Molecular Mechanisms in Calvarial and Suture Development

David Rice

Department of Orthodontics and Pedodontics, Institute of Dentistry, and, Developmental Biology Programme, Institute of Biotechnology, University of Helsinki, Finland.

Academic Dissertation

To be discussed publicly with the permission of the Faculty of Medicine of the University of Helsinki, in auditorium 1041, Viikki Biocenter 2 on 5th November 1999, at 12 noon.

Helsinki 1999 Supervised by:

Professor Irma Thesleff, University of Helsinki, Finland.

Reviewed by:

Professor Seppo Vainio, University of Oulu, Finland.

and

Professor Kalervo Väänänen, University of Turku, Finland.

Opponent:

Assistant Professor Lynne Opperman, Baylor School of Dentistry, Texas A & M University, U.S.A.

ISBN 951-45-8960-2 (PDF version) Helsinki 1999 Helsingin yliopiston verkkojulkaisut http://ethesis.helsinki.fi/

2 CONTENTS

3DJH /LVW RI 2ULJLQDO 3XEOLFDWLRQV 5 $EEUHYLDWLRQV 6 6XPPDU\ 7 5HYLHZ RI WKH /LWHUDWXUH 9

%RQH 'HYHORSPHQW DQG *URZWK 9 Origin and patterning of the craniofacial skeleton 9 Mesenchymal condensations 9 Intramembraneous 10 Endochondral ossification 11

&HOOXODU %LRORJ\ RI %RQH 11 , stromal cells and osteocytes 13 Osteoclasts 18 Proliferation of bone cells 22 Proliferation during calvarial bone and suture development 22 Apoptosis during embryogenesis 23 Apoptosis during bone and suture development 24

'HYHORSPHQW RI WKH &DOYDULD 24 Developmental anatomy 24 Sutures and 25 Suture morphology 26 Suture closure 27 Suture position 27 Sutural growth 28 Role of the dura 28 Biomechanical forces 30

3DWKRELRORJ\ RI &DOYDULDO 'LVRUGHUV 30 Premature suture fusion, 30 Delayed suture formation, cleidocranial dysplasia 32 Other conditions 33

$QLPDO 0RGHOV ZLWK D &DOYDULDO 3KHQRW\SH 34

5HJXODWLRQ RI (PEU\RQLF 'HYHORSPHQW 44

)LEUREODVW JURZWK IDFWRUV )*)V DQG WKHLU 5HFHSWRUV 45 FGFs and their receptors in embryogenesis 45 FGFs in bone and suture development 46 FGF receptors in bone and suture development 47

3 +HOL[/RRS+HOL[ SURWHLQV 48 TWIST and Inhibitors of differentiation (IDs) in embryogenesis 48 TWIST and IDs in bone and suture development 48

+HGJHKRJV 49 Hedgehogs in embryogenesis 49 Hedgehogs in bone development 49

06;V 50 MSX1 and 2 in embryogenesis 50 MSX1 and 2 in bone and suture development 51

7KH 7UDQVIRUPLQJ *URZWK )DFWRU β 7*)β 6XSHUIDPLO\ 51 The TGFβ superfamily in embryogenesis 51 The TGFβs in bone and suture development 52

1HOO 53

$LPV 54 0DWHULDOV DQG 0HWKRGV 55 5HVXOWV DQG 'LVFXVVLRQ 63 &RQFOXGLQJ 5HPDUNV 74 $FNQRZOHGJHPHQWV 75 5HIHUHQFHV 76 2ULJLQDO 3XEOLFDWLRQV ,,9 93

4 LIST OF ORIGINAL PUBLICATIONS

This thesis is based on the following original articles, which are referred to in the text by their Roman numerals. In addition, some unpublished data are also presented.

I Rice, D.P.C., Kim, H.J., and Thesleff, I. (1997). Detection of gelatinase B expression reveals osteoclastic bone resorption as a feature of early calvarial bone development. Bone 21, 479-486. http://www-east.elsevier.com:80/bone

II Rice, D.P.C., Kim, H-J., and Thesleff, I. (1999). Apoptosis in calvarial bone and suture development. Eur. Oral Sci. 106, 1-10. http://www.munksgaard.dk

III Kim, H-J.*, Rice, D.P.C.*, Kettunen, P.J., and 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. * Equal contribution. http://www.biologists.com/Development/125/07/dev1213.html

IV Rice, D.P.C., Åberg, T., Chan, C., Tang, Z., Kettunen, P.J., Pakarinen, L., Maxson, R.E.Jr., and Thesleff, I. (1999). Integration of FGF and TWIST signalling in calvarial bone and suture development. Submitted to Development. http://www.biologists.com

5 ABBREVIATIONS

bHLH Basic helix-loop-helix IHH Indian hedgehog BMP Bone morphogenetic protein LIF Leukaemia inhibitory factor BSA Bovine serum albumin OMIM Online Mendelian Inheritance BSP Bone sialoprotein in Man BrdU 5’-bromo-2’-deoxyuridine mRNA Messenger ribonucleic acid CAM Cell adhesion molecule MSX Vertebrate homologue of CBFA1 Core binding factor alpha 1 'URVRSKLOD muscle segment CCD Cleidocranial dysplasia (0VK) cDNA Complementary MMP Matrix metalloproteinase deoxyribonucleic acid OC Osteocalcin Col Collagen OPG Osteoprotegerin CSF Colony stimulating factor OSF2 specific factor 2 DEPC Diethylpyrocarbonate P Post natal day DIG Digoxigenin PBS Phosphate buffered saline DNA Deoxyribonucleic acid PDGF Platelet derived growth factor DTT Dithiothreitol PFA Paraformaldehyde E Embryonic day PTC Patched EGF Epidermal growth factor PTHrP Parathyroid hormone related FGF Fibroblast growth factor peptide FGFR Fibroblast growth factor RNA Ribonucleic acid receptor SHH Sonic hedgehog Gel B Gelatinase B TGF Transforming growth factor GDF Growth and differentiation TNAP Tissue non-specific alkaline factor phosphatase GLI Vertebrate homologue of the TRAP Tartrate resistant acid 'URVRSKLOD segment polarity phosphatase gene cubitus interruptus TUNEL Terminal deoxynucleotidyl HH Hedgehog transferase mediated nick end HLH Helix-loop-helix labelling HSPG Heparan sulphate UTP Uridine triphosphate proteoglycan WNT Vertebrate homologue of the ID Inhibitor of differentiation 'URVRSKLOD segment polarity IG Immunoglobulin gene wingless

Upper case letters refer to Humans. Lower case letters refer to other animals. Italics are used for . Non italics are used for proteins

In this thesis, unless it is specifically stated otherwise, I shall refer to the mouse as a model for discussion.

6 SUMMARY

The development and growth of the is a highly co-ordinated process involving many different tissues that interact to form a complex end result. When this normal development is disrupted debilitating pathological conditions, such as craniosynostosis and cleidocranial dysplasia, can result. The of the vault of the skull, or calvaria, are connected by joints called sutures and fontanelles. These joints normally close in a synchronized manner, allowing the underlying brain and the rest of the skull to reach its full size and shape. Craniosynostosis is a condition where the bones of the calvaria fuse prematurely, and as the brain and head continue to develop, so growth between the calvarial bones is restricted and deformity results. In contrast, cleidocranial dysplasia is characterised by a delay in suture closure. It is known that in the Fibroblast growth factor receptors 1, 2 and 3 (FGFR1, 2 and 3), as well as the transcription factors MSX2 and TWIST cause craniosynostosis and that mutations in the Core binding factor alpha 1 (CBFA1) causes cleidocranial dysplasia. However, relatively little is known about the development of the calvaria; about where and when these genes are active during normal calvarial development; how these genes may interact in the developing calvaria and the disturbances that may occur to cause these disorders. In this work we have attempted to address some of these questions from a basic biological perspective. We have put forward the developing mouse calvaria as a model system, in which one can study not only the processes of suture development but also the processes of intramembraneous bone formation and its subsequent modelling and remodelling. In comparison to the long bone, in which a cartilage framework is laid down prior to osteoblasts differentiating and finally forming bone, in the calvaria the majority of bone is formed intramembraneously. Here, osteoblasts differentiate directly form neighbouring mesenchymal or stromal cells and lay down matrix, which is subsequently mineralised without the formation of a cartilage substructure. This provides us with a relatively simplified model for studying osteoblast differentiation. By using both in-situ hybridisation and enzyme histochemical techniques we have described the distribution of osteoblasts and osteoclasts in the developing calvaria, and noted that the processes of bone deposition and resorption are highly integrated during calvarial bone and suture development and modelling. In addition, we have endeavoured to study bone cell turnover with regard to proliferation and cell death. we have found that cell death does indeed occur in both osteoblasts and osteoclasts and hypothesise that this may be one mechanism by which bone formation and resorption is regulated. Thus, the co-ordination of osteoblast and osteoclast differentiation, cell function and cell death appear to be central in maintaining suture patency, and hence the normal harmonized development of the calvaria. Consequently, when this co-ordination is disrupted craniosynostosis or cleidocranial dysplasia may result. In an effort to discover where and when some of the genes, either known or proposed the be involved in bone formation, as well as genes known to cause pathological conditions of the calvaria, are active during normal calvarial development; we have performed a detailed survey using in-situ hybridisation. This was in both sectional and

7 for the first time whole-mount mouse calvarial tissue. Rather than being expressed in a random fashion, we found that many of these genes were distributed in a unique spatio- temporal manner. From this baseline, we have attempted to investigate how these genes may interact. To this end, we have adapted a culture system, in which embryonic mouse calvarial development can be successfully monitored and manipulated in-vitro. Using a system of beads impregnated with a variety of different growth factors we have analysed, at a tissue level, how these molecules of interest, primarily those involved in osteoblast differentiation, may inter-relate. Finally, using these techniques we have shown that FGFs and Bone morphogenetic proteins together with Helix-loop-helix and homeobox containing MSX transcription factors, do indeed interact in an intricate signalling network during osteoblast differentiation that when disrupted could lead to pathological conditions of the skull. Interestingly, these groups of genes are the same as those involved in the development and regulation of not only long bones but also other tissues and organs, throughout the animal kingdom. Hence, findings from this study may not only help us to understand the normal development of the calvaria and pathological conditions resulting from its disruption, but also may give an insight into the mechanisms of development of other tissues and organs.

8 REVIEW OF THE LITERATURE

Bone Development and Growth

Origin and patterning of the craniofacial skeleton Most of the craniofacial skeleton is of neural crest cell origin (Le Lievré and Le Douarin, 1975; Noden, 1975). Neural crest cells are ectomesenchymal cells, so named because of their site of origin, arising from the crests of the neural folds. Early in foetal life the embryo undergoes formation of the neural tube, or neurulation (human days post ovulation 16; Carnegie stage 7; mouse embryonic day 8 (E8), Theiler stage 12; chick E1- 1.5; Hamburger and Hamilton stage 7). This starts with the formation of the neural folds, which occurs when presumptive epidermal cells move medially. These folds of neural epithelial cells then elevate to form the neural groove. This occurs because the cells in the midline are anchored to the underlying notochord. The neural groove then hinges and the apices of the folds approximate and subsequently adhere. Cells of each neural fold then merge thus forming the neural tube, the rudiment of the central nervous system (Gilbert, 1997 review). Neural crest has a great propensity for migration, and in mammals neural crest cells in the cranial region start to migrate before the neural tube is closed (Nichols, 1981; Tan and Morriss-Kay, 1985). Cranial neural crest cells move ventrally into predetermined locations to produce the craniofacial mesenchyme which differentiates into bone, cartilage, cranial neurons, glia and other connective tissue of the craniofacial region (Johnston, 1966; Lumsden et al., 1991). As they spread, they divide, forming a larger cell population at their destination than at their origin. Under local conditions they eventually generate a large number of different cell types including osteoblasts, chondroblasts, odontoblasts and fibroblasts (Le Douarin, 1982 review; Bronner-Fraser and Fraser, 1988; Shah et al., 1996). Except in the otic and occipital regions where the skull is derived from cephalic and somitic mesoderm, the bones of the face, skull vault and anterior cranial base are derived from neural crest cells. The limit of the neural crest derived skeleton is located in the sella turcica between the basipresphenoid and the basipostsphenoid regions, coinciding with the tip of the notochord. Thus the frontal and parietal bones and the sutures between are of neural crest origin (Couly et al., 1993).

Mesenchymal condensations Once undifferentiated mesenchymal cells of neural crest origin have arrived at the sites of a presumptive bone they form a cellular condensation. These cells then differentiate into either chondroblasts, in the case of endochondral bones, or directly into osteoblasts in the case of bones of intramembraneous origin. The condensations form templates for the future vertebrate bones (Hall and Miyake, 1992). Cranial neural crest cells possess some intrinsic patterning information that it conveys to the periphery (Noden, 1983). Thus the timing, location and form of a condensation, and therefore the future bone can be intrinsically controlled. The determination of the identity of cranial neural

9 crest derivatives results from a combination of the peripheral environment and regulatory genes such as Hox genes (Rijli et al., 1993; Couly et al., 1998). However, external environmental factors can alter the final shape and size of the bone (Kiliaridis, 1995). In the mouse, these condensations first appear at E10.5, though most appear at about E12.5 (Karsenty 1998 review; Kaufman and Bard, 1999 review). They express type I collagen and a splice variant of type II collagen, αI, that is not specific to chondroblasts (Lui et al., 1995). A variety of extracellular matrix and cell surface molecules including syndecans, neural-cell adhesion molecule (N-CAM) and N-cadherin are all thought to be important in mesenchymal condensate formation (Hall and Miyake, 1995). N-CAM is known to mediate cell adhesion and is therefore thought to play a role in the initiation of the condensation process, and has been localised in prechondrogenic condensations. However, as chondroblasts differentiate and express type II collagen, N-CAM is downregulated. Conversely, an upregulation of N-CAM is concomitant with osteoblast differentiation (Tavella et al., 1994; Fang and Hall, 1995). The mandibular bone condensation in the chick can be visualised with peanut agglutinin lectin at day 5.75. Interestingly, mandibular mesenchymal cells express the preosteoblast marker, alkaline phosphatase, prior to condensation (Dunlop and Hall, 1995). Increased mitotic activity (Fyfe and Hall, 1983; Hall and Miyake, 1992) and/or cellular aggregation (Thorogood and Hinchliffe, 1975) are thought to be of importance in the formation of cellular condensates. Interaction of the neural crest cells with the epithelium has been proposed as a mechanism whereby the skeletal cell fate is determined (Schowing, 1968; Bee and Thorogood, 1980; Hall, 1991). Indeed, the epithelium is vital over a critical period if osteogenesis is to be promoted in the mandibular mesenchyme (Tyler and Hall, 1977). There is evidence that signals from the epithelium are also important in establishing rostral-caudal polarity in the mandibular arch mesenchyme, and thereby committing cells to either to an odontogenic or skeletal pathway (Tucker et al., 1999). Although the mechanisms underlying osteogenesis promoting tissue interactions remain unclear, there is evidence that epithelia, or basal laminae derived from epithelia, might simply have a mitogenic effect on the adjacent mesenchyme (Thorogood, 1993 review). This could allow the formation of mesenchymal condensates but might not necessarily determine cell fate.

Intramembraneous ossification A large part of the cranial vault and face are formed from bones of intramembraneous origin, as are large portions of the mandible and . Other bones form by a combination of endochondral ossification and an element of intramembraneous ossification in the periosteum. During intramembraneous ossification there is no cartilaginous template; osteoblasts in mesenchymal condensations directly secrete bone extracellular matrix which then matures, thus forming an ossification centre. A single bone may have more than one ossification centre, in which case they coalesce to form the final bone (Kaufman and Bard, 1999 review). At the perimeter of the developing bone new osteoblasts continually differentiate, the ossification centre therefore expands and bone growth is achieved. Following vascular invasion and marrow space formation, a membrane bone with the structure of a bony core surrounded by periosteum

10 is finally, formed.

Endochondral ossification The majority of the bones of the skeleton are formed by endochondral ossification. Here the properties of both cartilage and bone are employed, enabling the individual to bear load and function, while the skeleton is still developing and growing. First a cartilage template is formed, which is subsequently converted into bone. Initially, mesenchymal cells condense, at a predetermined site, into roughly the shape of the future bone. These cells then differentiate into chondroblasts which undergo proliferation. They start to lay down extracellular matrix and as this matrix production progresses so the chondroblasts become embedded in their own matrix. These cells lie within their individual lacunae and are now called chondrocytes. Unlike osteocytes these cells continue to proliferate and then later enlarge and become hypertrophic. At this point the chondrocytes are lined up in columns, roughly parallel to the general direction of bone growth. This proliferation and hypertrophy is allowed, in part, by the gel like consistency of cartilage matrix. Vascularisation of the hypertrophic zone then occurs, with osteoclasts and osteoblasts being recruited to gradually replace the cartilage scaffold with bone. Bone matrix is laid down which then subsequently calcifies (Marks and Hermey, 1996 review). The fate of these chondrocytes is still controversial, many cells die, however some survive and it has been suggested that these cells differentiate into osteoblasts (Farnum et al., 1990; Thesingh et al., 1991; Roach et al., 1995). These cellular changes and the transformation of cartilage to bone in the growth plate are reflected as well demarcated by zones that can be visualised under light microscopy. Longitudinal bone growth is thus a balance between cartilage formation and its vascularisation and bone formation. As growth plates persist into later life they provide a mechanism whereby long bones may continue to adapt and grow while still allowing the individual to load a bone. The functional demands of the bones of the calvaria differ, with the majority of growth being completed earlier in life. In combination with endochondral ossification, bone is also formed on the surface of bones, in the periosteum. Here, there is no cartilaginous preformer, osteoblasts differentiating and secreting matrix in an intramembraneous manner. This process contributes to the growth in diameter and final shape of a bone.

Cellular Biology of Bone

Bone is a dynamic tissue made up of bone forming cells, osteoblasts and their lineage relatives, as well as bone resorbing cells, osteoclasts (Figure 1). Osteoblasts synthesise and regulate the deposition and later mineralisation of the extracellular matrix of bone. As osteoid, unmineralised bone matrix, is secreted, some osteoblasts may become surrounded by matrix which then mineralises, these cells thereby becoming encapsulated in bone. These osteocytes, are stellate in shape and spaced throughout mature bone at regular intervals. Each osteocyte occupies a lacuna within which its cell

11 body is located, from which radiate cytoplasmic extensions which connect with neighbouring cells by means of gap junctions (Doty, 1981). Although it is known that osteocytes have mechanoreception properties, are capable of matrix secretion and possibly resorption, osteocyte function is otherwise poorly understood (Nijweide et al., 1996 review). The bone surface is covered by a thin cellular layer of bone lining cells. These are flat cells of mesenchymal origin, osteoblast lineage, whose cytoplasm contains few organelles. It has been speculated that they may have the ability to differentiate into functioning osteoblasts (Marks and Hermey, 1996 review).

2VWHRFODVW 2VWHREODVW /LQHDJH /LQHDJH

Blood vessel Mesenchymal cells

Fusion of Osteo- osteoclast progenitors precursors Bone lining cell Preosteoblast

Osteoclast Osteoblast

Osteocyte

Osteoid Mineralized bone

Figure 1. Schematic illustration of the cell types in bone and their differentiation pathways. Osteoclasts, large multinucleated cells that resorb bone, develop from precursors of haematopoietic origin. Osteoblasts, differentiate from cells of mesenchymal origin, they lay down matrix, extracellularly, which subsequently mineralizes. Osteoblasts that become surrounded by bone are called osteocytes.

12 Osteoblasts, stromal cells and osteocytes Morphology and histochemistry Osteoblasts are post-proliferative, cuboidal cells that line bone matrix at sites of active bone production. They are strongly alkaline phosphatase positive. Also, they can be recognised by their ability to synthesise a number of cell specific macromolecules, including bone matrix proteins, hormone receptors, cytokines and growth factors. Preosteoblasts are the immediate precursors of osteoblasts and are found adjacent to, or one or two cell layers away from, osteoblasts lining the bone surface. Unlike osteoblasts and the less mature osteoprogenitors and undifferentiated mesenchymal cells, preosteoblasts are thought to be relatively limited in their ability to divide (Aubin and Liu, 1996 review). Reportedly, preosteoblasts, but not osteoprogenitors, stain for alkaline phosphatase (Farley et al., 1980; Weinreb et al., 1990). This tissue non-specific isoform of alkaline phosphatase (TNAP) is a ubiquitous plasma membrane-bound enzyme, synthesised at varying levels by many cell types (MacGregor et al., 1995; Henthorn, 1996 review). The precise role of TNAP is unknown, although it is thought to be involved in the mineralisation process. Indeed, individuals with a missense in the alkaline phosphatase gene suffer from hypophosphatasia, a disorder characterised by poorly mineralised bones (Weiss et al., 1988). Also, inhibition of alkaline phosphatase with levamisole blocks mineralisation in-vitro (Bellows et al., 1991). However, somewhat surprisingly, mice deficient in TNAP, while exhibiting some dental defects show only a minor skeletal phenotype (Waymire et al., 1995). It has also been speculated that alkaline phosphatases may play a role in osteoprogenitor and osteoblast migration (Chang et al., 1992; Aubin and Liu, 1996 review).

Osteoblast differentiation Osteoblasts can differentiate from undifferentiated pluripotent mesenchymal cells. Under different in-vitro conditions, these mesenchymal cells can differentiate into a variety of cell types, namely chondroblasts, adipoblasts, fibroblasts and myoblasts (Grigoriadis et al., 1988; Bennett et al., 1991). Osteoblast differentiation comprises commitment into the osteoblastic cell lineage, then maturation leading to a functional cell, fully expressing its differentiated characteristics. This multi-step process involves a variety of growth factors, cytokines and hormones (Triffitt, 1996 review). Although, mechanisms by which osteoblasts differentiate are starting to be understood, the existence of a pool of osteoblastic stem cells is still a controversial issue. The , defined as “A cell type which, in the adult organism, can maintain its own numbers in spite of physiological or artificial removal of cells from the population” (Lajtha, 1982, cited by Triffitt, 1996 review), by its hypothesized asymmetric cell division, enables continued proliferation of precursors which subsequently differentiate to supply physiological demands. The high capacity of bone to regenerate is indicative of the presence of stem cells but there is little evidence for their existence. Due to their indeterminate biochemical and morphological phenotype, as well as their low incidence and steady state, they have yet to be recognized. There is interest in osteogenic stem cell biology, not only as it leads to a deeper understanding of osteoblast differentiation but also in the potential of stem cell for skeletal disorders such as the osteogenesis imperfectas (Horwitz et al., 1999).

13 There is also speculation that the phenotype of the mature osteoblast is heterogeneous, with subpopulations of osteoblasts expressing only subsets of the known osteoblast markers, raising the possibility of multiple parallel differentiation pathways and perhaps even different progenitor pools (Aubin, 1998 review). A schematic representation of some of these markers is shown in Figure 2.

0HVHQFK\PDO 2VWHRSURJHQLWRU 0DWXUH FHOO RVWHREODVW

Alkaline phosphatase -++ -/ Bone sialoprotein --+ Cbfa1 +++ hXBP-1 -++ Id ++ - Msx1 ++ - Msx2 ++ - Osteocalcin --+ Osteopontin -++ PTH/PTHrp --/++ Type I collagen -?+ + + Twist ++ -

Figure 2. Selected osteoblast differentiation markers. Based on Aubin and Liu, 1996 review. Human X-box-binding protein 1 (hXBP-1) is a transcription factor of the basic region leucine zipper family that is important in the expression of the class II major histocompatibility complex genes.

14 Osteocalcin and Cbfa1 Osteocalcin (OC, bone gla protein, BGP) is a small protein, approximately 50 amino acid residues, accounting for nearly 10% of the non-collagenous protein of bone’s extracellular matrix (Gallop et al., 1980). Osteocalcin is highly expressed by mature osteoblasts with reportedly lower levels of expression in platelets and megakaryocytes (Bronckers et al., 1987; Weinreb et al., 1990). Like other gla proteins, osteocalcin contains glutamic acid residues that have been γ-carboxylated, a process that requires vitamin K. Through these modified gla elements, osteocalcin can bind Ca2+ ions and hydroxyapatite crystals, a key process in bone matrix maturation (Gundberg et al., 1984). Investigations aimed at understanding how the osteocalcin gene is regulated have proved invaluable in furthering our comprehension of osteoblast differentiation (Ducy and Karsenty, 1995; Karsenty, 1998 review). Core binding factor alpha 1 (Cbfa1) is a transcriptional activator of osteoblast differentiation (Ducy et al., 1997), activating most cell type specific genes expressed by osteoblasts. Cbfa proteins are the mammalian homologues of the Drosophila transcription factor runt, which is involved in neurogenesis and sexual differentiation (Gergen and Wieschaus, 1985; Duffy et al., 1991). Osf2 (Osteoblast specific factor 2, AML3) is a full-length transcript of Cbfa1 and is a binding factor of an osteoblast cis-acting element (OSE2) in the osteocalcin promoter (Ducy and Karsenty, 1995). Overexpression of Osf2 in non-osteoblastic cells leads to the expression of osteoblastic markers such as bone sialoprotein (BSP) and osteocalcin (Ducy et al., 1997). However, there is evidence that suggests that Osf2 binding to the BSP promoter is not essential for osteoblast-selective expression (Benson et al., 1999). Early in development, Osf2 is expressed by condensing mesenchymal cells which may differentiate into osteoblasts, chondroblasts or odontoblasts (Ducy et al., 1997; Komori et al., 1997; D’Souza et al., 1999). Later, Osf2 appears to show a more restricted expression pattern, with mRNA being detected in cells of osteoblast lineage and in ectodermally derived ameloblasts during the maturation phase of enamel formation (D’Souza et al., 1999; Jiang et al., 1999). As might be expected, Cbfa1 deficient mice exhibit a dramatic phenotype with a complete absence of osteoblasts and no osteogenesis (Komori et al., 1997; Otto et al., 1997). These mice also possess defects in chondrocyte differentiation and maturation (Inada et al., 1999; Kim et al., 1999). Mice heterozygotic for Cbfa1 exhibit hypoplasia of the calvarial bones as well of the , bones which, in the main, form directly from mesenchyme by intramembraneous ossification (Otto et al., 1997). This phenotype is similar to that of the cleidocranial dysplasia (Ccd -/+) mouse, a radiation induced mutant (Sillence et al., 1987) and also to the human condition of the same name, which will be described later. Furthermore, Cbfa1 has been shown to be partly deleted or its function partly lost in both the mouse and human conditions respectively. Thus the Cbfa1 gene is essential for osteoblast differentiation and bone formation (Mundlos et al., 1997; Otto et al., 1997). Despite the wealth of scientific activity surrounding Cbfa1, we are only now starting to discover details about its regulation and its downstream targets. Given the correct conditions, BMPs can induce osteoblast differentiation in non osteogenic mouse pluripotent cells (Katagiri et al., 1990; Piccolo et al., 1996). It has been shown that BMP7 induces Cbfa1 expression in these cells and that this is prior to the expression of

15 any other osteoblast marker genes (Ducy et al., 1997). BMP2, 4 and 7 have no such effect on Cbfa1 in isolated dental mesenchyme, while FGF4 stimulates its mRNA expression (D’Souza et al., 1999). Smads act at a number of sites during TGFβ superfamily signal transduction, and interestingly, Cbfa1 has been shown to be a downstream target of Smad2 in osteoblastic cell culture (Li et al., 1998). The hormone 1,25(OH) D abolishes Cbfa1 expression in primary mouse osteoblast cultures (Ducy et al., 1997) 2and3 this is in line with the finding that 1,25(OH) D inhibits mouse osteocalcin expression by abolishing DNA binding of Osf2 (Zhang et al.,2 3 1997). Osteoblast function can be further regulated by interaction with extracellular matrix proteins such as type I collagen (Lynch et al., 1995). Integrin α2-β1 is a cell adhesion promoting receptor which binds to type I collagen, is known, as are several other integrins, to be expressed by osteoblasts. Interestingly, it has been demonstrated that α2-integrin/type I collagen interaction is required for activation of Osf2 and induction of osteoblast specific gene expression (Horton and Davies, 1989 review; Clover et al., 1992; Xiao et al., 1998).

Other genes expressed by cells of the osteoblast lineage In addition to morphological and histochemical criteria, osteoblasts at different stages of maturation can be distinguished by studying their immunohistochemical and mRNA expression characteristics. This is useful, not only, in osteoblastic differentiation studies but also in a variety of investigations where precise information regarding cellular identity is of importance. To this end a variety of cellular markers have been employed, of which alkaline phosphatase, Cbfa1 and osteocalcin have previously been discussed. Figure 2 illustrates a number of these. In the following section selected key markers shall be discussed, while others, such as Msx2 and Twist, that are relevant to this study, shall be discussed later in their own section.

Sialoproteins Bone sialoprotein (BSP-II, BSP) and osteopontin (Spp, BSP-I, OPN) are highly sialylated, non-collagenous, bone matrix glycoproteins. They contain the amino acid sequence Arg-Gly-Asp which allows them to bind to the integrin class of cell surface receptors.

