primers and conditions as described above except for dNTP concentrations of 20 umol/L and no isotope. In parallel a fraction of the eluate was amplified in the presence of  35  S - A T P and electrophoresed, on a  sequence gel, next to the original DD-PCR product to ensure that the re-amplified product was the exact size of the band that was cut out from the gel. Four microliters of the above reaction was reamplified a second time under the same conditions for 40 cycles. The products of both amplifications were digested with Hindlll and EcoRI, than separated on a 2% agarose gel. A well was made below the bands of interest by excising a strip of agarose and the D N A bands were electrophoresed into the well containing 5x  39  T A E buffer. The D N A / T A E buffer was removed using a pasteur pipette and precipitated as above with sodium acetate. The resuspended D N A was used directly for cloning into the EcoRI/Hindlll site of a pBS+ vector (Stratagene, La Jolla, CA). For each excised band, a minimum of 5 white (recombinant) colonies was picked. Recombinant plasmid D N A was purified using the miniprep procedure described by Sambrook et al., (1989) and tested for inserts of appropriate size by excising the insert with EcoRl and Hindlll and viewing on a 2% agarose gel. D N A sequencing was performed by the Nucleic Acid Protein Service (UBC).  Database searches. The combined E M B L and GenBank databases were searched with BLAST.  40  Results  Tissue collection and RNA purification. A stage 24-frontonasal prominence yields approximately 0.16 pg of total RNA. In order to attain sufficient quantities of total RNA for northern analysis and differential display, approximately 500 RA treated and 400 control frontonasal prominences were collected. RNA extraction procedures yielded a sum of 200 u.g and 135 u.g of total RNA, respectively. Two separate preparations for each sample set were made and kept separately. The quality of the RNA samples was determined for each sample preparation by assessing the integrity of 18s and 28s bands when total RNA (20 pg) was electrophoresed on a MOPS-formaldehyde gel. In addition, northern hybridization using a cyclophilin or type I collagen probe indicated 'tight' bands. The absence of tailing indicated that the RNA was of high quality and was not partially degraded.  Display of differentially expressed sequences. Thirteen bands were identified as being differentially expressed (Fig. 2.1). Six bands were upregulated and seven downregulated following RA treatment. DNA eluted from the bands of interest was amplified by PCR using the original primers. Selected clones were taken from the top of the gel and determined to be 200-250 basepairs in length. Two of the 13 differential expressed products were subcloned into pBS+ vector and sequenced. The sequencing information obtained was used to compare with gene sequences in Genbank and EMBL databases, but unfortunately no significant homology to any reported sequence was found.  Isolated clones are ubiquitously expressed in the face. Radiolabeled probes were made from the two sequenced clones and used in sectioned in situ hybridization experiments. In 41  F i g u r e 2.1 A u t o r a d i o g r a p h y o f p a t t e r n s o f g e n e e x p r e s s i o n f o l l o w  DD-PCR. Left and  right lanes represent RA-treated and normal frontonasal tissues, respectively. DD-PCR was performed using the H-T11A anchored primer and E-JR-A arbitrary primer. Note that samples were not loaded evenly; there are greater amounts of normal PCR products. Despite the uneven loading, band numbers 3 and 4 were still upregulated in R A frontonasal mass tissue.  42  RA N  situs were performed on stage 20, 24 and 28 chicken face sectioned in the frontal and sagittal planes. Both genes were ubiquitously expressed.  43  Discussion  Our study shows that the differential display technique can be potentially used to identify genes that are differentially expressed between R A treated and normal frontonasal prominences.  We were able to detected 13 potential genes whose transcription was  altered in response to R A application.  We had proposed to confirm differential  expression of the remaining 11 clones using northern and in situ hybridization techniques. If the clones were differentially expressed in the R A treated face we would then screen a c D N A library to pull out a longer clone. This library-based clone would have no P C R errors in the sequence and would include coding sequence that can be used to identify the gene. Genes that are not differentially expressed in in situ hybridization screening but found to be expressed in the face may still be of interest as they may be novel genes.  There are however numerous limitations to differential display. preferentially amplifies abundant transcripts.  This technique  The differential expression of isolated  clones is often found to contain false positives and are not actually expressed at different levels in the two cell populations when examined using an independent technique (like northern hybridization; Donohue et al., 1994; Aiello et al., 1994).  The procedure  requires a lot of subcloning (11 of our clones still needed to be subcloned),-whichis a difficult procedure.  Clones are often ribosomal R N A or housekeeping genes and  therefore uninformative. Clones are predominantly, i f not exclusively, localized to the 3'-untranslated portion of a gene as a result of the oligo-dT-based c D N A primers. This is particularly a disadvantage when studying the chicken model system because the  44  untranslated flanking regions of chicken genes are very seldom reported in the D N A database. The aforementioned limitations in differential display lead us to reconsider whether it would in our best interest to continue with this project.  Despite all the problems mentioned, classical differential display still offers several advantages over screening methods, like subtractive-hybridization: 1) it requires only a relatively small amount of DNA; 2) it is a relatively simple technique and does not require special reagents or materials; 3) uniquely expressed RNAs  enriched or  underexpressed in two or more different tissues or cell types can be identified simultaneously,  in contrast to subtractive hybridization, which selects for those  overexpressed in a single tissue; and 4) it eliminates the danger of removal of novel genes sharing stretches of sequence with abundant members of gene families. A particularly advantageous modification to the classic differential technique uses degenerate oligos specific for conserved protein motif-encoding sequences instead of arbitrary primers (Stone and Warton, 1994). A similar modification was considered early in the project where the total R N A would be reverse transcribed, using a poly-T primer, and the resulting cDNA would be used for differential amplification by PCR using 2 arbitrary primers. This modification would have avoided the selective amplification of the 3'-end of a gene. This method, however, still preferentially amplifies abundant genes.  Examination of differential expression in the year 2001.  Differential display continues to be used but contains modifications to avoid the difficulties accompanying this procedure.  For example, a combination of differential  display and long-distance PCR protocols have been used to amplify and comparatively  45  display highly reproducible cDNA fragments ranging in size from 150 bp to 2 kb (Jurecic et al., 1998). amplified.  This combined method assures that the coding region of a gene is  The recent advent of monitoring gene expression using microarrays is in  many ways an extension of differential display. cDNA microarray technology provides a format to simultaneously compare the relative differences in expression levels of thousands of genes between samples, and in a single hybridization assay.  Numerous  improvements have been made to this procedure (microarray preparation, some human and mouse arrays are now commercially available, probe labeling, hybridization, software for image analysis and data analysis; Duggan et al. 1999; Cheung et al., 1999; Bassett et al., 1999; Loftus et al., 1999; Tamayo et al., 1999; Zweiger, 1999). Besides technological improvements, cDNA microarrays have been combined with other screening methods to facilitate the analysis of gene expression. For example, microarrays were used for expression analysis of cDNA clones from breast cancer cell lines that were generated by a modified subtractive hybridization procedure (Yang et al., 1999). By incorporating a preliminary subtraction step (called library normalization), this depletes sequences common to the target and driver samples. The application of this technology to development has been hampered by the lack of appropriate microarrays. Some have been reported for mouse embryos (Tanaka et al., 2000). require a chicken embryo cDNA microarray.  For our studies, we would  One was used to identify genes  downstream of Lyn, a member of the Src family of cytoplasmic signal transducers (Korade-Mirnics et al., 2000).  However, pre-made microarrays are generally of  utilitarian design and do not sufficiently accommodate organ specific research. The most important feature of a microarray is that it be species specific, then stage specific and  46  finally organ specific. Ideally we would require a chicken embryo, stage 20-28 face microarray. Microarrays have only just begun to be applied but they have already generated a wealth of information.  Contributions to this chapter  The scientific ideas for the preceding chapter were primarily the supervisor's.  I  contributed in selecting the procedure for isolating differentially expressed genes. The data presented in this chapter were almost completely collected by me. Dr. Todd Wilke performed the in situ experiment.  47  C H A P T E R III Bone Morphogenetic Proteins, B M P Receptors and Noggin: Expression d u r i n g C r a n i o f a c i a l Morphogenesis  Summary Bone Morphogenetic Proteins (BMPs) are secreted growth factors that may play a role in craniofacial morphogenesis. Here we correlate the expression of BMP ligands (Bmp-2, Bmp-4 and Bmp-7) with the expression of B M P receptors (BmpR-IA, BmpR-IB and BmpR-IT) and noggin, a B M P antagonist, in order to predict where endogenous B M P signaling occurs in the face.  In situ hybridization studies on stage 10-32 chicken  embryos showed that Bmps are expressed in facial epithelium and mesenchyme during outgrowth and patterning of the facial prominences.  Later, Bmps are expressed  peripheral to cartilage condensations and in the perichondrium, tongue and egg tooth. BmpR-IA and BmpR-II are ubiquitously expressed up to stage 30 when they are specifically excluded from differentiating cartilage. BmpR-IB transcripts are present in early head mesenchyme, and later in prechondrogenic condensations, the perichondrium and olfactory epithelium, suggesting.a more specialized function for this receptor. Bmp-7 is also expressed in the olfactory epithelium, overlapping the expression of BmpR-IB. Noggin is expressed - where it may act to suppress BMP activity - in the early head mesenchyme, frontonasal mass epithelium and differentiating cartilage.  We conclude  that the specificity of BMP signaling is more dependent on the localized expression of ligands and antagonists than it is on localized receptor expression.  48  Introduction  Bone Morphogenetic Proteins (BMPs) are secreted  signaling molecules  that are  structurally related to the transforming growth factor beta (TGFP) super family (Wozney et al., 1988). BMPs regulate an array of diverse embryonic processes, in part through reciprocal epithelial-mesenchymal interactions (limb, Niswander and Martin,  1993;  Pizette and Niswander, 1999; teeth, Vainio et al., 1993; Jernvall et al., 1998; Wang et al., 1999; lungs, Weaver et al., 2000; face, Barlow et al., 1999; Wang et al., 1999). Epithelial-mesenchymal  interactions  are  necessary  for  outgrowth  of  the  facial  prominences (Wedden, 1987; Richman and Tickle, 1989; 1992; Hu and Helms, 1999) and for induction of bone (Tyler and Hall, 1977). Bmp-2, Bmp-4 and Bmp-7 have been localized to the epithelium and mesenchyme of the developing facial prominences (Francis-West et al., 1994; Wall and Hogan, 1995; Wang et al., 1999; Shigetani et al., 2000). Only a few of the molecules involved in BMP signaling have been mapped and the data emphasizes expression during outgrowth of the facial prominences. Identifying the regions of the face that are responsive to B M P ligands (express B M P receptors) and modulate B M P activity (express B M P antagonists) would aid in our understanding of regional endogenous BMP signaling in the head. The comparative expression data would further explain biological differences following exogenous BMP application.  There are several key stages in avian head development; 1) neural crest cell production, specification and migration, 2) formation of branchial arches and the nasal placodes, 3) formation of the maxillary and mandibular divisions of the first branchial arch, 4) enlargement of the facial prominences, 5) fusion and merging of the upper facial  49  prominences to form the primary palate and 6) differentiation of skeletal tissues. Cranial neural crest cells arise and migrate from the dorsal margin of the neural tube ventrally through the paraxial mesoderm (somitomeres) into the face between the 6 and 9 somite stages (Tosney, 1982; Noden, 1975). Bmp-4 is expressed at high levels in r3 and r5 (Watanabe and Le Douarin, 1996) and may be locally restricting production of neural crest cells (Graham et al., 1994; but see also Fairlie et al. 1999). Derivatives of cranial neural crest cells include components of the sensory nervous system (there is also a placodal contribution to the ganglia and sensory neurons), melanocytes, cartilage and bone (Le Douarin, 1982; Hall and Horstadius, 1988). The somitomeric mesoderm gives rise to bones of the chordal skull and skeletal muscles (Couly et al, 1992, 1993; Noden, 1983; Hacker and Guthrie, 1998). The mesenchyme adjacent to the developing brain is unique compared to other areas of the body, since it is both neural crest- and mesodermally-derived. Hence it would be interesting to know where other Bmps are expressed during stages when cranial neural crest cells are migrating.  By stage 14, the branchial arches and the mesenchyme ventral to the forebrain and midbrain are filled with proliferating neural crest cells, some regions of which express Bmp-7 (Wall and Hogan, 1995).  The appearance of the nasal placodes at stage 15  demarcates the position of the frontonasal mass and the lateral nasal prominences, which form medial and lateral to the nasal placodes, respectively.  Expression of Bmp-2 and  Bmp-4, but not Bmp-7, has been described in the developing nasal pit (Francis-West et al., 1994). The ectoderm of the first branchial arch expresses Bmp-4 as soon as it is formed (Shigetani et al., 2000). At stage 18, the maxillary and mandibular prominences  50  become distinct divisions of the first branchial arch (Yee and Abbot, 1978; Tamarin et al., 1984) and all the facial prominences are established.  Bmp-2 and Bmp-4 are both  expressed in the presumptive maxillary region and in the maxillary prominence (FrancisWest etal., 1994).  Between stage 20 and 28, the facial prominences enlarge and take shape through cell proliferation and directional cell expansion (McGonnell et al., 1998; Peterka and Jelinek; 1983; Minkoff and Kuntz, 1977; 1978). Overlying epithelium is necessary for proper facial outgrowth while the mesenchyme contains the information for skeletal patterning (Wedden, 1987; Richman and Tickle, 1989; Hu and Helms 1999).  Frontonasal mass  epithelium also has unique polarizing capability when transplanted under the apical ectodermal ridge of the limb bud (Helms et al., 1997).  This organizing activity is  correlated with sonic hedgehog (Shh) expression in the frontonasal mass epithelium, a molecule that is also expressed in several other regions with polarizing activity (SanzEzquerro and Tickle, 2000; Riddle et al., 1993). Bmp-2 and Bmp-4 are expressed in all the facial prominences during outgrowth (Francis-West et al., 1994), while the expression of Bmp-7 has not been examined in all the facial prominences during this entire period (Wang et al., 1999; Wall and Hogan, 1995).  The frontonasal mass, lateral nasal prominences, and maxillary prominences merge and fuse to close the primary palate and establish the upper beak. The formation of the primary palate is a critical phase in facial morphogenesis. The most important step is the formation of the mesenchymal bridge between the corners of the frontonasal mass and  51  maxillary prominences.  Fusion requires sufficient  enlargement  of the  facial  prominences, removal of the epithelial seam between the prominences and migration of the underlying mesenchyme to form the mesenchymal bridge (McGonnell et al., 1998; Wang et al., 1995). While the molecular basis for the formation of the bridge is not well understood, it is possible that Bmps are involved. In support of this hypothesis is the expression of Bmp-2 and Bmp-4 in the contact region between the frontonasal mass and maxillary prominences prior to fusion (Francis-West et al., 1994). Moreover TGF-p3, a related family member, regulates fusion of the secondary palate by similar mechanisms to those used in the primary palate (Taya et al., 1999; Kaartinen et al., 1995; Proetzel et al., 1995).  In the avian embryo, fusion and merging of facial prominences coincide with the onset of cell differentiation. The mesenchyme of the frontonasal mass gives rise to the prenasal cartilage and the premaxillary bone, while the epithelium forms the egg tooth. The lateral nasal prominences form the nasal chonchae.  Derivatives of the maxillary prominences  include the pterygoid, quadratojugal and palatine bones. The mandibular prominences form Meckel's cartilage and the surrounding membrane bones of the lower beak. BMPs stimulate the formation of prechondrogenic condensations from mesenchyme and later regulate  the  differentiation  Niswander, 2000).  of chondroprogenitors  to  chondrocytes  (Pizette and  Unlike the limb skeleton, which is mesoderm-derived, the facial  skeleton is neural crest cell-derived. It is not known whether BMPs are expressed during facial skeletogenesis in similar patterns to those observed during limb differentiation.  52  Functional evidence to support a role for BMPs in patterning the chick face has been reported (Barlow and Francis-West, 1997; Ekanayake and Hall, 1997; Barlow et al., 1999; Shigetani et al., 2000). For example, application of BMP-2 and BMP-4 soaked beads to the maxillary prominence or to the medial mandibular prominence leads to changes in gene expression and the formation of a supernumerary palatine bone or branching of Meckel's cartilage (Barlow and Francis-West, 1997). Application of BMP-7 beads to isolated mandibular mesenchyme interferes with normal elongation of the cartilage (Wang et al., 1999); therefore BMP-7 is not able to replace the facial epithelium.  At present we did not know whether facial mesenchyme expresses Bmp  receptors (BMPRs) and which receptors are mediating the aforementioned responses to exogenous BMPs.  BMPs signal via transmembrane serine/ threonine kinases cell surface receptors, which are classified due to structural and functional differences, as type I and type II (BRK-3) receptors. Two type I receptors have been identified, BMPR-IA (ALK-3, BRK-1) and BMPR-IB (ALK-6, BRK-2; for review, see Massague, 1998). The type I receptors act as signal transducers by phosphorylating cytoplasmic targets, including members of the Smad family, on serine/threonine residues (Heldin et al., 1997; Massague, 1998; Hoodless and Wrana, 1998). BMPR-II enhances ligand binding. BMPs bind weakly to each receptor type when expressed alone, but high affinity binding is observed when both receptor types are present (Koenig et al, 1994; ten Dijke et al, 1994; L i u et al 1995; Nohno et al., 1995). In the presence of BMPR-II, BMP-4 and BMP-7 bind BMPR-IB more efficiently than BMPR-IA (Nohno et al., 1995; Rosenzweig et al., 1995). In the  53  limb, BmpRIA and BmpRII are ubiquitously expressed throughout early morphogenesis, while BmpRIB is more specifically localized in prechondrogenic aggregates, immature chondrocytes and the perichondrium (Kawakami et al., 1996; Zou et al., 1997; Merino et al., 1998).  The actions of endogenous BMPs are likely controlled by naturally occurring antagonists including noggin (Capdevila and Johnson, 1998; Merino et al., 1998), chordin (Streit et al., 1998), gremlin (Merino et al., 1999; Capdevila et al., 1999), cerberus (Zhu et al., 1999) and caronte (Yokouchi et al., 1999; Esteban et al., 1999). These antagonists bind to BMPs and prevent them from interacting with their receptors, thus blocking B M P signaling (Zimmerman et al., 1996; Piccolo et al., 1996; Yokouchi et al., 1999; Esteban et al., 1999). Noggin can induce secondary axis formation in Xenopus (Smith and Harland, 1992; Smith et al., 1993); inhibit neural crest cell delamination (Kanzler et al., 2000; Sela-Donenfeld and Kalcheim, 1999), inhibit joint formation (Brunet et al., 1998), inhibit programmed cell death (Pizette and Niswander, 1999; Capdevila and Johnson, 1998; Merino et al., 1998; McMahon et al., 1998), inhibit chondrogenesis (Capdevila and Johnson, 1998) and alter patterning of the neural tube and somites (McMahon et al., 1998). Biochemical studies indicate that noggin binds BMP-2 and BMP-4 strongly, and BMP-7 with lower affinity (Zimmerman et al., 1996). The Noggin knockout mouse exhibits craniofacial abnormalities thus suggesting there is noggin expression in the normal developing head. To date, there have been no detailed studies mapping expression of noggin in mouse or chicken head development.  54  In this chapter we correlate the expression of BMPs with the expression of BMPRs and noggin in order to predict the sites where endogenous B M P signaling occurs. Our expression data shows that noggin is expressed in specific locations in the facial epithelium and early head mesenchyme. We also demonstrate that facial mesenchyme and epithelium express BMPRs, suggesting mechanisms by which cells respond to exogenous and endogenous BMP protein during craniofacial development.  55  Materials and Methods  Chicken embryos Fertilized white leghorn chicken eggs were obtained from Coastline Chicks, Abbotsford, B.C. Eggs were incubated at 37.5°C until they reached the desired stage of development. Embryos were staged according to Hamburger and Hamilton (1951).  In situ hybridization probes The chicken cDNAs were kindly provided by the following individuals: BmpR-IA, BmpR-IB (L. Niswander; Zou and Niswander, 1996; Zou et al., 1997), BmpR-II (T. Nohno; Kawakami et al, 1996), Bmp-2 and Bmp-4 (P. Brickell; Francis-West et al., 1994), Bmp-7 (B. Houston; Houston et al., 1994), noggin (R. Johnson; Capdevila and Johnson, 1998) and shh (C. Tabin, Riddle et al., 1993). Plasmids were linearized with appropriate restriction enzymes in order to transcribe sense or antisense riboprobes labelled with [ S]-UTP (Rowe 35  etal., 1991).  Autoradiographic in situs Specimens used for in situ hybridization were fixed in 4% paraformaldehyde (PFA) and processed into wax. R N A in situ hybridization on wax sections was done according to Rowe et al., (1991)^  Serial sections were divided between several slides so that  expression patterns for different genes could be compared within the same embryo. Slides were left to develop for 1 to 3 weeks.  Photographs were taken on a Zeiss  compound microscope under darkfield illumination with Kodak Gold 100 print film or with a Minolta RD-175 digital camera.  Near-adjacent sections were stained with  56  Toluidine Blue and photographed with brightfield illumination. Plates were composed using Adobe Photoshop.  57  Results  Radioactive in situ hybridization was performed on a total of 17 stage 10, 24 stage 15, 16 stage 20, 18 stage 24, 16 stage 28, 5 stage 30 and 3 stage 32 embryos over 7 experiments. Adjacent sections were hybridized to Bmp-2, Bmp-4, Bmp-7, BmpR-IB, BmpR-IA, BmpR-II, and noggin in order to compare patterns of expression within the same embryo. We refer to the cranial-caudal or superior-inferior axis as being from the top of the head to the tail, the medio-lateral axis as being from the center of the embryo towards the eyes, dorsoventral axis as being from the back to the front of the embryo and the proximo-distal axis as being from the jaw-joint region to the tip of the beak (Fig. 1.1). Expression was compared to background levels observed in non-expressing tissue on the same slide.  Expression during neural crest cell migration At stage 10, Bmp-2 and Bmp-4 transcripts are not detected in early head mesenchyme (Fig. 3.1B,C). In contrast, Bmp-7 transcripts are found in the mesenchyme rostral to the mesencephalon-rhombencephalon isthmus, with expression fading caudal to r l (Fig. 3. ID). Noggin and BmpR-IB transcripts are expressed abundantly at all axial levels of the craniofacial mesenchyme (Fig. 3.1E,F) and in the notochord (data not shown). BmpR-IA and BmpR-IIaxe ubiquitously expressed in mesenchyme and epithelium (data not shown).  Epithelial expression is detected for the three ligands. Bmp-2 and Bmp-4 transcripts are detected throughout the head epithelium from the anterior neuropore and extends well into the hindbrain (Fig. 3.1B,C).  Bmp-7 expression is not observed in the anterior  neuropore but is seen in the presumptive lens ectoderm covering the optic stalks and in  58  Figure 3.1 Expression in stage 10 and 15 chicken embryos. A - F are frontal sections through the head of the embryo. A to E are adjacent sections from the same embryo while F is from a different embryo. G - L are parasagittal sections from a single stage 15 embryo. A , G are Toluidine Blue stained sections that were not hybridized. PT-J' and L ' (insets) are sagittal sections through the buccopharyngeal membrane, positive signal is seen as white silver grains in darkfield and black silver grains in bright field. J: arrow indicates expression in the mesenchyme proximal to the eye. K: Noggin did not appear to be expressed in stage 15 embryos. L : BmpR-IB continues to be expressed in the head mesenchyme, and is also in the epithelium in the roof of the stomodeum and caudal first branchial arch.—H'-J', L ' : Bmp-2, Bmp-4, Bmp-7 and BmpR-IB are all expressed in the buccopharyngeal membrane (arrows in H', I', J', L'). K E Y : anp - anterior neuropore; bl - first branchial arch; e - eye; f - forebrain; fg foregut; m - midbrain; mes - mesenchyme; np - nasal pit; r l - rhombomere 1; t telencephalon; tg - trigeminal nerve. Scale bars for stage 10 embryos = 50 urn; stage 15 embryos =100 um.  59  the remaining ectoderm extending caudally to the hindbrain (Fig. 3.ID).  BmpR-IB  transcripts are present caudal to the optic stalks and extend past the hindbrain (Fig. 3.IF). BmpR-IB transcripts in the surface epithelium overlap the expression of all three BMPs examined. Noggin transcripts are not detected in the head epithelium (Fig. 3.IE).  Localized expression of B M P signaling molecules in the neuroepithelium is also observed at stage 10. Bmp-2 is abundantly expressed in the anterior neuropore, the optic stalks and also in the future diencephalon (Fig. 3.IB). Bmp-7 is found at low levels in the neuroepithelium of the mesencephalon and anterior rhombencephalon (Fig. 3.ID). BmpR-IB transcripts are only detected in the anterior neuropore (data not shown). Noggin message is localized to the optic stalk regions (Fig. 3.IE).  Onset of face morphogenesis At stage 15, Bmp-2 and Bmp-7 are expressed in complementary patterns in the head mesenchyme. Bmp-2 transcripts are located proximal to the first branchial arch, within the first branchial arch, and adjacent to the ectoderm of the presumptive maxillary region (Fig. 3.1H). Bmp-4 is expressed at background levels in the head mesenchyme (Fig. 3.11).  Bmp-7 transcripts are more rostral than those of Bmp-2 and are located just  proximal to the eye (Fig. 3.1 J). This is the location of the future ophthalmic branch of the trigeminal nerve. 3.IK).  Noggin is transcribed at low levels in the head mesenchyme (Fig.  BmpR-IB message is abundant in mesenchyme ventral to the brain where the  cranial base will form (Fig. 3.1L). BmpR-IB transcripts are spread over a larger area of the head and overlap with Bmp-2, Bmp-4 and Bmp-7 transcripts. Bmp-2 transcripts also  60  extend into the first branchial arch and overlap with BmpR-IB mRNA (compare Fig. 3.1H and L). BmpR-IA and BmpR-II are expressed in both epithelium and mesenchyme but are not concentrated in any particular locations (data not shown).  Bmp-4 expression is predominantly in the surface epithelium with the highest levels in the first branchial arch, presumptive maxillary region, roof of the stomodeum, and in the epithelium covering the forebrain (Fig. 3.11; note that Bmp-4 transcripts are also in the epithelium around the forebrain at stage 10). Bmp-2, Bmp-4 and Bmp-7 are expressed in the epithelium of the first branchial arch and the presumptive maxillary region (Fig. 3.1H-J). Bmp-4 is also expressed in the presumptive palatal epithelium in mouse (E9.5; LaMantia 2000), similar to the roof of the stomodeum in stage 15 chicken embryos.  BmpR-IB and Bmp transcripts overlap in several epithelial regions at stage 15. BmpR-IB is highly expressed in the caudal surface of the first branchial arch and overlaps all three Bmps examined (Fig. 3.1L).  BmpR-IB transcripts are expressed in the roof of the  stomodeum in the same location as Bmp-4 and Bmp-7 (data not shown). The nascent nasal placode expresses Bmp-2, Bmp-4 and Bmp-7, but neither BmpRIB nor noggin (Fig. 3.1H-J,L). The buccopharyngeal membrane separates the stomodeum from the foregut and is only a transitory structure.  Shortly after stage 15, the membrane breaks down,  allowing communication between the gut and the oral cavity.  A l l three Bmps are  expressed in the buccopharyngeal membrane, as is BmpR-IB (Fig. 3.1Ff-J', L').  61  Expression in stage 20 facial prominences Frontonasal mass and lateral nasal prominence expression In the stage 20 frontonasal mass, the three Bmps are localized to the mesenchyme at the lateral edges of the frontonasal mass. Bmp-4 and Bmp-7 are also expressed in the center of the frontonasal mass, while Bmp-2 is not expressed (Fig. 3.2B-D and J-L; see also stage 20, Bmp-2 and Bmp-4 expression pattern in Francis-West et al., 1994). BmpRIB, BmpR-IA, and BmpR-II are all expressed in the lateral frontonasal mass mesenchyme and overlap the expression of the three ligands (Fig. 3.2E,F and data not shown). Noggin is not present in the frontonasal mass mesenchyme (Fig. 3.2G,H). The lateral nasal prominence mesenchyme does not express Bmp-2 or Bmp-4 (Fig. 3.2B,C and data not shown) but does express Bmp-7, BmpR-IB and BmpR-II (Fig. 3.2D-F). The frontonasal mass and lateral nasal prominences express all three Bmps, BmpR-II and BmpR-IA throughout the epithelium (Fig. 3.2B-D, F, J-K, N and data not shown). In contrast, transcripts for BmpR-IB are most abundant in the medial frontonasal mass epithelium and relatively lower in the lateral nasal prominence epithelium (Fig. 3.2E). Noggin is expressed in the epithelium covering the caudal edge of the frontonasal and extends into the nasal slits (Fig. 3.2G,H). The invaginating nasal pit epithelium does not express Bmp-2, Bmp-4 or noggin transcripts except at the extreme distal edges (Fig. 3.2G and data not shown).  Bmp-7 and BmpRII are expressed throughout the nasal pit  epithelium (Fig.3. 2D and F).  62  Figure 3.2 Expression in stage 20 embryos.  A and I are Toluidine Blue stained  sections that were not hybridized. A-F are frontal sections of a single embryo while G is from a different embryo.  Arrows in B and G highlight expression in the lateral  frontonasal mass regions. Arrow in D shows expression in the mandibular branch of the trigeminal nerve.  The angle at which embryos were cut in the frontal plane does not  include the maxillary prominences. Expression patterns in the maxillary prominence are displayed in parasagittal sections (H-N).  Sections I-N are from the same embryo. H :  Note restricted noggin expression to the narrow band in the frontonasal epithelium (arrow). Insets A ' , B', D ' , E ' and F' are frontal sections of the otic vesicle within the same embryo, photographed in dark field (except D which is in bright field). Planes of section are described by the lines in A and H. K E Y : fnm - frontonasal mass; lnp - lateral nasal prominence; m - midbrain; md mandibular prominences; mn - mandibular branch of trigeminal nerve; mx - maxillary prominence; p - posterior crista ampullaris; s - macula sacculi; sc - superior crista ampullaris; t - telencephalon; tg - trigeminal nerve. Scale bars for frontal and parasagittal sections =100 um; insets = 50 um.  63  Maxillary prominence expression The only B M P detected in the maxillary mesenchyme at stage 20 is Bmp-2. Bmp2 is also expressed in the distal cranial mesenchyme (Fig. 3.2J and Francis-West et al., 1994). BmpR-IB is localized to the caudal edge of the proximal maxillary prominence and thus is complementary to Bmp-2 (Fig. 3.2M). BmpR-IA and BmpR-II demonstrate local concentrations of intense signal in the maxillary mesenchyme (Fig. 3.2N; BmpR-IA not shown). In the epithelium, BmpR-IB transcripts are detected in the caudal edge of the maxilla, and thus are complementary to Bmp-2 and Bmp-4 (compare Fig. 3.2M with 3.2J,K). Noggin is not detected in the maxillary epithelium (Fig. 3.2H). BmpR-IA and BmpR-II are expressed throughout the maxillary epithelium (Fig. 3.2N and data not shown). Mandibular prominence expression At stage 20, Bmp-2 and Bmp-4 are expressed at low levels throughout the mandibular mesenchyme except adjacent to the midline, where Bmp-2 transcripts are much more abundant than in lateral regions (Fig. 3.2B,C). The mandibular branch of the trigeminal nerve expresses Bmp-7 transcripts (Fig. 3.2D). BmpR-IB is most abundant in the lateral mesenchyme of the mandibular prominences, complementary to Bmp-2 transcripts (compare Fig. 3.2B with E). BmpR-IA (not shown) and BmpR-II are both very abundant in the mandible (and maxilla) compared to the frontonasal mass at stage 20 (Fig. 3.2F,N).  Noggin message is not detected above background levels in the  mandibular prominence (Fig. 3.2G).  64  All three Bmps are also localized to the mandibular epithelium at stage 20. Bmp-2 and Bmp-4 (Fig. 3.2B,C) transcripts are more medial, while Bmp-7, BmpR-IA, BmpR-IB and BmpR-II are very abundant throughout the mandibular epithelium (Fig. 3.2D,F and data not shown). Noggin transcripts are not detected in the mandibular epithelium (Fig. 3.2G).  BMP expression during morphogenesis of the otic vesicle A description of where Bmp-4, Bmp-5, and Bmp-7 are expressed during sensory organ differentiation in the chicken inner ear can be found in other published work (Wu and Oh, 1996; Oh et al., 1996; Gerlach et al., 2000).  Here, we compare Bmp-4 and Bmp-7  localization to Bmp-2 and the B M P receptors at stage 20, when three foci of Bmp expression are observed.  The anterior, posterior and medial part of the otocyst  corresponded to regions that give rise to the superior and posterior, crista ampullaris, and macula sacculi, respectively. Bmp-2 is also transcribed in all three regions (Fig. 3.2B'), similar to other BMPs examined. We confirm that Bmp-7 expression is less restricted than the other Bmps (Fig. 3.2D'). BmpR-IB transcripts are localized to the medial otocyst region; expression begins just inferior to the anterior focus and extends to the posterior focus (Fig. 3.2E'). As in other tissues, BmpR-IA and BmpR-II are ubiquitously expressed (Fig. 3.2F' and data not shown).  Noggin transcripts are not detected in the stage 20  otocyst (also reported by Chang et al., 1999).  65  Expression in the stage 24 face The expression patterns of the examined genes at stage 24 are for the most part similar to those observed at stage 20. Here, we will highlight the differences at stage 24. In the frontonasal mass, Bmp-7 expression is in the mesenchyme along the inferior edge of the frontonasal mass and becomes more localized, rather than being spread all across as it was at stage 20 (Fig. 3.3D).  Bmp-2 is also more localized in the corners of the  frontonasal mass (Fig. 3.3B). BmpR-IB expression is more prominent in the mesenchyme of the frontonasal mass and in the region surrounding the cranial end of the nasal slit (Fig. 3.3F). BmpR-IB is no longer expressed in the medial frontonasal mass epithelium. Noggin is expressed in the caudal epithelium of the frontonasal mass just as at stage 20, and overlaps the most distal extent of shh expression (compare Fig. 3.3L and M).  In the lateral nasal prominences, Bmp-4 transcripts are mostly localized to the mesenchyme in the furrow between the lateral nasal and maxillary prominences (Fig. 3.3C).  Bmp-4 is also in the lateral epithelium but at lower levels than in the medial  epithelium (Fig. 3.3C). The expression patterns of Bmp-2, Bmp-7, BmpR-IB, and Noggin in the lateral nasal epithelium are unchanged from stage 20. Bmp-7 is expressed in the lateral regions of the lateral nasal prominences and overlaps BmpR-IB (compare Fig. 3.3D with F).  Bmp-4 and Bmp-7 transcripts are first detected in the maxillary mesenchyme at stage 24. Bmp-2 is expressed in the medial and rostral maxillary mesenchyme (Fig. 3.3B), while  66  Figure 3.3 Expression in stage 24 embryos.  Sections in frontal (A to K) and  parasagittal (L to S) planes. A , G , N - Toluidine Blue stained, non-hybridized sections. Adjacent sections within single embryos are as follows (A, C, D, F = embryol), (B,E = embryo 2), (G - J = embryo 3), (L-R = embryo 4) and (S = embryo 5). E : Note noggin expression in the epithelium around the lateral corners of the frontonsal mass (arrows). G - K are sections through the mandibular prominence.  H , K: Overlapping Bmp-2 and  noggin transcripts in the mandibular mesenchyme (arrows). J: Bmp-7 expression in the mandibular branch of the trigeminal nerve (arrow).  L , M : Noggin transcripts in the  frontonasal mass overlap and extend slightly distal (arrows in L) to the start of shh expression (arrowhead in L). P: Bmp-4 is highly expressed in the frontonasal mass epithelium (arrow) and overlaps with noggin transcripts (arrow in R). Planes of section are described by the lines in A and N . Sagittal sections in L , M were cut along line 1 in A. Sagittal sections in N,0,P,Q,R were cut along line 2 in A . Sagittal sections in S were cut along line 3 in A . K E Y : e, eye; fnm, frontonasal mass; lnp, lateral nasal prominence; m, midbrain; md, mandibular prominences; mx, maxillary prominence; t, telencephalon; tg, trigeminal nerve. Scale bars for frontal sections (A-F) = 200 urn; frontal sections of the mandible (G-K) = 100 um; sagittal sections (N-S) = 200 um and (L, M) = 200 urn.  67  Bmp-4 and Bmp-7 transcripts are present in the lateral and rostral maxillary mesenchyme (Fig. 3.3C,D). BmpR-IB message is expressed in the mesenchyme of the caudal half of the maxilla (Fig. 3.3F,S). Bmp-2 and Bmp-4 continue to be expressed in the maxillary epithelium, but transcripts are more prominent in the medial surface compared to the lateral sides (Fig. 3.3B,C Francis-West et al., 1994). Noggin message is not present in maxillary epithelium or mesenchyme (Fig. 3.3E).  In the mandibular prominences at stage 24, the expression of Bmp-2, Bmp-7, and noggin is restricted to the medial mandibular mesenchyme while BmpR-IB becomes restricted to the superior-lateral mesenchyme (Fig. 3.3F, H-K). Bmp-7 is expressed in the maxillary and mandibular branches of the trigeminal nerve (Fig. 3.3J, Q). Noggin is also expressed at low levels in the trigeminal nerve and in the otic vesicle (Fig. 3.3R).  Expression in stage 28 embryos At stage 28, Bmp-2, Bmp-4, and Bmp-7 are expressed in the epithelium and mesenchyme of the extreme corners of the frontonasal mass (now called the globular processes; Romanoff, 1960), and at the medial edge of the maxillary prominence where fusion takes place. The mesenchymal expression of Bmp-2 in the corners of the frontonasal mass is much higher than that of Bmp-4 (Fig. 3.4A,B). Bmp-4 expression is primarily epithelial in this zone of contact (Fig. 3.4B).  Bmp-7 is also expressed at high levels in the  frontonasal mass epithelium and subjacent mesenchyme (Fig. 3.4D). A l l three Bmps are also transcribed in the deep furrow between the merging lateral nasal and maxillary mesenchyme.  68  Figure 3.4 Expressions in stage 28 embryos.  A , B, D and E are adjacent frontal  sections within the same embryo while H is from a different embryo.  Arrows in E  highlight the absence of noggin transcripts in the corner epithelium of the frontonasal mass. C, G and I are parasagittal sections within the same embryo. F is a mid-sagittal section from a different embryo. F: Shows localized noggin expression in a narrow band of frontonasal mass epithelium (arrow). K E Y : e - eye; fnm - frontonasal mass; lnp - lateral nasal prominence; mc - Meckel's cartilage; md - mandibular prominences; mx - maxillary prominence.  Scale bar = 500  pm.  69  69a  Bmp-2, Bmp-4, and Bmp-1 are expressed in overlapping domains in the medial mesenchyme and epithelium of the maxillary process (Fig. 3.4A,B, D). Bmp-4 is located in a caudal domain within the maxillary prominence (Fig. 3.4C and Francis-West et al., 1994).  Noggin transcripts are localized to the caudal frontonasal mass epithelium but no longer extend around the globular processes (N=3/3; compare Fig. 3.3E with Fig. 3.4E). Midsagittal sections show that woggw-expressing epithelium separates the caudal edge of the frontonasal mass from the roof of the stomodeum (Fig. 3.4F). BmpR-IB is also present in the lateral mesenchyme of the globular processes (Fig. 3.4H). BmpR-IA and BmpR-II are ubiquitously expressed at stage 28 in the epithelium and mesenchyme (data not shown). Cartilagenous condensations Prechondrogenic condensations  are beginning to form in the center of the  frontonasal mass, the location of the future prenasal cartilage and nasal septum. BmpRIB transcripts are very abundant in the center of the frontonasal mass and define the chondrogenic region (Fig. 3.4H). Bmp-2 is expressed in mesenchymal cells lateral to the condensation (Fig. 3.4A and data not shown). Bmp-4 expression is found caudal to the condensation (Fig. 3.4B).  Bmp-7 is expressed in the mesenchyme adjacent to the  condensation on the distal, caudal and lateral sides (Fig. 3.4D and data not shown). Noggin transcripts are not present in the center of the frontonasal mesenchyme at stage 28 (data not shown). Condensing mesenchymal  cells in the medial edge of the lateral nasal  prominences and surrounding the cranial nasal pit form the nasal conchae.  The most  70  abundant transcripts found in the lateral nasal prominences are Bmp-4, Bmp-7, and BmpR-IB. Bmp-4 and Bmp-7 transcripts are localized to the lateral and caudal regions of the lateral nasal prominences and overlap BmpR-IB transcripts (Fig. 3.4B-D and G-I). In shallow frontal sections, BmpR-IB message is found in most of the lateral nasal mesenchyme corresponding to the superior choncha (data not shown). In deeper sections, the region of expression becomes restricted to the mesenchyme next to the nasal slits (Fig. 3.4H,I). Bmp-7 transcripts overlap with BmpR-IB (compare Fig. 3.4D,G with H , I). Noggin transcripts are localized to the mesenchyme in the lateral nasal- maxillary furrow (Fig. 3.4E). In the mandible, Meckel's cartilage condensation is further advanced than the frontonasal mass cartilages (Matovinovic and Richman, 1997).  Meckel's cartilage  expresses high levels of BmpR-IB (Fig. 3.4H,I) and noggin (Fig. 3.4E,F) transcripts. In contrast, the three BMPs are expressed in complementary patterns to that of noggin. Bmp-4 is expressed in a narrow region at the superior edge of the mandibular prominence (Fig. 3.4B,C and Francis-West et al., 1994).  Bmp-2 is expressed in the mesenchyme  superior and inferior to Meckel's cartilage.  Bmp-7 is expressed adjacent to the  condensations of Meckel's cartilage (Fig. 3.4G), in the presumptive perichondrium.  Expression in stage 30 embryos Expression in maxillary mesenchyme, sites of intramembranous ossification The maxillary prominences make membranous bone that condenses directly from mesenchyme, starting at stage 31 (day 7; Murray, 1963; Romanoff, 1960) without a cartilagenous template. The fate of discrete regions of the maxillary prominences is not  71  known, but the interesting patterns of gene expression suggest that some degree of molecular prepatterning exist as early as stage 28. These regional concentrations of gene expression are more noticeable at stage 30. Areas of transcript overlap in the maxillary prominence include lateral mesenchyme (Bmp-2, -4, -7, and noggin; Fig. 3.5E,G), medial and cranial mesenchyme (Bmp-2, mesenchyme (Bmp-2,  -4, -7, and BmpR-IB;  -4, -7 and BmpR-IA;  Fig. 3.5A,C, E) and caudal  Fig. 3.5A,C, K). The central mesenchyme  expresses Bmp-2, Bmp-4, and Bmp-7. Noggin transcripts were also detected in the caudal mesenchyme in deeper frontal sections (data not shown). Expression in cartilage and perichondrium By stage 30, cartilage differentiation is underway in the chondrocranium.  BmpR-  IB is expressed within cartilage elements and the developing perichondrium of the nasal septum, nasal conchae, and Meckel's cartilage (Fig. 3.5I,J).  Noggin message is very  abundant within the nasal septum, nasal conchae, and Meckel's cartilage, but nested within the larger BmpR-IB  expressing region (Fig. 3.5G). In the nasal conchae, Noggin  signal is prevalent in the more distal frontal sections and extends cranially and merges with noggin expression in the nasal septum (data not shown). Bmp-2, Bmp-4, and Bmp-7 are expressed in the perichondrium, bordering Noggin and in all the facial cartilages but overlapping the outer edges of the BmpR-IB  expression domain (compare Fig. 3.5G-J  with A-F). Bmp-2 but not Bmp-4 or Bmp-7, is present above background levels within the cartilage (Fig. 3.5A-F).  BmpR-IA  and BmpR-II  are expressed at low levels in the  cartilage compared to the surrounding mesenchyme (Fig. 3.5K-N; with the exception of BmpR-II  in Meckel's cartilage; Fig. 3.5M,N).  72  Figure_3.5 Expressions in stage 30 embryos.  Serial sections within embryo 1 were  made in the frontal (A,C,E,G,I,K,M) and embryo 2 in the mid-sagittal  plane  (B,D,F,H,J,L,N). The medial edge epithelium of the palate (white arrows in A) expresses Bmp2, Bmp-7 and BmpR-IB. Insets (B',D',F',H',J',L\N') are frontal sections of the tongue. O-T are brightfield, higher power views of darkfield frontal sections showing expression in the left nasal epithelium and nasal chonchae. Note the very abundant and specific expression of Bmp-7 (Q) and BmpR-IB (R) in the olfactory epithelium. Planes of section are described by the lines in A and D. Frontal sections in A , C , E , G , I , K , M were cut along line 2 in D. Frontal sections in A ' , C ' , E ' , G ' , I ' , K ' , M ' were cut along line 1 in D. K E Y : e - eye; eg - entoglossum; mid c - middle concha; ns - nasal septum; oe - olfactory epithelium; mc - Meckel's cartilage; mx - maxilla; ns - nasal septum; ps - palatal shelf; re - respiratory epithelium; sc - superior concha; t - tongue.  Scale bars for frontal and  sagittal sections = 500 um; nasal conchae sections = 250 pm; tongue insets =100 um.  73  73a  Expression in nasal epithelium The nasal epithelium gives rise to two distinct, sensory and non-sensory, epithelial domains.  The sensory olfactory epithelium arises in the cranial-proximal  regions of the nasal passage, while the respiratory epithelium forms more caudal-distally. BmpR-IA and BmpR-II are ubiquitously expressed throughout the nasal epithelium (Fig. 3.5K',M'). The olfactory epithelium expresses high levels of Bmp-7 and BmpR-IB (Fig. 3.5E',F).  Bmp-7 transcripts precisely overlap those of BmpR-IB in the superior and  middle nasal chonchae epithelium (compare Fig. 3.5E with I).  Bmp-2, Bmp-4, and  Noggin are not expressed in the nasal epithelium (Fig. 3.5A,C, G).  Maxillary and mandibular epithelium The medial edge of the maxillary prominences does not fuse in birds (unlike in mammals), hence the epithelium persists throughout life. Molecules that are restricted to the medial edge epithelium along the palatal shelves at stage 30 include Bmp-2 and BmpR-IB (Fig. 3.5A,I).  A l l other cDNAs used in the present study were expressed  throughout the maxillary epithelium, with the exception of noggin. Noggin message is not detected in the maxillary epithelium (Fig. 3.5G). . The mandibular ectoderm expresses BMPs, their receptors, and noggin. The cranial or oral side of the mandibular ectoderm expresses Bmp-2, Bmp-4, Bmp-7, noggin and BmpR-IB (Fig. 3.5A-I). Within the oral ectoderm there is slightly higher expression of Bmp-2, noggin and BmpR-IB in the medial compared to the lateral epithelium. Bmp-7, BmpR-IA and BmpR-II are expressed throughout the intraoral and extraoral mandibular epithelium (Fig. 3.5E.F, K-N).  74  Expression in the tongue Our in situ hybridization studies suggest that BMPs may also be involved in formation of the specialized tongue epithelium.  At stage 30, the BMP ligands display distinct  expression patterns within the tongue.  Bmp-2 transcripts are found in the epithelium,  predominantly in the medial regions, while Bmp-4 is in the mesenchyme underlying the epithelium (Fig. 3.5B,D and 3.5B',D'). Bmp-7 is expressed throughout the epithelium and underlying mesenchyme Fig. 3.5F and F'). BmpR-IA and BmpR-II are expressed in the entire developing tongue mesenchyme and epithelium (Fig. 3.5L,N and 3.5L',N'). Noggin message is not detected in the tongue epithelium but is expressed at low levels in the underlying mesenchyme (Fig. 3.5H and FT). BmpR-IB is only expressed in the mesenchyme and overlaps entoglossum condensations (data not shown).  Egg tooth formation at stage 32 The egg tooth is an epithelial specialization that forms at the tip of the upper beak at stage 30.  The egg tooth is keratinized at stage 32.  The chick uses the egg tooth during  hatching to break through the shell. The outer cornified layer of the egg tooth expresses Bmp-7, BmpRIA, and BmpR-II (¥ig. 3.6D,G, H). The intermediate layer expresses Bmp2, BmpR-IA, and BmpR-II (Fig. 3.6B,G, H). The basal cell layer expresses Bmp-2, Bmp4, Bmp-7, Noggin, BmpRIA, and BmpR-II (Fig. 3.6B-E, G, H).  75  Figure 3.6 Expression in the egg tooth of stage 32 embryos. Serial sections are in the coronal plane.  A is a non-hybridized, Toluidine Blue stained section photographed in  brightfield and B - H are darkfield photos. K E Y : b - basal cell layer; c - cornified layer; et - egg tooth; i - intermediate layer; pnc prenasal cartilage. Scale bar = 500 um.  76  ra  Discussion  Our data indicate that BmpR-II and BmpR-IA are ubiquitously expressed during craniofacial development. In contrast, BmpR-IB expression is much more specific and therefore suggests more specific functions. We identify novel regions where noggin is expressed in the early head mesenchyme and caudal frontonasal mass epithelium, not previously reported in chicken or mouse embryos.  The expression patterns suggest  various locations where interactions between ligands and receptors and between ligands and noggin are taking place in the developing head.  Noggin is expressed in head mesenchyme and possibly migrating neural crest cells We observe abundant noggin transcripts in stage 10 head mesenchyme, a finding not reported previously (Capdevila and Johnson, 1998; Tonegawa and Takahashi, 1998). At earlier stages of head development noggin is only expressed in the notochord (6 somites; Tonegawa and Takahishi, 1998). We observe a striking down-regulation of noggin in the mesenchyme between stage 10 and 15. Thus, there is a critical phase during neural crest production where high levels of noggin may be required.  Several investigators have  described noggin's role in the production of neural crest cells (Kanzler et al., 2000; Selleck et al., 1998). A n increase in noggin expression in neural crest cells destined for the second branchial arch leads to hypomorphic skeletal elements, due to inhibition of neural crest formation and/or migration (Kanzler et al., 2000). Overexpression of noggin next to the dorsal neural tube inhibits delamination of trunk neural crest cells (Kanzler et al., 2000; Sela-Donenfeld and Kalcheim, 1999). Interestingly, mice lacking noggin have increased deposition of cartilage and bone in the jaws, but show no obvious changes in  77  cranial neural crest cell migration or production (R. Harland, personal communication). In addition, Bmp2 knockout mouse embryos lack detectable migratory cranial neural crest cells and branchial arches (Kanzler et al., 2000; Zhang and Bradley, 1996), reinforcing the necessity for B M P signaling in neural crest cell production.  Noggin and Bmp-4 transcripts are concentrated in the dorsal neural tube (Liem et al., 1995, 1997; Sela-Donenfeld and Kalcheim, 1999; Sela-Donenfeld and Kalcheim, 2000). The B M P most likely to interact with Noggin in the head mesenchyme is BMP-7. Neither Bmp-2 nor Bmp-4 transcripts are detected in the paraxial mesoderm at stage 10, concurring with data from other groups (Farlie et al., 1999; Watanabe and Le Douarin, 1996; Liem et al., 1995).  Bmp-7 is however, expressed throughout the mesenchyme  cranial to the mid-hindbrain and might also be expressed in the migrating neural crest cells.  The effects of BMP-7 on migrating N C C may be attenuated by the presence of  Noggin.  BMP-7 can stimulate delamination of trunk neural crest but at higher  concentrations than BMP-4 (Sela-Donenfeld and Kalcheim, 1999). The effects of BMP 7 on cranial neural crest cell production have not been examined.  A role for BMPs in nasal pit invagination and differentiation of olfactory epithelium Bmp-2, Bmp-4, and Bmp-7 are all expressed in the nasal placode at stage 15. When the presumptive placodal epithelium from mice is explanted and grown in cell culture it remains as a single cell layer and fails to thicken (LaMantia et al., 2000). Formation of the nasal placodes requires signaling from the mesenchyme (LaMantia et al., 2000). The pattern of Bmp expression suggests a role in the formation or maintenance of the nasal  78  placode. At stage 16 when invagination begins, Bmp-2 and Bmp-4 are expressed at the edges of the nasal pit epithelium, but not in the center (Francis-West et al., 1994; Wu and Oh, 1996).  At later stages, the same pattern continues with Bmp-2 and Bmp-4 being  expressed at the outermost edges of the nasal pit.  It is possible that reduced B M P  expression in the depths of the nasal pit is associated with invagination. In contrast to Bmp-2 and Bmp-4, Bmp-7 is expressed throughout the nasal pit epithelium at stage 20 and may be involved in the determination of the undifferentiated (proximal) nasal epithelium to become olfactory epithelium.  At stage 30, Bmp-7 transcripts (not Bmp-2 and Bmp-4) are exclusively found in the olfactory epithelium and not in the respiratory epithelium. Other markers for olfactory epithelium such as N C A M , keratin sulfate and certain lectins are initially expressed throughout the placode and then become restricted to olfactory epithelium (Croucher and Tickle, 1989).  BmpR-IB is also expressed specifically in the olfactory epithelium,  however, unlike Bmp-7, BmpR-IB is not expressed in the placode. BmpR-IB is similarly expressed in the mouse olfactory epithelium (Zhang et al., 1998; Dewulf et al., 1995). In the chicken embryo, it is possible that BMP-7 signals are transduced via BMPR-IB to differentiate the olfactory epithelium from the respiratory epithelium. There are species differences  however, and in the mouse embryo Bmp-4 is expressed in olfactory  epithelium (Shou et al., 2000). B M P activity is also required for neurogenesis from the olfactory epithelium, since exogenous Noggin protein completely inhibits the formation of neural colonies in olfactory epithelium-stromal cell co-cultures (Shou et al., 2000).  79  In contrast, the respiratory epithelium does not express any of the BMPs examined here, but the sub adjacent mesenchyme expresses Bmp-2 and Bmp-4 at high levels.  The  absence of BMPs in the respiratory epithelium may specify or maintain the respiratory epithelium. Further support for this idea comes from our observation that chordin is exclusively expressed in respiratory epithelium at stage 30 (data not shown). It seems likely that complete absence of B M P activity is a necessary prerequisite for respiratory epithelium formation.  