Bone sialoprotein Bone sialoprotein is intensely expressed by osteoblasts and recently entombed osteocytes. It marks a late stage of osteoblast differentiation and an early stage of matrix production. Lower levels of expression are also seen in chondrocytes in hypertrophic cartilage as well as in odontoblasts, cementoblasts and osteoclasts. The only non-skeletal or non-dental cells that express Bsp are the placental trophoblasts and breast carcinoma cells, both of which form mineralised foci (Bianco et al., 1991; Chen et al., 1991; Bellahcene et al., 1994). From its expression pattern, its ability to bind collagen and calcium ions as well as promote hydroxyapatite crystal formation in-vitro, BSP is thought to have an intimate role in the mineralisation of connective tissues (Fujisawa et al., 1995; Goldberg et al., 1996). BSP is also capable of mediating attachment of several cell types including fibroblasts, osteoclasts, chondrocytes and osteoblasts, and this is most likely through the

16 vitronectin (α β ) receptor (Somerman et al., 1988; Helfrich et al., 1992; Loeser, 1993; Mintz et al., 1993).V 3 In addition, there is evidence that BSP possesses cell proliferative and migratory properties (Sung et al., 1998). Interestingly, BSP also stimulates in-vitro bone resorption, in part by enhancing osteoclast adhesion to bone (Raynal et al., 1996). Given this knowledge, BSP is still extensively used as a marker of osteoblasts in which it is abundantly expressed.

Osteopontin Osteopontin possesses a broader spectrum of expression than that of BSP. Osteopontin is detected in osteoblasts, as well as osteoclasts and hypertrophic chondrocytes. Expression has also been reported in preosteoblasts and stromal cells. In addition, osteopontin is found in a number of non mineralising tissues and is expressed by a number of different cell types including fibroblasts, lymphocytes and macrophages (Mark et al., 1988; Nomura et al., 1988; Patarca et al., 1989; Weinreb et al., 1990; Chen et al., 1991; Tezuka et al., 1992). Numerous functions have been ascribed to osteopontin, including the regulation of crystal size during mineralisation (Boskey et al., 1993), attachment of both osteoblasts to the extracellular matrix and osteoclasts to the resorption site (Somerman et al., 1987; Reinholt et al., 1990a; Dodds et al., 1995; Liaw et al., 1995), and a role in cell migration (Liaw et al., 1995). Interestingly, mice lacking osteopontin as well as double mutants lacking both osteopontin and vitronectin, with which it shares a receptor, had no reported skeletal phenotype. Intriguingly, the osteopontin mutants did have defective skin wound healing (Liaw et al., 1998).

Type I collagen Type I collagen is the major extracellular matrix protein of bone and constitutes approximately 90% of the total organic component of mature bone. It has a supercoiled triple-helical structure, consisting of two α1(I) chains and a structurally similar but genetically different α2(I) chain. Type I collagen forms the basic fibrillar network of bone matrix (Rossert and de Crombrugghe, 1996 review). Although synthesised by many cell types, as type I collagen is a primary product of osteoblasts during bone matrix production it has been extensively used as a marker of the osteoblast lineage. Type I collagen is expressed by osteoprogenitors, preosteoblasts and more notably by mature osteoblasts (Sandberg et al., 1988; Niederreither et al., 1995). Type I collagen is required for the adhesion, migration, proliferation, and differentiation of a number of cell types including osteoblasts. In addition, it has also been demonstrated that type I collagen/ integrin interaction is required for osteoblast specific gene expression (Takeuchi et al., 1996; Xiao et al., 1998). A large number of mutations in the two genes encoding the two distinct α chains of type I collagen have been identified in patients suffering from , which is characterised by bone deformity and fragility, and in some patients also by developmental dentine defects. Another connective tissue disorder, Ehlers-Danlos syndrome type VII, has also been assigned to the α1(I) gene (Sillence et al., 1987; Olsen, 1995 review).

17 Type III collagen In most tissues types I and III collagen are co-expressed (Niederreither et al., 1995). Unlike type I collagen, type III collagen is expressed by osteoprogenitors and preosteoblasts but not by mature osteoblasts (Sandberg et al., 1988; Niederreither et al., 1995). Type III collagen is also transiently expressed by odontoblasts (Lukinmaa et al., 1993). Interestingly, Type III collagen has been detected in the mid sutural mesenchyme of both calvarial and facial sutures, where it has been linked to mechanical stress (Sandberg et al., 1988; Yen et al., 1989).

Osteoclasts Morphology and histochemistry The cells primarily responsible for bone resorption are the osteoclasts. These are multinucleate giant cells attached to bone. Adjacent to the bone surface the plasma membrane is convoluted into finger-like projections, and beneath this characteristic ruffled border degradation of mineralised bone commences (Holtrop and King, 1977; Roodman, 1996 review). This region is sealed off by the cell membrane being closely opposed to the bone surface, creating a resorption region underneath the cell into which proteolytic enzymes and hydrogen ions can be released (Väänänen and Horton, 1995). Degradation products are endocytosed at the ruffled border to form vesicles which are transported through the cell and into the extracellular space (Salo et al., 1997). Mineral dissolution involves acidification of this resorption vacuole. Hydrogen ions are generated in the osteoclast by the enzyme carbonic anhydrase II, which are transported across the ruffled border by a vacuolar H+-ATPase proton pump (Väänänen et al., 1990). The resorption of bone involves not only removal of its mineral component but also degradation of its organic matrix. Osteoclasts express a number of proteolytic enzymes, namely lysosomal cysteine proteinases (cathepsins) and matrix metalloproteinases (MMPs) that are involved in this process (Hall and Chambers, 1996). Osteoclasts express high levels of the metalloproteinase gelatinase B (MMP-9, 92-kDa type IV collagenase) (Reponen et al., 1994; Tezuka et al., 1994; Okada et al., 1995). Gelatinase B is active against denatured collagen (gelatin) and specific locus type IV collagen, types III, V, VII, and XI collagen as well as elastin. It does not degrade native type I collagen, proteoglycan, or laminin, and it is hypothesised that gelatinase B has a role in the removal of denatured collagen fragments, following the action of other MMPs and cysteine proteinases as an early step in bone resorption (Everts et al., 1992; Murphy, 1995 review). In addition, it has been suggested that gelatinase B is obligatory for the migration of TRAP positive (pre)osteoclasts from the periosteum, and is a major component of the mechanism that determines where and when the osteoclasts will attack bone (Blavier and Delaissé, 1995). Gelatinase B is by no means exclusively expressed by osteoclasts and is frequently detected at sites of invasion, erosion or tissue remodelling (Salo et al., 1991; Canete-Soler et al., 1995). It is expressed by trophoblasts during embryo implantation, in macrophages associated with invading tumours, and by human T lymphocytes enabling them to exit the circulatory system and pass through the endothelial basement membrane (Pyke et al., 1993; Weeks

18 et al., 1993; Alexander et al., 1996). Homozygous mice with a null mutation in the gelatinase B gene exhibit an interesting phenotype consisting of delayed vascularisation and ossification in the growth plates of long bones (see section on endochondral ossification). The phenotype is rescued by the transplantation of wild type bone marrow, suggesting that these processes are mediated by cells of marrow origin, which the authors term chondroclasts (Vu et al., 1998). Tartrate-resistant acid phosphatase (TRAP, EC 3.1.3.2.) is a widely used enzyme histochemical marker of osteoclasts and mononuclear preosteoclasts (Minkin et al., 1982; Baron et al., 1986; Clark et al., 1989). TRAP activity is not exclusive to osteoclasts with alveolar macrophages of the lung also exhibiting TRAP activity (Yaziji et al., 1995). Care has also been advised if used as a marker of osteoclast differentiation in marrow culture (Hattersley and Chambers, 1989; Modderman et al., 1991). TRAP belongs to a group of purple acid phosphatases. It is a binuclear iron containing glycoprotein may be secreted into the resorptive vacuole underlying the osteoclast, where it may have a catalytic role in the removal of phosphate prior to hydroxyapatite solubilization (Lau et al., 1987; Reinholt et al., 1990b; Vincent and Averill, 1990; Halleen et al., 1998). Mice deficient in TRAP exhibit a mild form of osteopetrosis, with normal eruption, suggesting that TRAP is not essential for, but contributes to, bone resorption (Hayman et al., 1996). TRAP has also been used as an indicator of the bone turnover, since once osteoclasts have performed their resorptive function they secrete TRAP into the circulation. Here it can be measured and used in the diagnosis and management of certain clinical conditions (Lau et al., 1987; Halleen et al., 1999).

Osteoclast differentiation Osteoclasts are derived from haematopoietic precursors. Although the exact nature of this precursor is not known, Tondravi et al. (1997) have shown that the haematopoietic transcription factor PU.1 regulates the initial stages of osteoclast development. They generated mice lacking PU.1 and found that the development of both osteoclasts and macrophages, cells also of myeloid origin, were arrested. Furthermore, these mice exhibited an osteopetrotic phenotype which was rescued by bone marrow transplantation. The osteopetroses are a family of diseases in which bone resorption by osteoclasts is disrupted (Key and Ries, 1996 review). A review of other mutants exhibiting osteopetrotic phenotypes has increased our understanding of osteoclast development (Figure 3). The spontaneous osteopetrosis (op) mutant is caused by an absence of colony stimulating factor 1 (CSF1). Here, osteoclasts and their progenitors, as well as macrophages are indirectly targeted (Wiktor-Jedrzejczak et al., 1990; Yoshida et al., 1990). The inability of marrow transplantation to cure this form of osteopetrosis suggests that CSF1 is normally provided by cells external to the haematopoietic system (Erlebacher et al., 1995 review). NFκB is a family of dimeric transcription factors composed of various combinations of structurally related proteins: p50 (NFκB1); p52 (NFκB2); p65 (RelA); C-Rel (Rel) and relB. Double knockouts of the p50 and p52 subunits produced mice with a severe osteopetrotic phenotype, but here osteoclast differentiation is affected at a slightly later stage compared to the mice described above. In these mice macrophages develop but have impaired function, while osteoclasts as well as B cells fail to develop

19 (Franzoso et al., 1997). The targeted disruption of the proto-oncogene c-fos, a component of the Activating Protein-1 transcription factor expressed by differentiating osteoclasts, has an impact later in osteoclast differentiation with macrophages developing normally (Wang et al., 1992; Grigoriadis et al., 1994). Interestingly, c-fos is also thought to play a role in osteoblast and chondroblast differentiation and function (Grigoriadis et al., 1996). This phenotype, as well as that of the natural mutant, microphthalmic mouse (mi/mi) and the c-src knockout mouse is rescueable by bone marrow transplantation (Walker, 1975; Lowe et al., 1993). Both the mi/mi and the c-src knockout mice have osteoclasts present but these cells fail to resorb bone. The mi gene encodes a bHLH transcription factor that could act downstream of PU1 and c-fos (Hodgkinson et al., 1993a). The severe osteopetrotic phenotype in the c-src knockout mouse is due to a failure of osteoclasts to form a ruffled border and hence they malfunction (Soriano et al., 1991; Lowe et al., 1993).

OPGL

Monocyte Macrophage

CSF1

38 1)%κ FIRV FVUF Gelatinase B PL TRAP Carbonic anhydrase II H+-ATPase

Determination Polarization Resorption

Figure 3. Schematic illustration of osteoclast differentiation with key transcription factors and growth factors. Much of our knowledge of osteoclast differentiation comes from studying mutant mice exhibiting osteopetrotic phenotypes.

20 Overexpression of osteoprotegerin (OPG, also called OCIF), a secreted glycoprotein and member of the tumour necrosis factor (TNF) receptor family, leads to an increase in trabecular bone density due to a decrease in the number of mature osteoclasts (Simonet et al., 1997). These mice have normal numbers of osteoclast precursors suggesting that OPG may regulate late stages of osteoclast differentiation. This is reflected in their phenotype which, unlike other osteopetrotic mice, do not possess gross defects in bone size or shape, nor in tooth eruption (Simonet et al., 1997). Conversely, OPG knockout mice exhibit an osteoporotic phenotype with an increase in osteoclast recruitment and activation leading to a decrease in bone density, a thinning of the parietal bones and a high incidence of fractures (Bucay et al., 1998). OPG, therefore, inhibits osteoclast formation. The OPG ligand (OPGL, also called TRANCE, RANKL, ODF) is a member of the TNF family and is found both as a transmembrane receptor and as a soluble cleaved form (Lacey et al., 1998; Yasuda et al., 1998). OPGL is both an osteoclast differentiation and activation factor and is able to bind to the surface of osteoclast progenitors. It is not produced by osteoclasts, but by mesenchymal cells and by osteoblasts. In contrast to other cytokines, OPGL stimulates osteoclast differentiation directly, so that contact between osteoblasts/stromal cells and differentiating osteoclasts is needed for its effect to be exerted via the OPGL receptor (Figure 4). OPGL also stimulates osteoclast activity and this is dependent on CSF1. Interestingly, CSF1 can induce OPGL receptor expression in osteoclast precursors (Lacey et al., 1998). OPG (Also called RANK) is secreted by osteoblasts, it has only been found in secreted form and has no transmembrane region. It inhibits osteoclast differentiation and this is presumably by competing with OPGL receptor for OPGL (Yasuda et al., 1998).

Osteoclast progenitor Mature osteoclast

.H\

OPGL receptor

OPGL Osteoblasts and Mesenchymal cells OPG

21 Figure 4. Diagram illustrating a proposed model for the actions of OPG, OPGL and OPGL receptor in osteoclast differentiation and maturation. OPGL is produced by mesenchymal cells and by osteoblasts. It is an osteoclast differentiation and activation factor and acts via OPGL receptor found on osteoclast precursors. Thus, direct contact between osteoblasts/stromal cells and differentiating osteoclasts is needed for its effect to be exerted. OPG acts as an inhibitor of osteoclastogenesis. It is produced by osteoblasts, it has only been found in secreted form and has no transmembrane region. Presumably, it inhibits osteoclast differentiation by competing with OPGL receptor for OPGL.

Proliferation of bone cells Multi-potential mesenchymal cells, from which osteoblasts differentiate, as well as osteoprogenitors and, to a limited extent, preosteoblasts, possess the capacity to divide. Osteoblasts, osteocytes and bone lining cells are post-proliferative. During bone development and remodelling cellular proliferation involves the sequential expression of specific genes; these are mediated by a host of factors that include, but are not restricted to, growth factors, hormones and steroids. Only selected examples of these, affecting the very early stages, will be mentioned here. Leukaemia inhibitory factor (LIF) inhibits the differentiation of embryonic stem cells in general and permits their propagation as pure populations (Nichols et al., 1990). LIF may be an important factor in the control of the osteoblastic stem cell and early osteoblast determination and proliferation (Triffitt, 1996 review). Indeed, in the in-vitro calvarial cell model it has been shown that LIF inhibits osteoprogenitor differentiation and expression of osteoblast associated genes. Interestingly, dexamethasone, a potent stimulator of bone nodule formation, antagonises the action of LIF in this system (Malaval et al., 1995). Interleukin 6, 10 and 11 are also thought to be of importance in maintaining stem cell proliferation. Of the multitude of other growth factors having proliferative effects, Platelet derived growth factor (PDGF) and Epidermal growth factor (EGF) are among the most potent. Interestingly, mice where the PDGF α receptor has been deleted exhibit a calvarial bone phenotype (Table 3). Both FGF2 and TGFβ are mitogens for marrow stromal cells and may interact via the same cell surface proteoglycan, betaglycan. Both are also inhibitors of osteoblast differentiation (Triffitt, 1996 review).

Proliferation during calvarial bone and suture development Cell proliferation in the developing calvaria has been investigated by both 3H thymidine and 5'-bromo-2'-deoxyuridine (BrdU) incorporation. In E16 mouse calvaria BrdU uptake correlated well with the expression of Fgfr2 mRNA. And this was adjacent to, but mutually exclusive with, areas of osteoblast terminal differentiation. This was particularly pronounced at the osteogenic fronts, in osteogenic precursors, as well as in the outer layer of the developing meninges (Iseki et al., 1997). Similar results were seen in another elegant study, in which proliferating cells were mainly detected in the osteogenic fronts and in the periosteum with little BrdU incorporation in the mid-sutural mesenchymal cells of the sagittal suture of newborn and P4 mice (Liu et al., 1999). When the dura mater is removed from beneath the developing rat coronal suture (E19) and cultured for 5 days, the sutural space becomes partly obliterated by the encroaching frontal and

22 parietal bones. Interestingly, the rate of DNA synthesis is elevated in sutural cells between 3 and 4 experimental days, just prior to the obliteration of the sutural space, and thereafter declines in comparison to controls. The authors of this study therefore postulate that this abnormal proliferation may be a crucial step in sutural closure and that it may be due to the removal of a signal in the dura, which normally functions to inhibit sutural proliferation (Opperman et al., 1998). However, osteoblastic cells harvested from human sutures that have fused prematurely, have a similar proliferation rate to those harvested from a normal suture in the same patient (de Pollack et al., 1996). In light of the Opperman et al. study (1998), this is not altogether unsurprising, since they found that the proliferation rate returns almost to normal levels prior to sutural obliteration. The connection of cell proliferation and suture closure has been further studied in mice overexpressing Msx2. Gain of function mutations in MSX2 are known to cause premature calvarial bone fusion in humans and excessive and ectopic growth of the calvarial bones in transgenic mice. Mice overexpressing Msx2 have been found to exhibit an increase in the number of BrdU-labelled cells in the osteogenic front of the postnatal sagittal sutures. These are early stage osteoblasts or osteoblastic precursors, and this has lead to the hypothesis that the pool of potential osteoblasts is initially increased which is subsequently followed by a increase in bone formation (Liu et al., 1999).

Apoptosis during embryogenesis In contrast to necrosis, apoptosis is a form of cell death that is programmed. Apoptosis is essential for normal embryogenesis, that is to say, it is a non pathological process where the cell is thought to activate its own suicide (Raff et al., 1993). It is characterised by cell shrinkage, cleavage of genomic DNA by activation of endogenous proteases, nuclear and cytoplasmic condensation, as well as fragmentation of the nucleus and cytoplasm into membrane bound vesicles or apoptotic bodies (Martin et al., 1994). This is in contrast to necrosis, where the cell swells, mitochondrial changes occur, the organelles disintegrate and the cell membrane disrupts. Necrosis is often associated with tissue damage and inflammation (Schwartz and Osborne, 1993 review). Apoptosis is an important developmental mechanism, notably contributing to embryonic morphogenesis. The process of apoptosis may regulate the size of a cell population, eliminate cells that have fulfilled their transient purpose, remove cells that have migrated into the wrong place or select out ‘weak’ or malfunctioning cells (Raff, 1992; Raff et al., 1993). The role of apoptosis in the regulation of a size of a cell population is exemplified in the skin. Here, the removal of epidermal cells has an important function in establishing the final architecture of the human epidermis and its appendages (Polakowska et al., 1994). Examples where apoptosis may take part in tissue morphogenesis come from the tooth and also from the developing limb (Vaahtokari et al., 1996; Jernvall et al., 1998). In the tooth a collection of non-dividing cells in the dental epithelium, called the enamel knot, is thought to play a role in the folding of the dental epithelium and thus the determination of tooth shape. The enamel knot is known to express several growth factors including Fgfs, Bmps and Shh, and it is thought to function as a signalling centre. The enamel knot is removed by apoptosis; thus this process terminates its function as a signalling centre. Termination of function by the elimination

23 of cells through apoptosis is common to other signalling centres including the apical ectodermal ridge in the developing limb. Interestingly, in the limb, apoptosis is also necessary for the separation of the digits (Coelho et al., 1993).

Apoptosis during bone and suture development Apoptosis in bone and suture development has been less well studied than in other areas. Apoptosis has been reported both in isolated osteoclasts (Fuller et al., 1993; Hughes et al., 1995; Kameda et al., 1995; Selander et al., 1996), as well as in-vivo, in rodent long bones and mandibles (Boyce et al., 1995; Hughes et al., 1995; Bronckers et al., 1996; Hughes et al., 1996; Vaahtokari et al., 1996). Osteoblasts, too, have been observed to undergo apoptosis (McCabe et al., 1995; Hill et al., 1997), and there is some evidence that osteocytes undergo apoptosis and that at least some are subsequently phagocytosed by macrophages or osteoclasts, in which apoptotic mononuclear bodies are later seen (Elmardi et al., 1990; Bronckers et al., 1996). During bone development and growth many osteoblasts develop but a much smaller number end up as osteocytes. It is therefore speculated that apoptosis in osteoblasts may account, at least in part, for this discrepancy. However, at present there is little evidence for this. Apoptosis has been observed to occur in the sutural mesenchyme of both the rat and mouse (Ten Cate et al., 1977; Fürtwangler et al., 1985; Bourez et al., 1997). Apoptotic bodies were detected when the bony fronts of the suture came into close contact, thus this could prove to be an active process aimed at maintaining sutural integrity, thereby preventing the premature fusion of the two bones involved (Fürtwangler et al., 1985). Indeed, apoptosis of bone cells may prove to be important in the calvarial modelling and remodelling processes, as well as possibly elsewhere in the skeleton. The developing calvaria with its rapid cellular turnover might well provide a good model for studying apoptosis in mesenchymal cells, osteoblasts and osteoclasts. It may transpire that findings from this model may then be applied to other less easily studied areas and therefore more generally to the field of bone biology.

Development of the Calvaria

Developmental anatomy The bones of the skull can be divided up into those of the face, the viscerocranium, and those surrounding the brain, the neurocranium. The neurocranium can be subdivided into the base of the skull and the calvaria (cranial vault, vault of the skull or skull cap). The calvaria is principally comprised of two frontal bones, two parietal bones and an interparietal bone (Figure 5). There are also contributions from the squamous part of the temporal bone and the supraocciput. Two centres of ossification are seen within the interparietal bone which subsequently fuse across the midline. The interparietal bone fuses along its caudal border with the supraoccipital bone to form the squamous part of the occipital bone. Interestingly, the interparietal bone is formed by intramembraneous ossification while the supraoccipital

24 bone by endochondral ossification. The temporal bone also has a complex origin. Its tympanic and squamous portions are formed by intramembraneous ossification while the petrous part has a cartilage precursor (Kaufman and Bard, 1999 review). In the human foetus, the frontal, parietal and interparietal bones commence ossification between 30 and 37mm crown rump length corresponding to approximately 8-9 weeks of gestation. Each of these bones apparently possess two ossification centres (Lemire, 1986 review). In the chick, calvarial bone formation starts after the 12th day of incubation and in the mouse at E13.5 (Lengele et al., 1990).

Sutures and fontanelles Sutures are joints, between bones, that in vertebrates are found only in the craniofacial complex. As intramembraneous bones increase in size their borders approximate, and at this point, instead of fusing, they form a suture. Thus a suture is a joint with two bone fronts interposed with mesenchyme or fibrous tissue. Once a suture has been established either apposition or resorption can occur at the bone ends as the demands of each situation befit, therefore permitting adjustments in the size, shape and spatial orientation of contiguous parts of the craniofacial skeleton during development and growth. The sutures can also adapt to pathological disturbances such as hydrocephalus, in which the calvaria expands secondary to an increased intracranial pressure. Thus the growth of the calvaria and the underlying brain are highly co-ordinated. Sutures are generally named according to either their orientation or location, for example the coronal or sagittal, or the bones between which they lie, for example the frontonasal or the midpalatal suture. Those which will be concentrated on in this study are the coronal sutures (paired) between the frontal and parietal bones, the (single) between the two frontal bones (also termed metopic, particularly in humans), the sagittal suture (single) between the two parietal bones, and the lambdoid sutures (paired) between the parietal and interparietal bones. These sutures are illustrated in Figure 5. In the frontal and sagittal sutures the bone ends confront each other in an end-to-end manner, like a butt joint, while in the coronal suture the bones overlap. The lambdoidal suture is generally of an end-to-end morphology but its bones may also overlap. At the most lateral extension of the coronal suture both the frontal and parietal bones are overlapped by the temporal bone (Johansen and Hall, 1982). Fontanelles have a similar structure to sutures but are generally formed where three calvarial bones meet. They are generally wider than sutures and close to the width of a suture soon after birth. Sutures, and more notably, fontanelles allow some degree of movement between bones, and this allows some deformity of the calvaria during the birth process, which subsequently recovers post partum. In rodents there is a bony ridge, the jugum limitans olfactoriae, running paracoronally that divides the frontal bone into anterior and posterior regions; in sections it is seen as a thickening notably on the endocranial surface.

25 O S F I Q IV DQWI VLSSI Q I S MO

Figure 5. Schematic illustration of the bones, sutures and fontanelles of the calvaria in an E17 mouse. n nasal bone, f frontal bone (jl jugum limitans), fs frontal suture, p , ip interparietal bone, c coronal suture, s sagittal suture, l lambdoid suture, a.font anterior , pf posterior fontanelle.

Suture morphology As osteogenesis in the two parietal bones spreads out from their ossification centres, the first areas to confront each other, at the summit of the calvaria, are in the central region of the presumptive suture, if viewed from above. Hence, once the sagittal suture has been established, the central region is more advanced than anterior or posterior sections (Decker and Hall, 1985). The sagittal suture develops in a sulcus, deeper posteriorly than anteriorly, presumably because it is nestled between the two cerebral hemispheres. This means that before a butt joint morphology has been established, soon after birth, the osteogenic fronts bend down towards meninges. At the osteogenic fronts, preosteoblasts divide and differentiate into osteoblasts, they can be distinguished morphologically from the surrounding mesenchymal cells or fibroblasts. Mineralisation occurs extracellularly with the initial deposits of apatite crystals exhibiting no particular orientation with respect to either cellular or extracellular collagen fibres (Decker and Hall, 1985). The individual osteogenic fronts of the coronal suture have a similar appearance to those of the sagittal suture. However, the osteogenic fronts and later the bones themselves overlap, with the frontal bone endocranial to the parietal. This pattern is set prior to the overlap with the two presumptive bones lying in different planes. The frontal and parietal osteogenic fronts first approximate on the lateral aspect of the calvaria. Suture formation then progresses medially, toward the midline, in a zipper-like fashion. This morphogenesis is reflected in the histological maturation of the suture with more advanced maturation being exhibited laterally rather than medially (Johansen and Hall, 1982). Once the osteogenic fronts have approximated, the intervening mesenchymal tissue increases in thickness to form a highly cellular

26 ‘blastema’. Finally, a fibrous central zone appears between the two opposing bones, heralding the ‘mature’ suture (Markens, 1975b). Accumulations of cartilage can occur in the mesenchyme of developing sutures (Markens, 1975b). These are generally transient, eventually being transformed back into fibrous tissue, fibrocartilage or into bone. Rarely such cartilages may be found stenosing a suture (Kokich, 1986 review). These are often referred to as secondary cartilages having not been derived from the ‘primary’ cartilaginous skeleton (Schaffer, 1930). If a cartilage develops close to the sagittal suture, it may take the form of a rod, elliptical in cross section. The cartilage tends to reside in the endocranial sector of the mesenchyme above the sagittal venous sinus. It may be present just prior to birth and then disappear shortly post partum. Pritchard et al. (1965) suggested that in the rodent this cartilage may be a forward extension of the tectal region of the chondrocranium. ‘Chondroid tissue’ has also been described at the sutural edge, notably in the human metopic suture, where it has been linked to sutural fusion (Manzanares et al., 1988). Chondroid tissue is also detected in the primitive anlage of the calvarial bones of the chick, bone being deposited mainly on the external surface of this chondroid primordium (Lengele et al., 1990). It is known that stimuli such as mechanical stress, ischemia and anoxia can enhance mesenchymal cell differentiation into chondroblasts, while mechanical tension, a adequate blood supply and hyperoxia can favour differentiation into osteoblasts (Hall, 1970). These factors may be of importance in determining which route an uncommitted mesenchymal cell takes. Interestingly, greater amounts of cartilage can appear in connection with calvarial fractures, when the blood supply to an area is intentionally reduced (Pritchard et al., 1956).

Suture closure In the mouse, all calvarial sutures, except for the posterior section of the frontal suture and the occipito-interparietal, remain patent. This includes the coronal, lambdoidal, sagittal and the anterior section of the frontal suture. The posterior section of the frontal suture fuses between 25 and 45 days postnatal, and it does this in an anterior to posterior manner (Bradley et al., 1996). The Sprague-Dawley rat exhibits a similar pattern with the posterior section of the frontal suture fusing between 12 and 30 days postnatal. Here fusion starts on the endocranial side and progresses outwards (Moss, 1958; Opperman et al., 1997; Roth et al., 1997). However, in the rat, localised areas of , especially in the sagittal suture, can apparently occur at any time after the twenty first postnatal day (Pritchard et al., 1956). Except for the metopic (frontal) suture which is obliterated by the third year, most human calvarial sutures fuse in the third or forth decade of life. Interestingly, most facial sutures remain open late into adult life. This possibly reflects greater functional stimuli placed on them (Kokich, 1986 review).

Suture position The site where a suture forms is determined by the relative growth of adjacent cranial bones (Moss, 1954; Pritchard et al., 1956; Johansen and Hall, 1982). Some investigators have also suggested that the dura mater can dictate where a suture is

27 formed, and that this is in response to tension in the dura, as a result of neurocranial expansion, directed via the basicranial processes (Moss, 1975; Smith and Töndury, 1978). Also, the blastema has been proposed as playing a role in determining the location of a suture, though this appears to form after the position of the suture has been set (Markens, 1975b).