Potentialfunctions of BMPs in the facial prominences Depending on their concentration and time in development, exogenous BMPs can induce growth (Barlow and Francis-West, 1997), apoptosis (Zhao and Rivkees, 2000; Barlow and Francis-West, 1997; Ekanayake and Hall, 1997; Ganan et al., 1996; Graham et al., 1994), differentiation (Wu et al., 1997; Denker et al., 1999; Buckland 1998; Merino et al., 1998; Macias et al., 1997; Zou et al., 1997; Duprez et al., 1996a; Roark and Greer, 1994), alter patterning (Dahn and Fallon, 2000; Duprez et al., 1996b) or even suppress growth (Shigetani et al., 2000; Barlow and Francis-West, 1997; Niswander and Martin, 1993). It is likely that endogenous BMPs can also have a variety of effects during morphogenesis of the facial prominences.  An example of data that supports a patterning role for BMPs in the face is the unique expression of Bmp-2 in the mesenchyme of the presumptive maxillary region (our data and Francis West et al., 1994).  While there is no fate map of the stage 15 face, we  speculate that BMP-2 is involved in specifying the maxillary division of the first  80  branchial arch.  At stage 20, Bmp-2 expression persists in the maxillary mesenchyme  while all three Bmps are expressed in the maxillary epithelium. The specific expression patterns of the BMPs in the maxillary prominence at stage 24, 28, and 30 suggest that bone patterning may depend on BMPs. The first and second branchial arches fail to form in Bmp2-I- knockout mouse (Kanzler et al., 2000) therefore we are unable to draw conclusions about the role of BMP-2 in later maxillary prominence patterning. However, excess BMP applied to the maxillary prominence can cause an increase in proliferation and a duplication of the palatine bone (Barlow and Francis-West, 1997). It remains to be seen whether bones can also be respecified through localized inhibition of BMPs.  Bmp-2, Bmp-4, Bmp-7, and BmpR-IB are all expressed in the stomatodeal plate just prior to its breakdown. The expression of BMPs, known regulators of apoptosis (Pizette and Niswander, 1999; Macias et al., 1997; Zou and Niswander, 1994; Ganan et al., 1996; Graham et al., 1994), suggests that apoptosis may be the mechanism for removal of the buccopharyngeal membrane. Bmp expression at stage 24 and 28 also correlates with several areas of apoptosis in the face. Higher levels of programmed cell death are present at the corners of the frontonasal mass in both the epithelium and mesenchyme (McGonnell et al., 1998; Shen et al., 1997; Wedden, 1991). There are high levels of Bmp-2 transcripts in this region in both tissue layers.  There is also programmed cell  death in the midline mandibular epithelium (Shen et al., 1997) overlapping Bmp-4 expression.  The groove separating the lateral nasal prominence and maxillary  prominence has increased apoptosis (McGonnell et al., 1998; Shen et al., 1997) as well as Bmp-4 and Bmp-7 expression. Furthermore, the caudal edge of the mandible expresses  81  Bmp-2 and has increased apoptosis (Shen et al., 1997; Mina et al., 1995).  Apoptosis  appears to be regulated by different members of the BMP family in different regions of the face.  It has been suggested that BMPs are able to stimulate cell proliferation in the face (Barlow and Francis-West, 1997). Regions with high proliferation include the medial maxilla (McGonnell et al., 1998), lateral edges of the frontonasal mass (Peterka and Jelinek, 1983) and the medial-caudal edge of the mandible (McGonnell et al., 1998) - all of which specifically overlap regions of high Bmp-2 expression.  The expression data  argues in favour of a role for BMP-2 in stimulating regional growth in the developing face.  Noggin expression in caudal frontonasal mass epithelium is correlated with epithelial survival Noggin may be involved in survival of the frontonasal mass epithelium. In the limb, BMPs induce apical ectodermal ridge (AER) breakdown/regression, while Noggin promotes A E R survival/expansion (Pizette and Niswander, 1999). The frontonasal mass epithelium may be similar to the limb A E R .  Tissue recombination studies have  previously shown that the limb and facial epithelium can substitute for one another in epithelial-mesenchymal recombinations (Richman and Tickle, 1992). Noggin expression is specifically lost in areas where the epithelium does not survive (corners of the frontonasal mass) and where there is increased programmed cell death (Shen et al., 1997).  82  Presumably, the absence of Noggin allows for increased activity of BMP-2, BMP-4, and BMP-7, facilitates the breakdown of epithelium in the mesenchymal bridge region.  Shh is expressed in the frontonasal mass ectoderm and extends proximally into the intraoral ectoderm. Noggin transcripts in the frontonasal mass ectoderm overlap with Shh and extend just slightly distally. All three BMPs overlap the proximal ectoderm region of Shh expression. Noggin expression, and the potential suppression of BMP activity, may determine how far distally the expression of Shh extends. Exogenous Noggin can alter the response of neural tube cells to Shh (Liem et al., 2000). Therefore it is possible that noggin  is defining a unique region of frontonasal mass extraoral ectoderm and  distinguishing it from intraoral ectoderm.  Mesenchymal bridge formation in the primary palate Just prior to fusion of the frontonasal mass and maxillary prominences we detect an increase in Bmp-2, Bmp-4, and Bmp-7 expression in the area of contact.  Noggin  expression is present in the epithelium prior to fusion but shortly thereafter, is locally down regulated in the lateral corners of the frontonasal mass. Removal of the epithelial seam between the prominences and sufficient merging of the underlying mesenchyme are both necessary to establish the mesenchymal bridge.  Mesenchymal cells from the  frontonasal mass have been shown to migrate to the maxillary prominence (McGonnell et al., 1998), although the precise time at which this exchange of cells begins has not been determined. Our data suggest that the BMPs are especially active at stage 28, when mesenchymal bridge formation is occurring.  83  Noggin-BMP interactions in chondrogenesis Mesenchymal condensation is the earliest sign of the initiation of chondrogenesis. size of the condensation is pivotal to the size of the resulting skeletal element.  The If  condensations are smaller than a critical threshold, chondrogenesis may not begin. Similarly, an abnormally large condensation is associated with an overly large skeletal element (Hall and Miyake, 1992; Gruneberg, 1963). Condensation formation requires B M P signaling.  Overexpression of noggin reduces cartilage size by suppressing  condensation formation (Capdevila and Johnson, 1998; Pizette and Niswander, 2000).  During cartilage maturation we found high expression of noggin in the stage 30 and 32 facial cartilages, coinciding with the establishment of the final size. A similar correlation with high noggin expression and differentiating cartilage was observed in the chicken limb (Pathi et al., 1999). In homozygous -/- noggin mice, the cartilage elements and bones are enlarged. In the head, the mandible and cranial base also appear to be enlarged in the noggin nullizygous embryos (Brunet et al., 1998). Since elevated B M P activity suppresses later stages of cartilage differentiation (Zou et al., 1997; Duprez et al., 1996a), expression of noggin in facial cartilage may allow differentiation to proceed.  Receptor-ligand interactions in facial development As BMPR-II cooperates with type I receptors in order to bind BMPs it is therefore not surprising to find ubiquitous expression of BmpR-II.  The ubiquitous expression of  BmpR-IA, also reported in mice (Dewulf et al., 1995), indicates that all tissues are  84  potentially responsive to BMPs and that regions of endogenous B M P activity are not governed by BmpR-IA localization. Regions expressing all three receptors may exhibit cooperative signaling from both type-I receptors. Dual signaling from both BMPR-IA and BMPR-IB would likely have unique cellular responses compared to signaling with only BMPR-IA.  At present it is not known whether specific combinations of receptor  complexes can bind to distinct Smads in vivo (Nakayama et al., 2000).  There may be distinct roles for BMPR-IA and for BMPR-IB in embryo development. For example, decreased BMPR-IA activity inhibits programmed cell death (Yokouchi et al., 1996) while increased BMPR-IA signalling inhibits cartilage maturation (Zou et al., 1997) . The clearest role for BMPR-IB is in mediating the progression from cartilage condensation to differentiating chondroblasts (Baur et al., 2000; Y i et al., 2000). There is quite strong evidence that BMPR-IB is not required for programmed cell death. Evidence from the BMPR-IB knockout mouse shows that programmed cell death can occur in the absence of this receptor (Baur et al., 2000). Moreover, BMPR-IB is not expressed in the interdigital region at the time when programmed cell death is abundant (Baur et al., 2000; Zou etal., 1997).  There are numerous specific regions that express BmpR-IB, including the paraxial mesenchyme and early branchial arch mesenchyme. These data suggest that this receptor has other functions in addition to its role in chondrogenesis. At stage 10, Bmp-7 is expressed at higher levels than Bmp-2 and Bmp-4, and is the most likely ligand for BMPR-IB.  In stage 15 embryos, BmpR-IB transcripts overlap Bmp-2 in the maxillary  85  region and Bmp-7 in mesenchyme proximal to the eye. At stage 20, BmpR-IB is expressed in the medial otocyst epithelium of the developing ear. Bmp-2, Bmp-4, and Bmp-7 transcripts overlap with BmpR-IB in this region. The lateral otocyst also expresses all three Bmps overlapping BmpR-IA transcripts. The localization of the B M P ligands (Wu and Oh, 1996; Oh et al., 1996; Gerlach et al., 2000) and receptors may determine the fate of the individual regions of the otocyst. During primary palate formation, BmpR-IB is not expressed in the contact region between the fusing prominences and does not appear to be involved in fusion and merging of the upper beak facial prominences. Further studies interfering with individual receptor function would clarify the role of each of the type I receptors during development.  The individual role of the type I B M P receptors in regulating cell growth is not well understood. Outgrowth of the mandible occurs from proliferation in the medial and distal tips, outgrowth of the maxillary prominences is primarily medial and cranial, whereas expansion of the frontonasal mass occurs primarily at the lateral edges. A l l three of these regions express low levels of BmpR-IB, suggesting that BMPR-IA and BMPR-II complexes are mediating the involvement of BMPs in outgrowth.  Later, during cartilage differentiation, the expression patterns of the three Bmps and the expression patterns of BmpR-IB are largely mutually exclusive. The one exception is that the three BMPs and BMPRIB overlap in the perichondrium (also seen in the limb, Zou et al., 1997). Retrovirus-mediated expression of a constitutively active BMPR-IB construct leads to increased size of cartilage elements (Zou et al., 1997), consistent with a change  86  in the function of the perichondrium. Noggin is not expressed in the perichondrium and therefore interactions between BMPs and BMPR1B are permitted in this region.  Our expression data suggest new functions for the B M P receptors at early stages of development prior to cell differentiation, an intriguing role for Noggin in the formation of early head mesenchyme and during the fusion of the primary palate, and furthermore a role for BMP-7 and BMPR-IB in olfactory epithelial differentiation. Future experiments will address these interesting findings.  C o n t r i b u t i o n s to this chapter  The preceding chapter represents a manuscript that has been submitted to the journal Developmental Dynamics.  The scientific ideas of both my supervisor and myself  comprise this manuscript. Examining the expression of Bmp-2, Bmp-4, Bmp-7, BmpR-IA and BmpR-IB during facial outgrowth (stages 20, 24, 28) was Dr. Richman's idea. Including BmpR-II, noggin, chordin and tolloid genes in this study and also examining stages where neural crest cell migration (stage 10), initiation of facial prominences (stage 15), differentiation (stage 30, 32) occurs was my idea. The majority of the work and analysis presented in this chapter was done by me. The writing of this manuscript and assembly of accompanying figures was also a joint effort between Dr. Richman and me. I received technical assistance from Sandra Babich, Steve Ritchie and Les Grad who purified plasmids and linearized D N A used to make riboprobes.  87  CHAPTER IV  Bone morphogenetic proteins (BMPs) regulate primary palate closure  Summary The basis for cleft lip is the failure of individual components of the embryonic face to meet and fuse. However, the etiology of non-syndromic cleft lip with or without cleft palate (CL/P) is genetically complex and not well understood. Expression patterns of Bone Morphogenetic Proteins (BMPs) and antagonist Noggin suggested that there is fine control of BMP activity precisely where and when lip closure begins. Bmp-2, Bmp-4 and Bmp-7 are expressed in the zone of fusion between the frontonasal mass and maxillary prominences, and overlap regions of highest proliferation in the facial mesenchyme or are expressed in the adjacent epithelium. Noggin protein was applied to regions with high BMP expression in the zone of fusion and subsequently induced clefts as a result of reduced proliferation and outgrowth of the facial prominences.  Noggin expression is  restricted to the frontonasal mass epithelium and is downregulated at the corners of the frontonasal mass just prior to epithelial breakdown and fusion.  Noggin treatment  promotes epithelial survival. These experiments show for the first time that endogenous BMPs regulate two aspects of lip closure; outgrowth of facial prominences and thinning of the frontonasal epithelium prior to fusion.  We also examined the well-established  retinoic acid model for cleft lip to see whether BMPs were mediating this defect. Ectopic retinoic acid increases Bmp-2 and Bmp-4 expression prior to the induction of cleft lip in chickens. Applying BMP-2 to the face also induced clefts. These results are related to  88  genetic causes of non-syndromic human orofacial clefting and suggest that a loss- or gain-of-function mutation in B M P signaling molecules may be associated with cleft lip.  89  Introduction The basis for cleft lip is the failure of individual components of the embryonic face (facial prominences) to meet and fuse. Fusion in the avian upper beak occurs primarily between the corners of the frontonasal mass (medial nasal prominences in humans and mice) and the maxillary prominences at the base of the nasal slit. Failure of fusion to occur results in a cleft in the upper lip extending into the nostril. The midline of the upper lip is intact, rather the cleft is lateral to the midline. Cleft lip, with or without cleft palate (CL/P), is one of the most common birth defects in humans (approximately 1:800 live births) and is genetically distinct from isolated cleft palate. Unfortunately, human nonsyndromic C L / P is difficult to study because it is due to a combination of environmental and genetic factors (Schutte and Murray, 1999). Relatively few candidate genes have been linked to nonsyndromic CL/P (Schutte and Murray, 1999). Using mouse strains with an increased liability to form cleft lip has isolated at least 2 loci that are linked to cleft lip (Juriloff et al., 2001). The transgenic mouse model has not been particularly informative for the etiology of cleft lip. The majority of orofacial clefts in mutant embryos involve the secondary palate or are midline clefts between the medial nasal prominences (Zhang et al., 1996; Schorle et al., 1996; Lohnes et al., 1994; Richman and Mitchell, 1996; Francis-West et al., 1998). For these reasons another approach is required. We have turned to the chicken model system in which we can phenocopy cleft lip in a form that resembles human cleft lip in order to identify key genetic pathways that regulate lip closure. The fusion of apposed facial prominences is a complex process and requires 1) sufficient outgrowth in order for the prominences to obtain a critical size, 2) removal of  90  the epithelial seam between the apposing prominences, 3) breakthrough and invasion of the underlying mesenchyme to form the mesenchymal bridge (Wang et al., 1995) and 4) that these events occur within a certain period of development (Wang et al., 1995). The genes that regulate these individual processes are starting to be understood.  The  prominences proliferate at different levels and proliferation within the facial mesenchyme of each individual prominence is not uniform (McGonnell et al., 1998; Peterka and Jelinek, 1983; Patterson and Minkoff, 1985; Minkoff and Kuntz, 1977; 1978).  These  regional differences in proliferation as well as the direction of expansion of the facial mesenchyme (McGonnell et al., 1998) contribute to the changes in shape of the embryonic face as development proceeds. Ectopic Shh has been shown to increase the proliferation of facial mesenchyme (Hu and Helms, 1999). In its absence, using Shh antibodies, outgrowth of the facial prominences is reduced resulting in cleft lip (Hu and Helms, 1999). Shh regulates cell proliferation and cell death, thus it is not clear whether the application of Shh antibody inhibits outgrowth by suppressing mesenchymal proliferation, increasing cell death or both. Fgfs have been shown to be able to substitute for the frontonasal epithelium and support outgrowth of epithelium-stripped frontonasal mesenchyme (Richman et al., 1997). However, it isn't known whether FGF signaling is necessary for facial outgrowth. The epithelium between fusing prominences is removed by a combination of processes: the outer peridermal cells are removed via apoptosis while the inner basal epithelial cells undergo epithelial-mesenchymal transformation (Sun et al., 2000). In the mouse palate, exposure to type I collagen appears to trigger epithelial-mesenchymal transformation (Takigawa and Shiota, 2001) perhaps through the integrin mediated  91  activation of integrin-linked kinase (ILK; Somasiri et al., 2001).  However, the  disappearance of the peridermal cells precedes transformation of basal epithelial cells and thus may be necessary for transformation to proceed.  Nothing is known about the  molecular regulation of apoptosis in the peridermal cells of the primary palate. Several methods have been used to phenocopy cleft lip in chickens, including gene disruption experiments, stripping of facial ectoderm in situ (Shh, Hu and Helms, 1999) and exposure to teratogens, such as retinoic acid and cholesterol synthesis inhibitors (Young et al., 2000).  However retinoic acid treatment results in complete  failure of outgrowth of the frontonasal mass (Tamarin et al., 1984; Wedden and Tickle, 1986; Richman and Leon Delgado, 1995) therefore the morphology does not resemble human cleft lip. In addition few genes have been identified that specifically mediate the fusion process. Our goals were to identify genes that are expressed in the mesenchymal bridge region during the time of lip fusion (fusion occurs between stage 28.5-30; Hamburger and Hamilton, 1951; Sun et al., 2000) and then to study the function of these genes in lip fusion. Here we describe the expression of BMP's and the B M P antagonist Noggin in region of contact between the frontonasal mass and maxilla and demonstrate that BMPs are important regulators of outgrowth and of epithelial survival.  92  Materials and Methods In situ hybridization In situ hybridization was performed in wholemount as described in Shen et al., 1997 and for radioactive probe as in Rowe et al., 1991.  Bead preparation Affigel Blue beads 150 pm in diameter were soaked in 1 mg/ml of human recombinant BMP-2 (Genetics Institute) or 1.05 mg/ml of human recombinant Noggin (Regeneron). A single BMP soaked bead was implanted in the corner of the frontonasal mass at stage 24. Noggin treated embryos had a total of 4 beads placed. There were two beads placed in the frontonasal mass and maxilla at stage 22 and 2 more beads added to the same locations at stage 26. The stage 22 and 26 embryos were separated by 24 hours. Retinoic acid beads were prepared as in Richman and Leon Delgado (1995).  Skeletal staining Embryos were skinned and then fixed in 100% ethanol for 4 days followed by 100% Acetone for 4 days. Embryos were stained for up to 10 days in alizarin red/alcian blue solution (1 vol. 0.3% alcian blue 8GX in 70% ethanol, 1 vol. 0.1% alizarin red S in 95% ethanol, 1 vol. acetic acid, 17 vol. 70% ethanol) and then cleared in 20% glycerol/2% K O H solution followed by a series of glycerol/H 0 solutions (50%, 80%, 100%). 2  TUNEL: T U N E L staining was done as described in Shen et al., 1997.  BrdU staining  93  Embryos were treated with BrdU for 2 hours beginning 10 hours after the first 2 Noggin beads were placed. Sections were treated with 2 M HC1 30 minutes, 37°C and proteinase K 5ug/mL at 37°C, 10 minutes, blocked and then incubated for 1 hour at 37°C with antiBrdU (Becton-Dickinson, 1:30). The secondary biotinylated anti-mouse antibody (1:500, A B C kit, Vectastain) and avidin-biotin complex (ABC kit, Vectastain) were applied for 1 hour, each. Detection was performed with diaminobenzidine (DAB). Sections were counterstained with Hoechst 33258 (5 pg/ml) to visualize the nuclei. Cell counts were made on Dab stained and Hoechst stained views to calculate the percentage of proliferating cells. A paired t-test was performed to determine statistical significance.  Epithelial stripping:  Epithelium was stripped by first applying a solution of nile blue sulfate (Yang and Niswander, 1995) and using a tungsten needle.  A c r i d i n e O r a n g e staining  Embryos were rinsed in phosphate buffered saline (PBS), put into a solution of 500 pg/ml of Acridine Orange in PBS for 10 min at room temperature, rinsed in PBS for 10 min and the face was viewed underfluorescenceillumination.  94  RESULTS BMPs are expressed in the region of fusion In order to identify genes and pathways that may be involved in cleft lip we looked for expression of molecules in the mesenchymal bridge region during the time of lip fusion (fusion occurs between stage 28.5-30; Hamburger and Hamilton, 1951; Sun et al., 2000). Bmp-2, Bmp-4 and Bmp-7 are expressed at high levels in the medial maxillary prominence and corner of the frontonasal mass (globular process; Fig. 4. IB; also see Fig. 3.2, 3.3, 3.4; Romanoff, 1960).  Of these BMPs, Bmp-2 is most abundant in the  mesenchyme and overlaps regions of highest proliferation in the face. Bmp-7 and Bmp-4 are predominantly found in the overlying epithelium and my act in a paracrine manner on the mesenchyme (Fig. 3.4B,D).  We also examined the expression of noggin, a B M P  antagonist, in order to give us insight into the local control of BMP activity. Noggin is expressed specifically in a narrow strip of epithelium at the caudal edge of the frontonasal mass (Fig. 4.1A,D). Just prior to fusion noggin is down regulated at the corners of the frontonasal mass (Fig. 4.1C).  These changes in expression correlate closely with the  timing of fusion (Sun et al., 2000) and suggest that B M P activity is involved in the fusion of the lip.  Ectopic Noggin treatment suppress outgrowth of the facial prominences, resulting in cleft lip We locally applied Noggin protein to the developing face in order to address whether endogenous BMPs are involved in outgrowth of facial prominences. Noggin binds all three BMPs examined here (Zimmerman et al., 1996) although the affinity for Bmp-7 is  95  Figure 4.1 Expression of BMP-2 and Noggin at the time of fusion of the primary palate and induction of Bmps with retinoic acid. A Expression of noggin extends around the corners of the frontonasal mass (arrows). B Expression of Bmp-2 in region of contact between the frontonasal mass and maxillary prominences (arrows). C Noggin transcripts are no longer present in the fusing frontonasal mass epithelium (arrows). D Mid-sagittal section showing a narrow strip of epithelium expressing noggin (arrow). E,F,G embryos treated with 5 mg/ml retinoic acid placed in right nasal pit and fixed 6h later. E Increase in expression of Bmp-2 in right frontonasal mass. F deeper section through same embryo as in E showing upregulation in maxilla and lateral nasal prominence. G increased expression of Bmp-4 in right frontonasal mass (arrow). K E Y : fhm - frontonasal mass, lnp, lateral nasal prominence, md - mandible, mx maxilla. Scale bars = 400 pm A - C , E - G and 200 pm for D.  96  96a  10,000 fold less than for Bmp-4. Noggin treatment reduces activity of several BMPs at once thereby avoiding potential redundancy in BMP activity. We had determined that the placement of 4 beads over a 24 hour period produced extremely reliable clefts. Noggin treatment leads to unilateral clefts in the primary palate and deficiencies in the secondary palate (stage 30-38, N = 20/27, Fig. 4.2B,C) while control embryos are normal (N= 4/4; Fig. 4.2E,F).  Prior to extension of the beak (stage 30), the chicken cleft closely  resembles that of human; a gap between the frontonasal mass and the maxillary/lateral nasal prominences extends into the nasal pit (data not shown). The size of the maxilla on the treated side is initially reduced compared to the contralateral side and control embryos (Fig. 4.2A) whereas the globular process is only mildly affected. The clefts are therefore due to the inability of the frontonasal mass and maxillary prominences to make contact. Endogenous BMPs thus regulate outgrowth of the facial prominences.  Skeletal abnormalities are similar to those found in h u m a n cleft lip  In order to find addition evidence that our chicken clefts resemble those of humans we examined the skeletal abnormalities in Our Noggin treated embryos.  The skeletal  phenotype fits with the early changes in morphology in the maxillary prominence. On the side of treatment, the maxillary and nasal bones are completely absent while the palatine bone is either very thin or absent (10/10; Fig. 4.2B-D). Other derivatives of the maxilla are less affected (jugal, quadratojugal and pterygoid). In human cleft lip there are deficiencies in the maxillary bone often necessitating bone grafts (Eppley and Sadove, 2000). The palatine bones are also commonly deficient in human cleft palate, similar to our phenotype. All frontonasal mass derivatives form normally in Noggin treated  97  Figure 4.2 Effects of Noggin on facial morphogenesis. A A frontal S E M view of a  stage 28 embryo, 12h after the second set of beads was implanted. The decreased size of the maxilla on the treated side is evident (arrow). B a side view of the stage 36 beak showing a disruption in the normally smooth edge (arrow). The absence of palatal shelves can be seen on the treated side. The beak has extended fully and is capped with an egg tooth. C , D surface and cleared inferior views of  a specimen with a deficient palate.  Asterisk indicates absent palatal shelf and underlying palatine bone. E , F Control embryo with normal size gap between palatal shelves. White dashed lines outline palatine bone, green dashed line outlines maxilla. G , H Treated and control maxillary prominences, respectively, of BrdU labelled embryo showing decreased proliferation around bead 6h after bead implantation. I Decrease in expression in the frontonasal mass and maxillary prominence (arrow) 6h after the first set of beads compared to contralateral side (stage 24) and J 12h after the second set of beads (stage 28). Expression in frontonasal mass is partially restored on treated side, however expression in maxilla remains low (arrow). K Bright field view of embryo stained with acridine orange, 12h after second set of beads, stage 28 L Fewer dying cells are seen on left side of frontonasal mass (arrows) overlapping the position of the bead.  M Decrease in expression in epithelium and  mesenchyme 6h after the first set of beads (arrow). N Increase in size of expression domain near bead, 12h after second set of beads (arrow). O Treated side of stage 28 embryo 12h after second set of beads with thicker frontonasal mass epithelium (arrow). P Non-treated side of same embryo as in O with thin layer of epithelium. Scale bars = 0.5 mm for A,I,J,K,L; 1mm for B; 200 urn for G,H,M,0,P. K E Y : b - bead, et - egg tooth, fnm - frontonasal mass, j - jugal, qj - quadratojugal, lnp - lateral nasal prominence, md mandible, mx - maxilla, p - palatine, pmx - premaxilla, ps - palatal shelf.  98  embryos (premaxilla, prenasal cartilage and egg tooth, Fig. 4.2B,D). Similarly, nonsyndromic C L / P in humans does not affect the nasal septum and the premaxilla. Thus, the noggin-induced clefts are a close phenocopy of human CL/P.  Noggin treatment suppresses proliferation of the facial mesenchyme  The reduced growth of the maxillary and frontonasal mass prominences may potentially be due to either decreased cell proliferation or increased cell death. There is no increase in programmed cell death (stage 28-29, 7/7; data not shown) therefore reduced proliferation is a more plausible explanation. The maxillary prominences would be particularly affected because rate of proliferation remains high for a longer period of time than in the frontonasal mass (Minkoff and Kuntz 1977, 1978).  Proliferation is  significantly decreased 12h after treatment (stage 24, Fig. 4.2G,H; N = 3, 31% reduction, P = 0.043, frontonasal mass; 39% reduction, P = 0.002, maxilla). This data indicates that endogenous B M P positively control proliferation of the facial mesenchyme.  Msx-1  expression is reduced by Noggin  We looked for gene expression changes that are related to the decrease in cell proliferation and examined Msx-1. There are two other lines of evidence supporting a role for Msx-1 in cell proliferation in the face. First, areas with highest expression of Msx-1 overlap regions with relatively higher proliferation (McGonnell et al., 1998; Mina et al., 1995). Secondly, a reanalysis of the cleft palate of Msx-1 knockout mice shows a decrease in proliferation in palatal mesenchyme (Chen et al., 2001). Msx-1 is down regulated in the maxillary and frontonasal mass mesenchyme 6h after the first beads are  99  implanted (3/3, Fig. 4.21). The second set of beads maintains the down regulation of Msx1 in the maxillary prominence (3/3; Fig. 4.2J) however, Msx-1 expression is restored somewhat in the frontonasal mass. There is no change in embryos treated with PBSsoaked beads (6/6). The relative difference in Msx-1 expression at stage 28 between the maxilla and frontonasal mass correlates with the greater suppression of proliferation in the maxillary prominence. We conclude that Msx-1 is downstream of endogenous BMPs in facial mesenchyme and may be a downstream mediator of BMP-induced proliferation.  N o g g i n maintains survival of the frontonasal epithelium  Noggin expression in the corners of the frontonasal epithelium is downregulated prior to the induction of apoptosis (Fig. 4.1A,C), suggesting that elevated B M P activity may regulate epithelial thinning. Survival should be enhanced in Noggin-treated embryos. Acridine Orange staining was used to preferentially label dying epithelial cells in the primary palate.  Noggin treatment leads to a qualitative reduction in staining near the  bead (6/7; Fig. 4.2K,L).  P o t e n t i a l d o w n s t r e a m targets o f B M P s associated epithelial cell s u r v i v a l  Accompanying the inhibition of cell death, there is also down regulation of Msx-2 expression in the epithelium (Fig. 4.2M; 5/5).  The increased cell survival is evident in  the thickened epithelium on the treated side (3/4, Fig. 4.20) and is accompanied by an expansion of Fgf-8 expression (Fig. 4.2N, 6/7). Noggin also induces an expansion of the apical ectodermal ridge in limb buds (Pizette and Niswander, 1999). changes in Shh expression (data not shown).  There are no  Our data shows that the epithelium is  100  responding to decreased B M P signalling in the mesenchyme by increased Fgf-8 expression and increased survival of periderm cells.  Retinoic acid upregulates Bmp-2 and Bmp-4 expression  Our data with Noggin treatment suggests that BMPs are mediating outgrowth of the facial prominences. We wished to see whether BMPs are mediating cleft lip in another model system with selective inhibition of outgrowth - embryos treated with retinoic acid (RA). Irreversible changes occur by 16 hours (Richman and Leon Delgado, 1995) therefore we looked for changes in gene expression that take place at a shorter interval. Contrary to what we expected, the expression of Bmp-2 and Bmp-4 is rapidly up regulated in the frontonasal mass, maxilla and lateral nasal prominences adjacent to the bead (3/3, Fig. 4.1E-G). Expression of Bmp-7 and noggin is unchanged following retinoic acid treatment (data not shown). This is the earliest reported change in expression specifically in all facial prominences affected by R A (Brown et al., 1997; Shen et al., 1997; Helms et al., 1997). Therefore, increased B M P levels may mediate RA-induced inhibition of outgrowth in the maxillary and frontonasal mass prominences.  Ectopic B M P - 2 application induces clefts that resemble defects induced by Noggin  Based on the results with R A we wished to test in a direct way whether increased levels of B M P inhibit outgrowth. BMP-2 treated embryos had reduced outgrowth of the frontonasal mass (Fig. 4.3A, 3/3) ultimately resulting in a cleft in the upper beak (7/8). Despite the early effects on the morphology of the frontonasal mass, the frontonasal mass derivatives form in all specimens (Fig. 4.3B premaxilla, prenasal cartilage, nasal septum  101  F i g u r e 4.3 Effects of B M P - 2 on facial morphogenesis. A S E M showing the reduction  in size of the frontonasal mass (arrow). B Cleared specimen showing loss of the maxillary bone and distal tip of the palatine bone (large arrow). There is a notch in the side of the beak coinciding with the lost maxillary bone. C T U N E L stained section with increase in staining between the bead and the lateral edge of the frontonasal mass (arrows) and no staining medial to the bead. This embryo represents the highest amount of cell death in the mesenchyme at this time point. Epithelial cell death extends medially (arrowhead). D Peak T U N E L staining in mesenchyme but not extending into the midline. E , F Upregulation of expression in frontonasal mass and maxillary prominence (arrows). G Increase in cell death in the mesenchyme adjacent to stripped epithelium, epithelium has partially regrown lOh after stripping. H Decrease in expression of Shh near the bead. I Downregulation of expression in the nasal slit epithelium compared to the contralateral side (arrow). Scale bars = 0.5 mm for A,C-H,J,K; 50 um for I. K E Y : b - bead, firm frontonasal mass, lnp - lateral nasal prominence, md - mandible, mx - maxilla, ppalatine, pmx - premaxilla.  102  and egg tooth 8/8). Surprisingly, the skeletal defects are similar to Noggin treated embryos. The maxilla and the distal end of the palatine bone are usually absent (Fig. 4.3B; 7/8). Beads soaked in PBS or BMP-2 protein diluent affected neither outgrowth of the globular process nor craniofacial morphology (5/5).  Ectopic  BMP-2  expands  regions  of  apoptosis  a n d Msx-2  expression  in  the  frontonasal mesenchyme  We were curious about the effects on maxillary derivatives and the lack of effect of B M P on the frontonasal mass derivatives especially because the early deficit is localized to the frontonasal mass. In the frontonasal mass, apoptosis peaks 12h-post bead implantation and is predominantly in the lateral mesenchyme (Fig. 4.3C,D; Table 4.ID; 7/7 have > 100 dead cells in lateral mesenchyme). Cell death does not extend into the central region where cartilage will differentiate (7/7 have < 5 dead cells). This accounts for the normal development of frontonasal mass derivatives. Apoptosis does however extend into the maxilla (6h post bead placement, 4/6 show increased cell death; 2/4 had 5-10 dead cells; 2/4 had 10-20 dead cells) indicating that the BMP-2 protein is able to diffuse across from the frontonasal mass. The absence of maxillary bones appears to be due to increased cell death in the maxilla. BMP-soaked beads implanted directly into the maxilla delete maxillary derivatives and cause clefts similar to embryos with beads placed into the frontonasal mass (4/4; data not shown).  We examined expression of Msx-2, a gene that is induced by BMPs (Vainio et al., 1993; Barlow and Francis-West, 1997; Wang et al., 1999; Ekanayake and Hall, 1997) and can  103  Table 4.1 Number of TUNEL positive cells in BMP-2 treated embryos  Number of cells 0-5 5-10 10-50 50-100 > 100  Number of cells 0-5 5-10 10-50 50-100 > 100  Table 4.1 A: F N M corner epithelium Treated side (N = number of embryos) Untreated side (N = number of 4.5 6 hours 12 hours 24 hours 4.5 6 hours 12 hours hours hours 3 0 0 0 4 6 5 1 1 4 0 0 0 2 0 5 7 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0  Table 4.1B: F N M caudal epithelium Treated side (N = number of embryos) Untreated side (N = number of 4.5 6 hours 12 hours 24 hours 4.5 6 hours 12 hours hours hours 2 0 0 3 4 6 6 2 0 0 3 0 0 1 0 6 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0  embryos) 24 hours 4 1 1 0 0  embryos) 24 hours 5 0 1 0 0  Number of cells  Table 4.1C: F N M corner mesenchyme Treated side (N = number of embryos) 4.5 6 hours 12 hours 24 hours hours  distal region of prominence) Untreated side (N = number of embryos) 4.5 6 hours 12 hours 24 hours hours  0-5 5-10 10-50 50-100 > 100  1 2 1 0 0  4 0 0 0 0  0 0 6 0 0  0 0 1 6 0  0 6 0 0 0  4 1 1 0 0  2 3 2 0 0  3 3 0 0 0  Table 4.1D: Lateral third of F N M mesenchyme and future mesenchymal bridge region Number Treated side (N = number of embryos) Untreated side (N = number of embryos) of cells 4.5 hours 12 hours 24 hours 4.5 hours 12 hours 24 hours 0-5 5-10 10-50 50-100 > 100  0 0 6 0 0  0 0 0 0 7  0 0 0 0 6  6 0 0 0 0  2 5 0 0 0  4 2 0 0 0  104  increase programmed cell death (Graham et al., 1994), and find that expression is expanded in the lateral third of the frontonasal mass and extends into the maxillary prominence (3h 6/6, Fig. 4.3E; 6 hours, 3/3, Fig. 4.3F). We show that up regulation of Msx-2 precedes B M P induced cell death and that the spatial distribution of cell death correlates with the skeletal defects.  F r o n t o n a s a l mass epithelium maintains mesenchymal s u r v i v a l  There appears to be a sequential relationship between the cell death in the epithelium and mesenchyme. Programmed cell death in the epithelium occurs following a lag of 4-5 hr (Table 4.1). By 6-7h it is apparent that the frontonasal mass epithelium is particularly sensitive to B M P (Fig. 4.3C; Table 4.1). At the same time point, mesenchyme medial and directly contacting the bead lacks T U N E L positive cells (Fig. 4.3C, 6/6 have <5 dead cells medial to the bead). It is possible that initial cell death in the epithelium promotes cell death in the mesenchyme.  To test this we stripped frontonasal mass ectoderm.  Programmed cell death is increased on the side that was stripped (5/5; Fig. 4.3G; no increase in control embryos, 3/3). Embryos that are left to grow following epithelial stripping develop cleft lip due to reduced outgrowth of the globlular process (Hu and Helms, 1999). Here we show that reduced outgrowth is in part due to mesenchymal cell death.  B M P s suppress Shh a n d Fgf-8 expression  The effects of exogenous BMPs on epithelium and subsequently on mesenchymal cell death are similar to the effect of epithelial removal. The B M P experiment however,  105  allows us to study the molecular changes in the epithelium that may be important in maintaining cell survival in the mesenchyme.  BMP treatment may decrease expression  of genes in the epithelium that encourage outgrowth and survival. SHH and FGF-c? have been shown to promote cell survival in the head and limb bud (Ahlgren and BronnerFraser, 1999; Sanz-Ezquerro et al., 2000; Crossly et al. 1995; Trumpp et al., 1999). S H H in the frontonasal mass epithelium may also be promoting outgrowth of facial mesenchyme (Hu and Helms, 1999). In our study, Shh and Fgf-8 are down regulated by BMP-2 prior to the induction of programmed cell death in the epithelium (Shh 6/6, 3h post bead implantation, Fig. 4.3H; Fgf-8 3/3, 3-6h after bead implantation, Fig. 4.31). The loss of epithelial maintenance factors precedes and may facilitate epithelial apoptosis. The frontonasal epithelium is required for mesenchymal outgrowth (Wedden, 1987; Richman and Tickle, 1989; 1992; Hu and Helms, 1999) but our findings suggest that it is also required for mesenchymal survival.  106  Discussion C o n c e n t r a t i o n dependant effects of B M P s i n the face BMPs are versatile molecules that regulate many biological processes, including cell proliferation  (positively  or negatively),  cell  differentiation  and apoptosis  in a  concentration and stage specific manner. In the face, we have shown that endogenous BMPs regulate proliferation of the facial mesenchyme and apoptosis in the frontonasal mass epithelium. Noggin application to face suppressed proliferation of the mesenchyme by as much as 39%. Msx-1 was absent from the maxillary prominence suggesting that most of the BMPs were being antagonized. These findings indicate that BMPs act with other genes to regulate proliferation. These candidates genes include S H H and FGFs, which are not downregulated following Noggin treatment. Previous experiments where BMP-2 soaked beads were placed in the maxillary prominence did not cause clefts (Barlow and Francis-West, 1997). The reasons may be: 1) the protein concentrations were less than those used here, 2) the position of the bead was more posterior. Thus programmed cell death occurred near the bead, far from the region that fuses with the frontonasal mass (Barlow and Francis-West, 1997).  Despite close proximity with the mesenchyme, the epithelium undergoes cell death. This may be due to higher levels of BMPs in the epithelium as compared to the mesenchyme. The mesenchyme expresses Bmp-2, while the epithelium expressed Bmp-2, Bmp-4 and Bmp-7.  It isn't known whether this difference in expression levels translates into  differences in protein bioactivity however regional differences in Msx-1 and Msx-2 suggests this. BMPs induce cell proliferation al low levels and apoptosis at high levels,  107  which appear to be mediated by Msx-1 and Msx-2 respectively.  Thus at low levels  BMPs induce Msx-1 expression and at high levels induce both Msxl expression.  and Msx-2  Msx-2 must override the proliferative effects of Msx-1 in regions of co-  expression. Thus, in the frontonasal mass Msx-1 is expressed widely in the mesenchyme relative to the restricted region of Msx-2 expression in the immediate corner of the prominence. Alternatively, the epithelium may just be more sensitive to apoptosis as we showed in our study and thus there may not be any differences in BMP levels between the epithelium and mesenchyme.  Interplay between cell death and cell survival factors prior to fusion The epithelial seam normally expresses cell death (Bmp-2, Bmp-4, Bmp-7, Msx-2) and cell survival factors (Fgf-8, Shh, noggin) and the expression levels, with the exception of noggin, remain constant as the prominences contact each other (Figure 4.4). The ratio between B M P and FGF activities determines the balance between cell survival and cell death (Buckland et al., 1998). SHH should also be included in this equation as ectopic SHH can also inhibit BMP regulated apoptosis in the interdigital tissue (Sanz-Ezquerro, 2000). It is necessary to carefully orchestrate cell death and cell survival signals in order to facilitate fusion of the lip. When the decision is made to induce apoptosis, the localized downregulation of noggin allows the embryo to tip the balance towards cell death rather than survival (Fig. 4.4). Different signals may operate to control B M P activity in maxillary epithelium since Fgf-8 and Noggin are not expressed here.  108  Figure 4.4 Model of BMP function and pathways in the frontonasal mass. In the frontonasal mass mesenchyme, low B M P levels (only Bmp-2 is expressed) regulate proliferation, which may be mediated through Msx-1. R A and Shh may act upstream of Bmp-2. In the frontonasal mass epithelium, high B M P levels (Bmp-2, Bmp-4 and Bmp-7 are expressed) regulate apoptosis in the epithelium, which may be mediated through Msx-2.  This apoptotic signal is antagonized by Noggin in the caudal epithelium and  balanced by survival signals, such as Shh and Fgf-8 in the caudal and lateral epithelium, respectively.  Prior to fusion, B M P inhibtion is removed from the corner of the  frontonasal mass epithelium (noggin is downregulated) and thus the balance is shifted to favor cell death.  109  109a  BMPs are downstream of RA R A induces cellular effects by binding to its receptors that are transcription factors, and altering gene transcription. biological functions.  These downstream targets of R A thus mediate RA's  In the chicken, elevated R A levels reduces outgrowth of facial  prominences that give rise to the upper beak. Several genes have been identified whose transcription is altered following R A treatment, including Shh (Helms et al., 1997), FHF4 (Munoz-Sanjuan et al, 2001), Msx-1 and Msx-2 (Brown et al., 1997) and AP-2 (Shen et al., 1997). SHH has been shown to be particularly important since in inhibition results in clefts (Hu and Helms, 1999). However in R A treated embryos, Shh is downregulated very late, 30 hours after treatment leaving open the possibility that other genes are mediating this change in expression. Other genes downstream of R A may also be in part altered through B M P overexpression.  We have shown that ectopic BMP-2 can suppress Shh  expression. The fact that high R A concentrations inhibit Msx-1 and Msx-2 expression is somewhat perplexing since BMPs upregulate Msx-1 and Msx-2 expression in the face (Barlow and Francis-West, 1997; Wang et al., 1999; Ekanake and Hall, 1997). At low R A concentrations, R A is able to upregulate Msx-1 and Msx-2 expression (J. Hui, personal communication) and may be acting through BMPs.  However, at high  concentrations perhaps R A - R A R complexes are able to bind to Msx promoter regions and prevent Smad binding. Whatever the case, high R A levels can clearly over-ride BMP induced transcription of Msx genes.  Bmp-2 (and Bmp-4) however is upregulated in all the facial prominences of the upper beak. In addition, this upregulation is very rapid and thus may be an early response gene.  110  By 6 hours increased Bmp-2 transcription is very evident.  We did not examine the  earliest time at which Bmp-2 transcription is altered. Finally, we replicated this facial defect by apply ectopic BMP-2 protein to the mesenchyme where Bmp-2 is upregulated. One mechanism by which BMP-2 may mediate RA's suppression of outgrowth is by increasing cell death. Bmp-4 and Bmp-7 have been shown to mediate RA-induced cell death in the limb (Rodriguez-Leon, 1999).  BMP pathway association with human cleft lip We conclude that the 2 physiological roles of BMP during primary palate formation are 1) to control proliferation and outgrowth of the facial prominences; and 2) to thin the epithelium within the contact region via apoptosis. The question then arises, are B M P signalling pathways affected in mammalian orofacial clefts? Msx-1 and Msx-2 genes have been linked to orofacial clefting.  van den Boogaard et al. (2000) reported that  haploinsufficiency of human MSX-1, is correlated with cleft lip and palate. Msx-1 knockout mice have cleft palate but no cleft lip (Satokata and Maas, 1994) whereas Msx2 knockout mice have calvarial defects but no clefts (Satokata et al., 2000). The double knockout of Msx-1 and Msx-2 has bilateral cleft lip and palate (Maas, personnel communication). We predict that Msx-l/Msx-2 -/- embryos will have decrease in programmed cell death in the epithelium (due to lack of Msx-2, our Noggin data) and this will accompany decreased mesenchymal proliferation (due to lack of Msx-1; Chen et al., 2001).  It will be interesting to see whether the double knockout embryos have a  persistent epithelial seam in the primary palate. The data on Msx knockouts associates a part of the BMP pathway with clefts of the primary and secondary palate, however we are the first to demonstrate that endogenous BMPs are required for fusion.  Our findings  111  predict that elevated B M P activity may also cause clefting in humans. Directed overexpression of BMP-4 in the epithelium with a keratin promoter, which results in cleft palate (Blessing et al., 1993). It is not clear in the paper whether cleft lip is also observed in these trangenic mice. A loss-of-function mutation in the BMP antagonists would lead to increased BMP signalling. We identify for the first time, several critical roles for B M P signaling in fusion of the primary palate - by directly altering BMP activity we are able to phenocopy cleft lip. We suggest that a gain- or loss-of- function mutation in a B M P signaling molecule could be associated with human cleft lip.  Contributions to this chapter  The preceding chapter represents a manuscript that was submitted to the journal Science. Due to the space restraints in the journal, it is written in a very concise manner. These findings are presently being re-written for submission to the journal Development.  The scientific ideas from the initiation to the completion of this chapter were solely mine. The majority of the work presented in this chapter, data analysis and interpretation of results was done myself.  Dr. Richman did the skeletal analysis of stage 38 heads and  assisted with the T U N E L data analysis. I received technical assistance from Kathy Fu, who did the whole-mount in situs, T U N E L assay and BrdU labeling. Dr. Richman assisted in the writing of this manuscript and in the assembling the accompanying figures. Special thanks to Mary MacDonald, Sonia Donati, Kathy Fu, Jenny Hui, Jessica Kim and Mya Tuong for counting the BrdU labelled cells.  112  CHAPTER V  The type IA and type IB Bone morphogenetic protein receptors (BMPR) have similar functions in regulating chondrogenesis and feather formation in the chicken embryo  Summary  The facial skeleton is unique because it is neural crest derived. We have identified the expression of BMPs, BMP receptors and noggin during the different phases of cartilage morphogenesis in the face.  We sought to examine the role of the type IA or type IB  receptors' signaling. Our approach was to express dominant-negative (dn) and constitutively active (ca) B M P type I receptors prior to and during chondrogenesis. Ectopic expression of a dominant-negative form of BMPR-IB unilaterally in the face resulted in thinner, shorter cartilage elements and deviation of the upper and lower beaks towards the injection site. The phenotypes of caBMPR-IA and caBMPR-IB were similar - deviations of the beak away from the injection site caused by an overgrowth and fusion of cartilage elements. The embryos injected with ca viruses also lacked feather germs over much of the head and about 50% of specimens did not form an egg tooth. Both the ca viruses lead to ectopic expression of Msx-1 and Msx-2 in infected regions and upregulation Bmp-2 and Bmp-4 in adjacent regions. We conclude that the type IA and type IB receptors play similar roles in regulating cartilage and feather formation in the skull.  113  n  Introduction BMPs were originally purified on the basis of their ability to induce local formation of ectopic bone and cartilage when implanted under the skin or into the muscle of rats (Urist 1965; Urist et al., 1979). Since their discovery, the BMPs have been shown to be important signaling molecules in a range of developmental processes including apoptosis, chondrogenesis and embryo patterning (Hogan, 1996). In the developing chicken face, Bmp-2, Bmp-4 and Bmp-7 are expressed in both the ectoderm and mesenchyme (FrancisWest et al., 1994; Wall and Hogan, 1995; Wang et al., 1999; Ashique et al. submitted) in specific patterns. The addition of exogenous BMP-2, BMP-4 and BMP-7 protein to the developing chick face induces the formation of supernumerary bones in the maxillary region and secondary growth sites branching off of Meckel's cartilage (Barlow and Francis-West, 1997; Wang et al., 1999). These data argue for an important role for endogenous members of the B M P family in facial development but the mechanism of signal transduction is not clear.  BMPs bind to two type I receptors (IA also known as Alk-3, Brk-1; IB also known as Alk6, Brk2) and one type II receptor (Liu et al., 1995). Ligand binding can form heterotetrameric complexes with the type I and II transmembrane serine-threonine kinases (ten Dijke et al., 1996; Yamashita et al., 1994). The type I receptor phosphorylates downstream targets including members of the Smad family (Smad 1, 5, 8; Heldin et al., 1997; Kretzschmar and Massague, 1998; Hoodless and Wrana, 1998). Despite sequence divergence between Bmp-4 and Bmp-7, both of these growth factors share the same receptors (Koenig et al., 1994; ten Dijke et al., 1994). Other members of  114  the BMP family, growth/differentiation factors (GDF5, GDF7) also bind to B M P receptors (Yamashita et al., 1996; Nishitoh et al., 1996). Thus there is some promiscuity in ligand binding characteristics of BMP receptors.  Expression patterns of BmprIA and BmprIB in the developing craniofacial complex have been described and there are considerable areas of overlap (Ashique et al., submitted). However, in a small subset of tissues, the type IB receptor is expressed at high levels suggesting that the BMPR-IB receptor may have unique functions in craniofacial development. Knock-out mice lacking BmprIA  die early in development prior to  gastrulation (Mishina et al., 1995) confirming that widespread expression of IA receptor is necessary for early stages of embryogenesis. In contrast, the phenotypes of BmprIB -/mice are much more limited. Mice lacking the cartilage-specific isoform of BmprIB do not form the distal elements of the foot and handplate (Baur et al., 2000). A similar distal limb truncation has been reported in mice lacking exon 1 of BmprIB  (Yi et al., 2000).  The lack of abnormalities in the rest of the embryo suggests that the type IA receptor can take over for the IB receptor during differentiation of the craniofacial skeleton, most of the axial skeleton and long bones of the appendicular skeleton. It seems likely, therefore that the type I B M P receptors share similar functions in the differentiation of craniofacial skeletal tissue.  Another approach used to determine the redundant or unique functions of BMP receptors, has been to express constitutively active (ca) or dominant negative (dn) forms of BMPRIA and BMPR-IB in the chicken limb bud. The ca receptors can simulate Smadl  115  phosphorylation in the absence of a BMP ligand (Hoodless et al., 1996; Kretzschmar et al., 1997) whereas dn receptor constructs will bind ligand but are unable to phosphorylate downstream targets (Zou and Niswander, 1996). Retroviruses expressing mutated forms of BMPR-IA and BMPR-IB have been injected into the limb bud region of chick embryos (Zou and Niswander, 1996; Kawakami et al., 1996; Yokouchi et al., 1996; Zou et al., 1997). Infection of stage 20 limb buds showed that both ca receptors lead to increased chondrogenesis  indicating that at this stage of development they have  overlapping functions. In contrast injections done before stage 17 gave different results for the ca IA and calB receptors. The caBMPRIA delays endochondral bone formation with no apoptosis detected whereas caBMPR-IB virus gave rise to thin limbs due to increased programmed cell death (Zou et al., 1997).  Work from two different labs  showed that dn BMPRIA gave no phenotype (Zou and Niswander, 1996: Kawakami et al., 1996). These studies suggest that the type IB receptor is involved in the cell death pathway whereas the type IA receptor mediates chondrocyte differentiation (Zou et al., 1996).  The facial skeleton differs from the appendicular or axial skeleton in that most of the mesenchyme is derived from cranial neural crest (Kdntges and Lumsden, 1996; Couly et al., 1993; Le Lievre, 1978; Noden, 1978) cells rather than paraxial mesoderm. The difference in embryonic origin may mean that BMP receptors function differently in the face compared to elsewhere in the body. The three facial prominences, frontonasal mass, lateral nasal prominence and mandibular prominence differentiate into cartilage between stage 24 and 28 (Matovinovic and Richman, 1997). The frontonasal mass gives rise to the  116  interorbital septum, prenasal cartilage, premaxilla and egg tooth (Wedden, 1987; Richman and Tickle, 1989; 1992; Richman et a l , 1997), the lateral nasal prominence forms the nasal chonchae and the paired mandibular prominences form Meckel's cartilage and all the mandibular bones.  