Sutural growth Growth of the bones that make up a suture may occur in broadly equal amounts in each bone and may be at right angles to the suture line. This occurs in some midline sutures. However, data from studies where metal implants have been placed on either side of a suture, and growth monitored, indicate that this is not always the case (Persson, 1995 review). In the frontonasal suture, apposition on the frontal bone side is five fold greater than that on the nasal bone side. Indeed, one of the sutural margins may be resorptive. Also, in relation to each other, bones may either slide, as is seen in the nasopremaxillary suture in rats, or rotate about a sutural line. This is demonstrated by the two maxillary bones in the horizontal plane. Thus sutures can respond to functional demands placed upon them. Indeed this is exemplified when individual sutures fuse prematurely. Compensatory growth occurs at other sutures and this may be asymmetric with apposition at one bone end greater than its partner (Jane and Persing, 1986).

Role of the dura mater Attention has been focused on the possible role of the dura in calvarial bone development, sutural morphogenesis and sutural closure. Following the microsurgical removal of various parts of the embryonic chick brain and notochord, the neuroepithelium was shown to have an inductive role in cranial osteogenesis (Schowing, 1968). Calvarial bone osteogenesis is further linked to the meninges by way of the congenital hydrocephalus (ch) natural mutant mouse, in which both calvarial bone and meningeal defects are seen (Table 3). In the ch mouse, these abnormalities are part of a more general skeletal phenotype, including a shortened skull base. These mice fail to develop calvarial bones, and show abnormalities in the meningeal mesenchyme including defective arachnoid cells (Grüneberg, 1953; Grüneberg and Wickramaratne, 1974; Kume et al., 1998). The mouse mutation is caused by a nonsense mutation in the gene encoding winged helix/ forked head transcription factor, Mf1 (Kume et al., 1998; Hong et al., 1999). Mf1 is expressed in the developing head mesenchyme, meninges and blood vessels in the neuroepithelium (Hiemisch et al., 1998; Kume et al., 1998). In the absence of the underlying dura the developing coronal suture undergoes stenosis. This was demonstrated in-vivo by transplanting developing rat coronal sutures, with and without the dura mater, into prepared defects in the centre of adult parietal bones. These host sites still had their dural membrane intact and were considered to be free of putative biomechanical influence. After 3 weeks, in the absence of transplanted dura, sutures were obliterated by bone, while in the presence of dura, the sutures remained patent. The authors suggested that tissue interactions of a biochemical rather than biomechanical nature were required to maintain the coronal suture in a patent and

28 functioning state (Opperman et al., 1993b). Similar transplant experiments were performed where the periosteum ectocranial to the coronal suture was removed. Unlike the removal of the dura mater this did not cause sutural obliteration (Opperman et al., 1994). These findings were further supported by in-vitro experiments where developing coronal sutures were cultured in chemically defined serum free conditions for up to seven days. Some explants had had the underlying dura removed while others had the dura still intact. Explants with intact dura remained patent while those without underwent bony fusion. In addition, bony fusion was prevented when explants that had had the dura removed, were co-cultured with dura on the opposite side of a polycarbonate filter. Interestingly, coronal suture explants stripped of their underlying dura could be rescued, remained patent, if cultured in medium conditioned with a signal secreted by explants with the dura intact (Opperman et al., 1996). These data suggest that the influence of the dura mater on suture obliteration is mediated by soluble factors rather than cell-to-cell or cell-to-matrix interactions (Opperman et al., 1993a; Opperman et al., 1995). Interestingly, in this fusing suture model, an increase in cell proliferation and also collagen production prior to bony union has been observed (Opperman et al., 1998). The mouse posterior frontal suture fuses between 25 and 45 days postnatal. If posterior frontal sutures, from 24 day postnatal mice, are placed into in-vitro culture with their dura mater intact they undergo fusion in a manner that mimics the in-vivo situation. However, if the same culture is performed where the underlying dura has been removed the suture remains patent. This indicates that the dura mater is important, not only in keeping cranial sutures open but also during suture closure (Bradley et al., 1996). Similar conclusions can be drawn from in-vivo experiments performed on the rat posterior frontal suture. A procedure where a barrier of sialastic sheeting was surgically placed between the suture and the underlying dura mater resulted in the suture, which normally fuses, remaining patent (Bradley et al., 1996). The role of the dura mater was further investigated, in-vitro, when mouse sagittal suture explants were cultured together with dura from beneath the posterior frontal suture. The sagittal suture, which normally remains patent, fused. In the opposite experiment, when the posterior frontal suture was cultured with dura from beneath the sagittal suture, instead of the suture fusing it remained patent (Bradley et al., 1997). Thus there appears to be a signal from the dura underneath the posterior frontal suture that encourages stenosis. Alternatively, there could be a signal from the dura under the sagittal suture which maintains it in a patent state and that this signal is lost from the dura under the posterior frontal suture. These findings were confirmed by the same group, in- vivo, in a rat model (Levine et al., 1998). Furthermore, there is evidence that soluble, heparin-binding factors secreted by the dura, act to inhibit osteogenesis and suture fusion (Opperman et al., 1996). Indeed, an osteogenesis inhibiting factor maintaining suture patency has long since been proposed (Markens, 1975a). However, it was previously suggested that the intact foetal suture transmitted signals to the underlying dura mater, with the then imprinted dura being capable of regenerating a suture in, for example, cases of calvariectomy (Mabbutt and Kokich, 1979).

29 Biomechanical forces Biomechanical forces have been postulated to as contributory factors in the aetiology of craniosynostosis. It has been suggested that abnormalities in the cranial base can lead to tension in the dura mater which could then have an effect on the overlying suture (Moss, 1960; Smith and Töndury, 1978). Indeed, fronto-occipital deformation with boards, practised by Mayan Indians, has been found to result in sagittal suture synostosis and the formation of in the lambdoid suture (White et al., 1996). Also, calvarial deformation believed to be caused by sleeping posture is thought to result in deformational (Huang et al., 1996). Models studying the compressive and tensional forces on sutures have not only been of interest to those studying premature suture closure but also to orthodontists and others concerned with craniofacial growth and its possible manipulation. The application of tensile mechanical stress to the cranial sutures of newborn rabbits leads to an increase in total protein and collagen synthesis. Interestingly, Type III collagen, not normally found in large amounts in postnatal sutures, is upregulated in these joints when stressed (Meikle et al., 1979, 1982; Yen et al., 1989).

Pathobiology of calvarial disorders

I shall, for the sake of clarity, divide the following sections, broadly, into disorders typified by precocious calvarial bone formation and premature suture fusion (craniosynostosis), and into disorders with reduced calvarial bone formation and delayed suture formation and subsequent closure. However, it should be emphasised that some disorders, such as Apert syndrome, which is characterised by a lack of suture development as well as craniosynostosis later, do not fall neatly into these compartments.

Premature suture fusion, craniosynostosis Craniosynostosis is the process of premature cranial suture fusion, although the term is commonly used to describe the result, for which the term craniostenosis may also be used (Cohen, 1986a review). Though, less common, synostosis can also affect the facial skeleton with the sphenofrontal, frontoethmoidal and frontonasal sutures being most commonly affected (Cohen, 1986b review). Craniosynostosis occurs in approximately 1 in 2500 live births (Wilkie, 1997 review). It is a heterogeneous condition which may be the result of a number of different causes all leading to the same final pathological condition. It may be an isolated finding (non syndromic) or be part of a syndrome (syndromic) such as Apert, Carpenter or Crouzon syndromes. Craniosynostosis may also be secondary to another known disorder. Examples include external pressure or alternatively microcephaly and shunted hydrocephaly causing sutural compression, holoprosencephaly, metabolic conditions such as hyperthyroidism or rickets, as well as haematological disorders such as thalassemia or sickle cell anaemia. Over 100 craniosynostotic syndromes have been described, many of which are

30 inherited in a dominant fashion, such as Apert, Crouzon, Pfeiffer, Saethre-Chotzen and Jackson-Weiss syndromes. Others, such as Carpenter and Antley-Bixler syndromes, exhibit recessive inheritance (Cohen, 1986c review; Winter and Baraitser, 1996). The salient clinical features of some syndromal craniosynostoses are summarised in Table 1. Growth of the calvarial bones is tightly co-ordinated with growth of the underlying brain, such that if one or more sutures are disrupted, growth is redirected to accommodate the expanding brain. However, this can produce its own detrimental effects, including abnormalities in skull shape, an increase in intracranial pressure possibly leading to ophthalmic and neurological problems, as well as a host of other abnormalities, together with their concomitant psychosocial difficulties (Renier et al., 1982; Cohen, 1986b review). The changes in skull shape are, to some extent, predictable if the sutures involved are known, and also vice versa, so that the pattern of sutural involvement may be concluded from the skull shape. Also reflected in skull shape is the time at which individual sutures have undergone synostosis and the order in which it has occurred. Premature sutural fusion results in a cessation of growth in the affected suture, bone growth being, generally, arrested at right angles to the affected area, with compensatory growth in the same direction as the closed suture (Virchow, 1851; Cohen 1986b review). Thus, if synostosis of the sagittal suture occurs, lateral calvarial growth is limited and growth is redirected anteroposteriorly, resulting in an elongated, boat-shaped skull ( or ). Bilateral synostosis of the coronal sutures limits growth anteroposteriorly, with growth redirected laterally, resulting in a wide short skull (). If affected early enough, growth can be redirected upward to produce a pointed skull (, acrocephaly or turricephaly). Unilateral coronal or lambdoidal suture synostosis results in oblique growth, notably of the contralateral suture, causing an asymmetric skull (plagiocephaly), and bilateral lambdoidal synostosis results in flattening of the occipital region. Premature fusion of the metopic suture is rarer and results in ridging of the and a skull pointed anteriorly (). A variety of combinations of different sutural fusions may occur, producing a variety of shapes including a cloverleaf skull, where the coronal, sagittal and lambdoidal sutures are all affected. Not all abnormal skull shapes resulting from craniosynostosis follow Virchow’s explanations. For example, synostosed with minimal deformity have been observed (Bolk, 1915). Although, the calvaria grows to accommodate the brain, it is not known why some sutures respond and not others. In fact, following sagittal synostosis the skull shape could be normalised if compensatory growth were to occur at the temporo-parietal suture. Instead, the coronal and lambdoidal sutures grow distally and the metopic suture grows bilaterally, resulting in a long egg-shaped skull with frontal and occipital bossing. Similar scenarios can be postulated for other patterns of synostosis (Jane and Persing, 1986).

Aetiology Mutations in six genes have been found to cause craniosynostosis: fibrillin (FBN1); fibroblast growth factor receptors 1, 2 and 3 (FGFR1, FGFR2, FGFR3); muscle segment homeobox 2 (MSX2) and a basic helix-loop-helix transcription factor TWIST. Table 2 lists some of the mutations found in these genes in selected syndromes with calvarial bone phenotypes. One can see in Table 1 that many of these syndromes

31 are not only characterised by craniosynostosis but also by other, notably limb and facial, abnormalities. This is a reflection of the fact that many of the genes involved in limb development also have key roles in craniofacial development. This knowledge, together with an understanding at a molecular genetic level of the processes involved in the development of other animals, chiefly the Drosophila melanogaster and the mouse, has made it possible to use a candidate gene approach when searching for a causative human gene mutation. This approach also relies on an accurate diagnosis, good clinical records and a sound and willing pedigree of the individuals. Each of these genes causing craniosynostosis shall be discussed later in their individual sections.

Delayed suture formation Delayed suture formation and a subsequent delay in suture closure can occur in many disorders (Table 1) including trisomy 21 syndrome, Apert syndrome, cretinism and progeria, as well as when a general retardation of skeletal maturation occurs. However, in cleidocranial dysplasia, a retardation of calvarial bone and suture development, is pathognomonic for the condition. The aetiology of this disorder has started to be unravelled, and this has not only lead to a better understanding of the condition but also broadened our knowledge of osteoblastic differentiation. It will therefore be discussed in a little more detail.

Cleidocranial dysplasia Cleidocranial dysplasia also known as cleidocranial and CCD (caution should be employed when using the abbreviation CCD as it is also used for another condition, Central Core Disease) is a condition characterised by delayed fusion of the calvarial sutures, frontal and parietal bone bossing, supernumerary and unerupted teeth, as well as missing or hypoplastic clavicles. This condition, already reported in the 18th century (Meckel, 1760; Martin, 1765) was designated as cleidocranial dysostosis (Maire and Sainton, 1897) and was later renamed as cleidocranial dysplasia (McKusick and Scott, 1971). The condition has autosomal dominant inheritance and 20-40% present as new mutations (Fitchet, 1929).

Clinical features Individuals with cleidocranial dysplasia are of short stature and have a distinctive appearance (Gorlin et al., 1990 review). The skull is brachycephalic, large but short, with a cephalic index often in excess of 80. There is a delay in suture closure which may be prolonged. Indeed, there maybe a persistent wide opening from the nasion to the interparietal region with the metopic suture, anterior fontanelle and sagittal sutures being involved so that there is a palpable grove. In extreme cases the parietal bone may not be present at birth (Tan and Tan, 1981). Additional ossification centres may form in the sutures which give rise to Wormian bones. These islands of bone presumably coalesce. The cranial base is short but the width of the spheno-occipital synchondrosis is increased. The pronounced frontal and parietal bones with the hypoplastic maxillary, nasal and zygomatic bones make the face look small. The bridge of the nose is depressed and occurs. There is non-union of the mandibular and a high arched

32 palate which may be cleft (Kreiborg et al., 1981; Jensen and Kreiborg, 1993a, b). In general, there are supernumerary teeth and delayed exfoliation of primary teeth as well as delayed eruption in both the primary and secondary dentitions. Eruption is not aided by the extraction of deciduous teeth. The primary teeth and the crowns of the secondary teeth have been found to develop normally, even if delayed in eruption (Jensen and Kreiborg, 1990). Jensen and Kreiborg (1990) also noted that first permanent molars erupted, but permanent teeth with primary predecessors did so only occasionally. Cysts around unerupted teeth are common. Extracted teeth have been found to be deformed with hypoplastic enamel and defects in the cementum. However, perhaps most interesting is the observation that the supernumary teeth are morphologically similar to their predecessors in the secondary dentition, and also that all the teeth in the secondary dentition may give rise to a supernumary tooth such that a complete tertiary dentition is formed (Rushton, 1937; Rushton, 1956; Jensen and Kreiborg, 1990; Lukinmaa et al., 1995). The clavicles are either missing or hypoplastic with the lateral or acromial end being more commonly affected. The clavicle forms by a combination of endochondral (medial section) and intramembraneous ossification (lateral section) (Huang et al., 1997). Clavicular deficiency enables affected individuals approximate their to an abnormal degree. Other deformities include , a reduction in length, hypoplasia of the pubic bones, widening of the and occulta (OMIM 119600).

Aetiology CCD is caused by mutations in the CBFA1 gene (Lee et al., 1997; Mundlos et al., 1997). These include insertional and missense mutations as well as stop codons and deletions. The deletions are of particular note as they show that CCD is caused by haploinsufficiency. Inactivation of one Cbfa1 allele in mice is sufficient to produce skeletal defects that mimic the human condition (Otto et al., 1997). In addition, the Cbfa1 is deleted in a radiation induced mutant mouse (Ccd) whose phenotype also mimics the human condition (Selby and Selby, 1978; Sillence et al., 1987; Otto et al., 1997). Interestingly, in the Ccd mouse condensations of mesenchymal cells form, although reduced in size, but there is no differentiation of chondroblast precursors and altered differentiation of osteoblasts (Huang et al., 1997).

Other conditions Wormian or sutural bones, are small calvarial bones that develop from ossification centres in the sutures or fontanelles. They develop some distance from the calvarial bones within the calvarial mesenchyme, so that ossification centres are initiated de novo, osteoblasts differentiate and lay down bone matrix. In the human, Wormian bones most commonly occur in the lambdoid suture. They are seen in a number of conditions including cleidocranial dysplasia, all types of osteogenesis imperfecta, hypothyroidism and lateral meningocele syndrome (Gorlin et al., 1990 review; Jensen and Kreiborg, 1993a; Gripp et al., 1997). One case of Carpenter syndrome has been reported to have additional small bones in the anterior fontanelle, while in Apert syndrome the wide midline calvarial

33 defect closes by the coalescence of bony islands which form within the defect (Cohen and Kreiborg, 1996). In most cases of hydrocephalus Wormian bones can be detected, and here deforming stress has been proposed as a contributing factor.

Animal models with a calvarial phenotype

A number of both naturally occurring and genetically engineered animal mutants with calvarial bone phenotypes have been described. Those relevant to this study are listed in Table 3. Rabbit models exhibiting craniosynostosis are seen. One shows variability in phenotype with different sutures being involved resulting in a variety of skull shapes including oxycephaly, scaphocephaly and plagiocephaly (see section on craniosynostosis for terminology) (Greene and Brown, 1932). However, the whereabouts of this strain is apparently unknown. Another rabbit strain that exhibits craniosynostosis is well documented with premature fusion seen in the coronal suture and anterior fontanelle only (Mooney et al., 1993). The effects on craniofacial growth have been studied but use of the model in studying the pathogenesis of craniosynostosis has yet to be pursued. The Bulgy-eye or Bey mouse is generated by retroviral insertional mutagenesis by means of embryonic stem cells. The vector is inserted between the Fgf3/Fgf4 locus causing an upregulation of both Fgf3 and Fgf4. Heterozygous Bey mice exhibit facial shortening and premature closure of several calvarial sutures. The eyes are bulging and this is a manifestation of a shortened snout and maxilla and hypertelorism (Carlton et al., 1998). Mice homozygous for the LIM-homeodomain gene, Lmx1b exhibit numerous calvarial defects including coronal suture synostosis (Chen et al., 1998). Lmx1b is expressed by the developing neuroepithelium and in the developing cranial and limb mesenchyme. Interestingly, Lmx1b and Wnt7a are thought to control limb dorsoventral patterning, with Wnt7a inducing Lmx1b and loss of Wnt7a leading to ventralization of the distal limb (Karsenty, 1998 review). Mutations MSX2 and TWIST are known to cause craniosynostosis. Overexpression of Msx2/MSX2 and partial knockout of Twist lead to mouse phenotypes with enhanced calvarial bone growth and ectopic calvarial bone formation. The effects appear to be dosage and promoter dependent (Liu et al., 1995; Winograd et al., 1997a; Bourgeois et al., 1998; Liu et al., 1999). In the Msx2 overexpression model, excessive bone growth appears to be due to a rise in the number of osteoblastic precursors which leads to increase the pool of potential osteoblasts, with a later increase in osteogenesis. The challenge is now to make use of these mutant animal models in studying calvarial embryonic development and further our understanding of the pathogenesis of craniosynostosis.

34 wi – short toes toes – nd and 2 and st short – short and broad and short hypoplastic, – – finger th Hands Hands Space b/w 1 Terminal phalanges phalanges Terminal Nails feet Distal and middle phalangeal hypoplasia malsegmentation bone Carpal valgus Hallux 5 hypoplastic – large, furrowed large, straight; Open mouth; Open straight; – failure of eruption, of failure , delayed hypodontia, hypodontia, delayed hypodontia, – depressed bridge depressed bridge flat; palate - narrow - palate flat; bridge upward slanting palpebral slanting upward hypoplastic, micrognathia hypoplastic, – – – – – – – Eyes - hypertelorism - Eyes Teeth supernumaries Teeth rate caries low eruption, Tongue Nose fissures, hypertelorism, speckled irises, epicanthal folds Profile Nose Maxilla/zygomas Maxilla/zygomas Eyes Face Teeth eruption open – closes after – fontanelle found fontanelle ’ large, delayed large, – &DOYDULD )DFH /LPE frontal and parietal flat closure delayed, wide delayed, closure third – – ‘ – closure, closure, metopic and sagittal and metopic Bossing bones Wormian suture. sagittal in the suture Metopic 10 yrs Bachycephaly, microcephaly Occiput Brachycephy Sutures Fontanelles fontanelle Anterior Frontal and parietal bossing Craniosynostosis and hands of epiphyses Coned Phenotypic features of selected syndromes with calvarial phenotype.calvarial with syndromes selected of features Phenotypic 'HOD\HGVXWXUH IRUPDWLRQ 7DEOH  &OHLGRFUDQLDO G\VSODVLD 7ULVRP\  V\QGURPH 'RZQ 3URJHULD +XWFKLQVRQ *LOIRUG V\QGURPH 3UHPDWXUH VXWXUHIXVLRQ $GHODLGH 7\SH

35 bowing, – soft tissue soft – Phalangeal, tarsonavicular calcaneonavicular foot fusions short - feet and Hands Hands Syndactyly, symmetrical Longbones - short, bowed short, - Longbones Brachydactyly Radio-humeral synostosis Radio-humeral Ulnae and femora fractures downward – hypoplasia malformations – – straight narrow, lateral swellings lateral narrow, – bridge depressed bridge – proptosis hypertelorism – – – Midface Midface Craniofacial anomalies, defects Myopia/hyperopia - Eyes short - 80% - cleft Palate metatarsals First face small Micrognathia, Profile Eyes Eyes Palprebal fissures slanting Palate crowding Dental Eyes Nose depressed bridge - Nose st wide – small – : Craniosynostotic type with type : Craniosynostotic Craniosynostosis esp. coronal esp. Craniosynostosis lambdiodal and Brachycephaly coronal, esp Craniosynostosis Brachysphenocephalic acrocephaly in which defect midline Wide coalesce then form islands bony to close the defect Craniosynostosis, skull Cloverleaf Craniosynostosis, turribrachycephaly, forehead bossing, frontal retrusion, cloverleaf skull Dolicephaly Chondrocranium suture and fontanelle Ant. $OVR (rare) skull cloverleaf Craniosynostosis affecting all 1 lambdoid and sagittal sutures, $QWOH\%L[OHU $SHUW %HDUH 6WHYHQVRQ %RVWRQ7\SH &DPSRPHOLF '\VSODVLD &DUSHQWHU

36 – broad – a – bifid, Thumbs Thumbs bifid, – partial soft tissue soft partial – preaxial polysyndactyly preaxial – Feet Pfeiffer-like great toes but not Pfeiffer-like thumbs coned - Epiphyses Carpal and tarsal - fusions broad Hands and feet soft tissue syndactyly phalanges phalanges Thumbs and great toes toes great and Thumbs Hands perm molar perm st hypoplasia hypoplasia – – crowding, cross-bites, crowding, – beak shaped hypertelorism terminal toe great and Thumb proptosis, orbits shallow, occular shallow, orbits – – – – Midface Eyes Eyes Eyes Nose Eyes Eyes proptosis, hypertelorism proptosis, deafness Conductive Dental 1 of eruption ectopic then coronallater, cloverleaf bones Wormian skull, bone calvarial rounded Multiple defects, widespread hydrocephalus craniosynostosis, coronal, esp Craniosynostosis impressions digital Pronounced skull of Craniosynostosis cases, of 5% in Craniosynostosis variable sutures affected bossing, frontal Scaphocephaly, ridging sagittal coronal esp Craniosynostosis Crouzon, as phenotype Similar Pfeiffer and Saethre-Chotzen craniosynostosisCoronal Craniosynostosis esp coronal Midface - thimble-lik phalanges Middle &UDQLRODFXQLD DQG SDQV\QRVWRVLV &URX]RQ &URX]RQZLWK DFDQWKRVLV QLJULFDQV *ULHJ FHSKDORSRO\ V\QGDFW\O\ -DFNVRQ:HLVV 0XHQNH 3IHLIIHU

37 fing rd and 3 and nd contractures – partial soft tissue soft partial fragile hyperelastic – – – – fingers th 5 Hands syndactyly esp 2 syndactyly Toes Short stature, long bone deform bone long stature, Short Bones Bones Joints high arched – like Apert, soft tissue soft Apert, like – broad, flat with low set low with flat broad, – range from normal to blue normal from range crowding – range from normal to normal from range – ptosis exophthalmos small Femur - straight small Femur - curved – – – – – hypertrophy, pseudocleft palate pseudocleft hypertrophy, Maxillary/mandibular hypoplasia Maxillary/mandibular - low-set shelf Palatal hypertelorism; Palate Palate hypertelorism; Dental Asymmetry Forehead line hair Eyes Eyes Face Sclerae Teeth dentinogenesis imperfecta-like, cross-bites poorly – large – Craniosynostosis obligatory) (not esp coronal and sagittal esp Craniosynostosis dolichocephaly lambdoidal, Megalocephaly affecting Craniosynostosis, lambdoidal and sagittal coronal, skull Cloverleaf Face parietal frontal, bones, Wormian bossing temporal and Fontanelles bone Calvarial mineralised 6DHWKUH &KRW]HQ 6KSULQW]HQ *ROGEHUJ 7KDQDWRUSKRULF '\VSODVLD 7'  7\SH, 7KDQDWRUSKRULF '\VSODVLD 7'  7\SH,, 2WKHU GLVRUGHUVZLWK FDOYDULDO SKHQRW\SH 2VWHRJHQHVLV LPSHUIHFWD Based on: Cohen 1986; Gorlin et al., 1990; Cohen 1997; OMIM 1997; Cohen 1990; al., et Gorlin 1986; on: Cohen Based

38 P RVHG ? S activation activation activation activation activation 3UR Constitutive Constitutive Constitutive Constitutive 0HFKDQLV Gain of functi of Gain Haloinsufficien Haploinsufficie Haploinsufficie */, 62; 06; )*)5 )*)5 )*)5 )*)5 )*)5 0XWDWLRQ )*)5  06; " 114290 20,0 *HQH 7p13 175700 4p16 q25.1 10q26 101200 10q26 123500 10q26 123790 FGFR2 Constitutive 10q26 10q26 123150 /RFXV 17q24.3- 5q34-q35 123101 $OWHUQDWLYH QDPHV *HQH 0DS CRSAAcrocephalosyndactyly 2 and ACS1 II, and typeI type2, Craniosynostosis CRS2 4p16CMD1, CMPD1, SRY-BOX 9 600593 dysostosisCleidocranial 6p21Craniofacial dysostosis, 119600 CFDI typeI, CBFA1 GCPS, Polysyndactyly shape skull peculiar with Selected conditions with calvarial phenotypes with known gene mutation. gene known with phenotypes calvarial with conditions Selected 7DEOH  $GHODLGH W\SH $SHUW %HDUH6WHYHQVRQ %RVWRQ W\SH &DPSRPHOLF G\VSODVLD &OHLGRFUDQLDO G\VSODVLD &UDQLRODFXQLD DQG SDQV\QRVWRVLV &URX]RQ &URX]RQ  DFDQWKRVLV QLJULFDQV *ULHJ FHSKDORSRO\V\QGDFW\O\ -DFNVRQ:HLVV

39 non 2 cha – α of Ch 21 Ch of COL1A1 activation activation activation activation 1 or 1 or Constitutive Constitutive Constitutive Constitutive α Type I Type Types II-IV Types of type I collag I type of in functional allel functional Haploinsufficie qualitative chan qualitative or a critical por Structural disrup Triplicate state o )%1 7:,67 )*)5 )*)5 )*)5 )*)5 &2/$ aberration &2/$ RU )*)5 )*)5 (III) (IV) 259420 166220 101600 166200 (I) 166200 166210 (II) 7p21 101400 7q22.1 4p16.3 187600 4p16.3 602849 10q26, q22.05, 15q21.1 182212 17q21.31- 8p11.2-p11.1 moderately progressively perinatal lethal perinatal – – mild – – deforming IV Type severe Type I Type II Type Acrocephalosyndactyly Noack ACS5, typeV, syndrome Acrocephalosyndactyly ACS3 typeIII, Marfanoid SGS, Craniosynostosis Thanatorphoric TD, dwarfism syndromeDown 21q22.3 190685 Chromosomal Type III Type 0XHQNH 2VWHRJHQHVLV LPSHUIHFWDW\SHV,,9 3IHLIIHU 6DHWKUH&KRW]HQ 6KSULQW]HQ*ROGEHUJ 7KDQDWRUSKRULF G\VSODVLD 7ULVRP\V\QGURPH

40 Table 3. Selected animal models with calvarial phenotypes.

0XWDWLRQ W\SH  3KHQRW\SH 5HIHUHQFH FDXVH 3UHPDWXUH VXWXUHIXVLRQ 3RVWHULRU IURQWDO Posterior frontal suture Bradley et al., VXWXUH RI QRUPDO starts to fuse after 25 days 1996 UDW DQG PRXVH (mouse) 12 days (rat) Bradley et al., postnatal, unlike other 1997 cranial sutures which Opperman et al., remain patent 1997

+\SRSKRVSKDW Natural mutant, Craniosynostosis - coronal Roy et al., 1981 HPLF PRXVH X-linked dominant suture Neurocranium - short, wide and high Bony changes resembling rickets

5DEELW Natural mutant Craniosynostosis – various Greene and Autosomal sutures affected Brown, 1932, dominant cited by Cohen, 1986

5DEELW Natural mutant Craniosynostosis – Mooney et al., Autosomal coronal suture and ant 1993 dominant, non- fontanelle Mooney et al., syndromic Skull – brachycephalic 1994 Face – mid-face hypoplasia

%XOJ\H\H PRXVH Retroviral Craniosynostosis – frontal, Carlton et al., %H\ insertional mutation sagittal, coronal, lambdoid 1998 at the Fgf3/Fgf4 sutures locus Intermaxillary synostosis Skull – domed Eyes – proptosis, orbital skeletal changes, hypertelorism Snout, nasal bones, maxilla – short

/P[E PRXVH Knockout Sutures – narrow Chen et al., 1998 Coronal suture – synostosis Frontal bones – thickened Interparietal, supraoccipital bones – missing or small

41 ([SHULPHQWDO PDQLSXODWLRQ 5DEELW Gluing of sutures Secondary craniofacial Persson et al., defects 1979

5DW IGF1 subcutaneous Craniosynostosis – Thaller et al., injection anterior frontal suture 1993

0RXVH Mechanical Craniosynostosis – Koskinen- constraint due to coronal and squamosal Moffet, 1986 birth delay, caused by closing uterine cervix

6KHHS Suture excision, Craniosynostosis - coronal Stelnicki et al., followed by suture 1998 insertion of bone secondary craniofacial powder, BMP2 and defects TGFβ

5DEELWDQG Hypervitaminosis A Craniosynostosis - coronal Itabashi, 1969 0RQNH\ induced in the suture Yip et al., 1980 embryo, with retinoic acid and Vit A palmitate.