The present study addresses the function of type IA and type IB Bmprs in facial development by targeted viral misexpression of constitutively active or dominant negative B M P receptors. We compared the phenotype of embryos at several stages to determine whether there are differences in function of the type IA or type IB receptors during craniofacial development. We find that in most cases the phenotypes were similar and therefore our data suggests the type I receptors have similar roles in chondrogenesis and feather formation.  117  Materials and Methods Chicken Embryos Fertilized white leghorn chick eggs were obtained from Coastline Chicks, Abbotsford, B.C. and specific-pathogen-free Hyline chick fertilized eggs were obtained from Hyvac, Iowa.  In Situ Hybridization probes The cDNAs were kindly provided by the following individuals: RCASBP pol (L. Niswander), chick Msx-1 and Msx-2 (S.E. Wedden), chick type II collagen (P. Brickell; Devlin et al., 1988), chick BmprIA, chick BmprIB (L. Niswander; Zou and Niswander, 1996; Zou et al, 1997), chick Bmp-2 and Bmp-4 (P. Brickell; Francis-West et al., 1994), and chick Bmp-7 (B. Houston; Houston et al, 1994). Antisense riboprobes labelled with [ S]35  UTP were prepared as described in Rowe et al (1991).  Virus Infection Replication competent RCASBP virus was grown up using methods described in Morgan and Fekete (1997). RCASBP (A) virus containing the following genes were injected into embryos: human wild type (wt) BMPR-IA, human dominant-negative (DnA) BMPR-IA, human constitutively-active (CaA) BMPR-IA (human alk-3 cDNAs originally provided by P. ten Dijke), chick wtBMPR-lB, chick dnBMPR-lB (DnB), chick caBMPR-IB (CaB, Zou and Niswander, 1996; all BMP receptor viruses generously provided by L . Niswander). Human placental alkaline phosphatase (PLAP) cloned into RCASBP (E) was provided by S. Hughes. A Picospritzer (General Valve Corp.) with electrolytically pulled needles was used  118  to deliver the virus to the embryo. In stage 15 and 20 embryos the vitelline membrane was torn prior to injection.  A third group of virus infected embryos were incubated until they reached stage 38 and then skeletal morphology was assessed. Approximately 50% of embryos used for the characterization of the stage 38 phenotype were pathogen-free. There was no difference in the morphology between pathogen-free and pathogen-containing chick embryos.  Skeletal staining Stage 38 embryos were fixed in 10% neutral-buffered formal saline or 5% tricloroacetic acid (TCA) for at least 24 hours. The eyes and skin were then removed, and specimen was washed in water for two days. Specimens were stained for a week in alcian blue solution for one week (lOmg alcian blue 8GX (BDH), 70ml absolute alcohol and 30 ml glacial acetic acid (BDH)), and then destained in the above solution minus the alcian blue for 4 days.  De-staining was continued in 75% alcohol, 50% alcohol, 25% alcohol,  distilled water (2x) for two hour each and then 0.5% potassium hydroxide (KOH) until soft tissue was nearly transparent.  K O H solution was changed regularly.  Specimens  were then stained for bone with fresh K O H solution containing approximately 0.005% alizarin red for 24h.  De-staining was done using 25%, 50% and 75% glycerol in 1%  K O H for 5-7 days each and then 100% glycerol for storage.  119  Results  T a r g e t i n g v i r a l s p r e a d i n t h e face.  Retrovirus injections have not been done in the face previously.  Therefore, it was  necessary to test various injection sites in order to find several that gave reproducible viral spread. The RCASBP retrovirus containing a PLAP (placental lactogen alkaline phosphatase) insert was injected into the mesenchyme at multiple sites and stages (Fig. 5.1).  Histochemical detection of alkaline phosphatase activity indicates extent of virus  spread.  Injection sites that resulted in unilateral viral spread were of particular  importance allowing us to compare phenotypic effects with the contralateral side of the face/head.  Two sites (II and V) gave predominantly unilateral viral spread and were  chosen for subsequent injections using mutant B M P receptor constructs (Table 5.1). When injected into the right of the mesencephalon at stage 10 (site II), there is heavy infection of the right side of the face including the frontonasal mass, mandible, and maxilla at stage 28 or 96h post-injection (Fig. 5.2 and Table 5.1). Injections caudal to site II, right of the mid-hind brain junction (site IV), resulted in similar viral spread and thus were grouped together.  Virus suspension injected at stage 14 into the presumptive  maxilla/lateral nasal prominence (right side dorsal to the eye; site V) show similar virus spread but less infection to the mandible (Table 5.1). Thus, late injection allowed us to selectively infect the upper face and injection at stage 10 lead to infection of both the upper and lower facial primordia. Viruses used were replication-competent and therefore are capable of spreading throughout tissue. We detected minimal virus spread across the midline of the frontonasal mass or mandibular prominences 4-5 days post-injection (Fig.  120  Stage 10 embryo  Stage 14 embryo  Figure 5.1 Targeting virus to the developing facial promiences. Multiple areas were chosen for retroviral injection in order to determine sites that gave reproducible infection. The sites used for the present study were n, IV (A) and V (B).  These gave primarily  unilateral spread of the virus. The direction of the needle was found to be important in reproducing the viral spread. Key: M - mesencephalon; O - otic pit; P - prosencephalon; S - somite; 1-8 - rhombomeres 1 through 8.  121  Figure 5.2 Histochemical detection of alkaline phosphatase activity in chicken embryos infected with the RCASBP/alkaline phosphatase retrovirus.  Purple stain shows the  extent of viral spread. Viral suspension injected into the mesenchyme to the right of the mesencephalon at stage 10 resulted in infection of all facial prominences. The embryowas fixed at stage 28, 96 hour following injection.  Note how the virus has not spread  extensively across the midline of the frontonasal mass or mandibular prominences. Key: F - frontonasal mass; Mx - maxilla; M  D  - mandible. Scale bar = 1mm.  122  122a  63  CN  CO  u  8 Oh  X cu  o ce  ft)  2  DU  cu  s  -a d «  CN  Cl  CN  >/">  in  ^ H  CN  CD  "•C  "3 a  a o  CN  CJ  CU ISC  a  CO  13 S-l  -a I. C.  l l  CO  'I  S ft  P  I—I  IS cu  B  7  CN  £  s-  CN  CN  m  CN  in  £ .9 .fl  CA  — i1 1/3  o  o  * 2 8 CU  Cl  js CB © ft  cu  o  ^ 5  S-l  fl  SH  o cj  CO  c/5  fl © SCU  SH  Cl  Cl  °c3  Cl  SH  cu CU  (U  cu  o  CvS  3  cS  WD  fl  Cl  *o.  5  .9 fl  .a  d o  1  es «  I  o  as  r< cu  CU  !§  cu >-> CU  CU  Ch  CU  o S-i  o  <u  CU  o  CU  o Sh D  •a CU O  >n  IT)  c3  cu  CU  13  _ a CU  S3 cl  .His cl  %  C3  CU  CO hH  >  <U  a  o  HH  >  o 13 ti CO  O  SH +-» 00  3  5.2). There are some differences in the degree of infection or intensity of alkaline phosphatase staining which is likely due to variation in infection efficiency within individual injection.  Generally however, embryos examined for alkaline phosphatase  activity 5-6 days post infection showed more intense staining and increased spread throughout the primordia than those embryos analyzed 3 days post infection. There were never any malformations of the face produced in embryos injected with the PLAPcontaining virus indicating that the virus itself is not affecting morphogenesis (data not shown).  We also checked viral spread in tissue sections (Table 5.2). A riboprobe made to the pol region of the RCASBP virus was used to detect viral expression in injected embryos. These results show that at stage 24 (72 hrs post injection) there are foci of viral expression within the mesenchyme but by stage 29 much more of the mesenchyme on the right side of the head is expressing the virus (Fig. 5.6A). The pol probe shows the virus location. Sections hybridized to the BmpR-IB probe show only the position of the virus if exposed for a short period (6 days, Fig. 5.6B). The endogenous BmpR-IB signal does not normally develop for another 10-15 days. There was no difference in the extent of virus spread between the various viral constructs.  DnBMPR-IB virus leads to decrease in the size of cartilage condensations The uniform, bilateral production of developmental signals results in a symmetrical body pattern. We unilaterally altered B M P signaling in order to disrupt this symmetry and compare the phenotype resulting from ectopic B M P receptor expression with the normal  124  IT)  CN  fl 8  2  8 Oh  PL, c3  'Si  C/3  O fl  fl  3  -  CO  0) 00  m  >  i  a  u  Q  CD  fi  o o <L>  >  contralateral craniofacial region. In order to determine the role of the endogenous BMPs in craniofacial skeletogenesis we used two dn virus constructs; dnBMPR-IA and dn BMPR-IB.  There was no phenotype produced with the dnBMPR-IA in any embryo.  These results are similar to those reported in the limb bud (Zou and Niswander, 1996; Kawakami et al., 1996).  In contrast to the dn-BMPRIA virus, embryos injected with dnBMPR-IB. virus had a phenotype. Injection of dominant negative dn-BMPR-IB at stage 10 results in deviation of both lower and upper beaks towards the injection site (Fig. 5.3C,D; Table 5.3). Straight extension of lower beak is prevented because the infected side of Meckel's cartilage is shorter and thinner than the contralateral side (Fig. 5.4D,E,F). Approximately half of the embryos have a short segment of cartilage missing from the centre of Meckel's cartilage on the infected side (Fig. 5.4D). Components of hypobranchial complex are normal. The prenasal cartilage and distal half of the nasal septum in the upper beak is shorter and deviated to the side of injection. The cartilage differentiation is not as advanced as noninfected cartilage judging by alcian blue staining (Fig. 5.5 A,B,C). Similar defects were observed in the upper beak of embryos injected at stage 14. Truncations in the upper beak are also seen which is likely due to virus crossing the midline and affecting the entire F N M region (Fig. 5.3E,F and 5.4G,H,I). The lower beak was unaffected, consistent with the virus being targeted to the upper beak.  126  Figure  5.3  containing  Surface  BMP  morphology  of  heads  r e c e p t o r c o n s t r u c t s . (A,B)  following  injection  of  retroviruses  wtBMPR-IB injected at stage 14 has nearly  normal beak development except for a mild deviation of the upper beak towards the left. (C,D) dnBMPR-IB injected at stage 10 on left side resulted in lower beak deviation towards the left (arrow) and normal extension of upper beak. (E,F) dnBMPR-IB injected at stage 14 led to shortening of upper beak (arrow). Note presence of feather germs on head (arrowheads). (G,H) caBMPR-IB injected at stage 10 on right side. The lower beak is curved to right and upper beak is much thicker than normal. The egg tooth is small and there are many patches of ectoderm have no feather development (arrowheads). (I,J) caBMPR-IB injected at stage 14. The upper beak is curved severely to the left and lower beak is normal. Note denuded areas of ectoderm lacking feather germs (arrowheads). (K,L) caBMPR-IB injected at stage 14. The second phenotype was general spread of virus causing thickening of upper and lower beak cartilages. The upper beak has a bilateral partial cleft or notching of the tomium (arrows). The egg tooth is absent in this specimen (arrowhead) as are all feather germs. Keratinization has occurred randomly on the ectoderm instead of in the feathers. Key:  E T - egg tooth.  127  F i g u r e 5.4 M o r p h o l o g y o f t h e c h o n d r o c r a n i u m i n c l e a r e d s p e c i m e n s . Photographs in  A , G , M are superior views while those in B,D,J are inferior views. (A,B,C) wtBMPR-IB injected at stage 14. Mild curvature of the prenasal cartilage towards the right seen in superior and inferior views. (D,E,F) dnBMPR-IB injected at stage 10 on the right. Meckel's cartilage is thinner and has a small dehiscence on the right side causing the mandible to curve towards the right. There is a sharp bend in the cartilage on the left side (arrow). (G,H,I) dnBMPR-IB injected at stage 14. Prenasal cartilage is shorter and curved inferiorly (arrow) and the lower beak has extended normally. (J,K,L) caBMPR-IB injected at stage 10 on the right side. Extensive viral spread throughout the head has lead to the formation of many ectopic nodules and thickening of Meckel's cartilage. (M,N,0) caBMPR-IB injected at stage 14. The upper beak is curved sharply towards the left side (arrow) and is much thicker than normal. Meckel's cartilage has extended normally. Key: PNC - prenasal cartilage; M C - Meckel's cartilage; N C - nasal cartilage  128  F i g u r e 5.5. E f f e c t s o f B M P R v i r u s e s o n c a r t i l a g e d i f f e r e n t i a t i o n .  A, stage 38 dn-BMPR-IB injected embryo showing lack of Alcian Blue staining of Meckel's cartilage. B, right side enlargement of Meckel's cartilage showing one small region with normal Alcian Blue staining (arrow). C, normal staining of left Meckel's cartilage with Alcian Blue. D, frontal section of stage 32 embryo injected at stage 15 into the presumptive maxillary region with ca-BMPR-IB, hybridized to pol probe. Black silver grains (red arrow) indicate expression of virus primarily on left side of embryo. E , near adjacent section stained with alcian blue to show the cartilage. F, higher power of nasal septal cartilage showing closely packed chondrocytes in disorganized matrix. Scale bars = 200 pm for D,E; 50 pm for F; 100 pm for B,C; 2mm for A. Key: e - eye, ios - interorbital septum, mc - Meckel's cartilage.  129  fl I CO . CO  ^  8  [__, "  sw 3 CM GO O  O O  o  o  O CN  O  o  o  o  o  vo >o  T f OO  -a CD  o  o  o  o  vo iy->  T f 00  O  o  o  O ro  fl  IT)  fl  _o H 'to 2 co  o  fe  E  GD cu fl M &,  M O GO CO fe  T3 CD  a-a  3 CD  PQ  2  2  11  M  o  CD _ CD w cp eS "O CD rfl  o  -H  T3 CD  CD  13 <T  T3 c  —I T f  _ ti ti g  o  13  CN  S-c CD  O rfl  T3 O  a aa  O vo  o -fl  •a PQ M  2 rfl  1o  cu  S  V©  w  T3  T3 .-fl  fl m  CD CD > CD  OT  VO  f~H  >  !H  CD M  CN  a'? > >  (D  CD  CD CO  CD CO  vo r-  C o  o CD  o  CM  3fc fl .2  '-M  o CD  CM  O  CD  Q  I/O  O  TM  \ Tf t>- rM  +  ro h CN O vo  r-H r H •M  CM  CD  i—I r H  CD  CM  I  PQ PH  9  O  +  CD -*-»  co  o  ro vo ^ CN  >>+  fl  PH  O. <D 'co co  1 I  a o  +O -» fl  •>  'I  '§  CD  o  >  CD T 3  CaBMPR-IA and caBMPR-IB produce enlarged cartilage In general, the wtBMPR-IB virus produced a similar but much less severe phenotype as the caBMPR-IB infected embryos, therefore these results will not be described in detail. Injection of wtBMPR-IB virus particles, at stage 10 and at stage 14, results in a mild deviation of the upper beak away from the side of injection (Fig. 5.3A,B). Cleared embryo indicates a deviation of the prenasal cartilage at the nasal septum and a mild increase in the chondrogenic region on the injected side of the nasal septum (Fig. 5.4A,B,C).  In order to compare the potential functions of the type I receptors, caBMPRIA and caBMPRIB viruses were injected into the facial region. The effects of the two constitutively active type I viruses are very similar. There were two phenotypes produced with the embryos injected with the constitutively active viruses; one was due to an infection limited to the beak and the other phenotype was due to widespread infection of the entire head. Embryos with primarily beak infections have either very severe curvature of the upper beak away from the side of the injection or straight but thickened upper beaks (Fig. 5.3G,H). In no instance is the upper beak deviated towards the side of the injection. Cleared specimens indicate that cartilage elements are thickened, abnormally shaped and often fused together.  There is hypertrophy of the prenasal cartilage, nasal  septum, interorbital septum, and orbital rim on the side of the injection (Fig. 5.4J,K,L). The prenasal cartilage is occasionally fused with the nasal septum (data not shown). In contrast, embryos with deviated lower beaks, the deviation is nearly always deviated towards the side of the injection (Fig. 5.3G,H). The cleared specimens reveal that Meckel's cartilage is thicker and often bent on the injected side (Fig. 5.4J,K,L). The  131  deviation is therefore not due to lack of cartilage but rather because growth is not directed rostrally on the injected side resulting in tension on the opposite side of the mandible. The milder defects are predominantly observed when injections are done at stage 14/15, and either do not effect the mandible at all or only mildly (Fig. 5.3I,J, and 5.4 M,N,0).  Embryos with more widespread infection had enlarged heads compared to control embryos at similar stages (Fig. 5.3K,L). The earlier in development the infections are done the more severe the disruption of morphology. Virtually no individual cartilage elements could be identified in the embryos injected at stage 10 (data not shown, less severe defects are seen in Fig. 5.4J,K,L).  Despite the severe impact the virus has on  chondrogenesis a recognizable upper and lower beak still form (Fig. 5.3K,L).  Both ca viruses lead to a substantial increase in cartilage. Because of this similarity, only embryos infected with ca-BMPR-IB were further examined. Irregular shaped cartilages was the result of expansion in the cartilage elements and formation of ectopic condensations; the onsets of which are relatively rapid. By the time cartilage had differentiated (3-4 days after injection), the virus had disrupted the smooth outline of the cartilage (Fig. 5.5E). There are numerous closely packed chondrocytes in a disorganized matrix (Fig. 5.5F). There are also many extra condensations that coalesce to form large irregular masses of cartilage (Fig. 5.5E). The locations of the disruption in cartilage are closely correlated to the spread of the virus (Fig. 5.5D).  132  Constitutively active BMPR-IB leads to ectopic expression of Bmp-4, Bmp-7, Fgfr-2, Msx-1 and Msx-2 In order to understand the molecular basis by which B M P signaling controls facial skeletogenesis we examined the expression of several genes associated with cartilage formation. Surprisingly, in dn-BMPR-IB infected embryos we did not observe any expression changes in the genes examined (data not shown). Because of the similarities in phenotypes between the two ca-BMP type I receptors we performed gene expression studies only in ca-BMPR-IB injected embryos.  Muscle specific homeobox containing genes (Msx-1 and Msx-2) are expressed in the facial prominences in discrete patterns from stages 24-28, prior to fusion of the facial prominences (Nishikawa et al., 1994; Mina et al., 1995; Barlow and Francis-West, 1997). After fusion occurs, at stage 29, Msx-1 is expressed around the entire nasal passage (in the lateral edges of thefrontonasalmass and nearly the entire the lateral nasal prominences), medial edges of the maxillary prominences and cranial mandibular regions (Fig. 5.6G). At stage 29, Msx-2 is expressed below the nasal passage (lateral corners of thefrontonasalmass and medial region of the lateral nasal prominence) and in a small patch of mesenchyme just below the budding palatal shelves (Fig. 5.6H). Neither of the Msx genes are normally expressed in the center of thefrontonasalmass. In embryos infected with caBMPR-IB, Msxl and Msx-2 transcripts are ectopically expressed in the center of the frontonasal mass (Fig. 5.6E,F).  This  expression closely overlaps regions where ca-BMPR-IB is localized. Therefore, Msx-1 and Msx-2 are downstream of BMPR-IB.  133  Figure 5.6 Expression of genes in adjacent frontal sections of a stage 29 embryo injected with caBMPR-IB at stage 14. The expression in the mandibular prominence is normal for all genes but virus has changed gene expression in the frontonasal mass and maxillary prominences on the right side of the embryo. A , high levels of virus expression are seen in the right frontonasal mass and right superior maxillary prominence (red arrows). There is very little viral spread to the mandibular prominence. B, the expression of BmprIB is similar to the pol gene and no endogenous BmpRIB expression is visible yet in mandible. C, Bmp-4 is not normally expressed at high levels in the stage 29 facial mesenchyme but is considerably increased in the frontonasal mass (red arrow). D, Bmp-7 is normally expressed along the caudal edge of the frontonsasal mass (white arrows) but in this specimen there is additional expression in the center of the prominence (red arrows). E , expression of Msx-1 extends medially in the frontonasal mass and is spread in patches throughout the maxillary prominence on the right side (red arrows). F, expression of Msx-2 in the right maxillary prominence is present in small, disconnected areas of mesenchyme and there is an ectopic expression domain in the center of the frontonasal mass (red arrows). G, endogenous expression of Msx-1 at stage 29 with high expression in medial maxillary prominences, lateral nasal prominences and lateral edges of the frontonasal mass. H , endogenous expressions of Msx-2 in a discrete medial patch in the maxillary prominences and lateral edges of frontonasal mass (white arrows). I, Fgfr-2 is expressed in Meckel's cartilage condensations (white arrows). In the frontonasal mass expression in the midline cartilage condensation does not have well-defined borders. In the right maxillary prominence there is an expansion of the lateral expression into medial regions (red arrow). J, toluidine blue stained section from same specimen as in A-F,I. Scale bar = 200 pm. Key:  F N M - frontonasal mass; L N P - lateral nasal prominence; M d - mandibular prominence; Mx - maxillary prominence.  134  134a  In order to determine whether there is a feedback loop between BMPR-IB signaling and BMP ligands we examined the expression of Bmp-4 and Bmp-7. Both ligands are normally expressed along the caudal edge of the frontonasal mass mesenchyme (along with Bmp-2) but not in the central frontonasal mass (see chapter 2). The ca-BMPR-IB virus induces an ectopic region of Bmp-4 and Bmp-7 expression in the center of the frontonasal mass at stages 28 (data not shown) and 29 (Fig. 5.6C,D). The regions of ectopic expression did not overlap with the location of virus infection or regions of ectopic Msx-1 and Msx-2 expression, but rather was expressed in adjacent tissue where little (or no) infection was found.  We next examined the expression of Fgfr-2, a gene that is also expressed in facial mesenchyme and at high levels in stage 28 condensing cartilage (Wilke et al., 1997). We find that at stage 29, Fgfr2 continues to be expressed in chondrogenic regions in the centers of the frontonasal mass and paired-mandibular prominences.  In ca-BMPR-IB infected  frontonasal mass, the edge of Fgfr2 expression is very irregular and extends outside the chondrogenic region (3/3, Fig. 5.61).  CaBMPR-IA  and caBMPR-IB activity suppresses feather and egg tooth formation  Infection of the craniofacial region with caBMPR-IA or caBMPR-IB resulted in retardation of feather formation (Fig. 5.3G,H,I,J) or complete elimination of the feathers covering the head (Fig. 5.3K,L). Embryos with nearly complete absence of feathers generally had enlarged heads,  widespread changes in cartilage morphology and white granular  keritinizations on the surface of the skin (Fig. 5.3K,L; Table 5.3). The egg tooth is also an  135  ectodermal specialization and is reduced in size or absent in 50% of embryos with no cranial feathers (Fig. 5.3K,L; Table 5.3).  136  Discussion  Overlapping and unique function of type IA and type IB receptors in regulating the s of cartilage Expression of dnBMPRJA had no effect on cartilage differentiation whereas dnBMPRIB decreased the size of cartilage condensations. The dn type I receptor constructs are not designed to inhibit endogenous B M P signaling from a particular receptor type. Rather, the mutation is designed to inhibit all B M P signaling. Following infection, the mutation in the A T P binding domain still allows dimerization with the type II receptor and ligand binding. This abundant mutant complex would compete for available B M P ligands and subsequently inhibit BMP signaling. Therefore, in theory, similar defects should be observed with each dn receptor type but defects were only observed in dn-BMPR-IB infected embryos. I attribute this to the BMPR-IB receptor having greater receptor affinity for BMP-4, BMP-7 (ten Dijke et al., 1994; Nohno et al., 1995; Rosenzweig et al., 1995; and GDF-5, Nishitoh et al., 1996) than BMPR-IA.  No phenotype was also observed when dnBMPRIA was injected into the chicken limb bud (Kawakami et al., 1996; Zou et al., 1997) and in cultures of sternal cartilage (Enomoto-Iwamoto et al., 1998). A different dn-BMPR-IA construct, which lacks the entire cytoplasmic Ser/Thr kinase domain, inhibited interdigital BMP-induced apoptosis but only partially inhibited digit formation (Yokouchi et al., 1996). This construct appears to be a better inhibitor (than that used by us and Kawakami et al., 1996; Zou et al., 1997). Perhaps in the absence of the large cytoplasmic domain, dn-BMPR-IA can dimerize with BMPR-II receptors more easily and therefore partially inhibit BMP  137  signaling. Differences in ligand affinity between BMPR-IA and BMPR-IB may explain why this construct is able to inhibit apoptosis but not chondrogenesis. BMPR-IB is expressed in the distal phalages (Zou et al., 1997) and regulates chondrogenesis in this region (Baur et al., 2000; Y i et al., 2000). However, BMPR-IB is neither expressed in the interdigital mesenchyme (Zou et al., 1997) nor does it regulate apoptosis (Baur et al., 2000; Y i et al., 2000). Dn-BMPR-IB must be able to better compete for B M P ligands with whichever B M P receptor mediates  BMP-induced  apoptosis,  which awaits  identification. Now that the entire human genome has been sequenced we can search the database to determine how many BMP type I related receptors exist. A new B M P type I receptor was recently identified in zebrafish (Alk8; Bauer et al., 2001; Mintzer et al., 2001).  Under the conditions used here we do not entirely prevent the early stages of chondrogenesis with dnBMPRIB as has been seen in vitro (Kawakami et al., 1996; Zou et al., 1997). The difference may be that we infect only a proportion of the prechondrogenic cells allowing a smaller condensation to form. In some locations we observed a gap in the cartilage (Fig. 5.4D) suggesting that cells in this region were unable to organize into a condensation and thus failed to differentiate. The retroviral system causes very high expression of dnBMPRIB.  Despite this, the ectopic dn-BMPR-IB was unable to  completely inhibit chondrogenesis as was observed with overexpression of Noggin (Capdevilla and Johnson, 1998). These differences suggest that the endogenous type IB receptor is very abundant within cartilage condensations. The BMPRIB knockouts indicate that in vivo the role of the BMPRIB is much more limited than our dominant  138  negative virus experiments would suggest (Baur et al., 2000; Y i et al., 2000). Almost all the embryo develops normally with the exception of the absence of metacarpals and metatarsals.  Ca-type I receptors have been used to identify similarities and differences between BMPRIA and BMPR-IB signaling. One difference that is known is that ca-BMPR-IB can induce apoptosis while ca-BMPR-IA suppresses cartilage differentiation (Zou et al., 1997).  A  function that my results suggest is that both receptors share the ability to promote neural crest derived cartilage formation. This similarity may either be due to the receptors sharing similar functions or due to the induction of the other BMP receptor (Zou et al., 1997). Our studies may better explain why the type I receptors have similar functions. We showed that ca-BMPR-IB induces Bmp-4 and Bmp-7 expression. The elevated BMPs would then be able to  induce  localized BMP activity.  Others have  shown  that retrovirus-mediated  overexpression of BMP proteins (Duprez et al., 1996a) or exogenous BMP-4 (Buckland et al., 1998) and BMP-7 (Macias et al., 1997) applied to the early cartilage rudiments greatly increases the size of cartilage in the limb. The interesting feature of our finding is that the BMPs were upregulated in adjacent regions of low or no infection suggesting that the feedback loop involves neighbouring, and not infected cells. Thus, cells stimulated by BMPs (or activated receptor) stimulate neighboring cells to produce Bmp-4 and Bmp-7 in a paracrine mechanism. The upregulation of Bmp-4and Bmp-7 may be part of the mechanism by which chondrogenesis is promoted in embryos infected with constitutively active BMP receptors. Since the ectopic Bmp-4 and Bmp-7 are able to activate either of the B M P receptors this would prevent us from being able to identify specific signaling functions using  139  the ca-BMPR-IB receptor mutant. It is not known if ca-BMPR-IA also increases the expression of BMP ligands  We show that ca-BMPR-IB induces cartilage formation in ectopic regions. A possible mechanism to explain this observation may be ca-BMPR-IB increases cartilage formation through recruitment of non-chondrogenic precursors. Similar results were reported in the limb (Zou et al., 1997). These limb studies also showed that caBMPRIB virus increases the proliferation of chondrocytes. We did not examine proliferation in the facial cartilage but increased proliferation of chondrocytes may further explain why, in addition to ectopic cartilage, the individual cartilage elements were also much larger then usual.  The  expression of BmpR-IB in normally condensing and differentiating cartilage (chapter 2) suggests that the receptor plays a role in the proliferation of chondrocytes.  Msx genes are down stream of the BMPR-IB receptor The Msx-1 and Msx-2 genes are homeobox containing transcription regulators involved in cell proliferation (Odelberg et al., 2000; Chen et a l , 2001) and apoptosis (Ferrari et al., 1998), respectively. Both Msx genes are upregulated by BMPs. We are the first to show a relationship between BMP receptor signaling and Msx-1, Msx-2 gene expression. We identified that both Msx genes are downstream of BMPR-IB signaling. It is not known whether BMPR-IA can also regulate the expression of one or both Msx genes. Ectopic application of BMPs in two separate regions of the face induces both Msx genes (Barlow and Francis-West, 1997). Midline regions express BMPR-IA but not BMPR-IB while lateral regions express both BMPR-IA and BMPR-IB. Thus both type I receptor types can regulate  140  Msx gene expression. Examination of Msx-1 and Msx-2 gene expression following caBMPR-IA infection may help to determine whether Msx genes are downstream of both type I BMP receptors or only downstream of BMPR-IB.  Msx-1, Msx-2 and Fgfr2 are associated with chondrogenesis Msx-1 overexpression is associated with cellular proliferation (Chen et al., 2001) and dedifferentiation (Odelberg et al., 2000). Msx-1 is expressed at the edges of cartilage condensations and cartilage elements (Mina Mina et al., 1995) suggesting that Msx-1 may be involved in the proliferation of chondrocytes.  The ectopic overexpression of  Msx-1 has not been used to examine the effect of elevated Msx-1 protein in cartilage morphogenesis.  However, the MSX-1 knockout mouse has reduced size of cartilage  elements (Satokata and Maas, 1994). BMPs therefore may partially regulate the size of cartilage elements through the Msx-1 gene. In my study, ectopic Msx-1 expression may be associated with increased cartilage formation by stimulating proliferation.  Msx-2 associated with cell death in early development. Overexpression of Msx-2 stimulates ectopic apoptosis in the neural crest cells (Takahashi et al., 1998) and limb bud (Ferrari et al., 1998). Msx-2 knockout mice also have abnormal cartilage and endochondral bone formation. Msx-2 mutants have fewer resting, proliferative and hypertrophic chondrocytes (Satokata etal., 2000) suggesting that Msx-2 is required for chondrocyte proliferation and formation. Thus, the increased Msx-2 expression that we observed following ca-BMPR-IB infection may promote the formation of chondrocytes. Msx-2 has very different functions  141  during skeletogenesis than in early development. A proliferative role for Msx-2 is also observed in the calvariae of transgenic mice that overexpress Msx-2 (Liu et al., 1999).  I also detected an increase in Fgfr2 expression in response to increased BMPR-IB activity. Several lines of evidence suggest that both gain-of-function (reviewed in Webster and Donoghue, 1998) and loss-of-function mutations in FGF signaling (Celli et al., 1998) cause skeletal malformations in the skull and limb. The skeletal alterations observed in these embryos may be secondary to defects earlier in development.  Examining the effects of  FGFs directly on cartilage or chondrocytes would better address FGF function. In vitro, FGF-2 increased cartilage formation from facial mesenchyme (Richman and Crosby, 1990 and stimulated greater proliferation of porcine ear chondrocytes (Arevalo-Silva et al., 2001). In mouse, FGF4 can induce digit bifurcation (Ngo-Muller and Muneoka et al., 2000). Thus, ca-BMPR-IB infected tissue may stimulate cartilage formation through the upregulation of Fgfr2, and increased FGF signaling.  Increased BMP signaling inhibits formation of ectodermal specializations The ectopic expression of caBMPR-IA or IB reduces or eliminates feather formation and instead causes random keratin deposites in the epithelium. A similar result was obtained when BMP-2 was"applied to the healing epidermis of a lamb fetus. A scar "formed due to increased keratinization (Stelnicki et al., 1998). In an avian model, the application of BMP-4 suppresses differentiation of feather buds (Noramly and Morgan, 1998; Jung et al., 1998). Our data confirms that both BMP type I receptors are able to mediate this response. Overexpression of Noggin however results in abnormally large feather buds due to fusion of  142  adjacent placodes (Noramly and Morgan, 1998). We did not observe any change in feather morphology using our dn-BMPR-IB construct, which may not be able to sufficiently suppress BMP activity. The notch receptor and its ligands, delta and serrate are expressed in a dynamic fashion during patterning of the feather array and differentiation of the feather buds (Crowe et al., 1998). Misexpression of Delta-1 leads to a loss of feathers in virally infected skin and, contrary to what we would expect, inhibition of Bmp-2, Bmp-4 and Bmp-7 expression (Crowe et al., 1998). The loss of BMP ligand expression in feather buds is likely secondary to the absence of bud formation. Since similar phenotypes are produced with caBMPR-I and Delta-1 viruses it is possible that the two molecular pathways share common downstream signaling events.  Similar to feather formation, the BMPs are expressed in a dynamic pattern during egg tooth formation (see chapter 2). The egg tooth is a keratinization at the tip of the upper beak that the chick uses to break through the eggshell upon hatching. The pattern of BMP expression suggests that BMP signaling is involved in egg tooth morphogenesis. Since elevated B M P activity increases keratinization (our findings, Fig. 5.3K,L and Stelnicki et al., 1998) we did not anticipate that increased B M P signaling would inhibit egg tooth formation. However, the effects of BMPs  are highly dependent  on  concentration. For example, BMPs regulate outgrowth in the limb (Capdevilla and Johnson, 1998) and proliferation in face (my findings, chapter 4) mesenchyme but at high concentrations they induce cell death (Buckland et al., 1998; Ekanayake and Hall, 1997; Barlow and Francis-West, 1997; my findings, chapter 4). Inhibiting endogenous BMPs will determine whether they are indeed involved in morphogenesis of the egg tooth. We  143  did observe defects in egg tooth dn-BMPR-IB infected embryos (Fig. 5.3E).  This  observation was missed in preliminary analysis of our data because the defect was mild. A stronger inhibitor, like Noggin, may be more informative and generate a more obvious defect.  Evolutionary implications The BMPs may be the most highly conserved family of proteins in evolution with homologs that predate the evolution emergence of bone and cartilage.  After the  divergence of vertebrates and invertebrates the BMPs may have assumed the role for normal formation of skeletal elements as homologous genes are found in all mammals. Although some variation exists, the chondrocranial pattern is essentially conserved across the vertebrate taxa (de Beer, 1937). Specification of the primary form of the facial bony skeleton occurs epigenetically (Bosma 1976). Therefore, it has been suggested that much of the phylogenetic diversity of skull form can be accounted for by minor quantitative heterochronic shifts in the rate and timing of developmental events that regulate chondrocranial form (Alberch and Alberch, 1981; Wake 1980). The BMPs may be the genetic control resulting in phylogenetic variations of chondrocranial morphogenesis between and among species.  BMP mediated morphological changes have been reported that result in features resembling other animals or related species (Zou and Niswander, 1996; King et al., 1994). In our caBMP receptor injected embryos, two features are particularly striking. First, instead of the normally occurring long and narrow beak found in chicks, our  144  (bilaterally infected) caBMP receptor injected embryos had a much wider, thicker, rounded beak. This resulting beak displayed features much more characteristic of other birds such as parrots or budgies. Birds have wide variations in beak shape and this may be regulated by BMPs. Secondly, it was reported that infection of chick limbs with dnBMPR-IB resulted in a partial conversion of scales to feathers on the dorsal surface of the foot (Zho and Niswander, 1996). It appears that BMPs regulate a binary decision between the choice of scales or feathers. Our embryos with widespread caBMP receptor infection had complete absence of feathers and instead have patches of keratin. Perhaps other birds that lack feathers on their head and instead display keratinizations patches on the surface of their skin, such as turkeys, have elevated BMP signaling in the head.  Other intriguing genetic changes that result in morphological features resembling other animals or species have also been reported to be associated with BMPs and are worth mentioning.  Zou and Niswander (1996) found that infection of chick limbs with  dnBMPR-IB inhibited interdigital apoptosis resulting in a webbed chick foot resembling that seen in ducks. However, Bmp-2, Bmp-4 and Bmp-7 were all detected in the duck interdigit (Laufer et al., 1997). The interdigit region of ducks may not express the BMP receptor or downstream gene involved in BMP-induced apoptosis.  Moreover, in the  limb, BMPs regulate digit number (Drossopoulou et al., 2000) and digit patterning (Dahn and Fallon, 2000), therefore they could potentially regulate differences in digit formation among distinct species. Finally, compared to normal mice the se mouse (which has a mutation in the BMP5 gene) has a shorter nose, wider skull and shorter ears, resembling voles rather than house mice (King et al., 1994). Very short external ears are the most  145  characteristic feature of the se mouse. Could BMP-5 regulate morphological differences in ear size observed between chickens, mice, and rabbits? In situ hybridization studies determined that Bmp-5 is the only B M P gene that is expressed at high levels in the condensations of the developing ear (King et al., 1996).  Further studies should help  clarify the importance of BMP-5 in regulating region-specific morphological features in higher animals.  B M P signaling has been identified as far back as in c. elegans and  Drosophila, and thus the alteration in B M P function and activity may be a mechanism employed during evolutionary progression over time.  We conclude from these experiments that the embryo can use either the type IA or type IB receptor for chondrogenesis and formation of ectodermal specializations. In vivo however, there may be a preference for using one receptor over another in certain locations and at certain stages of development. BMP signaling is very important during embryogenesis; therefore it is necessary to build some redundancy into the system. We have demonstrated several overlapping functions of the type I receptors.  C o n t r i b u t i o n s to this c h a p t e r  The preceding chapter represents a portion of a manuscript that has been submitted to the International Journal of Developmental Biology.  The scientific ideas initiating these  studies were Dr. Richman's and I completed the study. Most of the introduction for this chapter was taken from our manuscript, which Dr. Richman wrote. I wrote the rest of the chapter. The majority of the work presented in this chapter, data analysis and  146  intrepretations were done by me. I received technical assistance from Sandra Babich in growing the virus and Steve Ritchie in assembling some of the figures.  147  CHAPTER VI General Discussion and Future Directions In this thesis I set forth to identify candidate genes associated with cleft lip and to test the function of these candidate genes in normal and abnormal facial development.  These  studies were undertaken at the organ (bead implantation experiments, viral expression experiments), tissue (epithelial stripping experiment) and molecular levels (changes in expression following bead implantation). First, the expression of B M P signaling molecules was described during most of the critical stages of craniofacial development in order to identify the locations where endogenous B M P signaling occurs. Second, functional experiments were performed that identified roles for BMP signaling in primary palate formation and the differentiation of cartilage, bone, feathers and egg teeth.  The expression studies in Chapter III are entirely descriptive. The presence of mRNA transcripts does not necessarily mean that protein is made - gene expression does not translate into gene function or bioactivity.  Despite this, expression studies are an  absolutely essential preliminary step in identifying where BMP signaling may potentially occur. The genes examined here are those protein products found extracellularly (BMP ligands, secreted antagonist) or at the cell surface (BMP receptor). Although my studies were quite extensive, there are numerous other BMP ligands, antagonists and receptors. All the main BMP receptors were examined with the caveat that there are two isoforms of the type IB receptor.  Mice were identified to have two BMPR-IB mRNA isoforms  derived from different promoters (Baur et al., 2000). It is not known if this applies to chickens, but my study would have detected all isoforms since the entire gene was used  148  as a probe. There may be other B M P receptors. For example, a new BMP type I receptor (Alk8) was recently identified in zebrafish (Bauer et al., 2001; Mintzer et al., 2001) and a pseudoreceptor, BAMBI, (BMP and activin membrane-bound inhibitor) was identified in Xenopus (Onichtchouk et al., 1999) and mice (Grotewold et al., 2001), which inhibits B M P signaling by binding to B M P type I receptors. Since the whole human genome is now sequenced we can search this database for other possible B M P type I receptors, as was done for F G F receptors (J. Rossant, personal communication). There are also numerous BMP antagonists (described in Chapter 2) that might modulate B M P signaling in the face. In addition to what I reported in Chapter 2, the expression of chordin was also examined. Chordin is ubiquitously expressed in the all head regions similar to what was reported in the mouse (Scott et al., 2001). I also found that the expression of tolloid (Reynolds et al., 2000), a metalloproteinase that cleaves Chordin, is highly expressed in the face mesenchyme but was absent in the neuroepithelium.  I did not examine the  numerous intracellular BMP signaling molecules, Smads for example, which remain to be investigated in the face. Smad 1 and Smad 5 are highly expressed in the mouse facial mesenchyme (Dick et al., 1998; Flanders et al., 2001).  Our expression studies were examined at numerous phases of facial morphogenesis; neural crest migration, initiation of facial prominences, outgrowth of the facial prominences, fusion, differentiation and skeletogenesis.  This data was used as a  springboard for our functional studies, which concentrated on the later phases mentioned.  149  I examined the involvement of BMPs in primary palate closure using two methods: 1) retrovirus mediated overexpression of mutant BMP type I receptors, and 2) the ectopic application of recombinant BMP-2 and Noggin proteins.  The retrovirus mediated  approach resulted in less consistent clefts (Table 5.3). We attribute this to not being able to consistently infect the facial prominences in the region of fusion and perhaps differences in the extent of viral spread. Approximately 10-15% of embryos examined prior to fusion localized virus expression in the corners of the frontonasal mass and maxillary prominences, which is the percentage of embryos that exhibited clefts.  Clefts  were observed in embryos infected with dn-BMPR-IB and ca-BMPR-IB, but not with dnBMPR-IA or ca-BMPR-IA. I attribute clefts in dn-BMPR-IB infected embryos to the BMPR-IB receptor having greater receptor affinity than BMPR-IA (Nohno et al., 1995; Rosenzweig et al., 1995), and therefore dn-BMPR-IB is better able to compete for available ligands. Phenotypic differences were observed with the two ca type I receptors. Clefts were only observed with the ca-BMPR-IB virus, once again in approximately 1015% of infected embryos. A difference between the two ca type I receptors is that caBMPR-IB can stimulate apoptosis in infected cells (Zou et al., 1997). It is possible that ca BMPR-IB increased apoptosis therefore reducing outgrowth of the facial prominences leading to clefts. Despite the lack of control of viral infection and spread, these preliminary data supported a role for BMP signaling in primary palate closure.  I found the bead-mediated application of ectopic protein to be a better method of targeting the regions of fusion and subsequently our results were much more consistent. For example, Noggin protein was implanted into the frontonasal mass and maxillary  150  mesenchyme in regions where high B M P expression was identified. These regions also corresponded with areas of highest cell proliferation.  By applying ectopic Noggin I  identified that endogenous BMPs regulate proliferation of the facial mesenchyme. Similarly, by targeting Noggin to near the epithelial surface I found that BMPs regulate epithelial survival. Combined with my gene expression studies, this finding suggests that epithelial apoptosis, normally during fusion, occurs by the downregulation of noggin. Using this method I was also able to examine gene interactions downstream of BMPs in the face.  At the molecular level, BMPs interact with FGFs and SHH in many different organisms and in many regions of the developing embryo. In the avian embryo there are many examples of mutations that affect both limb and face development (Schneider et al., 1999). This may be unexpected since these regions of the body are structurally quite dissimilar, have separate evolutionary origins, and have different embryonic origins. I find that BMPs act upstream of and define the region of fgf-8 expression in the face, similar to what is known in the limb apical ectodermal ridge (Pizette and Niswander, 1999). I find that elevated BMP activity also suppresses Shh expression in the facial ectoderm. Application of ectopic BMP protein to the limb also suppresses Shh expression in the limb mesenchyme (Zhang et al., 2000). However, the face is different from the limb as Shh is expressed in the ectoderm rather than in the mesenchyme (Helms et al., 1997) and we showed that ectopic BMP-2 applied to the mesenchyme suppresses Shh expression in the ectoderm. I also showed that inhibiting endogenous B M P activity, with ectopic Noggin, does not alter Shh expression suggesting that BMPs are not required to  151  maintain Shh expression. Other similarities involve the Msx-1 and Msx-2 genes. Ectopic B M P application increases Msx-1 and Msx-2 expression in the face (Barlow and FrancisWest, 1997; Wang et al., 1999) and limb (Merino et al., 1999; Ganan et a l , 1996). Msx-1 and Msx-2 have also been shown to be downstream of endogenous B M P signaling in the limb (Pizette and Niswander, 1999; Msx-2 only, Zou and Niswander, 1996). I am the first to identify this endogenous relationship in the face.  There are some differences in the signaling pathways that operate in the face and limb. The expression of noggin in the frontonasal epithelium is a feature unique to the face. No B M P antagonist has yet been localized to the limb apical ectoderm ridge (AER) or any other domain of the limb epithelium. In the limb, ectopic SHH upregulates noggin (Pizette and Niswander, 1999) and gremlin (Zuniga et al., 2000) expression but this relationship is localized to the limb mesenchyme. In the face, Shh overlaps with noggin expression in the frontonasal epithelium. Although I did not study this relationship, my findings suggest that Noggin may act to prevent BMPs from suppressing Shh expression.  The genes upstream of B M P signaling in the face were not extensively examined. However, in the limb it has been shown that SHH (Laufer et al., 1994; Yang et al., 1997), FGF-4 (Laufer et al., 1994) and Noggin (Capdevila and Johnson, 1998) can upregulate BMP ligand expression. By applying exogenous SHH and FGF-8, or antagonists to these proteins, we can better understand the upstream genes that regulate B M P expression. In this manner we can sort out gene hierarchy in the face. SHH may act partially through BMPs since Hu and Helms, (1999) reported that ectopic SHH protein upregulates Bmp-2  152  expression.  If this is true the simultaneous application of SHH and Noggin would  suppress SHH-induced proliferation and ectopic beak formation.  I touch on upstream signaling using our mutant viral constructs and showed the B M P signaling upregulates its own, B M P ligand (Bmp-4, Bmp-7), expression (also seen in the tooth, Vainio et al., 1993). I also show that exogenous R A acts upstream of BMPs, and upregulates Bmp-2 and Bmp-4. R A induces apoptosis in the face (Shen et al., 1997; McGonnell et al., 1998) and based on this data, I hypothesized that BMPs mediate R A induced apoptosis. I proposed to simultaneously apply R A and Noggin to the face. If BMPs  mediate RA-induced apoptosis Noggin would inhibit the BMP-mediated  apoptosis, and perhaps even prevent the R A induced cleft. Unfortunately, this experiment was performed in the interdigital region of the limb and published in Nature Cell Biology (Rodriguez-Leon, 1999) before I could begin the experiments. Programmed cell death normally occurs between the digits, however, when RA is applied, apoptosis is increased and interdigital tissue regression is accelerated. R A upregulates Bmp-7 and Bmp-4 in the limb. While I did see an upregulation of Bmp-2 and Bmp-4 immediately following R A treatment, I did not detect any change in Bmp-7 expression in the face. Bmp-7 may not be downstream of R A in the face.  Upregulation of Bmp-2 and Bmp-4 is the earliest  reported change in gene expression following R A treament (Munoz-Sanjuan et al., 2001; Shen et al,, 1997; Helms et al., 1997; Brown et al., 1997) therefore, BMPs may be an important early response genes in RA-induced cleft lip. Instead of performing the Noggin/RA cleft rescue experiment, I decided to focus on examining the endogenous role of BMP signaling during craniofacial development.  153  Retrovirus mediated overexpression of mutant BMP receptors was also used in attempts to identify biological differences between the type IA and IB receptor signalling. The effects of altered B M P signaling was examined on facial skeletogenesis.  Although I  reported some interesting functions of ca-BMP receptors on cartilage formation, I did not observe any obvious differences between the two type I receptors.  Zou et al, (1997)  found distinct roles for these ca-BMP receptors in limb skeletogenesis. The BMPR-IB was shown to be involved in the initiation of cartilage formation, while BMPR-IA regulated chondrocyte differentiation. My virus work is very preliminary and requires additional investigation in order to decipher the functional differences between BMPR-IA and BMPR-IB in facial cartilage morphogenesis. A n unexplored area of investigation is the roles of BMP type I receptor signalling on intramembraneous bone formation. Since the maxillary skeletal derivatives form by intramembraneous bone formation, in contrast to the limb where endochondral bones form, there is potential to identify some novel functions for B M P signaling. In order to address this, I injected the virus constructs specifically into the maxillary prominence.  Whole heads and sections from infected  embryos were stained for bone. Dr. Richman analyzed this data, which was not included in this thesis but can be found in the manuscript described in Chapter V .  My studies raise many new questions about B M P function. There are numerous other regions in the face that express BMPs, suggesting potential B M P function, but were beyond the scope of this thesis. For example, prior to tongue papillae formation BMPs are expressed in the epithelium and mesenchyme.  Expression and functional analysis  154  right at the time of papillae initiation will help us to better understand the function of BMPs during papillae formation. The restricted expression of Bmp-7 and BmpR-IB in the olfactory epithelium, and chordin in the respiratory epithelium suggests a possible role for BMPs in specifying the nasal epithelium.  B M P signaling molecules are also  expressed in the pharyngeal membrane, prior to its breakdown, during nasal pit epithelium differentiation and egg tooth morphogenesis. Little or nothing is known about the involvement of B M P signaling in the development of these structures.  Our  expression studies provided the basis from which numerous hypotheses can be formed. In addition, my differential display work suggests that there are numerous genes that may mediate R A teratogenicity in the face.  Screening a cDNA library with the clones I  isolated may identify novel genes in facial morphogenesis. My protein application work can also be further examined.  In conclusion my thesis work has identified many new functions for BMPs in craniofacial morphogenesis. The most significant contribution to science this thesis offers is the work involving avian primary palate formation. These findings raise many important questions. The most significant being: are B M P signaling molecules associated with human clefting?  B M P may have a central role in human primary palate closure as my data  identifies interactions with molecules upstream (RA, Lammer et al., 1985; R A R a , Chenevix-Trench et al., 1992) and downstream (Msx-1; van den Boogaard et al., 2000) of BMPs whom have been associated with human dysmorphogenesis.  Since I showed that  both gain-of-function and loss-of-function experiments induce cleft lip, numerous BMP  155  signaling molecules need to be added to the list of candidate genes that may cause human, non-syndromic orofacial clefting.  156  LITERATURE CITED Ahlgren SC and Bronner-Fraser M .  (1999). Inhibition of sonic hedgehog signaling in  vivo results in craniofacial neural crest cell death. Curr Biol 9:1304-14.  Aiello LP, Robinson GS, Lin Y W , Nishio Y, King G L . (1994).  Identification of  multiple genes in bovine retinal pericytes altered by exposure to elevated levels of glucose by using mRNA differential display. Proc Natl Acad Sci USA 91:6231-5.  Ardinger H H , Buetow K H , Bell GI, Bardach J, VanDemark DR, Murray JC. (1989). Association of genetic variation of the transforming growth factor-alpha gene with cleft lip and palate. Am J Hum Genet 45:348-53.  Ashique A M , Fu K, Richman JM. Developmental Dynamics. Submitted.  Attisano L and Wrana JL. (2000). Smads as transcriptional co-modulators. Curr Opin Cell Biol 12:235-43.  Bachiller D, Klingensmith J, Kemp C, Belo JA, Anderson R M , May SR, McMahon JA, McMahon AP, Harland R M , Rossant J, De Robertis E M . (2000). The organizer factors Chordin and Noggin are required for mouse forebrain-development. Nature 403:658-61.  Bachler M and Neubuser A. (2001). Expression of members of the Fgf family and their receptors during midfacial development. Mech Dev 100:313-6.  157  Bardot B, Lecoin L , Huillard E , Calothy G, Marx M . (2001). Expression pattern of the drm/gremlin gene during chicken embryonic development. Mech Dev 101:263-5.  Barlow A J and Francis-West PH. (1997). Ectopic application of recombinant BMP-2 and BMP-4 can change patterning of developing chick facial primordia. Development 124:391-8.  