5DEELW Hypervitaminosis D Craniosynostosis – Friedman and induced in the coronal, sagittal and Mills, 1969 embryo lambdoidal sutures with Vit D2

([FHVVRUHFWRSLF FDOYDULDOERQH 0V[ p148h mutant 0V[ Sutures – premature Liu et al., 1995 RYHUH[SUHVVLQJ under CMV or closure PRXVH TIMP1 promoter Ectopic calvarial bone

0V[ 0V[ (mouse) under Calvarial bone growth – Liu et al., 1999 RYHUH[SUHVVLQJ own promoter enhanced PRXVH Sutures – narrow Increased number of early stage osteoblasts

06; P148H mutant Interparietal bone – Winograd et al., RYHUH[SUHVVLQJ 06; or wild type missing 1997 PRXVH 06; (human) Mandible – hypoplastic under own promoter Secondary palate, median facial - cleft Neural tube – defects Perinatal lethality

42 7ZLVW Heterozygote Parietal and frontal bone Bourgeois et al., KHWHUR]\JRWLF growth – enhanced 1998 PRXVH Interfrontal bone - present Sutures – narrow Facial and supra occipital bones – delayed growth and small Extra toe

0XWDQWVZLWKDQ LQWHUIURQWDOERQH 7DLOVKRUWPRXVH Natural mutant Interfrontal bone – present Deol, 1961 7V chromosome 11 Frontal bones – wide Heterozygote Frontonasal suture – synostosis Facial cleft

([WUDWRHVPRXVH Natural mutant Maxilla – overgrowth Johnson, 1967 ;W *OL deletion Interfrontal bone – present Hiu and Joyner, Heterozygote Skull – wide, abnormal 1993 shape Facial cleft

3DWFKPRXVH 3K Natural mutant Interfrontal bone – present Grüneberg and 3'*)5α deletion Skull – wide, abnormal Truslove, 1960 Heterozygote shape Stephenson et Facial and palatal cleft al., 1991

'HILFLHQF\RI FDOYDULDOERQH &RQJHQLWDO Natural mutant Calvarial bones – only Grüneberg 1953 K\GURFHSKDOXV point mutation in small basal portion of Kume et al., FK PRXVH 0I frontal, parietal and 1998 Homozygote interparietal bones present Hong et al., Mandibular, zygomatic, 1999 squamosal bones – misshapen, Snout – short Basioccipital, exoccipital and hyoid bones – small Supraoccipital bone – Malformed – missing

&OHLGRFUDQLDO Radiation-induced Calvarial bones – small, Sillence et al., G\VSODVLD &FG mutant development delayed 1987 PRXVH &EIDdeletion Suture closure – delayed Otto et al., 1997 Heterozygote Clavicle – hypoplastic

43 &EIDx‚px‚ˆ‡ Homozygote - almost Huang et al., €‚ˆ†r complete absence of bone 1997 Heterozygote - similar to Komori et al., &FG mouse, Wormian 1997 bones Otto et al., 1997

&DUWx‚px‚ˆ‡ Homozygote Acrania, neural tube Zhao et al., 1996 €‚ˆ†r Rescued by folic closure defect acid Interparietal bone – missing Frontal and parietal bones - only small basal portion present Mandibular, maxillary and zygomatic bones – small

A htvyv‡h†‚††vˆ€ Natural mutant Model for Osteogenesis Guenet et al., IUR€‚ˆ†r Homozygote imperfecta Type III 1981 Sillence et al., 1993 0RY€‚ˆ†r pro-α1(I) collagen Model for Osteogenesis Bonadio et al., retroviral insert imperfecta Type I 1990

Regulation of Embryonic Development

The development of calvarial bone, like other tissues and organs, is based on co- ordinated interactions between cells or groups of cells. This signalling may be facilitated by direct cell-cell communication (juxtacrine) as well as via secreted soluble signalling molecules. These signalling molecules may act in an autocrine manner, effecting changes in the same cell that secreted them; in a paracrine manner, effecting changes in neighbouring cells or cells in close proximity to the secreting cell; or alternatively in an endocrine manner, effecting changes some distance from the secreting cell with the signal, or hormone, travelling via the circulation system. Paracrine factors are small proteins capable of diffusing over small distances, and are commonly referred to as growth and differentiation, or simply growth, factors. There are four key families of growth factors categorised on the basis of their structure. They are the fibroblast growth factors (FGFs), members of the transforming growth factor β (TGFβ) superfamily, the hedgehog (HH) family and the wingless family. These growth factors, together with transcription factors, mediate the development of most, if not all, body tissues including bone. Similar signalling systems are not only common to different body organs, but also homologous proteins are conserved between different species, throughout the animal kingdom. Thus, homologous genes may be responsible for the same organ development in different animals. For example, although the eye of insects and mammals look quite

44 different, the same gene, Pax6, regulates eye development in fruit flies, mice and in humans (Parr and McMahon, 1994 review; Gilbert, 1997 review; Hammerschmidt et al,. 1997 review; Wozney, 1998 review). The development of a tissue or organ may well involve several of these families of growth factors which may even signal from one tissue to influence another in an intricate network system. A well documented example of this is the signalling network regulating tooth development (Thesleff and Sharpe, 1997 review). Here not only do members of all four growth factor families, FGF, TGFβ, HH and WNT, interact, but signals pass from the oral epithelium to underlying mesenchyme and vice versa. This may occur several times in a signalling cascade before morphogenesis is complete. This interaction between tissues is a phenomenon often found during the development of many organs.

Fibroblast growth factors and their receptors

Fibroblast growth factors (FGFs) and their receptors in embryogenesis There are at least 19 FGFs which partake in a wide range of biological events, both in the embryo and in the adult, where they affect cell proliferation, differentiation, survival and motility (Yamaguchi and Rossant, 1995 review; Nishimura et al., 1999). Although many FGFs have similar effects and can often be substituted for one another, they each have unique expression patterns and functions (Yamaguchi and Rossant, 1995 review; Kettunen and Thesleff, 1998). Fgfs3, 4 and 5 are involved in early development both prior to and during gastrulation and neurulation (Wilkinson et al., 1988; Haub and Goldfarb, 1991; Niswander and Martin, 1992; Slack, 1994 review; Crossley and Martin, 1995). Later, Fgfs play a prominent role in most if not all organ development. For example, Fgfs2, 4, 8, 9 and 10 are integral to limb bud development (Martin, 1998 review). Fgf8 is involved in brain development (Crossley et al., 1996), and Fgfs 1, 2, 3, 4, 8, 9 and 10 are expressed in the developing tooth (Wilkinson et al., 1989; Cam et al., 1992; Kettunen and Thesleff, 1998, 1999). Each of the proteins encoded by the FGF genes contains a conserved core sequence of approximately 120 amino acids that permits binding to heparin and heparan sulphate proteoglycans (HSPGs), such as syndecans, glypican and perlecan (Zhu et al., 1991; Faham et al., 1996). As HSPGs are found on the cell surface and in the extracellular matrix, binding tends to restrict the distance FGFs are capable of travelling (Basilico and Moscatelli, 1992 review). Binding to HSPGs facilitates FGF signal transduction by presenting the FGF ligand to FGF high affinity receptors, the FGFRs. In vertebrates, there are four FGFRs which are high affinity tyrosine kinase transmembrane receptors (Dionne et al., 1990; Keegan et al., 1991; Lee et al., 1989; Partanen et al., 1991). When an FGF-cofactor complex binds to an FGFR, it causes the receptor to either homo- or heterodimerize, which in turn leads to phosphorylation and intracellular signal transduction (Givol and Yayon, 1992; Johnson and Williams, 1993). FGFRs can also be activated by ligands other than FGFs, however, the significance of these is at present unknown (Green et al., 1996). Like the individual FGFs, the FGFRs are also expressed in unique patterns

45 during development (Orr-Urtreger et al., 1991; Peters et al., 1992, 1993; Yamaguchi et al., 1992). The primary structure of FGFRs consists of a single peptide, three extracellular immunoglobulin-like loops (IgI, II and III), an acidic amino acid domain in the linker region between IgI and IgII, a transmembrane domain and intracellularly they possess a split kinase domain and a carboxy terminal. Each FGFR is capable of making a variety of isoforms by alternative splicing in the IgIII loop; these have differing expression patterns and binding. Therefore, FGFR2IIIb and FGFR2IIIc are two different receptors derived from the same gene. Indeed, in-vitro, the different splice forms of FGFRs possess different ligand binding affinities (Johnson and Williams, 1993; Cohen, 1997 review). Soluble secreted variants of the receptors can also form, and although these cannot partake in intracellular signalling, as they have no transmembrane domain, they could compete with the membrane bound form by mopping up ligand (Johnson et al., 1990; Katoh et al., 1992). Finally, both FGFR1 and 2 have a cell adhesion molecule (CAM) homology domain. This contains the sequences for N-CAM, N-Cadherin and L1, the former two having been implicated in mesenchymal cell adhesion and condensation. CAMs can signal via FGFRs, but whether it binds directly or in concert with FGFs is not known (Williams et al., 1994; Hall and Miyake, 1995; Cohen, 1997 review). Variations in the spatial and temporal distribution of the FGFR splice forms, their ligands and cofactors can all determine the specificity of a signal and therefore cellular action.

FGFs in bone and suture development FGF2 (basic FGF) protein is produced by osteoblasts, chondrocytes and the developing meninges, the later having regulatory role in calvarial bone formation (Gonzalez et al., 1990). Indeed, FGF2 protein and mRNA expression in both sutural osteoblasts and the underlying dura increases during calvarial sutural fusion (Mehrara et al., 1998; Most et al., 1998). In-vitro, FGF2 has been shown to be produced by cultured bovine osteoblasts which then accumulates in the extracellular matrix (Globus et al., 1989). Both FGF1 (acidic FGF), and FGF2, like many other FGFs, are potent mitogens for osteoblastic cells and in some systems suppress osteoblast differentiation as evidenced by a reduction in phenotypic markers such as type I collagen, alkaline phosphatase, type I collagen and osteocalcin mRNA (Canalis and Raisz, 1980; Canalis et al., 1988; Rodan et al., 1989; Tang et al., 1996). These data may indicate that FGF2 has independent effects on calvarial cell replication and differentiation. However, the effects of FGFs on osteoblastic cell differentiation and bone matrix formation in long term culture are more confusing, with data suggesting both stimulation and inhibition of osteoblastic phenotype and bone formation. Thus, FGF2 has been seen to increase the formation of mineralised nodules in prolonged (21 days) rat marrow cell culture (Noff et al., 1989), while both FGF1 and FGF2 have also been found to inhibit nodule formation and mineralisation and osteoblast differentiation (Tang et al., 1996). In-vivo, an increase in bone formation has been detected after administration of either FGF1 or FGF2 (Aspenberg and Lohmander, 1989; Mayahara et al., 1993). Also, beads soaked in FGF2 and placed onto the calvaria of embryonic mice alter the mRNA expression of the osteoblast marker, osteopontin (Iseki et al., 1997). Indeed, the effects of FGF2 may be differentiation stage specific,

46 with FGF2 causing an increase in matrix mineralisation and osteocalcin production in more mature cells (Debiais et al., 1998). Interestingly, FGF2 also stimulates osteoclast formation in marrow cultures (Hurley et al., 1998). The effects of FGF2 on bone cells may well be mediated by other growth factors, with FGF2 stimulating both TGFβ and Interleukin 6 expression in osteoblastic cell lines (Noda and Vogel, 1989; Hurley et al., 1996). FGF2 protein and mRNA are upregulated by FGF2 itself as well as by prostaglandins and parathyroid hormone. In addition, parathyroid hormone also stimulates FGFR1 and 2, in both calvarial tissue and osteoblastic cell cultures (Hurley et al., 1994; Sabbieti et al., 1999). FGF2 also appears to be dependent on proteoglycans for its effects on osteogenesis (Molteni et al., 1999). Mice ubiquitously overexpressing FGF2 exhibit chondrodysplasia, with shortened long bones and macrocephaly. Interestingly, their occipital bone, a bone formed partly intramembraneously and partly from endochondral ossification, appears to be enlarged (Coffin et al., 1995). FGF2 knockout mice apparently possess no bony phenotype, exhibiting neuronal and vascular defects and a disruption of wound healing (Ortega et al., 1998; Zhou et al., 1998).

FGFRs in bone and suture development The discovery of mutations in human FGFR 1-3 genes and the generation of knockout and transgenic mice for Fgfrs has added considerably to our understanding of bone development and pathological conditions arising from their disruption. Mutations in FGFR1, 2 and 3, most in FGFR2, have been described as causing both syndromic and non syndromic forms of craniosynostosis. These are summarised in Table 1. The majority of these mutations are missense, with a smaller number of in- frame insertions, deletions or indels (Wilkie, 1997 review). Interestingly, the same mutation in a different gene can cause a similar but distinct phenotype. For example, Pro→Arg mutations in the linker region between the IgII and III loops can cause Pfeiffer syndrome in FGFR1, Apert syndrome in FGFR2 and Muenke craniosynostosis if the mutation occurs in FGFR3. It appears that mutations causing craniosynostosis may confer reduced ligand dissociation, altered covalent cross linking and transmembrane hydrogen bonding, increased affinity for FGF ligands, as well as increased calvarial cell differentiation and bone matrix formation (Wilkie, 1997 review; Anderson et al., 1998; Lomri et al., 1998). Indeed, there is evidence indicating that these mutations cause ligand independent constitutive activation of receptors (Neilson and Friesel, 1995; Galvin et al., 1996; Mangasarian et al., 1997). Interestingly, the same syndrome may be caused by mutations in different genes. This is exemplified by Pfeiffer syndrome, in which many mutations in both the FGFR1 and FGFR2 genes have been detected. Fgfr1, 2 and 3 are found in differing but overlapping patterns of expression during both intramembraneous and endochondral bone formation, in both the developing somatic and craniofacial region (Orr-Urtreger et al., 1991; Peters et al., 1992, 1993; Yamaguchi et al., 1992; Szebenyi et al., 1995; Wuechner et al., 1996). Interestingly, signalling through FGFRs can upregulate an element in the osteocalcin promoter. The FGF specific component of the osteocalcin promoter is responsive to FGF2 but not to other mitogens such as Insulin-like growth factor 1, EGF, PDGF-BB and TGFβ. FGF2

47 stimulation of cultured osteoblastic cells seems to initiate two functionally distinct responses, the second being an induction of the human interstitial collagenase (MMP1) promoter (Newberry et al., 1996). Several genetically engineered mouse models affecting the Fgfrs have been generated, however none thus far have exhibited a calvarial phenotype. Deficiency of Fgfr1 and deletion of Fgfr2’s third Ig loop result in death before start of skeletal development (Deng et al., 1994; Yamaguchi et al., 1994; Xu et al., 1998). Interestingly, Fgfr1 has been found to have a role in anterior-posterior vertebral patterning (Partanen et al., 1998). Mice which secrete a soluble dominant negative FGFR, which has no kinase domain and therefore competes for wild type receptor, exhibit several craniofacial and limb deformities (Celli et al., 1998). Consistent with its proposed role as a negative regulator of endochondral ossification, knocking out Fgfr3 results in long bone overgrowth (Colvin et al., 1996; Deng et al., 1996). Indeed, mice overexpressing the mutation in the Fgfr3 gene, that causes achondroplasia, exhibit stunted growth (Naski et al., 1998). Although Fgfr4 is expressed, at low levels, by developing chondrocytes, it appears to be, perhaps, more important in skeletal muscle and endodermally derived organ development, such as the lung, liver and gut. In contrast to other Fgfrs, no or very low amounts of Fgfr4 mRNA are observed in the calvaria (Partanen et al., 1991; Korhonen et al., 1992). Indeed, mutations in FGFR4 have not been found in patients with craniosynostosis (Gaudenz et al., 1998).

Helix-Loop-Helix proteins

TWIST and IDs in embryogenesis Basic helix-loop-helix proteins (bHLH) act as regulators of DNA binding, that is, as transcription factors. Prior to DNA binding they form dimers; however, homodimers between two identical bHLHs do not bind well to DNA. Thus, heterodimers are formed with one of two potential partners, either a positive regulator, which stimulates transcription, or a negative regulator. Twist proteins are conserved bHLH transcription factors and Inhibitors of differentiation (IDs) are conserved dominant negative HLHs, which act as negative regulators of transcription. Thus, as Id1 lacks a DNA binding domain, it inhibits bHLH’s function by suppressing their heterodimerization through direct protein-protein interactions, thereby mopping up bHLHs (Pesce and Benezra, 1993). In the mouse, Twist is expressed in mesoderm outside the primative streak and then later in development in the somites as well as in the mesenchyme of the head, branchial arches, limb buds, tooth and heart (Füchtbauer, 1995). Both Twist and Ids have been implicated as regulators of mesoderm differentiation and myogenesis in both Drosophila and vertebrae (Benezra et al., 1990; Bate et al., 1991), though, in contrast to Drosophila twist, murine twist is thought to act as a suppressor rather than an activator of myogenesis (Hebrok et al., 1994; Baylies and Bate, 1996; Spicer et al., 1996).

Twist and Ids in bone and suture development Mutations in TWIST cause Saethre-Chotzen syndrome, characterised by

48 craniosynostosis and limb defects (El Ghouzzi et al., 1997; Howard et al., 1997; Paznekas et al., 1998) (Tables 1 and 2). In contrast to the FGFR and MSX2 mutations, these are largely deletions or nonsense mutations. Twist knockout mice die at E11.5 with a failure of the cranial neural folds to fuse and defects in head mesenchyme, somites and limb buds (Chen and Behringer, 1995). Mice heterozygote for Twist exhibit limb and calvarial phenotypes reminiscent of Saethre-Chotzen syndrome (Table 3) (Bourgeois et al., 1998). This, together with the known human mutations, suggests a loss of function pathology. Interestingly, Drosophila Twist is thought to be upstream of DFR1, a Drosophila FGF- receptor homologue (Shishido et al., 1993). It remains to be seen whether Twist and Fgf signalling are related in calvarial bone development. Twist may be a negative regulator of osteoblast differentiation. Indeed, early osteoblastic cell cultures have been shown to express both Twist and Id, with expression decreasing as maturity increases (Ogata and Noda, 1991; Murray et al., 1992). Twist is also expressed in the developing limb and mandible (Alborzi et al., 1996). Whether these HLHs perform similar regulatory functions in osteoblast differentiation as in the myoblast lineage, remains to be elucidated.

Hedgehogs

Hedgehogs (HHs) in embryogenesis The vertebrate hedgehog family comprises Indian hedgehog, Desert hedgehog and Sonic hedgehog (Ihh, Dhh and Shh). They are homologues of Drosophila Hedgehog, which regulates segmental and imaginal disc patterning. They encode secreted proteins that are involved in signalling between cells. In contrast, in vertebrates members of the hedgehog gene family are involved in the control of left-right asymmetry, cartilage differentiation, limb morphogenesis, myotome and sclerotome specification, hair follicle development, specification of different neuronal cell types and spermatogenesis (Hammerschmidt et al., 1997 review). HH ligand interaction and signal transduction is mediated by patched (Ptc) and smoothened (Smo), both of which are transmembrane proteins. Signalling via smo would proceed unabated if not for the action of Ptc which inhibits Smo. When HH binds to Ptc it prevents this inhibition thus allowing signal transduction. Ptc is itself activated downstream of HH, thus a negative feedback loop is established (Ingham, 1998). Sonic hedgehog (Shh) is expressed in several regions that have organiser activity including the notochord, enamel knot of the developing tooth and the zone of polarising activity in the limb bud. Predominantly, Shh is detected in epithelial structures involved in epithelial-mesenchymal interactions including the tooth, hair follicle, the lung and bladder (Bitgood and McMahon, 1995; Vaahtokari et al., 1996; Martin, 1998 review). In addition, Shh has a role in forebrain development and in the establishment of midline craniofacial structures (Marti et al., 1995; Chiang et al., 1996).

Hedgehogs in bone development Ihh is involved in chondrogenesis through interactions with both parathyroid

49 hormone related peptide (PTHrP) and possibly Bmps (Bitgood and McMahon, 1995; Karsenty, 1998 review). Ihh, which is secreted by the prehypertrophic chondrocytes, acts in a negative feedback loop as an activator of PTHrP via Ptc and Gli. PTHrP, in turn, negatively regulates the rate of chondrocyte differentiation by signalling through the PTH/PTHrP receptor. Thus, chondrocytes are prevented from moving down the differentiation pathway and from expressing Ihh. Consequently, overexpression of Ihh leads to a lack of hypertrophic chondrocytes and short, broad cartilage elements and an increase in PTHrP. Also, PTHrP-deficient mice exhibit a precocious differentiation of hypertrophic chondrocytes and exposure of bone explants to excess PTHrP delays their differentiation (Vortkamp et al., 1996; Karsenty, 1998 review). In addition to Ptc and Gli, Ihh’s action, PTHrP may also be mediated by another transcriptional target of HH signalling, Hedgehog-interacting protein (Chuang and McMahon, 1999). However, this remains to be clarified. A role for Shh in osteoblast differentiation and ectopic bone formation has also been postulated. Chicken fibroblasts transfected with Shh induced an increase in osteoblast differentiation in both fibroblasts and osteoblastic cell cultures. Also, intramuscular implantantion of the transfected cells induced ectopic bone formation (Kinto et al., 1997). Furthermore, early in embryonic development, chondrogenesis of the somitic mesoderm has been suggested to be mediated by Shh which make precursor cells responsive to Bmp signalling (Murtaugh et al., 1999). Indeed, in several different developing tissues including the limb and the gut, Bmps are know to act in response to Shh (Laufer et al., 1994; Roberts et al., 1995).

MSXs

MSX1 and 2 in embryogenesis Msx1 (Hox-7) and Msx2 (Hox-8) are members of a small family of diverged homeobox-containing genes related to the Drosophila muscle-segment homeobox (msh) gene. They are expressed in the developing vertebrate in a number tissues, including neural crest, cranial sensory placodes, bone and teeth, where they frequently have overlapping distributions (MacKenzie et al., 1991, 1992; Jowett et al., 1993). In particular, they have been associated with epithelial-mesenchymal interactions during organogenesis, where they are targets of BMP and FGF signalling (Satokata and Maas, 1994). For instance, in-vitro, BMP2 and 4 induce an upregulation of Msx gene expression in tooth explants as well as in rhombomeres (Vainio et al., 1993; Graham et al., 1994; Chen et al., 1996), and several FGFs induce the expression of Msx1 in dental mesenchyme (Kettunen and Thesleff, 1998). Msx1 and Msx2 have also been associated with the differentiation of neural crest-derived intramembraneous bones in the skull (Takahashi and Le Douarin, 1990; Takahashi et al., 1991). Msxl deficient mice exhibit an arrest in tooth development at the bud stage, while Msx2 deficient mice exhibit late defects in tooth development (Maas and Bei, 1997). It will be interesting to discover whether particularly the Msx2 deficient mice possess a calvarial phenotype.

50 MSX1 and 2 in bone and suture development Boston type craniosynostosis, observed in a single family, is caused by a Pro148His substitution in the MXS2 gene (Jabs et al., 1993). This mutation confers enhanced DNA binding affinity with a reduction in ligand dissociation, and this is without altering target specificity (Ma et al., 1996). Transgenic mice overexpressing the MSX2 mutation appear to have different phenotypes depending on which promoter is used. One strain exhibits precocious bone formation and accelerated suture closure (Liu et al., 1995), while another shows other craniofacial defects including, aplasia of the interparietal bone (Winograd et al., 1997a). It has been proposed that these differences are MSX2 dosage dependent, with a mild elevation in MSX2 causing an enhancement of calvarial bone formation (Wilkie, 1997 review). Msx2 is expressed in the developing calvaria and underlying meninges and is known to be involved in osteoblast differentiation and function (MacKenzie et al., 1991, 1992; Hodgkinson et al., 1993b). However, the level of Msx2 expression decreases with increased osteoblast differentiation and there is evidence that Msx2 negatively regulates collagen 1 and osteocalcin, genes expressed by terminally differentiated osteoblasts (Towler et al., 1994; Dodig et al., 1996; Hoffmann et al., 1996; Newberry et al., 1997). Indeed, in primary cultured chick calvarial osteoblasts, overexpression of Msx2 can inhibit terminal differentiation of osteoblasts and mineralisation of the extracellular matrix. In contrast, continued high expression of antisense Msx2 mRNA decreased proliferation and accelerated differentiation (Dodig et al., 1999). Thus, if the MSX2 mutation causing craniosynostosis appears to be a gain of function defect, how can this be reconciled with the knowledge that Msx2 is negative to osteoblast differentiation? Recently, transgenic mice overexpressing Msx2 under the control of the murine Msx2 promoter have been shown to exhibit an enhancement of calvarial bone growth and a narrowing of the calvarial sutures. These mice do not show premature fusion of the calvarial bones, although the oldest mice examined were only 6 days post natal in age. These animals also display an increase in the proliferative rate of osteogenic precursors in the osteogenic front. Thus, this precocious bone growth appears to be due to an increase in the pool of potential osteoblasts, this increase ultimately leading to an increase in osteogenesis (Liu et al., 1999).

The Transforming Growth Factor b Superfamily

The Tgfb superfamily in embryogenesis The transforming growth factor β superfamily (Tgfβ) is comprised of more than 20 members that mediate a wide spectrum of events in normal growth and development, ranging from cell differentiation to cell death. The superfamily includes the Tgfβs, inhibin, activin and many Bone morphogenetic proteins (Bmps) and Growth and differentiation factors (Gdfs). A role for Tgfβ superfamily members in tissue and organ development other than that of hard tissues is exemplified by the phenotype of the Bmp7 knockout mouse. This exhibits few skeletal deformities but has numerous defects in other tissues

51 including the kidneys and eyes (Dudley et al., 1995; Luo et al., 1995). Several other Bmps have been shown to posses functions in other non-skeletal organs including Bmp2 in heart development and Bmp4 in lung and testis development (Karsenty, 1998 review). Here, organogenesis proceeds through epithelial-mesenchymal interactions and by controlling cell proliferation and apoptosis.

The Tgfb superfamily in bone and suture development Several Bmps can initiate both endochondral and intramembraneous bone formation, they can regulate chemotaxis, cell proliferation and differentiation. Indeed, Bmps may be of key importance during mesenchymal condensation (Reddi, 1994 review). Bmps’ roles in chondrocyte and osteoblast differentiation has, in the main, be studied in cell lines and primary cultures. Here, it has been demonstrated that a variety of Bmps can stimulate the expression of mature chondrocyte and osteoblast markers, and that this is possibly via the induction of Msx genes (Davidson, 1995 review). At least 5 Tgfβ family members, Bmp4, 5, 7, Gdf5 and 6, have a role in the patterning of a number of skeletal elements. Disruption of Bmp5 causes the mouse mutation short-ear, which is characterised by reduction in precartilage mesenchymal condensations with multiple skeletal defects (Kingsley et al., 1992). Similarly, disruption of Gdf5 also causes a skeletal phenotype, brachypodism (Storm et al., 1994). Interestingly, no defects in chondrocyte or osteoblast differentiation could be detected in neither Bmp7 deficient mice nor in human GDF mutations leading to Hunter-Thompson chondrodysplasia (Karsenty, 1998 review). Three isoforms of Tgfβ exist which exhibit unique expression patterns (Pelton et al., 1991). Tgfβs influence the proliferation and differentiation of osteoblasts, osteoclasts and chondrocytes (Karsenty, 1998 review). A wealth of information with regard to craniofacial and bone development has been generated from the production of Tgfβ2 and 3 knockout and transgenic mice. Tgfβ2 deficient mice exhibit a wide range of defects including a small and dysmorphic mandible, cleft palate due to a failure of palatal shelf elevation, and a reduction in cranial bone size and ossification (Sanford et al., 1997). Mice overexpressing Tgfβ2 in osteoblasts results in increased matrix production and increased osteoclastic activity (Erlebacher and Derynck, 1996). Tgfβ3 knockout mice have cleft palate caused by a lack of adherence of the palatal shelves (Proetzel et al., 1995). A role for Tgfβ in suture biology has also been demonstrated (Opperman et al., 1997; Roth et al., 1997). Immunoreactivities of TGFβ1 and 2, and lower levels of TGFβ3, have been detected in the embryonic calvaria, notably, in the sutural mesenchymal cells and matrix, in the underlying dura and the overlying periosteum. Levels of TGFβ1 and 2 remain high during fusion of the posterior frontal suture while levels of TGFβ3 reduce. This has been further investigated in an in-vitro model, where by removing the dura mater underlying a developing, patent suture fusion can be induced. In the absence of the dura, elevated levels of TGFβ1 were detected, compared to those of both TGFβ2 and 3, and also TGFβ1 in non-fusing sutures. Indeed, levels of TGFβ3 appeared to decrease, thus confirming the in-vivo findings (Opperman et al., 1997). In addition, a high dose of recombinant TGFβ2 injected over a suture area, in-vivo, can cause both

52 ectopic bone formation as well as suture fusion (Roth et al., 1997). Also, increased levels of TGFβ2 and 3 have been demonstrated in sutures from children with persistent plagiocephaly (Lin et al., 1997). Thus, Tgfβ1 and 2 are associated with rat suture development and continue to be expressed later during suture closure, while Tgfβ3 has a role in keeping calvarial sutures patent. Interestingly, TGFβ is known to increase FGF2 mRNA and protein levels in osteoblastic cell cultures (Hurley et al., 1994). Although mutations in TGFβ genes have, to date, not been found in patients with craniosynostosis, TGFβ may act on sutural osteogenesis via an FGF/FGFR mediated pathway, in which mutations causing craniosynostosis have been found.

NELL-1

Using a differential display technique, the gene NELL-1 has been found to be upregulated in samples taken from sutures that were either undergoing or that had already undergone premature fusion, as compared to samples from normal patent sutures (Ting et al., 1999). NELL-1 is expressed in sutural mesenchymal cells and osteoblasts at the osteogenic front, and also in the developing brain. It contains six epidermal growth factor-like repeats as well as Calcium ion binding sequences. However, the secreted protein does not effectively bind to any known EGF receptor. Further investigation will tell us whether this potential signalling molecule is indeed involved in known or novel signalling networks and whether it has a causal role in craniosynostosis.

53 AIMS

The aim of this study was to investigate the molecular mechanisms involved in calvarial bone and suture development. It was attempted specifically to:

1.detail the role of osteoclasts and osteoblasts in the developing calvaria, using mouse as the experimental model.

2.study cell turnover during calvarial bone and suture development, and notably the role of apoptosis.

3.analyse the expression of signalling molecules considered to be of relevance to calvarial bone and suture development.

4.develop, and use, an in-vitro culture system whereby the molecular signalling involved in calvarial bone and suture development may be investigated.

54 MATERIALS AND METHODS

Preparation of tissues Whole heads of mice (CBA x NMRI) aged between embryonic day 10 and embryonic day 15 (E10 – E15), as well as calvariae from mice aged between E15 and postnatal day 6 (P6) were dissected in Dulbecco’s phosphate buffered saline (pH 7.3) under a stereomicroscope, with Vannas ophthalmic scissors (Heiss, Germany). The approximate line of dissection is illustrated in Figure 6. The age of the embryos was determined by the day of the appearance of the vaginal plug (day 0) and confirmed by morphological criteria. Material for tissue sections was cut at 90o to the suture under investigation. Thus, if the sagittal or anterior frontal sutures were under study, the dissection was made paracoronally. If however, the coronal or lambdoid sutures were under investigation, the dissection was made in a parasagittal orientation. The tissue samples were composed of the bones and the interposed sagittal suture as well as the overlying skin. Following overnight fixation at 4oC in 4% paraformaldehyde (PFA) in PBS, the tissues were dehydrated in an ethanol series, stained with eosin to aid tissue orientation in paraffin, treated in xylene and embedded in paraffin. Sections of 5-7 µm were cut and mounted on triethoxysilane treated slides, dried overnight at 37oC and stored at 4oC. Postnatal material for tissue sections was decalcified in 12.5% ethylenediaminetetra-acetic acid (EDTA)/2.5% PFA in PBS, changed every 4th day, for about 7-14 days, at 4oC. Some histological sections of both in-vivo and cultured tissues were studied after haematoxylin and eosin, Toluidine blue, von Kossa and van Gieson staining. Whole-mount material was fixed overnight at 4oC in 4% PFA, then dehydrated in a methanol series with DEPC treated PBS and stored in absolute methanol at -20oC until use. Material of age E15 and older was denuded of the overlying skin and underlying brain in order to improve fixation and probe penetration of the tissue under question.

In-vitro culture of calvariae Calvariae, consisting of the developing two parietal bones, the interparietal bone and part of the frontal bones, were dissected from mice aged E15 for all experiments except from the dura mater experiments, for which we used calvariae from mice aged E16.5 and P1. Here, the dura mater was carefully removed from the sagittal suture. Explants were cultured in a Trowell-type organ culture system (Figure 6). Explants were placed on 0.1 µm pore size Nuclepore polycarbonate filters (Pleasanton, CA, USA), supported by metal grids and cultured in Dulbecco’s minimal essential medium (Sigma) supplemented with 10% foetal bovine serum (Gibco, UK) and penicillin/streptomycin in a humidified atmosphere of 5% CO2 in air at 37oC. Ascorbic acid, 100 µg/ml, was supplemented, and culture medium and ascorbic acid was changed every other day. Explants were cultured for 24-96 hours after which they were examined as either whole- mount or sectioned material, and processed as described above. Whole-mount and histological sections of E15 in-vitro cultures are illustrated in Figure 7.

55 $ % Culture medium

Explant Metal grid Filter

Bead

' &

Bone Bone

Figure 6. Microdissection and in-vitro culture of mouse calvaria. (A) Schematic illustration showing of the approximate line of dissection for mouse whole-mount calvaria. (B and C) Trowell- type in-vitro culture. The calvarial explant is placed on a filter, which rests on a metal grid. Unlike conventional tissue culture systems, where the explant is immersed in liquid, in the Trowell technique the explant lies in the gas/culture medium interface permitting the tissue draw nutrients from the medium below and allowing gaseous exchange with the air above. As the explant is situated in the liquid meniscus, the tissue remains wet. (D) Calvarial explant as viewed through a stereomicroscope, showing the two parietal bones and the developing sagittal suture between them. Scale bar in D 1mm.

Bead preparation and in-vitro experiments Approximately 100 beads (Table 4) were placed into an non-stick Eppendorf tube. The beads were then spun briefly to form a pellet and the buffer solution removed with a drawn Pasteur pipette. The appropriate growth factor or Bovine serum albumin (BSA) was then added and incubated for 1 hour at 37oC. Beads which were not used immediately were stored at 4oC. Before use, beads were rapidly washed in culture medium, then two or three beads were placed with a mouth-controlled capillary pipette onto the desired location of the freshly dissected calvarial explants: osteogenic front; mid-sutural mesenchyme of sagittal suture; the centre of the parietal bone. These are illustrated in figure 6.

56 Table 4. Proteins and beads used during in-vitro culture of calvaria.

3URWHLQ &RQFHQWUDWLRQ %HDG FGF2, recombinant human (R&D 25 ng/µl Heparin acrylic (Sigma Systems Ltd, UK) H-5263)

FGF4, recombinant human (British 25 ng/µl Heparin acrylic (Sigma Biotechnology Products, UK) H-5263)

BMP2, recombinant human 100 ng/µl Affi-Gel agarose, 100-200 (Genetics Institute, MA, USA) mesh, 17-150 µm diameter (Biorad)

BMP4, recombinant human 100 ng/µl Affi-Gel agarose, 100-200 (Genetics Institute, MA, USA) mesh, 17-150 µm diameter (Biorad)

Enzyme histochemistry (TRAP) TRAP staining of paraffin sections was performed according to Thompson (1966). Tissues were fixed in 4% PFA and 7 µm sections cut from paraffin blocks as detailed above. Deparaffinized, rehydrated sections were then incubated in 50 mM tartrate in acetic buffer for 2 hours at 37oC, followed by incubation in acid phosphatase substrate buffer (25% Michaelis acetate buffer, 0.16% pararosanilin, 0.16% NaNo , 0.05% Naphthol AS-BI phosphate), pH5.0 with 20 mM tartrate for 2 hours at 37oC. Sections2 were then washed in distilled water, counterstained with 1% aqueous toluidine blue solution for 10 seconds and mounted in Aquamount (BDH, UK). TRAP staining of whole calvariae was carried out according to Holt et al. (1994). Tissues were fixed in 95% ethanol/5% glacial acetic acid, then washed in 1mg/ml BSA in PBS. They were then incubated in 1 mM naphthol AS-BI phosphoric acid (Sigma), 0.3 mg/ml Fast Garnet GBC salt (Sigma) in 100 mM acetic acid and 26.8 mM L (+) tartaric acid (Sigma), with pH adjusted to 5.2, at 37oC for 15 minutes (mins). After staining, the reaction was stopped by washing in BSA-PBS, and tissues were cleared in 50% then 70% glycerol in PBS. Negative controls were performed by omitting the substrate.

57 $ p f

In-vivo s ip E15 f p

% &

In-vitro 1 day filter

' ( p

In-vitro 4 days p p p

Figure 7. In-vitro culture of E 15 mouse calvaria. (A) Whole-mount in-situ hybridisation showing mRNA expression of the extracellular matrix molecule Bsp, in E15 calvaria. The calvaria has been dissected free from the underlying brain and overlying skin. (B and D) Whole-mount in-situ hybridisation showing Bsp mRNA expression in E15 calvaria cultured for 1 and 4 days. Not only do explants remain viable with osteoblasts continuing to produce mRNA for Bsp, but also the calvarial bones expand into the mesenchymal tissue. Where they meet to form sutures. (C and E) Haematoxylin and eosin stained sections of 1 and 4 day calvarial explants. Sections are cut at similar planes to the dotted lines seen in B and D. The osteogenic fronts (arrows) of the calvarial bones approximate but do not fuse. Thus mimicking the in-vivo situation. Not only do the calvarial bones come closer together but there is also an increase in the thickness of the calvarial bones (white dotted lines). f, frontal bone; ip, interparietal bone; p, parietal bone; s, sagittal suture. Scale bars: whole-mounts 1mm; sections 200µm.

58 In-situ hybridisation In-situ hybridisation was carried out on both tissue sections and whole-mounts using techniques based on protocols described by Wilkinson and Green (1990 review) and Vainio et al. (1991) with modifications. cDNA containing plasmids were linearised and single stranded antisense and sense RNA probes generated by in-vitro transcription. Before precipitation, the Fgfr probes were passed through Sephadex G-50 NICK-columns (Parmacia) in order to remove unincorporated nucleotides. In-situ hybridisation on tissue sections was performed using either 35S-uridine triphosphate (UTP) or digoxigenin-UTP (DIG) labelled riboprobes, and on whole-mounts using DIG labelled riboprobes. The mRNA probes used are illustrated in Table 5.

Table 5. cDNAs used in this study.

3UREH /HQJWK 3ODVPLG 3DSHUXVHG ES LQ Bmp2 240 pGEM3 (Promega) III Bmp4 285 pGEM3 (Promega) II, III Bmp7 220 pGEM3 (Promega) II Bsp1 1000 pCRII (Invitrogen) II, IV Cbfa1 (Osf2) 336 pBluescript KS- (Stratagene) Thesis Fgf2 579 pGEMZ2 (Promega) IV Fgf4 620 Bluescript (Stratagene) III Fgf9 (rat) 254 pAMP1 (Cloneamp System, Gibco) III Fgfr1b 133 PAMP1 (Cloneamp System, Gibco) IV Fgfr1c 125 PAMP1 (Cloneamp System, Gibco) IV Fgfr2b 161 Bluescript II KS+ (Stratagene) IV Fgfr2c 139 PAMP1 (Cloneamp System, Gibco) III, IV Fgfr3b 108 PAMP1 (Cloneamp System, Gibco) IV Fgfr3c 128 PAMP1 (Cloneamp System, Gibco) IV Gelatinase B 323 pSP64 (sense) / pSP65 (antisense) I Id1 900 pBluescript-SK (Stratagene) IV Msx1 600 pSP72 III Msx2 740 pSP72 II, III Proα1(II) Collagen 405 pBluescript II KS- (Stratagene) II, III Ptc 841 pBluescript II KS+ (Stratagene) III Shh 2600 pBluescript SK+ (Stratagene) III Twist 400 pT7T3 19U (Ambion) IV

All mouse unless stated

In-situ hybridisation on histological sections 35S-UTP labelled probes were precipitated with ethanol, air-dried and dissolved in 1M DL-Dithiothreitol (DTT) (Promega) and hybridisation buffer at a dilution of 1 to 9. The hybridisation buffer used was: 60% deionised formamide, 300 mM NaCl, 20 mM Tris-HCl pH 8.0, 5 mM EDTA pH 8.0, 10% dextran sulphate, 0.5 mg/ml yeast tRNA,

59 1×Denhardt’s solution consisting of 0.02% BSA, 0.02% ficollTM and 0.02% polyvinylpyrolidone in diethyl pyrocarbonate, in DEPC treated water. The final probe concentration was 40-60,000 counts/min/µl. Pretreatments were carried out which included digestion with proteinase K (7-10 µg/ml; Sigma) of the deparaffinised sections to enhance probe penetration. Following denaturation for 2mins at 80oC, sufficient probe (20-100 µl) was placed on each slide and then covered in parafilm. Overnight hybridisation was carried in a humidified sealed box at 52oC, followed by high stringency washes with 50% formamide and 20 mM DTT at 65oC. Slides were then prepared for autoradiography. The dehydrated slides were dipped into photographic emulsion (Kodak NTB-2), dried and exposed for between 10 and 21 days at 4oC. The slides were then developed (Kodak D-19), fixed (Kodak Unifix), then counterstained with Delafield’s haematoxylin and coverslips mounted with DePeX (BDG). For in-situ hybridisation of tissue sections with digoxigenin-UTP labelled riboprobes, similar pre hybridisation treatment used for 35S-UTP labelled riboprobes was employed. After measuring probe concentration and purity, hybridisation was carried out overnight at 65oC with a probe concentration of 0.5-1.0 µg/ml. Post treatment high stringency washes were carried out before antibody blocking and antibody staining with anti-DIG. Prior to colour development, slides were washed in MABT (0.1 M maleic acid, 0.15 M NaCl, 0.1% Tween 20, pH adjusted to 7.5) then NTMT (0.1 M Tris-HCL pH 9.5, 0.1 M NaCl, 50 mM MgCL, 0.1% Tween 20, 2-5 mM levamisole). Colour development was carried out in the dark for between 30 mins and 4 hours, the solution consisted of 4.5 µl NBT and 3.5 µl BCIP in 2 ml NTMT (NBT: 4- nitroblue tetrazolium chloride (BM 1087479), 75 mg/ml 70% dimethylformamide. BCIP: 5-bromo-4-chloro-3-inodyl-phosphate (BM 760994), 50 mg/ml dimethylformamide). Colour development was stopped by washing in PBS with 0.1% tween 20 (PBST). Following the colour reaction, sections were counterstained with van Gieson stain without its haematoxylin component, then dehydrated, washed twice in xylene for 10 mins, then coverslips mounted with DePeX (BDH).

In-situ hybridisation of whole-mounts Prehybridisation treatment for whole-mounts was as follows. Tissues were rehydrated, washed in PBST 3 times for 5 mins, then treated with prewarmed 7 µg/ml proteinase K (Sigma P-4914) in Tris/EDTA buffer (TE), for between 15 and 75 mins at 37oC, depending on the age of the tissue. They were then washed in PBST twice for 5 mins and fixed with freshly prepared 4% PFA, 0.2% glutaraldehyde for 20 mins, then rinsed in DEPC treated water and washed twice in PBST for 5 mins. Tissues were prehybridised in the hybridisation buffer for 5 hours at 55oC. The probe was diluted to 0.5 - 0.8 µg/ml in hybridisation buffer. The hybridisation buffer used for whole-mount tissues was: 50% deionised formamide, 1.3×SSC, 5mM EDTA, 0.5% CHAPS, 0.1% tween 20, 1% blocking powder (Boehringer), 100 µg/ml yeast tRNA, 50 µl/ml heparin, in DEPC treated water. Following overnight hybridisation at 55oC tissues were rinsed then washed twice for 60 mins in 50% formamide, 2×SSC, 1% tween 20 at 55oC (Wash 1). Tissues were

60 further washed twice in Wash 1 supplemented 1:1 with MABT for 40 mins at +55oC. Next they were washed twice in MABT for 30 mins at room temperature (RT), followed by 60 mins at RT in MABT plus 2%BBR (Boehringer Blocking Reagent in MAB). Tissues were then preblocked in 20% normal goat serum, 2% BBR in MABT for 2-4 hours at RT. Following this, tissues were incubated overnight at 4oC in alkaline phosphatase-coupled anti-DIG antibody which had been preabsorbed with embryo powder as reported earlier. Next, tissues were washed five times in MABT for 1 hour at RT, then in MABT overnight at 4oC. Three washes were carried out in NTMT for 30 mins at RT, prior to colour development. All post hybridisation washes were carried out with gentle rocking or rolling (MABT). Colour development solution was the same as that used for DIG labelled in-situ hybridisation on tissue sections, after which tissues were fixed in 4% PFA in PBS overnight, then washed in PBST before clearing in 50% then 70% glycerol in PBST.

Terminal Deoxynucleotidyl Transferase Mediated Nick End Labelling (TUNEL) To detect apoptotic cells, from both tissue sections and whole-mounts, TUNEL modified from Gavrieli et al. (1992) was employed. In TUNEL, a labelled nucleotide, in this case DIG, is incorporated into the 3’ end of DNA. As dying cells contain fragmented DNA they become heavily stained. These DIG labelled cells can be specifically detected using anti-DIG antibodies that are conjugated to alkaline phosphatase. Substrates for alkaline phosphatase can be detected using a NBT / BCIP colour reaction (See section on DIG labelled in-situ hybridisation). The TUNEL employed is outlined hereafter, and commences with tissue rehydration. Following this, treatment with acetic acid, hydrogen peroxide and then proteinase K was performed to enhance reagent penetration. To stop the proteinase K and refix the tissue 4% PFA treatment was carried out followed by incubation with the DIG-11-uridine labelling mixture in the presence of terminal transferase for 1 hour at 37oC. For negative controls, terminal transferase was omitted from the labelling mixture, and to check that the staining method was working correctly, in some tissues DNA was deliberately degraded with DNAase prior to incubation. Following the labelling, tissues were treated with 10% goat or sheep serum in order to block unspecific antibody binding. Fab fragments from anti-DIG antibodies conjugated to alkaline phosphatase were preabsorbed with embryo powder and then incubated with the tissues, after which the colour reaction was performed. Some tissue sections were counterstained with van Gieson stain without its haematoxylin component prior to being dehydrated, washed in xylene and mounted with DePeX (BDH, UK). If tissue sections were not counterstained coverslips were mounted directly with Aquamount (BDH, UK). TUNEL was performed on whole calvariae in a similar way as it was for sections, except labelling was carried out overnight at 37oC with rocking, and following the colour reaction, tissues were cleared in 50% followed by 70% glycerol in PBS.

61 Cell proliferation assay Cell proliferation was analysed by labelling calvarial explants with 5'-bromo-2'- deoxyuridine (BrdU; Boehringer-Mannheim). Following culture, tissues were incubated with BrdU (1:1000) for 60 mins in the culture medium, washed with 100% methanol and fixed with 100% methanol overnight. Samples were analysed as whole-mounts. After washing in PBS, incorporated BrdU was detected by an indirect immunoperoxidase method with monoclonal antibodies against BrdU and biotinylated secondary antibodies (Jackson Laboratories). The colour reaction was performed with a Vectastain ABC Elite Kit (Vector Laboratories).

Nile blue staining Lipophilic Nile Blue Sulphate is a vital dye that enters cells undergoing apoptosis (Ohkuma and Poole, 1981). Although its mechanism of action is unknown it correlates well with histomorphological criteria of apoptosis (Milaire, 1992). Following dissection, the unfixed tissues were stained with filtered, water-saturated Nile Blue Sulphate diluted to 1:1000 in PBS. The reaction was stopped by washing several times in PBS.

Image analysis Images were taken using a Cohu 4912-5000 CCD (Cohu, San Diego, CA, USA) camera and a Scion LG-3 Frame Grabber card (Scion, Frederick, MD, USA) with an Olympus BX50 microscope and a Macintosh PPC computer. Images were then processed with the NIH Image 1.61 program (US National Institute of Health, available from the Internet from zippy.nimh.nih.gov) and Adobe Photoshop 4.0 software (Adobe Systems, San Jose, CA, USA). To aid interpretation of gene expression using 35S labelled in-situ hybridisation, both dark and bright field images were taken. Silver grains in the dark field image were selected, coloured red and then superimposed onto the bright field image.

62 RESULTS AND DISCUSSION

Co-ordinated calvarial development is a balance between bone apposition and resorption Our first aim was to investigate the role of osteoblasts in calvarial bone and suture development. We therefore performed a detailed survey, to localise osteoblasts in the developing mouse calvaria, between the ages E10 and P4. By in-situ hybridisation, using both whole-mount and sectioned tissue, we detected messenger ribonucleic acid (mRNA) of two osteoblast markers, Bone sialoprotein (Bsp) and Core binding factor alpha 1 (Cbfa1). Cbfa1 as an early marker and Bsp as a late marker. Bsp is an extracellular matrix protein, produced by osteoblasts, that is thought to have an intimate role in the process of bone mineralisation. It was thus used as a marker of mature osteoblasts (Articles II and IV). We first detected Bsp messenger ribonucleic acid at E12 just lateral to the temporal cartilages, in a strip medial and superior to the eye extending occipitally (Figure 8). From these ossification centres of the frontal and parietal bones, the expression spread toward the apex of the cranium where the osteogenic fronts approximate to form a suture, two osteogenic fronts and intervening mesenchyme, at E15. Until E17 Bsp was expressed throughout the calvarial bones, most notably on their outer surfaces. In contrast, osteoclasts were found mainly on the endocranial surfaces (Article I). Thus, as the calvaria expands there is an intimate balance between bone apposition and resorption, thereby maintaining bone thickness and shape. After E17 transcripts became more restricted to areas of high osteogenic activity, notably the sutures (Figure 8). Cbfa1 is a transcription factor. Osf2 is a full length transcript of Cbfa1 that binds to the osteocalcin promoter, thus regulating osteoblast differentiation. Cbfa1 is known to be expressed by osteoblasts, and thought be an early marker. Our aim was to investigate whether Cbfa1 was indeed expressed early during osteoblastic differentiation, in-vivo. We used the calvarial model, where, unlike the long bone, bone forms intramembraneously without a cartilage performer . Here, osteoblasts and mesenchymal cells, from which osteoblasts differentiate can be clearly located. We first detected Cbfa1 at E11 in the head mesenchyme inferior-lateral to the developing brain, just lateral to the temporal cartilages. Like Bsp, Cbfa1 expression then spread rapidly through the calvarial mesenchyme toward the apex of the cranium. Here, the two parietal and two frontal bones confronted each other to form the interfrontal and sagittal sutures respectively (Figure 8). Once osteogenesis had been established, E12 onwards, Cbfa1 was expressed by both osteoblastic progenitors and also by mature osteoblasts. Thus, although distinctive, this expression pattern overlapped with that of Bsp. The localisation in osteoblastic progenitor cells was manifest as a wider area of expression than that of Bsp. This was clearly seen at E15, where Cbfa1 expression extended beyond the osteogenic fronts, where preosteoblasts are situated, into the mid-sutural mesenchyme (compare Figure 8 H and I).

After localising osteoblasts in the developing calvaria, our next objective was to investigate the role of osteoclasts in calvarial bone and suture development. To achieve

63 this we used a combination of in-situ hybridisation and enzyme histochemistry to localise two potential osteoclast markers, gelatinase B (Gel B) and tartrate-resistant acid phosphatase activity (TRAP) (Article I). Using calvariae from mice aged between E16 and P6, the expression patterns of Gel B mRNA and TRAP activity showed a remarkable cell and stage specificity. Some cells expressed Gel B mRNA but were TRAP negative, while others demonstrated an overlap in enzyme profile exhibiting both TRAP activity and expressing Gel B mRNA. During more advanced development, most osteoclasts exhibited TRAP activity but did not express gelatinase B mRNA. The distribution of these cells differed, TRAP positive cells being detected in a widespread pattern at all ages while Gel B transcripts were increasingly concentrated to areas of new and rapid bone growth, notably around the sutures. Indeed we propose that osteoclasts may also function in the maintenance of suture width. We detected TRAP + cells within the developing suture, for example, lining up on the endocranial surface of the parietal bone immediately opposite the encroaching frontal bone osteogenic front. Gel B + cells were also detected at this location. The maintenance of suture width or patency is of crucial importance whilst the cranium is still growing and premature suture closure, by fusion of osteogenic fronts, can lead to gross skull deformity, notably in syndromic and non syndromic craniosynostoses. This hypothesis is supported by work in an ex-vivo suture fusion model (Winograd et al., 1997b). The presence of Gel B + osteoclasts and TRAP activity already at E16 suggests that modelling starts early in bone development and hence that osteoclasts play an integral part from the outset. These results and conclusions are supported by more recent data (Jemtland et al., 1998). Using the foetal mouse hind limb as a model system, they, too, found osteoclasts to exhibit considerable heterogeneity in gene expression patterns. The observation that osteoclasts expressing Gel B mRNA may be TRAP negative indicates that the abundance of osteoclasts during early development may have gone unnoticed in previous studies in which TRAP activity has been localised. Indeed, Gel B + cells were detected in the growth plates of c-Fos deficient mice, in which TRAP + osteoclasts were absent (Grigoriadis et al., 1994). The Gel B knockout mouse has recently been shown to present an interesting bony phenotype (Vu et al., 1998). The long bones are approximately 10% shorter with ‘chondroclasts’ failing to resorb the zone of hypertrophic chondrocytes in the growth plate. The authors propose that Gel B is a regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes.

Thus, we have seen that not only osteoblasts but also osteoclasts are active during very early bone formation, and that intramembraneous bone development, at a cellular level, appears to be not a random but, instead, a highly co-ordinated process. When the balance between apposition and resorption is disrupted, pathological conditions affecting the calvaria, such as cleidocranial dysplasia and craniosynostosis, will presumably result. This baseline data, of cell type and position, forms a start point from which further studies of cell turnover and molecular signalling may be carried out.

64 Cell turnover in calvarial bone and suture development One way in which bone apposition or resorption may be controlled is by varying the number of functioning cells, namely osteoblasts and osteoclasts. The number of cells present is dependent on their formation and proliferation, and on their death and ultimate removal. We therefore undertook a study to look at cell turnover in the developing calvaria and sutures. Cell proliferation was studied in E15 calvarial explants by 5'-bromo-2'-deoxyuridine (BrdU) incorporation (Article III). It was found that there was indeed BrdU labelling in calvarial mesenchymal and osteoprogenitor cells, and that beads soaked in FGF4, placed on the osteogenic fronts, induced an increase in this labelling. A more extensive survey was not performed as several other investigators have carried out more detailed studies using both BrdU uptake and tritiated thymidine incorporation (Iseki et al., 1997; Opperman et al., 1998; Liu et al., 1999). They have elegantly shown that there is indeed extensive cell proliferation in the developing calvaria, most notably in osteogenic progenitor cells close to or at the osteogenic fronts. This proliferation may well be influenced by the underlying dura mater. In order to study cell death in the developing calvaria, Terminal Deoxynucleotidyl Transferase Mediated Nick End Labeling (TUNEL) was performed on whole-mount and sectioned calvaria from mice aged between E14 and P6 (Article II). In an attempt to identify the cell types labelled positive for TUNEL, we performed TRAP staining, to identify osteoclasts and in-situ hybridisation to detect Bsp mRNA to identify osteoblasts. TUNEL-positive cells were found from E16 onwards in the calvarial bones, and intervening sutures and fontanelles. TUNEL-positive osteoblasts and preosteoblasts were identified at or close to the osteogenic fronts, areas of intense osteogenic activity with TUNEL-positive mesenchymal cells located in the mid-sutural mesenchyme. TUNEL-positive osteoclasts and osteocytes were also observed in a sporadic fashion, as well as TUNEL-positive dural cells. The distribution of TUNEL-positive cells was not random, but rather appeared to be developmentally regulated. Hence, it appears that apoptosis is part of normal cranial development. We also analysed, by in-situ hybridisation of both whole-mount and sectioned tissue, the expression of Msx2, Bmp4 and Bmp7 genes, which are known to act in conserved signalling pathways leading to apoptosis. Transcripts of Msx2 were detected in the mesenchymal tissue of the anterior and posterior fontanelles, sagittal, coronal and lambdoidal sutures and the underlying dura. Bmp4 transcripts were detected in the osteogenic fronts of the coronal, sagittal and lambdoid sutures and also in the dura and weakly in the mid-sutural mesenchyme. Transcripts of Bmp7 were found intensely in the frontal and parietal bones, notably at the osteogenic fronts, as well as in the overlying skin including developing hair follicles and underlying meningeal layers. They were also seen with weaker intensity in the mid-sutural mesenchyme. The expression of Msx2, Bmp4 and Bmp7 correlated with TUNEL-positivity in some but not all locations. We have detected several cell types involved in bone formation and resorption undergoing apoptosis in the developing calvaria, and we propose that this might be a mechanism whereby the rate of osteogenesis is controlled. Furthermore, from our analysis of Msx2, Bmp4 and Bmp7 mRNA expression we suggest that this apoptosis may be via a BMP/MSX pathway, and speculate that this is involved in normal suture formation and

65 the maintenance of suture patency.