Barlow AJ, Bogardi JP, Ladher R, Francis-West PH. (1999). Expression of chick Barx-1 and its differential regulation by FGF-8 and B M P signaling in the maxillary primordia. Dev Dyn 214:291-302.  Bassett D E Jr, Eisen M B , Boguski MS. (1999). Gene expression informatics—it's all in your mine. Nat Genet 21:51-5.  Baur ST, Mai JJ, Dymecki SM. (2000). Combinatorial signaling through B M P receptor IB and GDF5: shaping of the distal mouse limb and the genetics of distal limb diversity. Development 127:605-619.  Bennett JHj Hunt P, Thorogood P.  (1995).  Bone morphogenetic protein-2 and -4  expression during murine orofacial development. Archs. Oral Biol 40:847-854.  158  Beppu H , Kawabata M , Hamamoto T, Chytil A , Minowa O, Noda T, Miyazono K. (2000).  B M P type II receptor is required for gastrulation and early development of  mouse embryos. Dev Biol 221:249-58.  Berggren K , McCaffery P, Drager U , Forehand CJ. (1999). Differential distribution of retinoic acid synthesis in the chicken embryo as determined by immunolocalization of the retinoic acid synthetic enzyme, RALDH-2. Dev Biol 210:288-304.  Blessing M , Nanney L B , King L E , Jones C M , Hogan B L . (1993). Transgenic mice as a model to study the role of TGF-beta-related molecules in hair follicles.  Genes Dev  7:204.  Botchkarev V A , Botchkareva N V , Roth W, Nakamura M , Chen L H , Herzog W, Lindner G, McMahon JA, Peters C, Lauster R, McMahon AP, Paus R.  (1999). Noggin is a  mesenchymally derived stimulator of hair-follicle induction. Nat Cell Biol 1:158-64.  Bouwmeester T, Kim S, Sasai Y , L u B, De Robertis E M .  (1996). Cerberus is a head-  inducing secreted factor expressed in the anterior endoderm of Spemann's organizer. Nature 382:595-601.  Briscoe J and Ericson J. (1999). The specification of neuronal identity by graded Sonic Hedgehog signalling. Semin Cell Dev Biol 10:353-62.  159  Britto JM, Tannahill D, Keynes RJ. (2000). Life, death and sonic hedgehog. Bioessays 22:499-502.  Brown JM, Robertson K E , Wedden SE, 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 195:203-7.  Brummel TJ, Twombly V , Marques G, Wrana JL, Newfeld SJ, Attisano L , Massague J, O'Connor M B , Gelbart W M . (1994). Characterization and relationship of Dpp receptors encoded by the saxophone and thick veins genes in Drosophila. Cell 78:251 -61.  Brunet L J , McMahon JA, McMahon AP, Harland R M . (1998).  Noggin, cartilage  morphogenesis, and joint formation in the mammalian skeleton. Science 280:1455-7.  Buckland RA, Collinson JM, Graham E , Davidson DR, Hill RE. (1998). Antagonistic effects of FGF-4 on BMP induction of apoptosis and chondrogenesis in the chick limb bud. MechDev 71:143-150.  Capdevila J and Johnson RL.  (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.  160  Capdevila J, Tsukui T, Rodriquez Esteban C, Zappavigna V , Izpisua Belmonte JC. (1999). Control of vertebrate limb outgrowth by the proximal factor Meisl and distal antagonism of BMPs by Gremlin. Mol Cell 4:839-849.  Carette MJ, Ferguson M W . (1992). The fate of medial edge epithelial cells during palatal fusion in vitro: an analysis by Dil labelling and confocal microscopy. Development 114:379-88.  Celeste A J , Iannazzi JA, Taylor RC, Hewick R M , Rosen V , Wang E A , Wozney J M . (1990).  Identification of transforming growth factor beta family members present in  bone-inductive protein purified from bovine bone. Proc Natl Acad Sci USA 87:9843-7.  Chang W, Nunes FD, De Jesus-Escobar JM, Harland R, Wu DK. (1999). Ectopic noggin blocks sensory  and nonsensory  organ morphogenesis  in the chicken inner ear.  Developmental Biology 216:369-381.  Chen YP, Zhang Z, Song Y , Zhao X , Zhang X . (2001). IADR conference. Abstract no. 01555.  Chenevix-Trench G , Jones K, Green A C , Duffy D L , Martin N G . (1992). Cleft lip with or without cleft palate: associations with transforming growth factor alpha and retinoic acid receptor loci. Am J Hum Genet 51:1377-85.  161  Cheung V G , Morley M , Aguilar F, Massimi A , Kucherlapati R, Childs G .  (1999).  Making and reading microarrays. Nat Genet 21:15-9.  Chimal-Monroy J and Diaz de Leon L . (1999).  Expression of N-cadherin, N - C A M ,  fibronectin and tenascin is stimulated by T G F - p i , p2, P3 and P5 during the formation of precartilage condensations. Int J Dev Biol 43:59-67.  Clarke PG. (1990). Developmental cell death: morphological diversity and multiple mechanisms. Anat Embryol 181:195-213.  Cohlan SQ. (1953). Excessive intake of vitamin A as a cause of congenital anomalies in the rat. Science 117:535-536.  Couly GF, Coltey PM, Le Douarin N M . (1992). The developmental fate of the cephalic mesoderm in quail-chick chimeras. Development 114:1-15.  Couly GF, Coltey PM, Le Douarin N M . (1993). The triple origin of skull in higher vertebrates: a study in quail-chick chimeras. Development 117:409-429.  _  Crossley PH, Minowada G, Mac Arthur C A , MartinGR. (1996). Roles for FGF8 in the induction, initiation, and maintenance of chick limb development. Cell 84:127.  162  Crossley P H and Martin GR.  (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.  Croucher SJ and Tickle C. (1989). Characterization of epithelial domains in the nasal passages of chick embryos: spatial and temporal mapping of a range of extracellular matrix and cell surface molecules during development of the nasal placode. Development 106:493-509.  Dahn RD and Fallon JF. (2000). Interdigital regulation of digit identity and homeotic transformation by modulated BMP signaling. Science 290:271-27'4.  Daopin S, Piez K A , Ogawa Y , Davies DR. (1992). Crystal structure of transforming growth factor-p2: an unusual fold for the superfamily. Science 257:369-73.  Davideau JL, Demri P, Gu T T , Simmons D, Nessman C, Forest N , MacDougall M , Berdal A . (1999). Expression of DLX5  during human embryonic craniofacial  development. Mech Dev 81:183-6.  Dealy C N , Roth A , Ferrari D, Brown A M , Kosher RA. (1993). Wnt-5a and Wnt-7a are expressed in the developing chick limb bud in a manner suggesting roles in pattern formation along the proximodistal and dorsoventral axes. Mech Dev 43:175-86.  163  Denker A E , Haas AR, Nicoll SB, Tuan RS. (1999). Chondrogenic differentiation of murine C3H10T1/2 multipotential mesenchymal cells: I. Stimulation by bone morphogenetic protein-2 in high-density micromass cultures. Differentiation 64:61-16.  Devlin CJ, Brickell PM, Taylor ER, Hornbruch A , Craig R K and Wolpert L. (1988). In situ hybridization reveals differential spatial distribution of mRNAs for type I and type II collagen in the chick limb bud. Development 103, 111-118.  Dewulf N , Verscheuren K, Lonnoy O, Moren A , Grimsby S, Vande Spiegle K , Miyazono K, Huylebroeck D, ten Dijke P. (1995). Distinct spatial and temporal expression patterns of two type I receptors for bone morphogenetic proteins during mouse embryogenesis. Endocrinology 136:2652-2663.  Diewert V M and Wang K Y .  (1992). Recent advances in primary palate and midface  morphogenesis research. Critical Reviews in Oral Biology & Medicine 4:111-30.  Dick A , Risau W, Drexler H .  (1998).  Expression of Smadl and Smad2 during  embryogenesis suggests a role in organ development. Dev Dyn 211:293-305.  Dixon MJ, Dixon J, Raskova D, Le Beau M M , Williamson R, Klinger K , Landes G M . (1992).  Genetic and physical mapping of the Treacher Collins syndrome locus:  refinement of the localization to chromosome 5q32-33.2. Hum Mol Genet 1:249-53.  164  Dolle P, Fraulob V , Kastner P, Chambon P. (1994) Developmental expression of murine retinoid X receptor (RXR) genes. Mech Dev 45:91-104.  Dolle P, Ruberte E , Kastner P, Petkovich M , Stoner C M , Gudas LJ, Chambon P. (1989). Differential expression of genes encoding alpha, beta and gamma retinoic acid receptors and C R A B P in the developing limbs of the mouse. Nature 342:702-5.  Dolle P, Ruberte E , Leroy P, Morriss-Kay G , Chambon P.  (1990). Retinoic acid  receptors and cellular retinoid binding proteins. I. A systematic study of their differential pattern of transcription during mouse organogenesis. Development 110:1133-51.  Donohue PJ, Alberts GF, Hampton BS, Winkles JA. (1994).  A delayed-early gene  activated by fibroblast growth factor-1 encodes a protein related to aldose reductase.  J  Biol Chem 269:8604-9.  Drossopoulou G , Lewis K E , Sanz-Ezquerro JJ, Nikbakht N , McMahon AP, Hofmann C, 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.  Dudley A T , Lyons K M , Robertson EJ. (1995). A requirement for bone morphogenetic protein-7 during development of the mammalian kidney and eye. Genes Dev 9:2795-807.  165  Duggan DJ, Bittner M , Chen Y , Meltzer P, Trent JM. (1999). Expression profiling using cDNA microarrays. Nat Genet 21:10-4.  Dunn NR, Winnier G E , Hargett L K , Schrick JJ, Fogo A B , Hogan B L M . (1997). Haploinsufficient  phenotypes in Bmp4 heterozygous null mice and modification by  mutations in Gli3 and Alx4. Dev Biol 188, 235-247.  Duprez D, Bell EJ, Richardson M K , Archer CW, Wolpert L, Brickell PM, Francis-West PH. (1996a). Overexpression of BMP-2 and BMP-4 alters the size and shape of developing skeletal elements in the chick limb. Mech Dev 57:145-157.  Duprez D M , Kostakopoulou K, Francis-West PH, Tickle C, Brickell P M . (1996b). Activation of Fgf-4 and HoxD gene expression by BMP-2 expressing cells in the developing chick limb. Development 122:1821-1828.  Ekanayake S and Hall B K . (1997). The in vivo and in vitro effects of Bone Morphogenetic Protein-2 on the development of the chick mandible. Int J Dev Biol 41:67-81.  Eppley B L and Sadove A M . (2000). Management of alveolar cleft bone grafting-state of the art. Cleft Palate Craniofac J37:229.  166  Erickson D M , Harris SE, Dean DD, Harris M A , Wozney JM, Boyan B D , Schwartz Z. (1997). Recombinant bone morphogenetic protein (BMP)-2 regulates costochondral growth plate chondrocytes and induces expression of BMP-2 and BMP-4 in a cell maturationdependent manner. J Orthop Res 15:371-380.  Esteban CR, Capdevila J, Economides A N , Pascual J, Ortiz A , Belmonte JCI. 1999. The novel Cer-like protein Caronte mediates the establishment of embryonic left-right asymmetry. Nature 401:243-251.  Estevez M , Attisano L, Wrana JL, Albert PS, Massague J, Riddle DL. (1993). The daf-4 gene encodes a bone morphogenetic protein receptor controlling C. elegans dauer larva development. Nature 365:644-9.  Fairlie PG, Kerr R, Thomas P, Symes T, Minichiello J, Hearn CJ, Newgreen D. 1999. A paraxial exclusion zone creates patterned cranial neural crest cell outgrowth adjacent to rhombomeres 3 and 5. Dev Biol 213:70-84.  Fang J and Hall BK.  (1997). Chondrogenic cell differentiation from membrane bone  periostea. Anat Embryol 196:349-62.  Farrall M and Holder S. (1992). Familial recurrence-pattern analysis of cleft lip with or without cleft palate. American Journal of Human Genetics 50:270-7.  167  Feijen A , Goumans MJ, van den Eijnden-van Raaij A J . (1994). Expression of activin subunits, activin receptors and follistatin in postimplantation mouse embryos suggests specific developmental functions for different activins. Development 120:3621-37.  Ferguson E L and Anderson K V . (1992). Decapentaplegic acts as a morphogen to organize dorsal-ventral pattern in the Drosophila embryo. Cell 71:451-61.  Fernhoff P M and Lammer EJ. (1984). Craniofacial features of isotretinoin embryopathy. Journal of Pediatrics 105:595-7.  Ferrari D, Lichtler A C , Pan ZZ, Dealy C N , Upholt WB, Kosher R A . (1998). Ectopic expression of Msx-2 in posterior limb bud mesoderm impairs limb morphogenesis while inducing BMP-4 expression, inhibiting cell proliferation, and promoting apoptosis. Dev Biol 197:12.  Fitchett JE and Hay ED. (1989). Medial edge epithelium transforms to mesenchyme after embryonic palatal shelves fuse. Dev Biol 131:455-74.  Flanders K C , Kim ES, Roberts A B . (2001). Immunohistochemical expression of Smads 1-6 in the 15-day gestation mouse embryo: signaling by BMPs and TGF-betas. Dev Dyn 220:141-54.  168  Francis-West P H , Tatla T, Brickell P. (1994). Expression patterns of the bone morphogenetic protein genes Bmp-4 and Bmp-2 in the developing chick face suggest a role in the outgrowth of the primordia. Dev Dyn 201:168-178.  Francis-West P, Ladher R, Barlow A , Graveson A. (1998). Signalling interactions during facial development. Mech Dev 75:3-28.  Ganan Y , Macias D, Duterque-Coquillaud M , Ros M A , Hurle JM. (1996). Role of TGFps and BMPs as signals controlling the position of the digits and the areas of interdigital cell death in the developing check limb autopod. Development 122:2349-2357.  Georgi L L , Albert PS, Riddle D L . (1990). daf-1, a C. elegans gene controlling dauer larva development, encodes a novel receptor protein kinase. Cell 61:635-45.  Gerlach L M , Hutson MR, Germiller JA, Nguyen-Luu D, Victor JC, Barald K F . (2000). Addition of the BMP4 antagonist, noggin, disrupts avian inner ear development. Development 127:45-54.  Giguere V , Ong ES, Segui P, Evans R M .  (1987). Identification of a receptor for the  morphogen retinoic acid. Nature 330:624-9.  169  Golden JA, Bracilovic A , McFadden K A , Beesley JS, R Rubenstein JL, Grinspan JB. (1999). Ectopic bone morphogenetic proteins 5 and 4 in the chicken forebrain lead to cyclopia and holoprosencephaly. Proc Natl Acad Sci U SA 96:2439-44.  Gong S-G, White NJ, Sakasegawa A Y . (2000). The Twirler mouse, a model for the study of cleft lip and palate. Arch Oral Biol 45:87.  Graham A , Francis-West P, Brickell P, Lumsden A . (1994). The signalling molecule BMP4 mediates apoptosis in the rhombencephalic neural crest. Nature 372:684-686.  Graham A , Francis-West P, Brickell P, Lumsden A . (1994). The signalling molecule BMP4 mediates apoptosis in the rhombencephalic neural crest. Nature 372:684.  Granan Y , Macias D, Duterque-Coquillaud M , Ros M A , Hurle M . (1996). Role of TGFfls and BMPs as signals controlling the position of the digits and the areas of interdigital cell death in the developing chick limb autopod. Development 122:2349-2357.  Griffith D L , Keck PC, Sampath T K , Rueger D C , Carlson WD.  (1996).  Three-  dimensional structure of recombinant human osteogenic protein 1: structural paradigm for the transforming growth factor beta superfamily, Proc Natl Acad Sci 93:878-83.  Grotewold L , Plum M , Dildrop R, Peters T, Ruther U . (2001). Bambi is coexpressed with Bmp-4 during mouse embryogenesis. Mech Dev 100:327-30.  170  Griineberg H . (1963). The pathology of development:  a study of inherited skeletal  disorders in animals. Blackwell, Oxford.  Gruneberg H . (1963).  The pathology of development.  A study of inherited skeletal  disorders in animals. Oxford: Blackwells Scientific Publications.  Hacker A and Guthrie SA. (1998). Distinct developmental programme for the cranial paraxial mesoderm in the chick embryo. Development 125:3461-3472.  Hacker A , Guthrie SA (1998). Distinct developmental programme for the cranial paraxial mesoderm in the chick embryo. Development 125:3461-3472.  Hall B K and Hdrstadius S. (1988). The neural crest. Oxford University Press, New York.  Hall B K and Miyake T.  (1992).  The membranous skeleton: the role of cell  condensations in vertebrate skeletogenesis. Anat Embryol 186:107-24.  Hall B K and Miyake T. (1995). Divide, accumulate, differentiate: cell condensation in skeletal development revisited. IntJDev Biol 39:881-93.  Hall B K and Miyake T. (2000).  A l l for one and one for all: condensations and the  initiation of skeletal development. Bioessays 22:138-47.  171  Hall BK, Horstadius S. (1988). The neural crest. Oxford University Press, New York.  Hall B K , Miyake T. (1992). The membranous skeleton: the role of cell condensations in vertebrate skeletogenesis. Anat Embryol 186:107-124.  Hamburger V , Hamilton H. (1951). A series of normal stages in the development of the chick embryo. JMorphol 88:49-92.  Hassell JR, Greenberg JH, Johnston M C . (1977). Inhibition of cranial neural crest cell development by vitamin A in the cultured chick embryo. Journal of Embryology & Experimental Morphology 39:267-71.  Hazama M , Aono A , Ueno N , Fujisawa Y. (1995). Efficient expression of a heterodimer of bone morphogenetic protein subunits using a baculovirus expression system. Biochem Biophys Res Commun 209:859-66.  Heikinheimo M , Lawshe A , Shackleford G M , Wilson D B , MacArthur CA. (1994). Fgf8 expression in the post-gastrulation mouse suggests roles in the development of the face, limbs and central nervous system. Mechanisms of Development 48:129-38.  Heldin C H , Miyazono K , ten Dijke P. (1997). TGFp\ signalling from cell membrane to nucleus through S M A D proteins. Nature 390:465-71.  172  Helms JA, Kim C H , Hu D, Minkoff R, Thaller C, Eichele G. (1997). Sonic hedgehog participates in craniofacial morphogenesis and is downregulated by teratogenic doses of retinoic acid. Dev Biol 187:25-35.  Hemmati-Brivanlou A , Kelly O G , Melton D A . (1994).  Follistatin, an antagonist of  activin, is expressed in the Spemann organizer and displays direct neuralizing activity. Cell 77:283-95.  Hogan B L .  (1996).  Bone morphogenetic proteins: multifunctional regulators of  vertebrate development. Genes & Development 10:1580-94.  Hoodless PA, Wrana JL. (1998). Mechanism and function of signaling by the T G F beta superfamily. Curr Top Microbiol Immunol 228:235-72.  Houston B, Thorp B H , Burt DW. (1994). Molecular cloning and expression of bone morphogenetic protein-7 in the chick epiphyseal growth plate. J Mol Endocrinol 13:289301.  Hsu DR, Economides A N , Wang X , Eimon P M , Harland R M .  (1998).  The Xenopus  dorsalizing factor Gremlin identifies a novel family of secreted proteins that antagonize BMP activities. Mol Cell 1:673-83.  173  Hu D and Helms JA. (1999).  The role of sonic hedgehog in normal and abnormal  craniofacial morphogenesis. Development 126:4873-84.  Iamaroon A and Diewert V M . Distribution of basement membrane components in the mouse primary palate. J Craniofac Genet Dev Biol 16:48-51.  Ichiro S and Maas R. (1994). Msxl deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development. Nat Genet 6:348-56.  Izpisua-Belmonte JC, De Robertis E M , Storey KO, Stern CD. (1993). The homeobox gene goosecoid and the origin of organizer cells in the early chick blastoderm. Cell 74:645-659.  Jelinek R and Kistler A.  (1981).  Effect of retinoic acid upon the chick embryonic  morphogenetic systems. I. The embryotoxicity dose range. Teratology 23:191-5.  Jernvall J, Aberg T, Kettunen P, Keranen S, Thesleff I. (1998). The life history of an embryonic signaling center: BMP-4 induces p21 and is associated with apoptosis in the mouse tooth enamel knot. Development 125:161-169.  Jones C M , Dale L , Hogan B L , Wright C V , Smith JC. (1996). Bone morphogenetic protein-4 (BMP-4) acts during gastrula stages to cause ventralization of Xenopus embryos. Development 122:1545-54.  174  Jones M C . (1988). Etiology of facial clefts: prospective evaluation of 428 patients. Cleft Palate J25:16-20.  Jung H-S, Francis-West PH, Widelitz RB, Jiang T-X, Ting-Berreth S, Tickle C, Wolpert L , Chuong C - M . (1998). Local inhibitory action of BMPs and their relationships with activators in feather formation: Implications for periodic patterning. Dev Biol 196:11 -23.  Jurecic R, Nachtman R G , Colicos SM, Belmont JW. (1998). Identification and cloning of  differentially  expressed  genes  by  long-distance  differential  display.  Anal Biochem 259:235-44.  Juriloff D M . (1995). Genetic analysis of the construction of the AEJ.A congenic strain indicates that nonsyndromic CL(P) in the mouse is caused by two loci with epistatic interaction. Journal of Craniofacial Genetics & Developmental Biology 15:1-12.  Juriloff D M , Harris MJ, Brown CJ. (2001). Unravelling the complex genetics of cleft lip in the mouse model. Mamm Genome 12:426-35.  Kaartinen V , Voncken JW, Shuler C, Warburton D, Bu D, Heisterkamp N , Groffen J. 1995. Abnormal lung development and cleft palate in mice lacking TGF-beta 3 indicates defects of epithelial-mesenchymal interaction. Nat Genet 11:415-421.  175  Kalter H . (1960).  The teratogenic effects of hypervitaminosis A upon the face and  mouth of inbred mice. Ann NY Acad Sci 85:42-55.  Kanzler B, Foreman RK, Labosky PA, Mallo M . (2000). BMP signaling is essential for development of skeletogenic and neurogenic cranial neural crest. Development 127:10951104.  Katagiri T, Boorla S, Frendo JL, Hogan B L , Karsenty G. (1998). Skeletal abnormalities in doubly heterozygous Bmp4 and Bmp7 mice. Dev Genet 22:340-8.  Kawakami Y , Ishikawa T, Shimabara M , Tanda N , Enomoto-Iwamoto M , Iwamoto M , Kuwana T, Ueki A , Noji S, Nohno T. (1996). BMP signaling during bone pattern determination in the developing limb. Development 122:3557-3566.  Kay RR and Smith JC.  (1989).  The molecular basis of positional signalling:  introduction. Development 107 Suppl:l-2.  Keynes RJ, Stern C D . (1988). Mechanisms of vertebrate segmentation. Development 103:413-29.  Kim H-J, Rice DPC, Kettunen PJ, Thesleff I. (1998). FGF-, BMP- and Shh-mediated signalling pathways in the regulation of cranial suture morphogenesis and calvarial bone development. Development 125:1241-1251.  176  King JA, Marker PC, Seung K J , Kingsley D M . (1994).  BMP5 and the molecular,  skeletal, and soft-tissue alterations in short ear mice. Dev Biol 166:112-22.  Kingsbury JW, Allen V G , Rotheram BA. (1953). Histological structure of the beak in the chick. Anat Rec 116:95-115.  Kingsley D M .  (1994). The TGF-beta superfamily: new members, new receptors, and  new genetic tests of function in different organisms. Genes & Development 8:133-46.  Koenig BB, Cook JS, Wolsing DH, Ting J, Tiesman JP, Correa PE, Olson CA, Pecquet A L , Ventura F, Grant R A , Chen G-X, Wrana JL, Massague J, Rosenbaum JS. 1994. Chracterization and cloning of a receptor for BMP-2 and BMP-4 from NIH 3T3 cells. Mol Cell Biol 14:5961-5974.  Korade-Mirnics Z, Burnside J, Corey S.  (2000).  Dna microarray analysis of g-csf  dependent, lyn-dependent genes. Exp Hematol 28:1497.  Kontges G and Lumsden A . (1996).  Rhombencephalic neural crest segmentation is  preserved throughout craniofacial ontogeny. Development 122:3229-42.  Kosher R A , Kulyk W M , Gay SW.  (1986).  Collagen gene expression during limb  cartilage differentiation. J Cell Biol 102:1151-6.  177  Kretzschmar M , Massague J. (1998). SMADs:  Mediators and regulators of T G F - a  signaling. Current Opinion in Genetics & Development 8:103-111.  LaMantia A , Bhasin N , Rhodes K , Heemskerk J. (2000). Mesenchymal/Epithelial induction mediates olfactory pathway formation. Neuron 28:411-425.  Ladher RK, Church V L , Allen S, Robson L , Abdelfattah A , Brown N A , Hattersley G, Rosen V , Luyten FP, Dale L , Francis-West PH. (2000). Cloning and expression of the Wnt antagonists Sfrp-2 and Frzb during chick development. Dev Biol 218:183-98.  Lamb T M , Knecht A K , Smith WC, Stachel SE, Economides A N , Stahl N , Yancopolous GD, Harland R M . (1993). Neural induction by the secreted polypeptide noggin. Science 262:713-8.  Lammer EJ, Chen DT, Hoar R M , Agnish ND, Benke PJ, Braun JT, Curry CJ, Fernhoff P M , Grix A W Jr, Lott IT, et al. (1985). Retinoic acid embryopathy.  N Engl J Med.  313:837-41.  Laufer E , Nelson C E , Johnson RL, Morgan B A , Tabin C. (1994). Sonic hedgehog and Fgf-4 act through a signaling cascade and feedback loop to integrate growth and patterning of the developing limb bud. Cell 79:993-1003.  178  Le Douarin N M . 1982. The Neural Crest. London: Cambridge University Press.  Le Lievre CS. (1978). Participation of neural crest-derived cells in the genesis of the skull in birds. J Embryol Exp Morphol 47:17-37.  Le Lievre CS. Le Lievre CS, Le Douarin N M . (1975). Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos.  J Embryol Exp Morphol  34:125-54.  Li E , Sucov H M , Lee K F , Evans R M , Jaenisch R. (1993). Normal development and growth of mice carrying a targeted disruption of the alpha 1 retinoic acid receptor gene.  Proc Natl Acad Sci US AWA 590-4.  Liang P and Pardee A B . (1992). Differential display of eukaryotic messenger R N A by means of the polymerase chain reaction. Science 257:967-71.  Liang P and Pardee A B .  (1998).  Differential display. A general protocol. Mol  Biotechnol 10:261-7.  Liang P, Averboukh L , Pardee A B . (1993).  Distribution and cloning of eukaryotic  mRNAs by means of differential display: refinements and optimization. Nucleic Acids Res 21:3269-7'5.  179  Lidral A C , Romitti PA, Basart A M , Doetschman T, Leysens NJ, Daack-Hirsch S, Semina E V , Johnson LR, Machida J, Burds A , Parnell TJ, Rubenstein JL, Murray JC. Association of MSX1  (1998).  and TGFB3 with nonsyndromic clefting in humans. Am J Hum  Genet 63:557-68.  Liem K F Jr, Tremrnl G, Jessell T M . (1997). A role for the roof plate and its resident TGFbeta-related proteins in neuronal patterning in the dorsal spinal cord. Cell 91:127138.  Liem K F Jr, Tremrnl G, Roelink H , Jessell T M . 1995. Dorsal differentiation of neural plate cells induced by BMP-mediated signals from epidermal ectoderm. Cell 82:969-979.  Liem K F , Jessell T M , Briscoe J. (2000). Regulation of the neural patterning activity of sonic hedgehog by secreted B M P inhibitors expressed by notochord and somites. Development  127:4855-4866.  Liu F, Ventura F, Doody J, Massague J. (1995).  Human type II receptor for bone  morphogenic proteins (BMPs): extension of the two-kinase receptor model to the BMPs. Mol  Cell Biol 15:3479-86.  Loftus SK, Chen Y, Gooden G, Ryan JF, Birznieks G, Hilliard M , Baxevanis A D , Bittner M , Meltzer P, Trent J, Pavan W. (1999).  Informatic selection of a neural crest-  melanocyte cDNA set for microarray analysis. Proc Natl Acad Sci USA 96:9277-80.  180  Lohnes D, Kastner P, Dierich A , Mark M , LeMeur M , Chambon P. (1993). Function of retinoic acid receptor gamma in the mouse. Cell 73:643-58.  Lohnes D, Mark M , Mendelsohn C, Dolle P, Dierich A , Gorry P, Gansmuller A , Chambon P. (1994). Function of the retinoic acid receptors (RARs) during development (I). Craniofacial and skeletal abnormalities in R A R double mutants. Development 120:2723-48.  Lufkin T, Lohnes D, Mark M , Dierich A , Gorry P, Gaub MP, LeMeur M , Chambon P. (1993). High postnatal lethality and testis degeneration in retinoic acid receptor alpha mutant mice. Proc Natl Acad Sci USA 90:7225-9.  Lumsden A , Sprawson N , Graham A. (1991). Segmental origin and migration of neural crest cells in the hindbrain region of the chick embryo. Development 113:1281-91.  Luo G, Hofmann C, Bronckers A L , Sohocki M , Bradley A , Karsenty G. (1995). BMP-7 is an inducer of nephrogenesis, and is also required for eye development and skeletal patterning. Genes Dev 9:2808-20.  Maas R., Personal communication.  181  Macias D, Ganan Y, Sampath TK, Piedra M E , Ros M A , Hurle JM. (1997). Role of BMP-2 and OP-1 (BMP-7) in programmed cell death and skeletogenesis during chick limb development. Development 124:1109-1117.  Maden M , Gale E , Kostetskii I, Zile M . (1996). Vitamin A-deficient quail embryos have half a hindbrain and other neural defects. Curr Biol 6:417-26.  Maden M , Graham A , Zile M , Gale E . (2000). Abnormalities of somite development in the absence of retinoic acid. IntJDev Biol 44:151-9.  Mangelsdorf DJ, Ong ES, Dyck JA, Evans R M . (1990). Nuclear receptor that identifies a novel retinoic acid response pathway. Nature 345:224-9.  Marazita M L , Goldstein A M , Smalley SL, Spence M A . (1986). Cleft lip with or without cleft palate: reanalysis of a three-generation family study from England. Genet Epidemiol 3:335-42.  Marazzi G, Wang Y, Sassoon D. (1997). Msx2 is a transcriptional regulator in the BMP4mediated programmed cell death pathway. Dev Biol 186:127-138.  Marcelle C , Stark MR, Bronner-Fraser M . (1997). Coordinate actions of BMPs, Wnts, Shh and noggin mediate patterning of the dorsal somite. Development 124:3955-63.  182  Martinez-Alvarez C, Tudela C, Perez-Miguelsanz J, O'Kane S, Puerta J, Ferguson M W . (2000). Medial edge epithelial cell fate during palatal fusion. Dev Biol 220:343-57.  Massague J. (1990). The transforming growth factor-beta family. Annu Rev Cell Biol 6:597-641.  Massague J. (1998). TGF-beta signal transduction. Annu Rev Biochem 67:753-91.  Matovinovic E and Richman JM. (1997). Epithelium is required for maintaining FGFR2  expression  levels in facial  mesenchyme  of the developing  chick  embryo.  Developmental Dynamics 210:407-16.  Matzuk M M , L u N , Vogel H , Sellheyer K, Roop DR, Bradley A . (1995). Multiple defects and perinatal death in mice deficient in follistatin. Nature 374:360-3.  