As the calvaria forms largely from intramembraneous ossification and is easily accessible , it may well prove to be a good model for studying apoptosis of bone cells. This would be at a tissue level, as opposed to isolated cell culture, where the effects of an variety of growth factors and drug aimed at altering the process of apoptosis in bone cells may be tested. The process of apoptosis is common to many developing tissues. Here, we have seen that apoptosis, like in many other tissues, may play a role in the regulation of a size of a cell population, notably in the number of osteoblasts, but also osteoclasts, differentiating and functioning. Thus, affecting the balance between bone formation and degradation. We also know that in other tissue apoptosis may function in establishing the morphology or shape of a particular tissue. Here too, in the developing calvaria, with the knowledge that apoptosis occurs in cells of both osteoblast and osteoclast lineages, it may be extrapolated that a change in morphology of a bone may be evoked if the amount of apoptosis in either cell population varies. In the main, this might have an affect on bone thickness but may possibly also affect the overall size and shape of an individual bone. However, it must be remembered that the overall size and shape of a particular bone is largely determined by other over overriding factors. Of course, if apoptosis is proved to influence the balance between bone formation and resorption in the calvaria, notably in the sutures, then ultimately a change in the whole head shape may occur if premature fusion of the calvarial bones takes place with consequent skull deformity.

Figure 8. Localisation by in-situ hybridisation of mRNA of Cbfa1 and Bsp, an early and late marker of osteoblastic differentiation, during calvarial bone and suture development. (A) Schematic diagram to illustrate calvarial bone growth and suture formation. Formation of the frontal and parietal bones starts on the lateral side of the head as a condensation of mesenchymal cells, at around E12. These cells differentiate into osteoblasts, which start to lay down extracellular matrix, which mineralizes into bone. Ossification spreads upward toward the apex of the cranium to meet in the midline where a suture is formed. Growth of the calvarial bones is co-ordinated between the expanding underlying brain and overlying skin. (D,G and J) Whole-mounts showing the distribution of Bsp mRNA from E14 to E17. (B,E,H and K) Frontal sections showing the localisation of Bsp transcripts from E12 to E17. Bsp is expressed by mature osteoblasts, being seen only in cells close to the bone surface. (C,F,I and L) Frontal sections showing the distribution of Cbfa1 transcripts in the developing calvaria. Cbfa1 is expressed early during osteoblastic differentiation. Thus, osteoprogenitor cells, pre-osteoblasts and mature osteoblasts are all seen to express Cbfa1. This is illustrated by its slightly area of expression when compared to Bsp. In I Cbfa1 transcripts can clearly be seen stretching beyond the osteogenic fronts, into the intersutural mesenchyme (compare to H). White dotted line in J illustrates the plane of section (frontal) in the histological sections shown. br, brain; ca, cartilage; ep, epithelium; f, frontal bone; ip, interparietal bone; of, osteogenic front; mes, sutural mesenchyme; p, parietal bone. Scale bars: whole-mounts 1mm; sections 200µm.

66 %63 %63 &EID Suture

Parietal bone Parietal bone E12 E12

$ E12 % & E17

ep

E14 br ca

' ( )

p p p E15 of of

p * + , mes

E17

- . /

67 P51$H[SUHVVLRQ $ )JIUF % )JI

of of

mes

& 0V[ ' 7ZLVW

)XQFWLRQDOVWXGLHV ( 7ZLVW ) 7ZLVW S S S

)*) S EHDGV )*) EHDGV

* 0V[ + %VS

S %03 RI EHDGV )*)EHDG

68 Figure 9. Summary of some of the key findings of the thesis. Genes known to cause craniosynostosis were localised in the developing calvaria by in-situ hybridisation. These show unique but overlapping expression patterns. (A) Fgfrs1, 2 and 3 are expressed at the osteogenic fronts of the developing calvarial bones, Fgfr2c extending into the sutural mesenchyme, all be it with decreased intensity. (B) Fgf2 is mainly expressed in the mesenchyme but also in the calvarial bones. (C) Msx2 is also expressed in the sutural mesenchyme but also by cells at the osteogenic front. (D) Twist is expressed in the mesenchyme and as osteoblastic cells increase in maturity so Twist expression diminishes. Thus it would appear that Twist is expressed early in osteoblast differentiation then Msx2 is switched on and then the Fgfrs. On the basis of these findings bead assay experiments in in-vitro culture with the following findings. (E and F) FGF2 beads upregulate Twist in both whole-mount (E) and sectioned tissue (F). (G) BMP2 as well as 4 upregulate Msx2 in the calvarial mesenchyme. (H) FGF beads were found to accelerate sutural closure when placed on the parietal bone osteogenic fronts. However, curiously although FGF stimulated Bsp expression in mesenchymal cells it also inhibited Bsp expression in mature osteoblasts. Thus, FGF may have a dual effect during osteoblastic differentiation. of, osteogenic front; mes, sutural mesenchyme; p, parietal bone. Scale bars: whole-mounts 1mm; sections 200µm. All sections are the same scale.

Molecular signalling in calvarial bone and suture development

Gene expression patterns To elucidate the roles of FGFs, BMPs and SHH as well as their receptors and transcriptional targets in calvarial bone and suture development, we have performed a detailed survey of the gene expression of the splicing alternatives of Fgfr1, 2 and 3, and some potential ligands Fgf2, 4, 8 and 9. Also Bmp2, 4 and 7, Shh and its receptor Ptc, as well as the transcription factors Msx1 and 2, Twist and Id were analysed in mice aged between E10 and P6 (Figure 9, and articles II, III and IV). Distinct patterns of expression were found. Except for Fgfr1 IIIb, all Fgfrs studied were expressed in the sutural osteogenic fronts of the calvarial bones. Notably, transcripts of Fgfr2 IIIc were detected with high intensity (Figure 9A). Fgfr3 IIIb and IIIc were chiefly found in cartilage, while Fgfr2 IIIb was detected in the periostea and perichondria. In contrast to the receptor expression, Fgf2 and Fgf9 mRNA were found in the sutural mesenchyme between the calvarial bones, as well as weakly in the developing calvarial bones and underlying meninges. Fgf2 had a more discrete distribution than Fgf9. These results correlate well with more recently published gene expression patterns of FGFR in human calvaria, Fgf2 also correlates with immuno-distribution studies in rats (Gonzalez et al., 1990; Delezoide et al., 1998). Msx1 and Msx2 were expressed in the sutural mesenchyme and the dura mater. Bmp2 and Bmp4 were intensely expressed in the osteogenic fronts and Bmp4 also in the mesenchyme of sagittal suture. Bmp7 was detected at similar locations as Bmp4 but with noticeably stronger intensity in the meninges and overlying epidermis. Interestingly, Shh and Ptc started to be expressed in patched pattern along the osteogenic fronts only at the end of embryonic development, and at this time the expression of Bmp4, and

69 sequentially those of Msx2 and Bmp2 were dramatically reduced and they also acquired patched expression patterns. In addition, the expression of Msx2 gene in the dura mater disappeared after birth. Early in head development (E10-11) both Twist and Id1 were intensely and extensively expressed in mesenchyme of the first and second branchial arches, as well as in the mesenchyme surrounding the developing eye and cranial mesenchyme just beneath the epithelium. Id1 also was expressed in epithelia adjacent to these mesenchymal areas. Expression then became more restricted so that by E14 transcripts were seen bordering areas of condensing calvarial mesenchyme. While Twist continued to be expressed in the calvarial mesenchyme until postnatal stages, by E15 levels of Id mRNA had decreased. Thus, Twist and Id1 were not expressed by preosteoblasts at the osteogenic front but by neighbouring osteoprogenitor cells, in patterns that overlapped those of the Fgf ligands studied. As these osteoprogenitor cells differentiated expression of Twist and Id1 was lost. This is in agreement with expression studies in osteoblastic cell culture, where Twist and Id are expressed very early in osteoblast differentiation but decrease with increasing maturity (Murray et al., 1992). A schematic summary of some of the genes both we and others have localised is shown in Figure 10.

2VWHRJHQLF IURQW  )JIU )JI 0V[  %PS

%RQH %RQH

6XWXUDO PHVHQFK\PH 'XUD  0V[ %PS )JI )JI 7ZLVW 0V[

Figure 10. Schematic diagram illustrating some of the genes active during suture development.

70 In summary, the genes investigated have differing, but overlapping, mRNA expression patterns. This can be illustrated by looking at Fgfr2c, one of its ligands Fgf2, Msx2 and Twist, that is to say some of the genes known to cause craniosynostosis (Figure 9A-D). Fgfr2c was chiefly expressed at the osteogenic fronts of the developing calvarial bones. Fgf2 is mainly expressed in the mesenchyme but also at the osteogenic fronts and in the calvarial bones. Msx2 also is expressed in the sutural mesenchyme and cells at the osteogenic front. Whilst Twist is expressed in the mesenchyme and as osteoblastic cells increase in maturity so Twist expression diminishes, that is to say it is not expressed at the osteogenic fronts. Thus it would appear that Twist is expressed early in osteoblast differentiation then Msx2 is switched on and then the Fgfrs. On the basis of these findings functional experiments were performed.

Functional studies We have devised an in-vitro model system to enable us to investigate molecular signalling in the developing calvaria (Articles III and IV). At E15 osteogenesis in the mouse calvaria has commenced. However, the sutural spaces are still wide, and the tissue thin enough to permit its development in-vitro. Explants developed in a manner consistent with suture development in-vivo (Figure 7). The opposing parietal bone osteogenic fronts approximated but did not fuse. The osteogenic fronts usually turned back on themselves away from the mid-sutural mesenchyme, exhibiting a morphology somewhat reminiscent of the osteogenic fronts in an E15 sagittal suture being downturned endocranially before establishing the characteristic butt joint morphology. Occasionally the confronting osteogenic fronts would over-shoot each other, resulting in an overlapping suture. This is not entirely surprising, as the culture system does not have the same mechanical pressures on the developing tissues as would be the case in the developing foetus, as the rapidly expanding underlying brain and the overlying skin have been removed. Not only did the two parietal bones come together to form a suture, but there was also an increase in the thickness of each bone, illustrated in Figure 7C and E. The approximation of the calvarial bones was also demonstrated by performing in-situ hybridisation, for the osteoblast marker Bsp, on whole-mount explants cultured for 1 to 4 days (Figure 7B and D). The FGF and BMP signalling pathways were examined in-vitro (Figure 9E-H, and articles III and IV). Mouse calvaria at E15.5 were cultured as explants and beads soaked in FGF4 or BMP4 protein were placed on the sutural tissue. Interestingly, FGF4 beads when placed only on the osteogenic fronts accelerated sutural closure, whereas BMP4 beads caused an increase in tissue volume when placed both on the osteogenic fronts and on the mid-sutural area. FGF4 induced the expression of Msx1 gene in sutural tissue, while BMP induced both Msx1 and Msx2 (Figure 9G). FGF4 stimulated cell proliferation around the beads. We suggest that the local application of FGF on the osteogenic fronts, accelerating suture closure in-vitro, mimics the pathogenesis of human craniosynostosis syndromes in which mutations in the FGF receptor genes apparently cause constitutive activation of the receptors. These data not only highlight the role of FGF, but also that of both BMPs and MSXs in calvarial development. As the expression patterns of Fgf2 and Twist were found to be overlapping,

71 signalling was further investigated in-vitro. In E15 mouse calvarial explants, beads soaked in FGF2 induced Twist. This is in accordance with an earlier report by McDougall et al. (1997), who found a similar upregulation in osteoblast cell cultures. There is evidence that this induction maybe via NF-κB signalling. Inhibition of the transcription factor NF- κB in the developing limb results in a reduction of Twist (Bushdid et al., 1998). Interestingly, NF-κB is thought to be a negative regulator of osteoblast function and as the transcription of Rel/NF-κB is regulated by FGF4, it can be speculated that NF-κB regulates FGF mediated Twist induction. In the developing limb bud, inhibition of the NF- κB has also been shown to cause a derepression, or an increase in, Bmp4 (Bushdid et al., 1998). Interestingly, a homologous pathway has been found during Drosophila mesoderm induction and dorso-ventral patterning. Here, high nuclear concentrations of the protein Dorsal, the NF-κB homologue, activate twist and repress decapentaplegic, the BMP homologue (Gilbert, 1997 review). We also found that in calvarial mesenchyme BMP, known to stimulate osteoblastic differentiation, upregulated Id1. Id1 is a dominant negative HLH thought to inhibit basic HLH, such as Twist by directly competing for Twist binding sites. Thus in the mouse osteoblast differentiation model, we propose that NF-κB signalling inhibits Bmp and thus decreases Id’s repression of Twist. Hence NF-κB may serve to maintain osteoblasts in a relatively undifferentiated state by acting both directly on twist and indirectly via Bmp. This model is summarised in Figure 11. In Drosophila, the Fgf receptor homologue, DFR1, is known to be downstream of Drosophila Twist, null embryos showing a reduction in DFR1 expression in mesodermal cells. We therefore investigated whether this signalling pathway may be conserved in the mouse and analysed the protein distribution of Fgfrs in a Twist heterozygote background. We found that Fgfr2 and Fgfr3 protein were distributed at an ectopic location, in the central region of the sutural mesenchyme, interestingly where Twist mRNA is normally expressed. Fgfr2c mRNA is normally intensely expressed by differentiating osteoblasts at the osteogenic front and at much lower levels in the intervening mesenchyme. It would appear that as Twist levels decrease, Fgfr distribution is altered (Article IV). Together with these findings, we have shown that Twist is expressed in early osteoblastic cells and that its expression is lost as these cells differentiate. Also it is known that a loss of Twist/TWIST function leads to enhanced bone formation (heterozygote phenotype) and also craniosynostosis. In addition, we found that Fgfrs are expressed later in osteoblast development. Also, gain of FGFR/FGF function leads to a similar phenotype as that seen by a loss of TWIST function, namely enhanced bone formation and craniosynostosis. Taken together, it seems reasonable to propose that Twist and Fgfr lie in the same pathway, with Twist maintaining cells in a relatively undifferentiated state in a similar manner to that seen observed during vertebrate muscle development but differing from that seen in the Drosophila embryo, in that murine Twist appears to inhibit and not stimulate FGFR/FGF signalling (Hebrok et al., 1994; Bayliss and Bate, 1996). Thus, Twist would appear to be upstream of Fgfr/Fgf signalling, though whether it is inhibitory or stimulatory can not yet be definitively concluded, and whether other intermediaries such as Msx2 lie in between, remains to be seen.

72 The effects of FGF are not, however, crystal clear. Enhanced calvarial bone formation and craniosynostosis are caused by constitutive activation of the receptors in humans and excess FGF ligand in the Bey mouse. In addition, in our own experiment FGF placed on the osteogenic fronts causes an acceleration of suture closure. However, FGF is also known to inhibit osteoblast differentiation and promote proliferation. We also see that FGF can upregulate Twist, an event early in osteoblast differentiation, and that this may promote osteoblastic cells to proliferate rather than terminally differentiate. These discrepancies can, at present, not be reconciled. However, it should be noted that FGF2 beads placed on the calvaria has both stimulatory and inhibitory effects on Bsp expression. Close to the bead Bsp is induced in mesenchymal cells while further away, in more mature cells, Bsp is inhibited (Figure 9H). This fascinating observation suggests that FGF2 beads appear to have a dual effect on Bsp expression. This may be a concentration dependent response, similar to the response of Drosophila twist to Dorsal, or a response dependent on the stage of osteoblast differentiation, or possibly a combination of the two. In summary, we propose a model of osteoblast differentiation integrating Twist and Fgf in the same pathway, in which Fgf acts both at early and late stages (Figure 11). Disruption of this network may lead to craniosynostosis.

%PS ,G )JI

1).% 7HUPLQDO )JI 7ZLVW )JIU %VS 'LIIHUHQWLDWLRQ

2VWHREODVW

'LIIHUHQWLDWLRQ 0HVHQFK\PDO 0DWXUH FHOO RVWHREODVW

Figure 11. Schematic illustration showing a possible model of osteoblast differentiation with proposed integration of Twist and FGF.

73 CONCLUDING REMARKS

Over the past few years we have seen our knowledge of calvarial bone and suture development and the causes of pathological conditions affecting the calvaria expand rapidly. Foremost, has been the discovery of mutations in human genes that cause craniosynostoses. These findings, however, are merely a first step in understanding the aetiology and pathogenesis of these disorders. The challenge is now to combine this knowledge with more classical ideas and experimentation of suture biology, as well as making use of the array of more modern techniques open to us. We have endeavoured to study, in normal tissue, the expression some of the genes known or thought to be involved in the developing calvaria. We have attempted to combine this information to formulate working hypotheses on how these genes may interact, and then test these ideas in an in-vitro tissue culture system, as well as in the calvaria of mice known to have mutations that cause craniosynostosis. We put forward the developing mouse calvaria as useful model system, for studying, not only, osteoblast differentiation during intramembraneous ossification, but also other aspects of bone and suture development, such as bone modelling and remodelling. Although this system does not fulfill all our criteria, it nonetheless does offer us a good tool to work with and provides us with valuable information. For example, we have not studied craniosynostosis per se; however, data gained can certainly contribute to our understanding of the molecular basis of craniosynostosis. Indeed, one of the challenges in the future will be to apply knowledge gained in this system, when studying animals which exhibit a calvarial phenotype. These investigations, integrated into a background of data from osteoblastic cell culture work, together with data from a variety of in-vivo studies, may well yield significant advances in our understanding of the pathogenesis of calvarial disorders. Also, much can be gained from studying how other systems and organs develop and applying this knowledge of molecular signalling networks to the developing calvaria. Indeed, we have seen how the same genes involved in calvarial bone and suture formation are the same as those involved in many other systems in the body. And one of the key unanswered questions in developmental biology is, what specifies one tissue to form and not another? It should be remembered that the developing and growing skull including the calvaria is a dynamic system, constantly adjusting to changing demands. Underpinning this development are the basic biological processes of bone formation and resorption, including osteoblast and osteoclast differentiation, proliferation, function, interaction and ultimately cell death. Only by having a sound comprehension of these fundamental processes, as well as a better understanding of the molecular signalling involved in calvarial bone and suture development, both physiological and pathological, can we start to develop effective treatment regimes for patients with conditions affecting the calvaria such as craniosynostosis.

74 ACKNOWLEDGEMENTS

The production of a doctoral thesis in Finland is a fruitful but long process. This way has been paved smooth by many individuals to whom I am indebted. I am sincerely grateful to my supervisor Professor Irma Thesleff, Head of the Department of Orthodontics and Pedodontics, and Director of the Developmental Biology Programme. I am grateful for taking me into her group and providing me with a first class working environment, which has allowed me a small degree of controlled freedom, which I have relished. Irma has been a continual source of inspiration, and although Irma commands an elevated position of international renown she always has had her door open and time for even the most lowly doctoral student, with the most basic of questions. I have the deepest admiration for Irma and feel fortunate to have had her as my mentor. Thank you. Professor Sinikka Pirinen, Department of Orthodontics and Pedodontics, is a colleague and a friend, for whom I hold the highest regard and to whom I owe a debt of gratitude. She has been extremely supportive and taken good care of my academic, clinical, professional and, at times, social well being. I thank Professor Mart Saarma, Director of the Institute of Biotechnology. He has not only provided me with excellent work facilities but has always been encouraging and interested in my work. I thank the reviewers of this work, Professor Seppo Vainio and Professor Kalervo Väänänen for their constructive criticism. I express my heartfelt thanks to Ritva Härönen, Janna Waltimo and Thomas Åberg. Working in the ‘bone group’ has been, and I hope will continue to be, both productive and great fun. Working with individuals whom you respect is good, but working with individuals whom you respect and whom are among your closest of friends is fantastic. I am indebted to technical and organisational staff of both the lab and the clinic, particularly Kaija Kettunen, Merja Mäkinen, Riikka Santalahti and Helena Zilliacus- Kauhanen. Their help is duely acknowledged. Throughout my career, my father has been an invaluable source of advice, a tower of strength and a trusted confidant. He has given me unwavering support and I feel very lucky to have him as a father. I only wish I could have inherited some of his organisational skills. The European Orthodontic Society, the Sigrid Jusélius Foundation and the Wellcome Trust have all supported this work. This is greatly appreciated. Finally, I wish to thank all my colleagues, family, friends and fellow BI Knockouts salybandy team members for their help and support.

David Rice Helsinki, October 1999

75 REFERENCE LIST

Alborzi, A., Mac, K., Glackin, C.A., Murray, S.S., and Zernik, J.H. (1996). Endochondral and intramembranous fetal bone development: osteoblastic cell proliferation, and expression of alkaline phosphatase, m- twist, and histone H4. J. Craniofac.Genet. Dev. Biol. 16, 94-106. Alexander, C.M., Hansell, E.J., Behrendtsen, O., Flannery, M.L., Kishnani, N.S., Hawkes, S.P., and Werb, Z. (1996). Expression and function of matrix metalloproteinases and their inhibitors at the maternal- embryonic boundary during mouse embryo implantation. Development 122, 1723-1736. Anderson, J., Burns, H.D., Enriquez-Harris, P., Wilkie, A.O.M., and Heath, J.K. (1998). Apert syndrome mutations in fibroblast growth factor receptor 2 exhibit increased affinity for FGF ligand. Hum. Mol. Genet. 7, 1475-1483. Aspenberg, P. and Lohmander, L.S. (1989). Fibroblast growth factor stimulates bone formation. Bone induction studied in rats. Acta Orthop. Scand. 60, 473-476. Aubin, J.E. (1998). Bone stem cells. J. Cell Biochem. 73-82. Aubin, J.E., and Liu, F. (1996). The osteoblast lineage. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. ed. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 5. Baron, R., Neff, L., Tran, V.P., Nefussi, J.R., and Vignery, A. (1986). Kinetic and cytochemical identification of osteoclast precursors and their differentiation into multinucleated osteoclasts. Am. J. Pathol. 122, 363- 378. Basilico, C., and Moscatelli, D. (1992). The FGF family of growth factors and oncogenes. Adv. Cancer Res. 59, 115-165. Bate, M., Rushton, E., and Currie, D.A. (1991). Cells with persistent twist expression are the embryonic precursors of adult muscles in Drosophila. Development 113, 79-89. Baylies, M.K. and Bate, M. (1996). twist: a myogenic switch in Drosophila. Science 272, 1481-1484. Bee, J., and Thorogood, P. (1980). The role of tissue interactions in the skeletogenic differentiation of avian neural crest cells. Dev. Biol. 78, 47-62. Bellahcene, A., Merville, M.P., and Castronovo, V. (1994). Expression of bone sialoprotein, a bone matrix protein, in human breast cancer. Cancer Res. 54, 2823-2826. Bellows, C.G., Aubin, J.E., and Heersche, J.N. (1991). Initiation and progression of mineralization of bone nodules formed in vitro: the role of alkaline phosphatase and organic phosphate. Bone Miner. 14, 27- 40. Benezra, R., Davis, R.L., Lassar, A., Tapscott, S., Thayer, M., Lockshon, D., and Weintraub, H. (1990). Id: a negative regulator of helix-loop-helix DNA binding proteins. Control of terminal myogenic differentiation. Ann. N.Y. Acad. Sci. 599, 1-11. Bennett, J.H., Joyner, C.J., Triffitt, J.T., and Owen, M.E. (1991). Adipocytic cells cultured from marrow have osteogenic potential. J. Cell Sci. 99, 131-139. Benson, M.D., Aubin, J.E., Xiao, G., Thomas, P.E., and Franceschi, R.T. (1999). Cloning of a 2.5 kb murine bone sialoprotein promoter fragment and functional analysis of putative Osf2 binding sites. J. Bone Miner. Res. 14, 396-405. Bianco, P., Fisher, L.W., Young, M.F., Termine, J.D., and Robey, P.G. (1991). Expression of bone sialoprotein (BSP) in developing human tissues. Calcif. Tissue Int. 49, 421-426. Bitgood, M.J., and McMahon, A.P. (1995). Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev. Biol. 172, 126-138. Blavier, L., and Delaissé, J.M. (1995). Matrix metalloproteinases are obligatory for the migration of preosteoclasts to the developing marrow cavity of primitive long bones. J. Cell Sci. 108, 3649-3659. Bolk, (1915). On the premature synostosis of the cranial sutures. Acta Ophthalmol. 51 Suppl., 1-174. Cited by Cohen, M.M.Jr. (1986). Perspectives on craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 2. Bonadio, J., Saunders, T.L., Tsai, E., Goldstein, S.A., Morris-Wiman, J., Brinkley, L., Dolan, D.F., Altschuler, R.A., Hawkins, J.E.Jr., Bateman, J.F., et al., (1990). Transgenic mouse model of the mild dominant form of osteogenesis imperfecta. Proc. Natl. Acad. Sci. USA 87, 7145-7149. Boskey, A.L., Maresca, M., Ullrich, W., Doty, S.B., Butler, W.T., and Prince, C.W. (1993). Osteopontin- hydroxyapatite interactions in vitro: inhibition of hydroxyapatite formation and growth in a gelatin- gel. Bone Miner. 22, 147-159. Bourez, R.L., Mathijssen, I.M., Vaandrager, J.M., and Vermeij-Keers, C. (1997). Apoptotic cell death during normal embryogenesis of the coronal suture: early detection of apoptosis in mice using annexin V. J. Craniofac. Surg. 8, 441-445. Bourgeois, P., Bolcato-Bellemin, A.L., Danse, J.M., Bloch-Zupan, A., Yoshiba, K., Stoetzel, C., and Perrin- Schmitt, F. (1998). The variable expressivity and incomplete penetrance of the twist-null heterozygous mouse phenotype resemble those of human Saethre-Chotzen syndrome. Hum. Mol. Genet. 7, 945-957. Boyce, B.F., Wright, K., Reddy, S.V., Koop, B.A., Story, B., Devlin, R., Leach, R.J., Roodman, G.D., and Windle,

76 J.J. (1995). Targeting simian virus 40 T antigen to the osteoclast in transgenic mice causes osteoclast tumors and transformation and apoptosis of osteoclasts. Endocrinology 136, 5751-5759. Bradley, J.P., Levine, J.P., McCarthy, J.G., and Longaker, M.T. (1997). Studies in cranial suture biology: regional dura mater determines in vitro cranial suture fusion. Plast. Reconstr. Surg. 100, 1091-1099. Bradley, J.P., Levine, J.P., Roth, D.A., McCarthy, J.G., and Longaker, M.T. (1996). Studies in cranial suture biology: IV. Temporal sequence of posterior frontal cranial suture fusion in the mouse. Plast. Reconstr. Surg. 98, 1039-1045. Bronckers, A.L., Gay, S., Finkelman, R.D., and Butler, W.T. (1987). Developmental appearance of Gla proteins (osteocalcin) and alkaline phosphatase in tooth germs and bones of the rat. Bone Miner. 2, 361-373. Bronckers, A.L., Goei, W., Luo, G., Karsenty, G., D’Souza, R.N., Lyaruu, D.M., and Burger, E.H. (1996). DNA fragmentation during bone formation in neonatal rodents assessed by transferase-mediated end labeling. J. Bone Miner. Res. 11, 1281-1291. Bronner-Fraser, M., and Fraser, S.E. (1988). Cell lineage analysis reveals multipotency of some avian neural crest cells. Nature 335, 161-164. Bucay, N., Sarosi, I., Dunstan, C.R., Morony, S., Tarpley, J., Capparelli, C., Scully, S., Tan, H.L., Xu, W., Lacey, D.L., Boyle, W.J., and Simonet, W.S. (1998). Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev. 12, 1260-1268. Bushdid, P.B., Brantley, D.M., Yull, F.E., Blaeuer, G.L., Hoffman, L.H., Niswander, L., and Kerr LD. (1998). Inhibition of NF-kappaB activity results in disruption of the apical ectodermal ridge and aberrant limb morphogenesis. Nature 392, 615-618. Cam, Y., Neumann, M.R., Oliver, L., Raulais, D., Janet, T., and Ruch, J.V. (1992). Immunolocalization of acidic and basic fibroblast growth factors during mouse odontogenesis. Int. J. Dev. Biol. 36, 381-389. Canalis, E., Centrella, M., and McCarthy, T. (1988). Effects of basic fibroblast growth factor on bone formation in vitro. J. Clin. Invest. 81, 1572-1577. Canalis, E., and Raisz, L.G. (1980). Effect of fibroblast growth factor on cultured fetal rat calvaria. Metab. Clin. Exp. 29, 108-114. Canete-Soler, R., Gui, Y.H., Linask, K.K., and Muschel, R.J. (1995). Developmental expression of MMP-9 (gelatinase B) mRNA in mouse embryos. Dev. Dyn. 204, 30-40. Celli, G., Larochelle, W.J., Mackem, S., Sharp, R., and Merlino, G. (1998). Soluble dominant-negative receptor uncovers essential roles for fibroblast growth factors in multi-organ induction and patterning. EMBO J. 17, 1642-1655. Chang, W.S., Serikawa, K., Allen, K., and Bentley, D. (1992). Disruption of pioneer growth cone guidance in vivo by removal of glycosyl-phosphatidylinositol-anchored cell surface proteins. Development 114, 507- 519. Chen, H., Ovchinnikov, D., Pressman, C.L., Aulehla, A., Lun, Y., and Johnson, R.L. (1998). Multiple calvarial defects in mutant mice. Dev. Genet. 22, 314-320. Chen, J.K., Shapiro, H.S., Wrana, J.L., Reimers, S., Heersche, J.N., and Sodek, J. (1991). Localization of bone sialoprotein (BSP) expression to sites of mineralized tissue formation in fetal rat tissues by in situ hybridization. Matrix 11, 133-143. Chen, Y., Bei, M., Woo, I., Satokata, I., and Maas, R. (1996). Msx1 controls inductive signaling in mammalian tooth morphogenesis. Development 122, 3035-3044. Chen, Z.F., and Behringer, R.R. (1995). Twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev. 9, 686-699. Chiang, C., Litingtung, Y., Lee, E., Young, K.E., Corden, J.L., Westphal, H., and Beachy P.A. (1996). Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature 383, 407-413. Chuang, P.T., and McMahon, A.P. (1999). Vertebrate Hedgehog signalling modulated by induction of a Hedgehog- binding protein. Nature 397, 617-621. Clark, S.A., Ambrose, W.W., Anderson, T.R., Terrell, R.S., and Toverud, S.U. (1989). Ultrastructural localization of tartrate-resistant, purple acid phosphatase in rat osteoclasts by histochemistry and immunocytochemistry. J. Bone Miner. Res. 4, 399-405. Clover, J., Dodds, R.A., and Gowen, M. (1992). Integrin subunit expression by human osteoblasts and osteoclasts in situ and in culture. J. Cell Sci. 103, 267-271. Coffin, J.D., Florkiewicz, R.Z., Neumann, J., Mort-Hopkins, T., Dorn, G.W., Lightfoot, P., German, R., Howles, P.N., Kier, A., and O’Toole, B.A. (1995). Abnormal bone growth and selective translational regulation in basic fibroblast growth factor (FGF-2) transgenic mice. Mol. Biol. Cell 6, 1861-1873. Coelho, C.N., Upholt, W.B., and Kosher, R.A., (1993). The expression pattern of the chicken homeobox- containing gene GHox-7 in developing polydactylous limb buds suggests its involvement in apical ectodermal ridge-directed outgrowth of limb mesoderm and in programmed cell death. Differentiation 52, 129-137. Cohen, M.M.Jr. (1986a). History, terminology, and classification of craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 1. Cohen, M.M.Jr. (1986b). Perspectives on craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and