McGonnell IM, Clarke JD, Tickle C. (1998). Fate map of the developing chick face: analysis of expansion of facial primordia and establishment of the primary palate.  Developmental Dynamics 212:102-18.  McMahon JA, Takada S, Zimmerman L B , Fan C M , Harland R M , McMahon AP. (1998). Noggin-mediated antagonism of B M P signaling is required for growth and patterning of the neural tube and somite. Genes Dev 12:1438-52.  183  McPherron A C and Lee SJ. (1993).  GDF-3 and GDF-9: two new members of the  transforming growth factor-beta superfamily containing a novel pattern of cysteines. J  Biol Chem 268:3444-9.  Mendelsohn C, Mark M , Dolle P, Dierich A , Gaub MP, Krust A , Lampron C, Chambon P. (1994). Retinoic acid receptor beta 2 (RAR beta 2) null mutant mice appear normal. Dev Biol 166:246-58.  Merino R, Ganan Y , Macias D, Economides A N , Sampath K T , Hurle J M . (1998). Morphogenesis of digits in the avian limb is controlled by FGFs, TGFPs, and Noggin through BMP signaling. Dev Biol 200:35-45.  Merino R, Macias D, Ganan Y , Economides A N , Wang X , Wu Q, Stahl N , Sampath K T , Varona P, Hurle J M . (1999). Expression and function of Gdf-5  during digit  skeletogenesis in the embryonic chick leg bud. Dev Biol 206:33-45.  Mina M , Gluhak J, Upholt WB, Kollar EJ, Rogers B. 1995. Experimental analysis of Msx-1 and Msx-2 gene expression during chick mandibular morphogenesis. Dev Dyn 202:195-214.  Minkoff R, Kuntz AJ. 1977. Cell proliferation during morphogenetic change; analysis of frontonasal morphogenesis in the chick embryo employing D N A labelling indices. J Embryol Exp Morphol 40:101-113.  184  Minkoff R, 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.  Mishina Y , Suzuki A , Ueno N , Behringer RR. (1995). Bmpr encodes a type I bone morphogenetic  protein receptor that is essential for gastrulation during mouse  embryogenesis. Genes Dev 9:3027-37.  Mitchell L E and Risch N . (1992). Mode of inheritance of nonsyndromic cleft lip with or without cleft palate: a reanalysis. American Journal of Human Genetics 51:323-32.  Mitchell L E . (1997). Transforming growth factor alpha locus and nonsyndromic cleft lip with or without cleft palate: a reappraisal. Genet Epidemiol 14:231-40.  Mizutani H and Urist M R . (1982). The nature of bone morphogenetic protein (BMP) fractions derived from bovine bone matrix gelatin. Clin Orthop 171:213-23.  Morgan, B.A. and Fekete, D . M . (1996). Manipulating gene expression with replicationcompetent retroviruses. In: Methods in Cell Biology, Vol. 51:185-233.  Morriss-Kay G, Ruberte E, Fukiishi Y . (1993). Mammalian neural crest and neural crest derivatives. Anat Am 175:501-7.  185  Morriss-Kay G M and Ward, SJ.  (1999).  Retinoids and mammalian development.  International review of cytology 188:73-131.  Munoz-Sanjuan, M K Cooper, PA Beachy, JF Fallon, J Nathans. (2001). Expression and regulation of chicken fibroblast growth factor homologous factor (FHF)-4 during craniofacial morphogenesis. Dev Dyn 220:238-45.  Murray JC, Nishimura D Y , Buetow K H , Ardinger H H , Spence M A , Sparkes RS, Falk RE, Falk P M , Gardner RJ, Harkness E M , et al.  (1990).  Linkage of an autosomal  dominant clefting syndrome (Van der Woude) to loci on chromosome Iq. Am J Hum Genet 46:486-91.  Murray PDF. 1963. Adventitious (secondary) cartilage in the chick embryo, and the development of certain bones and articulations in the chick skull. AustJZool 11:368-430.  Nakayama T, Cui Y , Christian JL. (2000). Regulation of BMP/Dpp signaling during embryonic development. Cell Mol Life Sci 57:943-956.  Nellen D, Affolter M* Basler K. (1994). Receptor serine/threonine kinases implicated in the control of Drosophila body pattern by decapentaplegic. Cell 78:225-37.  186  Ng LJ, Wheatley S, Muscat G E , Conway-Campbell J, Bowles J, Wright E , Bell D M , Tarn PP, Cheah KS, Koopman P. (1997). SOX9 binds D N A , activates transcription, and coexpresses with type II collagen during chondrogenesis in the mouse. Dev Biol 183:10821.  Niswander L and Martin GR. (1992). Fgf-4 expression during gastrulation, myogenesis, limb and tooth development in the mouse. Development 114:755-68.  Niswander L, Martin GR. (1993). FGF-4 and BMP-2 have opposite effects on limb growth. Nature 361:68-71.  Noden D M .  (1975). A n analysis of migratory behavior of avian cephalic neural crest  cells. Dev Biol 42:106-130.  Noden D M . (1983a). The embryonic origins of avian cephalic and cervical muscles and associated connective tissues. Am J Anat 168:257-276.  Noden D M . (1983b). The role of the neural crest in patterning of avian cranial skeletal, connective, and muscle tissues. Dev Biol 96:144-65.  Noden D M . (1978). The control of avian cephalic neural crest cytodifferentiation. I. Skeletal and connective tissues. Dev Biol 67:296-312.  187  Nohno T, Ishikawa T, Saito T, Hosokawa K, Noji S, Wolsing D H , Rosenbaum JS. (1995). Identification of a human type II receptor for bone morphogenetic protein-4 that forms differential heteromeric complexes with bone morphogenetic protein type I receptors. J Biol Chem 270:22522-6.  Noji S, Koyama E , Myokai F, Nohno T, Ohuchi H , Nishikawa K, Taniguchi S. (1993). Differential expression of three chick FGF receptor genes, FGFR1, FGFR2 and FGFR3, in limb and feather development. Prog Clin Biol Res 383:645-54.  Noramly S and Morgan BA. (1998). BMPs mediate lateral inhibition at successive stages in feather tract development. Development 125:3775-87.  Odelberg SJ, Kollhoff A, Keating M T . (2000). Dedifferentiation of mammalian myotubes induced by msxl. Cell 103:1099.  Oh SH, Johnson R, Wu DK. 1996. Differential expression of bone morphogenetic proteins in the developing vestibular and auditory sensory organs. J Neuroscience 16:6463-6475.  Ohuchi H, Yoshioka H , Tanaka A , Kawakami Y , Nohno T, Noji S. (1994). Involvement of  androgen-induced growth factor (FGF-8) gene in mouse  embryogenesis  and  morphogenesis. Biochemical & Biophysical Research Communications 204:882-8.  188  Onichtchouk D, Chen Y G , Dosch R, Gawantka V , Delius H , Massague J, Niehrs C. (1999).  Silencing of TGF-beta signalling by the pseudoreceptor B A M B I .  Nature  401:480-5.  Ozkaynak E , Rueger DC, Drier E A , Corbett C, Ridge RJ, Sampath T K , Oppermann H . (1990).  OP-1 cDNA encodes an osteogenic protein in the TGF-P family.  EMBO  J  9:2085-93.  Padgett RW, Wozney JM, Gelbart W M .  (1993). Human B M P sequences can confer  normal dorsal-ventral patterning in the Drosophila embryo. Proc Natl Acad Sci USA 90:2905-9.  Path! S, Rutenberg JB, Johnson R L , Vortkamp A . 1999. Interaction of Ihh and BMP/Noggin signaling during cartilage differentiation. Dev Biol 209:239-253.  Pearse R V and Tabin CJ. (1998). The molecular ZPA. J Exp Zool 282:677-90.  Pearce JJ, Penny G, Rossant J. (1999). A mouse cerberus/Dan-related gene family. Dev Biol 209:98-110.  Peterka M , Jelinek R. 1983. Origin of hydrocortisone induced orofacial clefts in the chick embryo. Cleft Palate J20:35-46.  189  Petkovitch M , Brand NJ, Kurst A , Chambon P. (1987). A human retinoic acid receptor which belongs to the family of nuclear receptors. Nature 330:444-450.  Piccolo S, Sasai Y , Lu B, De Robertis E M . (1996). Dorsoventral patterning in Xenopus: inhibition of ventral signals by direct binding of chordin to BMP-4. Cell 86:589-598.  Pizette S, Niswander L . (1999). BMPs negatively regulate structure and function of the limb apical ectodermal ridge. Development 126:883-894.  Pizette S, Niswander L. (2000). BMPs are required at two steps of limb chondrogenesis: formation of prechondrogenic condensations and their differentiation into chondrocytes. Dev Biol 219:237-249.  Proetzel G, Pawlowski SA, Wiles M V , Yin M , Boivin GP, Howies PN, Ding J, Ferguson M W , Doetschman T. (1995). Transforming growth factor-P3 is required for secondary palate fusion. Nat Genet 11:409-14.  Ren P, Lim CS, Johnsen R, Albert PS, Pilgrim D, Riddle D L . (1996). Control of C. elegans larval development by neuronal expression of a TGF-P homolog. Science 274:1389-91.  190  Reynolds SD, Zhang D, Puzas JE, O'Keefe RJ, Rosier R N , Reynolds PR.  (2000).  Cloning of the chick BMPl/Tolloid cDNA and expression in skeletal tissues. Gene 248:233-43.  Richman JM and Delgado JL. (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. Journal of Craniofacial Genetics & Developmental Biology 15:190-204.  Richman J M and Tickle C.  (1989).  Epithelia are interchangeable between facial  primordia of chick embryos and morphogenesis is controlled by the mesenchyme. Developmental Biology 136:201 -10.  Richman J M and Tickle C.  (1992).  Epithelial-mesenchymal interactions in the  outgrowth of limb buds and facial primordia in chick embryos. Developmental Biology 154:299-308.  Richman JM, Herbert M , Matovinovic E , Walin J. (1997). Effect of fibroblast growth factors on outgrowth of facial mesenchyme. Developmental Biology 189:135-47.  Richman JM and Mitchell PJ. (1998). Craniofacial development: knockout mice take one on the chin. Curr Biol 6:364-7.  191  Riddle RD, Johnson R L , Laufer E , Tabin C. 1993. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 75:1401-1416.  Ripamonti TJ and Reddi A H .  (1994). Periodontal regeneration: potential role of bone  morphogenetic proteins. J Periodontal Res 29:225-35.  Roark E F , Greer K. 1994. Transforming growth factor-beta and bone morphogenetic protein-2 act by distinct mechanisms to promote chick limb cartilage differentiation in vitro. Dev Dyn 200:103-116.  Rodriguez-Leon J, Merino R, Macias D, Ganan Y , Santesteban E , Hurle JM.  (1999).  Retinoic acid regulates programmed cell death through B M P signalling. Nat Cell Biol 1:125-6.  Romanoff A L . (1960). The Avian Embryo. Structure and Functional Development. Macmillan, New York.  Romitti PA, Lidral A C , Munger RG, Daack-Hirsch S, Burns T L , Murray JC. (1999). Candidate genes for nonsyndromic cleft lip and palate and maternal cigarette smoking and alcohol consumption: evaluation of genotype-environment  interactions from a  population-based case-control study of orofacial clefts. Teratology 59:39-50.  192  Rosenzweig B L , Imamura T, Okadome T, Cox G N , Yamashita H , ten Dijke P, Heldin C H , Miyazono K. (1995). Cloning and characterization of a human type II receptor for bone morphogenetic proteins. Proc Natl Acad Sci USA 92:7632-6.  Rowe A , Richman J M , Brickell P M . (1992).  Development of the spatial pattern of  retinoic acid receptor-beta transcripts in embryonic chick facial primordia. Development 114:805-13.  Rowe A , Richman J M , Brickell P M . (1991). Retinoic acid treatment alters the distribution of retinoic acid receptor-P transcripts in the embryonic chick face. Development 111:1007-1016.  Ruberte E , Dolle P, Chambon P, Morriss-Kay G. (1991). Retinoic acid receptors and cellular retinoid binding proteins. II. Their differential pattern of transcription during early morphogenesis in mouse embryos. Development 111:45-60.  Ruberte E , Dolle P, Krust A , Zelent A , Morriss-Kay G, Chambon P. (1990).  Specific  spatial and temporal distribution of retinoic acid receptor gamma transcripts during mouse embryogenesis. Development 108:213-22.  Saber G M , Parker SB, Minkoff R.  (1989).  Influence of  epithelial-mesenchymal  interaction on the viability of facial mesenchyme in vitro. Anat Rec 225:56-66.  193  Sampath T K and Reddi A H . (1981).  Dissociative extraction and reconstitution of  extracellular matrix components involved in local bone differentiation. Proc Natl Acad Sci USA 78:7599-603.  Sampath TK, Rashka K E , Doctor JS, Tucker RF, Hoffmann F M .  (1993). Drosophila  transforming growth factor beta superfamily proteins induce endochondral bone formation in mammals. Proc Natl Acad Sci USA 90:6004-8.  Sanz-Ezquerro JJ and Tickle C. (2000).  Autoregulation of Shh expression and Shh  induction of cell death suggest a mechanism for modulating polarising activity during chick limb development. Development 127:4811-23.  Sargent T D .  (1987).  Isolation of differentially expressed genes. Methods Enzymol  152:423-32.  Sasai Y , L u B, Steinbeisser H, Geissert D, Gont L K , De Robertis E M . (1994). Xenopus chordin: a novel dorsalizing factor activated by organizer-specific homeobox genes. Cell 19:119-90.  Sasai Y , L u B, Steinbeisser H , De Robertis E M . (1995). Regulation of neural induction by the Chd and Bmp-4 antagonistic patterning signals in Xenopus. Nature 376:333-6.  194  Satokata T and Mass R.  (1994).  Msxl  deficient mice exhibit cleft palate and  abnormalities of craniofacial and tooth development. Nature Genet 6:348-55.  Satokata I, M a L , Ohshima H , Bei M , Woo I, Nishizawa K, Maeda T, Takano Y , Uchiyama M , Heaney S, Peters H , Tang Z , Maxson R, Maas R.  (2000).  Msx2  deficiency in mice causes pleiotropic defects in bone growth and ectodermal organ formation. Nature Genet 24:391-5.  Schlunegger MP, Cerletti N , Cox D A , McMaster GK, Schmitz A , Grutter M G . (1992). Crystallization and preliminary X-ray analysis of recombinant human transforming growth factor beta 2. FEBSLett 303:91-3.  Schneider RA, Hu D, Helms JA. (1999). From head to toe: conservation of molecular signals regulating limb and craniofacial morphogenesis. Cell Tissue Res 296:103-9.  Schorle H, Meier P, Buchert M , Jaenisch R, Mitchell PJ. (1996). Transcription factor AP2 essential for cranial closure and craniofacial development. Nature 381:235.  Schutte B C and Murray JC. (1999). The many faces and factors of orofacial clefts. Hum Mol Genet 8:1853-9.  195  Scott IC, Steiglitz B M , Clark T G , Pappano W N , Greenspan DS. (2000). Spatiotemporal expression patterns of mammalian chordin during postgastrulation embryogenesis and in postnatal brain. Dev Dyn. 217(4):449-56.  Scott IC, Steiglitz B M , Clark T G , Pappano W N , Greenspan DS. (2000). Spatiotemporal expression patterns of mammalian chordin during postgastrulation embryogenesis and in postnatal brain. Dev Dyn 217:449-56.  Sela-Donenfeld D, Kalcheim C. (1999). Regulation of the onset of neural crest migration by coordinated activity of BMP4 and Noggin in the dorsal neural tube. Development 126:4749-4762.  Sela-Donenfeld D, Kalcheim C. (2000). Inhibition of noggin expression in the dorsal neural tube by somitogenesis: a mechanism for coordinating the timing of neural crest emigration. Development 127:4845-4854.  Selleck M A , Garcia-Castro MI, Artinger K B , Bronner-Fraser M . (1998). Effects of Shh and Noggin on neural crest formation demonstrate that B M P is required in the neural tube but not ectoderm. Development 125:4919-4930.  Shen H , Wilke T, Ashique A M , Narvey M , Zerucha T, Savino E , Williams T, Richman JM.  (1997).  Chicken transcription factor AP-2: cloning, expression and its role in  outgrowth of facial prominences and limb buds. Developmental Biology 188:248-66.  196  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 228:73-85.  Shou J, Murray RC, Rim PC, Calof A L . (2000). Opposing effects of bone morphogenetic proteins on neuron production and survival in the olfactory receptor neuron lineage. Development 127:5403-5413.  Shuler C F . (1995).  Programmed cell death and cell transformation in craniofacial  development. Critical Reviews in Oral Biology & Medicine 6:202-17.  Smith S M and Eichele G. (1991). Temporal and regional differences in the expression pattern of distinct retinoic acid receptor-beta  transcripts in the chick embryo.  Development 111:245-52.  Smith W C and Harland R M . (1992). Protein, Nucleotide Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70:829-840._  Smith W C , Knecht A K , Wu M , Harland R M . (1993). Secreted noggin protein mimics the Spemann organizer in dorsalizing Xenopus mesoderm. Nature 361:547-5499.  197  Smith WC. (1999). TGF beta inhibitors. New and unexpected requirements in vertebrate development. Trends Genet 15:3-5.  Solloway MJ, Dudley A T , Bikoff E K , Lyons K M , Hogan B L , Robertson EJ.  (1998).  Mice lacking Bmp6 function. Dev Genet 22:321-39.  Solloway MJ, Robertson EJ.  (1999). Early embryonic lethality in Bmp5;Bmp7 double  mutant mice suggests functional redundancy within the 60A subgroup. Development 126:1753-68.  Somasiri A , Howarth A , Goswami D, Dedhar S, Roskelley CD. (2001). Overexpression of the integrin-linked kinase mesenchymally transforms mammary epithelial cells. J Cell Sci 114:1125-36.  SperberGH. (2001). Craniofacial development. B C Decker Inc. Hamilton, London.  Stanley E G , Biben C, Allison J, Hartley L , Wicks IP, Campbell IK, McKinley M , Barnett L , Koentgen F, Robb L , Harvey RP. (2000). Targeted insertion of a lacZ reporter gene into  the  mouse  embryogenesis.  Cerl  locus  reveals  complex  and dynamic  expression  during  Genesis 26:259-64.  198  Storm E E and Kingsley D M . (1996).  Joint patterning defects caused by single and  double mutations in members of the bone morphogenetic protein (BMP) family. Development 122:3969-79.  Stratford T, Logan C , Zile M , Maden M . (1999). Abnormal anteroposterior and dorsoventral patterning of the limb bud in the absence of retinoids. Mech Dev 81:115-25.  Streit A , Lee KJ, Woo I, Roberts C, Jessell T M , Stern C D . (1998). Chordin regulates primitive streak development and the stability of induced neural cells, but is not sufficient for neural induction in the chick embryo. Development 125:507-519.  Stringa E and Tuan RS. (1996). Chondrogenic cell subpopulation of chick embryonic calvarium: isolation  by peanut  agglutinin affinity  chromatography  and in vitro  characterization. Anat Embryol 194:427-37.  Sun D, Baur S, Hay E D .  (2000).  Epithelial-mesenchymal transformation is the  mechanism for fusion of the craniofacial primordia involved in morphogenesis of the chicken lip. Dev Biol 228:337-49.  Suzuki A , Kaneko E , Maeda J, Ueno N . (1997). Mesoderm induction by BMP-4 and -7 heterodimers. Biochem Biophys Res Commun 232:153-6.  199  Takigawa T and Shiota K.  (2001).  Study on the differentiation of medial edge  epithelium in the developing mouse fetal palate in vitro. Dev Growth and Differ 4 3 Suppl:S55 Abstract S3-P17.  Takahashi K , Nuckolls G H , Tanaka O, Semba I, Takahashi I, Dashner R, Shum L , Slavkin H C . (1998). Development 125:1627.  Tamayo P, Slonim D, Mesirov J, Zhu Q, Kitareewan S, Dmitrovsky E, Lander ES, Golub TR. (1999). Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation. Proc Natl Acad Sci USA  96:2907-12.  Tamarin A , Crawley A , Lee J and Tickle C. (1984). Analysis of upper beak defects in chicken embryos following with retinoic acid. Journal of Embryology & Experimental Morphology 84:105-23.  Tanaka TS, Jaradat SA, Lim M K , Kargul GJ, Wang X , Grahovac MJ, Pantano S, Sano Y , Piao Y , Nagaraja R, Doi H , Wood W H 3rd, Becker K G , Ko MS. (2000). Genome-wide expression profiling of mid-gestation placenta and embryo using a 15,000 mouse developmental cDNA microarray. Proc Natl Acad Sci U SA 97:9127-32.  Tassabehji M , Read AP, Newton V E , Harris R, Balling R, Grass P, Strachan T. (1992). Waardenburg's syndrome patients have mutations in the human homologue of the Pax-3 paired box gene. Nature 355:635-6.  200  Tavella S, Raffo P, Tacchetti C, Cancedda R, Castagnola P. (1994). N - C A M and N cadherin expression during in vitro chondrogenesis. Exp Cell Res 215:354-62.  Taya Y , Kane S, Ferguson MWJ. (1999). Pathogenesis of cleft palate in TGF-p3 knockout mice. Development 126:3869-3879.  ten Dijke P, Yamashita H , Sampath T K , Reddi A H , Estevez M , Riddle D L , Ichijo H , Heldin C H , Miyazono K.  (1994).  Identification of type I receptors for osteogenic  protein-1 and bone morphogenetic protein-4. J Biol Chem 269:16985-8.  Tolarova M M and Cervenka, J. (1998).  Classification and birth prevalence of orofacial  clefts. Am J Med Genet 75:126-37.  Tonegawa A and Takahashi Y. (1998). Somitogenesis controlled by Noggin. Dev Biol 202:172-182.  Tosney KW. (1982). The segregation and early migration of cranial neural crest cells in the avian embryo. Dev Biol 89:13-24.  Trainor PA, Tarn PP. (1995). Cranial paraxial mesoderm and neural crest cells of the mouse embryo: co-distribution in the craniofacial mesenchyme but distinct segregation in branchial arches. Development 121:2569-82.  201  Trasler D G . (1968). Pathogenesis of cleft lip and its relation to embryonic face shape in A-J and C57BL mice. Teratology 1:33-49.  Trumpp A , Depew M J , Rubenstein JL, Bishop JM, Martin GR. (1999). Genes Dev. 13:3136.  Tyler MS, Hall BK. (1977). Epithelial influences on skeletogenesis in the mandible of the embryonic chick. Anat Rec 188:229-239.  Urist M R , Grant T T , Lindholm TS, Mirra JM, Hirano H , Finerman G A .  (1979).  Induction of new-bone formation in the host bed by human bone-tumor transplants in athymic nude mice. J Bone Joint Surg Am 61:1207-16.  Urist MR, Lietze A , Mizutani H, Takagi K , Triffitt JT, Amstutz J, DeLange R, Termine J, Finerman G A . (1982).  A bovine low molecular weight bone morphogenetic protein  (BMP) fraction. Clin Orthop 162:219-32.  Urist MR. (1965). Bone: formation by autoinduction. Science 150:893-9.  Urist MR. (1973).  A bone morphogenetic system in residues of bone matrix in the  mouse. Clin Orthop 91:210-20.  202  Vainio S, Karavanova I, Jowett A , 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 MJ, Dorland M , Beemer FA, van Amstel HK. (2000). MSX1 mutation is associated with orofacial clefting and tooth agenesis in humans. Nature Genet 24:3423.  Vanderas AP. (1987). Incidence of cleft lip, cleft palate, and cleft lip and palate among races: a review. Cleft Palate Journal 24:216-25.  Wall N A and Hogan B L . (1995). Expression of bone morphogenetic protein-4 (BMP-4), bone morphogenetic protein-7 (BMP-7), fibroblast growth factor-8 (FGF-8) and sonic hedgehog (SHH) during branchial arch development in the chick. Mechanisms of Development 53:383-92.  Wang E A , Rosen V , Cordes P, Hewick R M , Kriz M J , Luxenberg DP, Sibley BS, Wozney JM.  (1988). Purification and characterization of other distinct bone-inducing  factors. Proc Natl Acad Sci USA 85:9484-8.  Wang K Y and Diewert V M . (1992). A morphometric analysis of craniofacial growth in cleft lip and noncleft mice. Journal of Craniofacial Genetics & Developmental Biology 12:141-54.  203  Wang K Y , Juriloff D M , Diewert V M . (1995). Deficient and delayed primary palatal fusion and mesenchymal bridge formation in cleft lip-liable strains of mice. J Craniofac Genet Dev Biol 15:99-116.  Wang Y H , Rutherford B, Upholt WB, Mina M .  (1999). Effects of BMP-7 on mouse  tooth mesenchyme and chick mandibular mesenchyme. Dev Dyn 216:320-335.  Watanabe Y and Le Douarin N M . (1996). A role for BMP-4 in the development of subcutaneous cartilage. Mech Dev 57:69-78.  Weaver M , Dunn NR, Hogan B L . 2000. Bmp4 and FgflO play opposing roles during lung bud morphogenesis. Development 127:2695-704.  Wedden SE and Tickle C. (1986). Quantitative analysis of the effect of retinoids on facial morphogenesis. J Craniofac Genet Dev Biol 2:169-78.  Wedden SE. (1987). Epithelial-mesenchymal interactions in the development of chick facial primordia and the target of retinoid action. Development 99:341-51.  Wedden, S. E . (1991). Effects of retinoids on chick face development. J. Craniofac. Genet. Dev Biol 11:326-337.  204  Weeks D L , Melton DA. (1987). A maternal mRNA localized to the vegetal hemisphere in Xenopus eggs codes for a growth factor related to TGF-beta. Cell 51:861-7  Welsh J, Chada K, Dalai SS, Cheng R, Ralph D, McClelland M . (1992). Arbitrarily primed PCR fingerprinting of RNA. Nucleic Acids Res 20:4965-70.  Wharton K A , Ray RP, Gelbart W M . (1993). A n activity gradient of decapentaplegic is necessary for the specification of dorsal pattern elements in the Drosophila embryo. Development 117:807-22.  Wiley M J , Cauwenbergs P, Taylor IM.  (1983).  Effects of retinoic acid on the  development of the facial skeleton in hamsters: early changes involving cranial neural crest cells. Acta Anatomica 116:180-92.  Wilke T A , Gubbels S, Schwartz J, Richman JM.  (1997).  Expression of fibroblast  growth factor receptors (FGFR1, FGFR2, FGFR3) in the developing head and face. Developmental Dynamics 210:41-52.  Wilson PA and Hemmati-Brivanlou A . (1995). Induction of epidermis and inhibition of neural fate by Bmp-4. Nature 376:331 -3.  Winnier G, Blessing M , Labosky PA, Hogan BL. (1995). Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev 9:2105-16.  205  Wolpert L . (1989). Positional information revisited. Development 107 Suppl:3-12.  Wozney JM, Rosen V , Celeste AJ, Mitsock L M , Whitters MJ, Kriz RW, Hewick R M , Wang E A . (1988). Novel regulators of bone formation: molecular clones and activities. Science 242:1528-34.  Wozney JM. (1989). Bone morphogenetic proteins. Prog Growth Factor Res 1:267-80.  Wozney JM. (1990).  Using purified protein to clone its gene. Methods Enzymol  182:738-51.  Wrana JL, Attisano L , Wieser R, Ventura F, Massague J.  (1994).  Mechanism of  activation of the TGF-p receptor. Nature 370:341-7.  Wu DK, Oh SH. (1996). Sensory organ generation in the chick inner ear. J Neuroscience 16:6454-6462.  Wu L N , Ishikawa Y, Genge BR, Sampath TK, Wuthier RE. (1997). Effect of osteogenic protein-1 on the development and mineralization of primary cultures of avian growth plate chondrocytes: modulation by retinoic acid. J Cell Biochem 67:498-513.  206  Wyszynski D F and Beaty T H . (1996). Review of the role of potential teratogens in the origin of human nonsyndromic oral clefts. Teratology 53:309-17.  Xie T, Finelli A L , Padgett RW. (1994).  The Drosophila saxophone gene: a serine-  threonine kinase receptor of the TGF-beta superfamily. Science 263:1756-9.  Yamaguchi TP, Bradley A , McMahon AP, Jones S. (1999). A Wnt5a pathway underlies outgrowth of multiple structures in the vertebrate embryo. Development 126:1211-23.  Yee GW, Abbott U K . (1978). Facial development in normal and mutant chick embryos. I. Scanning electron microscopy of primary palate formation. J Exp Zool. 206:307-321.  Yi SE, Daluiski A , Pederson R, Rosen V , Lyons K M . (2000). The type I B M P receptor BMPRIB is required for chondrogenesis in the mouse limb. Development 127:621-630.  Yip JE, Kokich V G and Shepard TH. (1980). The effect of high doses of retinoic acid on prenatal craniofacial development in Macaca nemestrina. Teratology 21:29-38.  Yokouchi Y, Sakiyama J, Kameda T, Iba H, Suzuki A , Ueno N , Kuroiwa A. (1996). BMP2/-4 mediate programmed cell death in chicken limb buds. Development 122:3725-3734.  207  Yokouchi Y , Vogan K J , Pearse R V , Tabin CJ. (1999). Caronte, a novel  Antagonistic signaling by  Cerberus-related gene, establishes left-right  asymmetric  gene  expression. Cell 98:573-83.  Young D L , Schneider R A , Hu D, Helms JA.  (2000).  Genetic and teratogenic  approaches to craniofacial development. Crit Rev Oral Biol Med 11:304-17.  Zhang D, Mehler M F , Song, Q, Kessler J. (1998). Development of bone morphogenetic protein receptors in the nervous system and possible roles in regulating trkC expression. Neuroscience 18:3314-3326.  Zhang H and Bradley A. (1996).  Mice deficient for BMP2 are nonviable and have  defects in amnion/chorion and cardiac development. Development 122:2977-86.  Zhang J, S. HagopianDonaldson, G. Serbedzija, J. Elsemore, D. plehn-Dujowich, A.P. McMahon, R.A. Flavell, T. Williams. (1996). Neural tube, skeletal and body wall defects in mice lacking transcription factor AP-2. Nature 381:238-41.  Zhao Q, Eberspaecher H , Lefebvre V , De Crombrugghe B. (1997). Parallel expression of Sox9 and Col2al in cells undergoing chondrogenesis. Dev Dyn 209:377-86.  Zhao Z , Rivkees SA. (2000). Programmed cell death in the developing heart: regulation by BMP4 and FGF2. Dev Dyn 217:388-400.  208  Zhu L , Marvin MJ, Gardiner A , Lassar A B , Mercola M , Stern C D , Levin M . (1999). Cerberus regulates left-right asymmetry of the embryonic head and heart. Curr Biol 9:931-938.  Zimmerman L B , De Jesus-Escobar JM, Harland R M . (1996). The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. Cell 86:599-606.  Zou H and Niswander L .  (1996).  Requirement for B M P signaling in interdigital  apoptosis and scale formation. Science 272:738-41.  Zou H , Wieser R, Massague J, Niswander L . (1997).  Distinct roles of type I bone  morphogenetic protein receptors in the formation and differentiation of cartilage. Genes Dev 11:2191-203.  Zuniga A , Haramis A P , McMahon AP, Zeller R.  (1999).  Signal relay by B M P  antagonism controls the SHH/FGF4 feedback loop in vertebrate limb buds.  Nature  401:598-602.  Zweiger G. (1999). Knowledge discovery in gene-expression-microarray data: mining the information output of the genome. Trends Biotechnol 17:429-36.  209  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items