77 Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 2. Cohen, M.M.Jr. (1986c). Syndromes with craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 13. Cohen, M.M., Jr. (1997). Transforming growth factor beta s and fibroblast growth factors and their receptors: role in sutural biology and craniosynostosis. J. Bone Miner. Res. 12, 322-331. Cohen, M.M.J., and Kreiborg, S. (1996). Suture formation, premature sutural fusion, and suture default zones in Apert syndrome. Am. J. Med. Genet. 62, 339-344. Cohn, D.H., Byers, P.H., Steinmann, B., and Gelinas, R.E. (1986). Lethal osteogenesis imperfecta resulting from a single nucleotide change in one human pro alpha 1(I) collagen allele. Proc. Natl. Acad. Sci. USA 83, 6045-6047. Colvin, J.S., Bohne, B.A., Harding, G.W., McEwen, D.G., and Ornitz, D.M. (1996). Skeletal overgrowth and deafness in mice lacking fibroblast growth factor receptor 3. Nat. Genet. 12, 390-397. Couly, G., Grapin-Botton, A., Coltey, P., Ruhin, B., and Le Douarin, N.M. (1998). Determination of the identity of the derivatives of the cephalic neural crest: incompatibility between Hox gene expression and lower jaw development. Development 125, 3445-3459. Couly, G.F., Coltey, P.M., and Le Douarin, N.M. (1993). The triple origin of skull in higher vertebrates: a study in quail-chick chimeras. Development 117, 409-429. Crossley, P.H., and Martin, G.R. (1995). The mouse Fgf8 gene encodes a family of polypeptides and is expressed in regions that direct outgrowth and patterning in the developing embryo. Development 121, 439-451. Crossley, P.H., Martinez, S., and Martin, G.R. (1996). Midbrain development induced by FGF8 in the chick embryo. Nature 380 , 66-68. Davidson, D. (1995). The function and evolution of Msx genes: Pointers and paradoxes. Trends Genet. 11, 405- 411. De Pollack, C., Renier, D., Hott, M., and Marie, P.J. (1996). Increased bone formation and osteoblastic cell phenotype in premature cranial suture ossification (craniosynostosis). J. Bone Miner. Res. 11, 401-407. Debiais, F., Hott, M., Graulet, A.M., and Marie, P.J. (1998). The effects of fibroblast growth factor-2 on human neonatal calvaria osteoblastic cells are differentiation stage specific. J. Bone Miner. Res. 13, 645-654. Decker, J.D., and Hall, S.H. (1985). Light and electron microscopy of the new born sagittal suture. Anat. Rec. 212, 81-89. Delezoide, A.L., Benoist-Lasselin, C., Legeai-Mallet, L., Le Merrer, M., Munnich, A., Vekemans, M., and Bonaventure, J. (1998). Spatio-temporal expression of FGFR 1, 2 and 3 genes during human embryo- fetal ossification. Mech. Dev. 77, 19-30. Deng, C.X., Wynshaw-Boris, A., Shen, M.M., Daugherty, C., Ornitz, D.M., and Leder, P. (1994). Murine FGFR- 1 is required for early postimplantation growth and axial organization. Genes Dev. 8, 3045-3057. Deng, C., Wynshaw-Boris, A., Zhou, F., Kuo, A., and Leder P. (1996). Fibroblast growth factor receptor 3 is a negative regulator of bone growth. Cell 84, 911-921. Deol, M.S. (1961). Genetical studies on the skeleton of the mouse. XXVIII. Tail-short. Proceedings of the Royal Society B155, 78-95. Dionne, C.A., Crumley, G., Bellot, F., Kaplow, J.M., Searfoss, G., Ruta, M., Burgess, W.H., Jaye, M., and Schlessinger, J. (1990). Cloning and expression of two distinct high-affinity receptors cross-reacting with acidic and basic fibroblast growth factors. EMBO J. 9, 2685-2692. Dodds, R.A., Connor, J.R., James, I.E., Rykaczewski, E.L., Appelbaum, E., Dul, E., and Gowen, M. (1995). Human osteoclasts, not osteoblasts, deposit osteopontin onto resorption surfaces: an in vitro and ex vivo study of remodeling bone. J. Bone Miner. Res. 10, 1666-1680. Dodig, M., Kronenberg, M.S., Bedalov, A., Kream, B.E., Gronowicz, G., Clark, S.H., Mack, K., Liu, Y.H., Maxon, R., Pan, Z.Z., Upholt, W.B., Rowe, D.W., and Lichtler, A.C. (1996). Identification of a TAAT-containing motif required for high level expression of the COL1A1 promoter in differentiated osteoblasts of transgenic mice. J. Biol. Chem. 271, 16422-16429. Dodig, M., Tadic, T., Kronenberg, M.S., Dacic, S., Liu, Y.H., Maxson, R., Rowe, D.W., and Lichtler, A.C. (1999). Ectopic Msx2 overexpression inhibits and Msx2 antisense stimulates calvarial osteoblast differentiation. Dev. Biol. 209, 298-307. Doty, S.B. (1981). Morphological evidence of gap junctions between bone cells. Calcif. Tissue Int. 33, 509-512. D’Souza, R.N., Åberg, T., Gaikwad, J., Cavender, A., Owen, M., Karsenty, G. and Thesleff, I. (1999). Cbfa1 is required for epithelial-mesenchymal interactions regulating tooth development in mice. Development 126, 2911-2920. Ducy, P., and Karsenty, G. (1995). Two distinct osteoblast-specific cis-acting elements control expression of a mouse osteocalcin gene. Mol. Cell. Biol. 15, 1858-1869. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A.L., and Karsenty, G. (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89, 747-754. Dudley, A.T., Lyons, K.M., and Robertson, E.J. (1995). A requirement for bone morphogenetic protein-7 during development of the mammalian kidney and eye. Genes Dev. 9, 2795-2807. Duffy, J.B., Kania, M.A., and Gergen, J.P. (1991). Expression and function of the Drosophila gene runt in early

78 stages of neural development. Development 113, 1223-1230. Dunlop, L.L., and Hall, B.K. (1995). Relationships between cellular condensation, preosteoblast formation and epithelial-mesenchymal interactions in initiation of osteogenesis. Int. J. Dev. Biol. 39, 357-371. El Ghouzzi, V., Le Merrer, M., Perrin-Schmitt, F., Lajeunie, E., Benit, P., Renier, D., Bourgeois, P., Bolcato- Bellemin, A.L., Munnich, A., and Bonaventure, J. (1997). Mutations of the TWIST gene in the Saethre- Chotzen syndrome. Nat. Genet. 15, 42-46. Elmardi, A.S., Katchburian, M.V., and Katchburian, E. (1990). Electron microscopy of developing calvaria reveals images that suggest that osteoclasts engulf and destroy osteocytes during bone resorption. Calcif. Tissue Int. 46, 239-245. Erlebacher, A., and Derynck, R. (1996). Increased expression of TGF-beta 2 in osteoblasts results in an osteoporosis-like phenotype. J. Cell Biol. 132, 195-210. Erlebacher, A., Filvaroff, E.H., Gitelman, S.E., and Derynck, R. (1995). Toward a molecular understanding of skeletal development. Cell 80, 371-378. Everts, V., Delaissé, J.M., Korper, W., Niehof, A., Vaes, G., and Beertsen, W. (1992). Degradation of collagen in the bone-resorbing compartment underlying the osteoclast involves both cysteine-proteinases and matrix metalloproteinases. J. Cell Physiol. 150, 221-231. Faham, S., Hileman, R.E., Fromm, J.R., Linhardt, R.J., and Rees, DC. (1996). Heparin structure and interactions with basic fibroblast growth factor. Science 271, 1116-1120. Falk, C.T., Schwartz, R.C., Ramirez, F., and Tsipouras, P. (1986). Use of molecular haplotypes specific for the human pro alpha 2(I) collagen gene in linkage analysis of the mild autosomal dominant forms of osteogenesis imperfecta. Am. J. Hum. Genet. 38, 269-279. Fang, J., and Hall, B.K. (1995). Differential expression of neural cell adhesion molecule (NCAM) during osteogenesis and secondary chondrogenesis in the embryonic chick. Int. J. Dev. Biol. 39, 519-528. Farley, J.R., Ivey, J.L., and Baylink, D.J. (1980). Human skeletal alkaline phosphatase. Kinetic studies including pH dependence and inhibition by theophylline. J. Biol. Chem. 255, 4680-4686. Farnum, C.E., Turgai, J., and Wilsman, N.J. (1990). Visualization of living terminal hypertrophic chondrocytes of growth plate cartilage in situ by differential interference contrast microscopy and time-lapse cinematography. J. Orthop. Res. 8, 750-763. Fitchet, S.M. (1929). Cleidocranial dysostosis: Hereditary and familial. J. Bone Joint Surg. 11, 838-866. Foster, J.W., Dominguez-Steglich, M.A., Guioli, S., Kowk, G., Weller, P.A., Stevanovic, M., Weissenbach, J., Mansour, S., Young, I.D., Goodfellow, P.N., et al., (1994). and autosomal sex reversal caused by mutations in an SRY-related gene. Nature 372, 525-530. Franzoso, G., Carlson, L., Xing, L., Poljak, L., Shores, E.W., Brown, K.D., Leonardi, A., Tran, T., Boyce, B.F., and Siebenlist, U. (1997). Requirement for NF-kappaB in osteoclast and B-cell development. Genes Dev. 11, 3482-3496. Friedman, W.F. and Mills, L.F. (1969). The relationship between vitamin D and the craniofacial and dental anomalies of the supravalvular aortic stenosis syndrome. Pediatrics 43, 12-18. Fujisawa, R., Nodasaka, Y., and Kuboki, Y. (1995). Further characterization of interaction between bone sialoprotein (BSP) and collagen. Calcif. Tissue Int. 56, 140-144. Fuller, K., Owens, J.M., Jagger, C.J., Wilson, A., Moss, R., and Chambers, T.J. (1993). Macrophage colony- stimulating factor stimulates survival and chemotactic behavior in isolated osteoclasts. J. Exp. Med. 178, 1733-1744. Füchtbauer, E.M. (1995). Expression of M-twist during postimplantation development of the mouse. Dev. Dyn. 204, 316-322. Fürtwangler, J.A., Hall, S.H., and Koskinen-Moffett, L.K. (1985). Sutural morphogenesis in the mouse calvaria: the role of apoptosis. Acta Anat. 124, 74-80. Fyfe, D.M., and Hall, B.K. (1983). The origin of the ectomesenchymal condensations which precede the development of the bony scleral ossicles in the eyes of embryonic chicks. J. Embryol. Exp. Morphol. 73, 69-86. Gallop, P.M., Lian, J.B., and Hauschka, P.V. (1980). Carboxylated calcium-binding proteins and vitamin K. N. Engl. J. Med. 302, 1460-1466. Galvin, B.D., Hart, K.C., Meyer, A.N., Webster, M.K., and Donoghue, D.J. (1996). Constitutive receptor activation by mutations in fibroblast growth factor receptor (FGFR)2 and FGFR2/ Neu chimeras. Proc. Natl. Acad. Sci. USA 93, 7894-7899. Gaudenz, K., Roessler, E., Vainikka, S., Alitalo, K., and Muenke, M. (1998). Analysis of patients with craniosynostosis syndromes for a Pro246Arg mutation of FGFR4. Mol. Genet. Metab. 64, 76-79. Gavrieli, Y., Sherman, Y., and Ben-Sasson, S.A. (1992). Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J.Cell Biol. 119, 493-501. Gergen, J.P., and Wieschaus, E.F. (1985). The localized requirements for a gene affecting segmentation in Drosophila: analysis of larvae mosaic for runt. Dev. Biol. 109, 321-335. Gilbert, S.F. (1997). Developmental Biology. 5th Edition. Sinauer Associates Inc, Sunderland, Massachusetts, USA. Chapter 7.

79 Givol, D., and Yayon, A. (1992). Complexity of FGF receptors: genetic basis for structural diversity and functional specificity. FASEB J. 6, 3362-3369. Globus, R.K., Plouet, J., and Gospodarowicz, D. (1989). Cultured bovine bone cells synthesize basic fibroblast growth factor and store it in their extracellular matrix. Endocrinology 124, 1539-1547. Goldberg, H.A., Warner, K.J., Stillman, M.J., and Hunter, G.K. (1996). Determination of the hydroxyapatite- nucleating region of bone sialoprotein. Connect. Tissue Res. 35, 385-392. Gonzalez, A.M., Buscaglia, M., Ong, M., and Baird, A. (1990). Distribution of basic fibroblast growth factor in the 18-day rat fetus: localization in the basement membranes of diverse tissues. J. Cell Biol. 110, 753- 765. Gorlin, R.J., Cohen, M.M.Jr., and Levin, L.S. (1990). Syndromes of the Head and Neck, 3rd Edition, Oxford University Press. Oxford, UK. Gorry, M.C., Preston, R.A., White, G.J., Zhang, Y., Singhal, V.K., Losken, H.W., Parker, M.G., Nwokoro, N.A., Post, J.C., and Ehrlich, G.D. (1995). Crouzon syndrome: mutations in two spliceoforms of FGFR2 and a common point mutation shared with Jackson-Weiss syndrome. Hum. Mol. Genet. 4, 1387-1390. Graham, A., Francis-West, P., Brickell, P., and Lumsden, A. (1994). The signalling molecule BMP4 mediates apoptosis in the rhombencephalic neural crest. Nature 372, 684-686. Green, P.J., Walsh, F.S., and Doherty, P. (1996). Promiscuity of fibroblast growth factor receptors. Bioessays 18, 639-646. Greene, H.S.N., and Brown, W.H. (1932). Hereditary variations in the skull of the rabbit. Science 76, 421-422. Cited by Cohen, M.M.Jr. (1986). Etiology of craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 3. Grigoriadis, A.E., Heersche, J.N., and Aubin, J.E. (1988). Differentiation of muscle, fat, cartilage, and bone from progenitor cells present in a bone-derived clonal cell population: effect of dexamethasone. J. Cell Biol. 106, 2139-2151. Grigoriadis, A.E., Wang, Z.Q., Cecchini, M.G., Hofstetter, W., Felix, R., Fleisch, HA, and Wagner, E.F. (1994). c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 266, 443-448. Grigoriadis, A.E., Wang, Z-Q. and Wagner, E.F. (1996). Regulation of bone cell differentiation and bone remodelling by the c-Fos/AP-1 transcription factor. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 5. Gripp, K.W., Scott, C.I.Jr., Hughes, H.E., Wallerstein, R., Nicholson, L., States, L., Bason, L.D., Kaplan, P., Zderic, S.A., Duhaime, A.C., Miller, F., Magnusson, M.R., and Zackai, E.H. (1997). Lateral meningocele syndrome: three new patients and review of the literature. Am. J. Med. Genet. 70, 229-239. Grüneberg, H. (1953). Genetical studies on the skeleton of the mouse. VII congenital hydrocephalus. J. Genet. 51, 327-360. Grüneberg, H. (1966). The molars of the tabby mouse, and a test of the ‘single-active X-chromosome’ hypothesis. J. Embryol. Exp. Morphol. 15, 223-244. Grüneberg, H., and Truslove, G.M. (1960). Two closely linked genes in the mouse. Genetical Research 1, 69-90. Grüneberg, H., and Wickramaratne, G.A. (1974). A re-examination of two skeletal mutants of the mouse, vestigial-tail (vt) and congenital hydrocephalus (ch). J. Embryol. Exp. Morphol. 31, 207-222. Gundberg, C.M., Hauschka, P.V., Lian, J.B., and Gallop, P.M. (1984). Osteocalcin: isolation, characterization, and detection. Methods Enzymol. 107, 516-544. Hall, B.K. (1970). Cellular differentiation in skeletal tissues. Biol. Rev. Camb. Philos. Soc. 45, 455-484. Hall, B.K. (1991). Cellular interactions during cartilage and bone development. J. Craniofac. Genet. Dev. Biol. 11, 238-250. Hall, B.K., and Miyake, T. (1992). The membranous skeleton: the role of cell condensations in vertebrate skeletogenesis. Anat. Embryol. 186, 107-124. Hall, B.K., and Miyake, T. (1995). Divide, accumulate, differentiate - cell condensation in skeletal development revisited. Int. J. Dev. Biol. 39, 881-893. Hall, T.J., and Chambers, T.J. (1996). Molecular aspects of osteoclast function. Inflammation Res. 45, 1-9. Halleen, J.M., Kaija, H., Stepan, J.J., Vihko, P., and Väänänen, H.K. (1998). Studies on the protein tyrosine phosphatase activity of tartrate-resistant acid phosphatase. Arch. Biochem. Biophys. 352, 97-102. Halleen, J.M., Karp, M., Viloma, S., Laaksonen, P., Hellman, J., Kakonen, S.M., Stepan, J.J., Holmes, S., Väänänen, H.K., and Pettersson, K. (1999). Two-site immunoassays for osteoclastic tartrate-resistant acid phosphatase based on characterization of six monoclonal antibodies. J. Bone Miner. Res. 14, 464- 469. Hammerschmidt, M., Brook, A., and McMahon, A.P. (1997). The world according to hedgehog. Trends in Genetics 13, 14-21. Hattersley, G., and Chambers, T.J. (1989). Generation of osteoclastic function in mouse bone marrow cultures: multinuclearity and tartrate-resistant acid phosphatase are unreliable markers for osteoclastic differentiation. Endocrinology 124, 1689-1696. Haub, O., and Goldfarb, M. (1991). Expression of the fibroblast growth factor-5 gene in the mouse embryo.

80 Development 112, 397-406. Hayman, A.R., Jones, S.J., Boyde, A., Foster, D., Colledge, W.H., Carlton, M.B., Evans, M.J., and Cox, T.M. (1996). Mice lacking tartrate-resistant acid phosphatase (Acp 5) have disrupted endochondral ossification and mild osteopetrosis. Development 122, 3151-3162. Hebrok, M., Wertz, K., and Füchtbauer, E.M. (1994). M-twist is an inhibitor of muscle differentiation. Dev. Biol. 165, 537-544. Helfrich, M.H., Nesbitt, S.A., Dorey, E.L., and Horton, M.A. (1992). Rat osteoclasts adhere to a wide range of RGD (Arg-Gly-Asp) peptide-containing proteins, including the bone sialoproteins and fibronectin, via a beta 3 integrin. J. Bone Miner. Res. 7, 335-343. Henthorn, P.S. (1996). Alkaline Phosphatase. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 15. Hiemisch, H., Monaghan, A.P., Schutz, G., and Kaestner, K.H. (1998). Expression of the mouse Fkh1/Mf1 and Mfh1 genes in late gestation embryos is restricted to mesoderm derivatives. Mech. Dev. 73, 129-132. Hill, P.A., Tumber, A., and Meikle, M.C. (1997). Multiple extracellular signals promote osteoblast survival and apoptosis. Endocrinology 138, 3849-3858. Hodgkinson, C.A., Moore, K.J., Nakayama, A., Steingrimsson, E., Copeland, N.G., Jenkins, N.A., and Arnheiter, H. (1993a). Mutations at the mouse microphthalmia locus are associated with defects in a gene encoding a novel basic-helix-loop-helix-zipper protein. Cell 74, 395-404. Hodgkinson, J.E., Davidson, C.L., Beresford, J., and Sharpe, P.T. (1993b). Expression of a human homeobox- containing gene is regulated by 1,25(OH)2D3 in bone cells. Biochim. Biophys. Acta 1174, 11-16. Hoffmann, H.M., Beumer, T.L., Rahman, S., McCabe, L.R., Banerjee, C., Aslam, F., Tiro, J.A., van Wijnen, A.J., Stein, J.L., Stein, G.S., and Lian, J.B. (1996). Bone tissue-specific transcription of the osteocalcin gene: role of an activator osteoblast-specific complex and suppressor hox proteins that bind the OC box. J. Cell Biochem. 61, 310-324. Hollway, G.E., Phillips, H.A., Ades, L.C., Haan, E.A., and Mulley, J.C. (1995). Localization of craniosynostosis Adelaide type to 4p16. Hum. Mol. Genet. 4, 681-683. Holt, I., Davie, M.W., Braidman, I.P., and Marshall, M.J. (1994). Interleukin-6 does not mediate the stimulation by prostaglandin E2, parathyroid hormone, or 1,25 dihydroxyvitamin D3 of osteoclast differentiation and bone resorption in neonatal mouse parietal bones. Calcif. Tissue Int. 55, 114-119. Holtrop, M.E., and King, G.J. (1977). The ultrastructure of the osteoclast and its functional implications. Clin. Orthop. Relat. Res. 123, 177-196. Hong, H.K., Lass, J.H., and Chakravarti, A. (1999). Pleiotropic skeletal and ocular phenotypes of the mouse mutation congenital hydrocephalus (ch/Mf1) arise from a winged helix/forkhead transcription factor gene. Hum. Mol. Genet. 8, 625-637. Horton, M.A., and Davies, J. (1989). Perspectives: adhesion receptors in bone. J. Bone Miner. Res. 4, 803-808. Horwitz, E.M., Prockop, D.J., Fitzpatrick, L.A., Koo, W.W.K., Gordon, P.L., Neel, M., Sussman, M., Orchard, P., Marx, J.C., Pyeritz, R.E., and Brenner, M.K. (1999). Transplantability and therapeutic effects of bone marrow-derived mesenchymal cells in children with osteogenesis imperfecta. Nature Medicine 5, 309-313. Howard, T.D., Paznekas, W.A., Green, E.D., Chiang, L.C., Ma, N., De Luna, R.I.O., Delgado, C.G., Gonzalez- Ramos, M., Kline, A.D., and Jabs, E.W. (1997). Mutations in TWIST, a basic helix-loop-helix transcription factor, in Saethre-Chotzen syndrome. Nat. Genet. 15, 36-41. Huang, L.F., Fukai, N., Selby, P.B., Olsen, B.R., and Mundlos, S. (1997). Mouse clavicular development: analysis of wild-type and cleidocranial dysplasia mutant mice. Dev. Dyn. 210, 33-40. Huang, M.H., Gruss, J.S., Clarren, S.K., Mouradian, W.E., Cunningham, M.L., Roberts, T.S., Loeser, J.D., and Cornell, C.J. (1996). The differential diagnosis of posterior plagiocephaly: true lambdoid synostosis versus positional molding. Plast. Reconstr. Surg. 98, 765-774. Hughes, D.E., Dai, A., Tiffee, J.C., Li, H.H., Mundy, G.R., and Boyce, B.F. (1996). Estrogen promotes apoptosis of murine osteoclasts mediated by TGF-beta. Nature Medicine 2, 1132-1136. Hughes, D.E., Wright, K.R., Uy, H.L., Sasaki, A., Yoneda, T., Roodman, G.D., Mundy, G.R, and Boyce, B.F. (1995). Bisphosphonates promote apoptosis in murine osteoclasts in vitro and in vivo. J. Bone Miner. Res. 10, 1478-1487. Hurley, M.M., Abreu, C., Gronowicz, G., Kawaguchi, H., and Lorenzo, J., (1994). Expression and regulation of basic fibroblast growth factor mRNA levels in mouse osteoblastic MC3T3-E1 cells. J. Biol. Chem. 267, 26031-26037. Hurley, M.M., Abreu, C., Marcello, K., Kawaguchi, H., Lorenzo, J., Kalinowski, J., Ray, A., and Gronowicz, G. (1996). Regulation of NFIL-6 and IL-6 expression by basic fibroblast growth factor in osteoblasts. J. Bone Miner. Res. 11, 760-767. Hurley, M.M., Lee, S.K., Raisz, L.G., Bernecker, P., and Lorenzo, J. (1998). Basic fibroblast growth factor induces osteoclast formation in murine bone marrow cultures. Bone 22, 309-316. Inada, M., Yasui, T., Nomura, S., Miyake, S., Deguchi, K., Himeno, M., Sato, M., Yamagiwa, H., Kimura, T., Yasui, N., Ochi, T., Endo, N., Kitamura, Y., Kishimoto, and Komori, T. (1999). Maturational disturbance

81 of chondrocytes in Cbfa1-deficient mice. Dev. Dyn. 214, 279-290. Ingham, P.W. (1998). Transducing Hedgehog: the story so far. EMBO J. 17, 3505-3511. Iseki, S., Wilkie, A.O., Heath, J.K., Ishimaru, T., Eto, K., and Morriss-Kay, G.M. (1997). Fgfr2 and osteopontin domains in the developing skull vault are mutually exclusive and can be altered by locally applied FGF2. Development 124, 3375-3384. Itabashi, M. (1969). Experimental production of congenital malformations in Japanese white rabbits by maternal treatment with hypervitaminosis A. Cong. Anom. 9, 143-155. Cited by Cohen, M.M.Jr. (1986). Etiology of craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 3. Jabs, E.W., Li, X., Scott, A.F., Meyers, G., Chen, W., Eccles, M., Mao, J.I., Charnas, L.R., Jackson, C.E., and Jaye, M. (1994). Jackson-Weiss and Crouzon syndromes are allelic with mutations in fibroblast growth factor receptor 2. Nat. Genet. 8, 275-279. Jabs, E.W., Muller, U., Li, X., Ma, L., Luo, W., Haworth, I.S., Klisak, I., Sparkes, R., Warman, M.L., and Mulliken, J.B. (1993). A mutation in the homeodomain of the human MSX2 gene in a family affected with autosomal dominant craniosynostosis. Cell 75, 443-450. Jane, J.A., and Persing, J.A. (1986). Neurosurgical treatment of craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 11. Jemtland, R., Lee, K., and Segre, G.V. (1998). Heterogeneity among cells that express osteoclast-associated genes in developing bone. Endocrinology 139, 340-349. Jensen, B.L., and Kreiborg, S. (1990). Development of the dentition in cleidocranial dysplasia. J. Oral Path. Med. 19, 89-93. Jensen, B.L., and Kreiborg, S. (1993a). Development of the skull in infants with cleidocranial dysplasia. J. Craniofac. Genet. Dev. Biol. 13, 89-97. Jensen, B.L., and Kreiborg, S. (1993b). Craniofacial abnormalities in 52 school-age and adult patients with cleidocranial dysplasia. J. Craniofac. Genet. Dev. Biol. 13, 98-108. Jernvall, J., Åberg, T., Kettunen, P., Keränen, S., and 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. Jiang, H.P., Sodek, J., Karsenty, G., Thomas, H., Ranly, D., Chen, J.K. (1999). Expression of core binding factor Osf2/Cbfa-1 and bone sialoprotein in tooth development. Mech. Dev. 81, 169-173. Johansen, V.A., and Hall, S.H. (1982). Morphogenesis of the mouse coronal suture. Acta Anat. 114, 58-67. Johnson, D.E., and Williams, L.T. (1993). Structural and functional diversity in the FGF receptor multigene family. Adv. Cancer Res. 60, 1-41. Johnson, D.E., Lee, P.L., Lu, J., and Williams, L.T. (1990). Diverse forms of a receptor for acidic and basic fibroblast growth factors. Mol. Cell. Biol. 10, 4728-4736. Johnson, D.R. (1967). Extra-toes: a new mutant gene causing multiple abnormalities in the mouse. J. Embryol. Exp. Morphol. 17, 543-581. Johnston, M.C. (1966). A radioautographic study of the migration and fate of cranial neural crest cells in the chick embryo. Anat. Rec. 156, 143-155. Jowett, A.K., Vainio, S., Ferguson, M.W., Sharpe, P.T., and Thesleff, I. (1993). Epithelial-mesenchymal interactions are required for msx 1 and msx 2 gene expression in the developing murine molar tooth. Development 117, 461-470. Kameda, T., Ishikawa, H., and Tsutsui, T. (1995). Detection and characterization of apoptosis in osteoclasts in vitro. Biochem. Biophys. Res. Commun. 207, 753-760. Karsenty, G. (1998). Genetics of skeletogenesis. Dev. Genet. 22, 301-313. Katagiri, T., Yamaguchi, A., Ikeda, T., Yoshiki, S., Wozney, J.M., Rosen, V., Wang, E.A., Tanaka, H., Omura, S., and Suda, T. (1990). The non-osteogenic mouse pluripotent cell line, C3H10T1/2, is induced to differentiate into osteoblastic cells by recombinant human bone morphogenetic protein-2. Biochem. Biophys. Res. Commun. 172, 295-299. Katoh, M., Hattori, Y., Sasaki, H., Tanaka, M., Sugano, K., Yazaki, Y., Sugimura, and Terada, M. (1992). K-sam gene encodes secreted as well as transmembrane receptor tyrosine kinase. Proc. Natl. Acad. Sci. USA 89, 2960-2964. Kaufman, M.H., and Bard, J.B.L. (1999). The Anatomical Basis of Mouse Development. Academic Press, San Diego and London. Keegan, K., Johnson, D.E., Williams, L.T., and Hayman, M.J. (1991). Isolation of an additional member of the fibroblast growth factor receptor family, FGFR-3. Proc. Natl. Acad. Sci. USA 88, 1095-1099. Kettunen, P., and Thesleff, I. (1998). Expression and function of FGFs-4, -8, and -9 suggest functional redundancy and repetitive use as epithelial signals during tooth morphogenesis. Dev. Dyn. 211, 256-268. Kettunen, P., Nobuyuki, I., and Thesleff, I. (1999). Associations of FGF-3 and FGF-10 in signaling networks regulating tooth morphogenesis. Dev. Genet. Submitted. Key, L.L., and Ries, W.L. (1996). Osteopetrosis. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 67.

82 Kiliaridis, S. (1995). Masticatory muscle influence on craniofacial growth. Acta Odontol. Scand. 53, 196-202. Kim, I.S., Otto, F., Zabel, B., Mundlos, S. (1999). Regulation of chondrocyte differentiation by Cbfa1. Mech. Dev. 80, 159-170. Kingsley, D.M., Bland, A.E., Grubber, J.M., Marker, P.C., Russell, L.B., Copeland, N.G., and Jenkins, N.A. (1992). The mouse short ear skeletal morphogenesis locus is associated with defects in a bone morphogenetic member of the TGF beta superfamily. Cell 71, 399-410. Kinto, N., Iwamoto, M., Enomoto-Iwamoto, M., Noji, S., Ohuchi, H., Yoshioka, H., Kataoka, H., Wada, Y., Yuhao, G., and Takahashi, H.E., Yoshiki, S., and Yamaguchi, A. (1997). Fibroblasts expressing Sonic hedgehog induce osteoblast differentiation and ectopic bone formation. FEBS Letters 404, 319-323. Kokich, V.G. (1986). The biology of sutures. In Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 4. Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y., Bronson, R.T., Gao, Y.H., Inada, M., Sato, M., Okamoto, R., Kitamura, Y., Yoshiki, S., and Kishimoto, T. (1997). Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755-764. Korhonen, J., Partanen, J., and Alitalo, K. (1992). Expression of FGFR-4 mRNA in developing mouse tissues. Int. J. Dev. Biol. 36, 323-329. Koskinen-Moffet, L. (1986). In vitro experimental model for prenatal craniosynostosis. J. Dent. Res. 65, Special issue, abstract 980. Kreiborg, S., Jensen, B.L., Bjork, A., and Skieller, V. (1981). Abnormalities of the cranial base in cleidocranial dysostosis. Am. J. Orthod. 79, 549-557. Kume, T., Deng, K.Y., Winfrey, V., Gould, D.B., Walter, M.A., and Hogan, B.L. (1998). The forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation congenital hydrocephalus. Cell 93, 985- 996. Lacey, D.L., Timms, E., Tan, H.L., Kelley, M.J., Dunstan, C.R., Burgess, T., Elliott, R., Colombero, A., Elliott, G., Scully, S., Hsu, H., Sullivan, J., Hawkins, N., Davy, E., Capparelli, C., Eli, A., Qian, Y.X., Kaufman, S., Sarosi, Shalhoub, V., Senaldi, G., Guo, J., Delaney, J., and Boyle, W.J. (1998). Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 165-176. Lajeunie, E., Ma, H.W., Bonaventure, J., Munnich, A., Le Merrer, M., and Renier, D. (1995). FGFR2 mutations in Pfeiffer syndrome. Nat. Genet. 9, 108. Lajeunie, E., El Ghouzzi, V., Le Merrer, M., Munnich, A., Bonaventure, J., and Renier, D. (1999). Sex related expressivity of the phenotype in coronal craniosynostosis caused by the recurrent P250R FGFR3 mutation. J. Med. Genet. 36, 9-13. Lajtha, (1982). cited by Triffitt, J.T. (1996). The stem cell of the osteoblast. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 4. Lau, K.H., Freeman, T.K., and Baylink, D.J. (1987). Purification and characterization of an acid phosphatase that displays phosphotyrosyl-protein phosphatase activity from bovine cortical bone matrix. J. Biol. Chem. 262, 1389-1397. Laufer, E., Nelson, C.E., Johnson, R.L., Morgan, B.A., and 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. Le Douarin, N.M. (1982). The Neural Crest. Le Douarin, N.M., ed. Cambridge University Press, Cambridge U.K. p259. Le Lievré, C.S., and Le Douarin, N.M. (1975). Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos. J. Embryol. Exp. Morphol. 34, 125-154. Lee, B., Thirunavukkarasu, K., Zhou, L., Pastore, L., Baldini, A., Hecht, J., Geoffroy, V., Ducy, P., and Karsenty, G. (1997). Missense mutations abolishing DNA binding of the osteoblast-specific transcription factor OSF2/CBFA1 in cleidocranial dysplasia. Nat. Genet. 16, 307-310. Lee, P.L., Johnson, D.E., Cousens, L.S., Fried, V.A., and Williams, L.T. (1989). Purification and complementary DNA cloning of a receptor for basic fibroblast growth factor. Science 245, 57-60. Lemire, R.J. (1986). Embryology of the skull. In Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 5. Lengele, B., Dhem, A., and Schowing, J. (1990). Early development of the primitive cranial vault in the chick embryo. J. Craniofac. Genet. Dev. Biol. 10, 103-112. Levine, J.P., Bradley, J.P., Roth, D.A., McCarthy, J.G., and Longaker, M.T. (1998). Studies in cranial suture biology: regional dura mater determines overlying suture biology. Plast. Reconstr. Surg. 101, 1441- 1447. Li, J., Tsuji, K., Komori, T., Miyazono, K., Wrana, J.L., Ito, Y., Nifuji, A., and Noda, M. (1998). Smad2 overexpression enhances Smad4 gene expression and suppresses CBFA1 gene expression in osteoblastic osteosarcoma ROS17/2.8 cells and primary rat calvaria cells. J. Biol. Chem. 273, 31009-31015. Liaw, L., Birk, D.E., Ballas, C.B., Whitsitt, J.S., Davidson, J.M., and Hogan, B.L. (1998). Altered wound healing in mice lacking a functional osteopontin gene (spp1). J. Clin. Invest. 101, 1468-1478.

83 Liaw, L., Skinner, M.P., Raines, E.W., Ross, R., Cheresh, D.A., Schwartz, S.M., and Giachelli, C.M. (1995). The adhesive and migratory effects of osteopontin are mediated via distinct cell surface integrins. Role of alpha v beta 3 in smooth muscle cell migration to osteopontin in vitro. J. Clin. Invest. 95, 713-724. Lin, K.Y., Nolen, A.A., Gampper, T.J., Jane, J.A., Opperman, L.A., and Ogle, R.C. (1997). Elevated levels of transforming growth factors beta 2 and beta 3 in lambdoid sutures from children with persistent plagiocephaly. Cleft Palate Craniofac. J. 34, 331-337. Liu, Y.H., Kundu, R., Wu, L., Luo, W., Ignelzi, M.A.J., Snead, M.L., and Maxson, R.E., Jr. (1995). Premature suture closure and ectopic cranial bone in mice expressing Msx2 transgenes in the developing skull. Proc. Natl. Acad. Sci. USA 92, 6137-6141. Liu, Y.H., Tang, Z., Kundu, R.K., Wu, L., Luo, W., Zhu, D., Sangiorgi, F., Snead, M.L., and Maxson, R.E. (1999). Msx2 gene dosage influences the number of proliferative osteogenic cells in growth centers of the developing murine skull: a possible mechanism for MSX2-mediated craniosynostosis in humans. Dev. Biol. 205, 260-274. Loeser, R.F. (1993). Integrin-mediated attachment of articular chondrocytes to extracellular matrix proteins. Arthritis & Rheumatism 36, 1103-1110. Lomri, A., Lemonnier, J., Hott, M., de Parseval, N., Lajeunie, E., Munnich, A., Renier, D., and Marie, P.J. (1998). Increased calvaria cell differentiation and bone matrix formation induced by fibroblast growth factor receptor 2 mutations in Apert syndrome. J. Clin. Invest. 101, 1310-1317. Lowe, C., Yoneda, T., Boyce, B.F., Chen, H., Mundy, G.R., and Soriano, P. (1993). Osteopetrosis in Src-deficient mice is due to an autonomous defect of osteoclasts. Proc. Natl. Acad. Sci. USA 90, 4485-4489. Lui, V.C., Ng, L.J., Nicholls, J., Tam, P.P., and Cheah, K.S. (1995). Tissue-specific and differential expression of alternatively spliced alpha 1(II) collagen mRNAs in early human embryos. Dev. Dyn. 203, 198-211. Lukinmaa, P.L., Jensen, B.L., Thesleff, I., Andreasen, J.O., and Kreiborg, S. (1995). Histological observations of teeth and peridental tissues in cleidocranial dysplasia imply increased activity of odontogenic epithelium and abnormal bone remodeling. J. Craniofac. Genet. Dev. Biol. 15, 212-221. Lukinmaa, P.L., Vaahtokari, A., Vainio, S., Sandberg, M., Waltimo, J., and Thesleff (1993). Transient expression of type III collagen by odontoblasts: developmental changes in the distribution of pro-alpha 1(III) and pro-alpha 1(I) collagen mRNAs in dental tissues. Matrix 13, 503-515. Lumsden, A., Sprawson, N., and Graham, A. (1991). Segmental origin and migration of neural crest cells in the hindbrain region of the chick embryo. Development 113, 1281-1291. Luo, G., Hofmann, C., Bronckers, A.L., Sohocki, M., Bradley, A., and Karsenty, G. (1995). BMP-7 is an inducer of nephrogenesis, and is also required for eye development and skeletal patterning. Genes Dev. 9, 2808- 2820. Lynch, M.P., Stein, J.L., Stein, G.S., and Lian, J.B. (1995). The influence of type I collagen on the development and maintenance of the osteoblast phenotype in primary and passaged rat calvarial osteoblasts: modification of expression of genes supporting cell growth, adhesion, and extracellular matrix mineralization. Exp. Cell Res. 216, 35-45. Ma, L., Golden, S., Wu, L., and Maxson, R. (1996). The molecular basis of Boston-type craniosynostosis: the Pro148—>His mutation in the N-terminal of the MSX2 homeodomain stabilizes DNA binding without altering nucleotide sequence preferences. Hum. Mol. Genet. 5, 1915-1920. Maas, R., and Bei, M. (1997). The genetic control of early tooth development. Crit. Rev. Oral Biol. Med. 8, 4- 39. Mabbutt, L.W., and Kokich, V.G. (1979). Calvarial and sutural re-development following craniectomy in the neonatal rabbit. J. Anat. 129, 413-422. MacGregor, G.R., Zambrowicz, B.P., and Soriano, P. (1995). Tissue non-specific alkaline phosphatase is expressed in both embryonic and extraembryonic lineages during mouse embryogenesis but is not required for migration of primordial germ cells. Development 121, 1487-1496. MacKenzie, A., Ferguson, M.W., and Sharpe, P.T. (1991). Hox-7 expression during murine craniofacial development. Development 113, 601-611. MacKenzie, A., Ferguson, M.W., and Sharpe, P.T. (1992). Expression patterns of the homeobox gene, Hox-8, in the mouse embryo suggest a role in specifying tooth initiation and shape. Development 115, 403- 420. Maire, P., and Sainton, P. (1897). Observation d’hydrocéphalie héréditaire (pére et fils) par vice de développment du crâne et du cerveau. Bull. Soc. Méd. Hôp. Paris 14, 706-712. Cited in Gorlin, R.J, Cohen M.M.Jr. and Levin L.S. (1990). Syndromes of the Head and Neck. 3rd Edition. Oxford University Press, Oxford, UK. Chapter 9. Malaval, L., Gupta, A.K., and Aubin, J.E. (1995). Leukemia inhibitory factor inhibits osteogenic differentiation in rat calvaria cell cultures. Endocrinology 136, 1411-1418. Mangasarian, K., Li, Y., Mansukhani, A., and Basilico, C. (1997). Mutation associated with Crouzon syndrome causes ligand-independent dimerization and activation of FGF receptor-2. J. Cell Physiol. 172, 117- 125. Manzanares, M.C., Goret-Nicaise, M., and Dhem, A. (1988). Metopic sutural closure in the human skull. J. Anat.

84 161, 203-215. Mark, M.P., Butler, W.T., Prince, C.W., Finkelman, R.D., and Ruch, J.V. (1988). Developmental expression of 44-kDa bone phosphoprotein (osteopontin) and bone gamma-carboxyglutamic acid (Gla)-containing protein (osteocalcin) in calcifying tissues of rat. Differentiation 37, 123-136. Markens, I.S. (1975a). Transplantation of the future coronal suture on the dura mater of 3- to 4-month-old rats. Acta Anat. 93, 29-44. Markens, I.S. (1975b). Embryonic development of the coronal suture in man and rat. Acta Anat. 93, 257-273. Marks, S.C.Jr., and Hermey, D.C. (1996). The structure and development of bone. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 1. Marti, E., Takada, R., Bumcrot, D.A., Sasaki, H., and McMahon, A.P. (1995). Distribution of Sonic hedgehog peptides in the developing chick and mouse embryo. Development 121, 2537-2547. Martin, G.R. (1998). The roles of FGFs in the early development of vertebrate limbs. Genes Dev. 12, 1571- 1586. Martin, S.J., Green, D.R., and Cotter, T.G. (1994). Dicing with death: dissecting the components of the apoptosis machinery. Trends Biochem. Sci. 19, 26-30. Martin (1765). Sur un déplacement natural de la clavicule. J. Méd.Chir. Pharmacol. 23, 456-460. Cited in Gorlin, R.J, Cohen M.M.Jr. and Levin L.S. (1990). Syndromes of the Head and Neck. 3rd Edition. Oxford University Press, Oxford, UK. Chapter 9. Mayahara, H., Ito, T., Nagai, H., Miyajima, H., Tsukuda, R., Taketomi, S., Mizoguchi, J., and Kato, K. (1993). In vivo stimulation of endosteal bone formation by basic fibroblast growth factor in rats. Growth Factors 9, 73-80. McCabe, L.R., Kockx, M., Lian, J., Stein, J., and Stein, G. (1995). Selective expression of fos- and jun-related genes during osteoblast proliferation and differentiation. Exp. Cell Res. 218, 255-262. McDougall, S., Chen, S., Glackin, C., and Fang, M.A. (1997). The role of basic fibroblast growth factor and EGR- 1 in regulating ?2(I) collagen and Twist gene expression in osteoblasts. J. Bone Miner. Res. 12, S280. McKusick, V.A., and Scott, C.I. (1971). A nomenclature for constitutional disorders of bone. J. Bone Joint Surg. - American Volume 53, 978-986. Meckel, (1760). Cited by Cohen, M.M.Jr. (1986a). History, terminology, and classification of craniosynostosis. In: Craniosynostosis: Diagnosis, Evaluation and Management. Cohen, M.M.Jr. ed. Raven Press, New York. Chapter 1. Mehrara, B.J., Mackool, R.J., McCarthy, J.G., Gittes, G.K., and Longaker, M.T. (1998). Immunolocalization of basic fibroblast growth factor and fibroblast growth factor receptor-1 and receptor-2 in rat cranial sutures. Plast. Reconstr. Surg. 102, 1805-1817. Meikle, M.C., Heath, J.K., Hembry, R.M., and Reynolds, J.J. (1982). Rabbit cranial suture fibroblasts under tension express a different collagen phenotype. Arch. Oral Biol. 27, 609-613. Meikle, M.C., Reynolds, J.J., Sellers, A., and Dingle, J.T. (1979). Rabbit cranial sutures in vitro: a new experimental model for studying the response of fibrous joints to mechanical stress. Calcif. Tissue Int. 28, 137-144. Meyers, G.A., Day, D., Goldberg, R., Daentl, D.L., Przylepa, K.A., Abrams, L.J., Graham, J.M.J., Feingold, M., Moeschler, J.B., Rawnsley, E., Scott, A.F., and Jabs, E.W. (1996). FGFR2 exon IIIa and IIIc mutations in Crouzon, Jackson-Weiss, and Pfeiffer syndromes: evidence for missense changes, insertions, and a deletion due to alternative RNA splicing. Am. J. Hum. Genet. 58, 491-498. Meyers, G.A., Orlow, S.J., Munro, I.R., Przylepa, K.A., and Jabs, E.W. (1995). Fibroblast growth factor receptor 3 (FGFR3) transmembrane mutation in Crouzon syndrome with acanthosis nigricans. Nat. Genet. 11, 462-464. Milaire, J. (1992). A new interpretation of the necrotic changes occurring in the developing limb bud paddle of mouse embryos based upon recent observations in four different phenotypes. Int. J. Dev. Biol. 36, 169- 178. Minkin, C., Mintz, K.P., Grzesik, W.J., Midura, R.J., Robey, P.G., Termine, J.D., and Fisher, L.W. (1982). Bone acid phosphatase: tartrate-resistant acid phosphatase as a marker of osteoclast function purification and fragmentation of nondenatured bone sialoprotein: evidence for a cryptic, RGD-resistant cell attachment domain. Calcif. Tissue Int. 34, 285-290. Mintz, K.P., Grzesik, W.J., Midura, R.J., Robey, P.G., Termine, J.D., and Fisher, L.W. (1993). Purification and fragmentation of nondenatured bone sialoprotein: evidence for a cryptic, RGD-resistant cell attachment domain. J. Bone Miner. Res. 8, 985-995. Modderman, W.E., Tuinenburg-Bol, R.A., and Nijweide, P.J. (1991). Tartrate-resistant acid phosphatase is not an exclusive marker for mouse osteoclasts in cell culture. Bone 12, 81-87. Moloney, D.M., Slaney, S.F., Oldridge, M., Wall, S.A., Sahlin, P., Stenman, G., and Wilkie, A.O. (1996). Exclusive paternal origin of new mutations in Apert syndrome. Nat. Genet. 13, 48-53. Molteni, A., Modrowski, D., Hott, M., and Marie, P.J. (1999). Alterations of matrix- and cell-associated proteoglycans inhibit osteogenesis and growth response to fibroblast growth factor-2 in cultured rat mandibular condyle and calvaria. Cell Tissue Res. 295, 523-536.

85 Mooney, M.P., Losken, H.W., Siegel, M.I., Lalikos, J.F., Losken, A., Smith, T.D., and Burrows, A.M. (1994). Development of a strain of rabbits with congenital simple nonsyndromic coronal suture synostosis. Part I: Breeding demographics, inheritance pattern, and craniofacial anomalies. Cleft Palate Craniofac. J. 31, 1-7. Mooney, M.P., Losken, H.W., Tschakaloff, A., Siegel, M.I., Losken, A. and Lalikos, J.F. (1993). Congenital bilateral coronal suture synostosis in a rabbit and comparisons with experimental models. Cleft Palate Craniofac. J. 30, 121-128. Moss, M.L. (1954). Growth of the calvaria in the rat. Am. J. Anat. 94, 333-362. Moss, M.L. (1958). Fusion of the frontal suture in the rat. Am. J. Anat. 102, 141-166. Moss, M.L. (1960). Inhibition and stimulation of sutural fusion in the rat calvaria. Anat. Rec. 136, 457-467. Moss, M.L. (1975). Functional anatomy of cranial synostosis. Childs Brain 1, 22-33. Most, D., Levine, J.P., Chang, J., Sung, J., McCarthy, J.G., Schendel, S.A., and Longaker, M.T. (1998). Studies in cranial suture biology: Up-regulation of transforming growth factor-beta 1 and basic fibroblast growth factor mRNA correlates with posterior frontal cranial suture fusion in the rat. Plast. Reconstr. Surg. 101, 1431-1440. Muenke, M., Schell, U., Hehr, A., Robin, N.H., Losken, H.W., Schinzel, A., Pulleyn, LJ, Rutland, P., Reardon, W., and Malcolm, S. (1994). A common mutation in the fibroblast growth factor receptor 1 gene in Pfeiffer syndrome. Nat. Genet. 8, 269-274. Muenke, M., Gripp, K.W., McDonald-McGinn, D.M., Gaudenz, K., Whitaker, L.A., Bartlett, S.P., Markowitz, R.I., Robin, N.H., Nwokoro, N., Mulvihill, J.J., Losken, HW, Mulliken, J.B., Guttmacher, A.E., Wilroy, R.S., Clarke, L.A., Hollway, G., Ades, L.C., Haan, E.A., Mulley, J.C., Cohen, M.M.J., Bellus, G.A., Francomano, C.A., Moloney, D.M., Wall, S.A., and Wilkie, A.O. (1997). A unique point mutation in the fibroblast growth factor receptor 3 gene (FGFR3) defines a new craniosynostosis syndrome. Am. J. Hum. Genet. 60, 555-564. Mundlos, S., Otto, F., Mundlos, C., Mulliken, J.B., Aylsworth, A.S., Albright, S., Lindhout, D., Cole, W.G., Henn, W., Knoll, J.H., Owen, M.J., Mertelsmann, R., Zabel, BU, and Olsen, B.R. (1997). Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell 89, 773-779. Murphy, G. (1995). Matrix metalloproteinases and their inhibitors. Acta Orthop. Scand. 266 Suppl., 55-60. Murray, S.S., Glackin, C.A., Winters, K.A., Gazit, D., Kahn, A.J., and Murray, E.J. (1992). Expression of helix- loop-helix regulatory genes during differentiation of mouse osteoblastic cells. J. Bone Miner. Res. 7, 1131-1138. Murtaugh, L.C., Chyung, J.H., and Lassar, A.B. (1999). Sonic hedgehog promotes somitic chondrogenesis by altering the cellular response to BMP signaling. Genes Dev. 13, 225-237. Myers, J.C., Dickson, L.A., Pope, F.M., Korhonen, V.R., Nicholls, A., Prockop, D.J., and Pihlajaniemi, T. (1985). A homozygous frameshift mutation in the pro-alpha-1(I) collagen COOH-propeptide results in osteogenesis imperfecta. Ann. N. Y. Acad. Sci. 460, 482-485. Naski, M.C., Colvin, J.S., Coffin, J.D., and Ornitz, D.M. (1998). Repression of hedgehog signaling and BMP4 expression in growth plate cartilage by fibroblast growth factor receptor 3. Development 125, 4977- 4988. Neilson, K.M., and Friesel, R.E. (1995). Constitutive activation of fibroblast growth factor receptor-2 by a point mutation associated with Crouzon syndrome. J. Biol. Chem. 270, 26037-26040. Newberry, E.P., Boudreaux, J.M., and Towler, D.A. (1996). The rat osteocalcin fibroblast growth factor (FGF)- responsive element: an okadaic acid-sensitive, FGF-selective transcriptional response motif. Mol. Endocrinol. 10, 1029-1040. Newberry, E.P., Latifi, T., Battaile, J.T., and Towler, D.A. (1997). Structure-function analysis of Msx2-mediated transcriptional suppression. Biochemistry 36, 10451-10462. Nichols, D.H. (1981). Neural crest formation in the head of the mouse embryo as observed using a new histological technique. J. Embryol. Exp. Morphol. 64, 105-120. Nichols, J., Evans, E.P., and Smith, A.G. (1990). Establishment of germ-line-competent embryonic stem (ES) cells using differentiation inhibiting activity. Development. 110, 1341-1348. Niederreither, K., D’Souza, R., Metsaranta, M., Eberspaecher, H., Toman, P.D., Vuorio, E., and de Crombrugghe, B. (1995). Coordinate patterns of expression of type I and III collagens during mouse development. Matrix Biol. 14, 705-713. Nijweide, P.J., Burger, E.H., Klein Nulend J., and Van der Plas, A. (1996). The osteocyte. In: Bilezikian, J.P., Raisz, L.G., and Rodan G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, 1996; Chapter 9. Nishimura, T., Utsunomiya, Y., Hoshikawa, M., Ohuchi, H., and Itoh, N. (1999). Structure and expression of a novel human FGF, FGF-19, expressed in the fetal brain. Biochim. Biophys. Acta 1444, 148-151. Niswander, L., and Martin, G.R. (1992). Fgf-4 expression during gastrulation, myogenesis, limb and tooth development in the mouse. Development 114, 755-768. Noda, M., and Vogel, R. (1989). Fibroblast growth factor enhances type beta 1 transforming growth factor gene expression in osteoblast-like cells. J. Cell Biol. 109, 2529-2535.

86 Noden, D.M. (1975). An analysis of migratory behavior of avian cephalic neural crest cells. Dev. Biol. 42, 106- 130. Noden, D.M. (1983). The role of the neural crest in patterning of avian cranial skeletal, connective, and muscle tissues. Dev. Biol. 96, 144-165. Noff, D., Pitaru, S., and Savion, N. (1989). Basic fibroblast growth factor enhances the capacity of bone marrow cells to form bone-like nodules in vitro. FEBS Letters. 250, 619-621. Nomura, S., Wills, A.J., Edwards, D.R., Heath, J.K., and Hogan, B.L. (1988). Developmental expression of 2ar (osteopontin) and SPARC (osteonectin) RNA as revealed by in situ hybridization. J. Cell Biol. 106, 441- 450. Ogata, T., and Noda, M. (1991). Expression of Id, a negative regulator of helix-loop-helix DNA binding proteins, is down-regulated at confluence and enhanced by dexamethasone in a mouse osteoblastic cell line, MC3T3E1. Biochem. Biophys. Res. Commun. 180, 1194-1199. Ohkuma, S., and Poole, B. (1981). Cytoplasmic vacuolation of mouse peritoneal macrophages and the uptake into lysosomes of weakly basic substances. J. Cell Biol. 90, 656-664. Okada, Y., Naka, K., Kawamura, K., Matsumoto, T., Nakanishi, I., Fujimoto, N., Sato, H., and Seiki, M. (1995). Localization of matrix metalloproteinase 9 (92-kilodalton gelatinase/type IV collagenase = gelatinase B) in osteoclasts: implications for bone resorption. Lab. Invest. 72, 311-322. Oldridge, M., Wilkie, A.O., Slaney, S.F., Poole, M.D., Pulleyn, L.J., Rutland, P., Hockley, A.D., Wake, M.J., Goldin, J.H., and Winter, R.M. (1995). Mutations in the third immunoglobulin domain of the fibroblast growth factor receptor-2 gene in Crouzon syndrome. Hum. Mol. Genet. 4, 1077-1082. Olsen, B.R. (1995). New insights into the function of collagens from genetic analysis. Curr. Opin. Cell Biol. 7, 720-727. OMIM, Online Mendelian Inheritance in Man, Data Base, www.ncbi.nlm.nih.gov/Omim Opperman, L.A., Chhabra, A., Nolen, A.A., Bao, Y., and Ogle, R.C. (1998). Dura mater maintains rat cranial sutures in vitro by regulating suture cell proliferation and collagen production. J. Craniofac. Genet. Dev. Biol. 18, 150-158. Opperman, L.A., Nolen, A.A., and Ogle, R.C. (1997). TGF-beta 1, TGF-beta 2, and TGF-beta 3 exhibit distinct patterns of expression during cranial suture formation and obliteration in vivo and in vitro. J. Bone Miner. Res. 12, 301-310. Opperman, L.A., Passarelli, R.W., Morgan, E.P., Reintjes, M., and Ogle, R.C. (1995). Cranial sutures require tissue interactions with dura mater to resist osseous obliteration in vitro. J. Bone Miner. Res. 10, 1978- 1987. Opperman, L.A., Passarelli, R.W., Nolen, A.A., Gamper, T.J., Lin, K.Y.K., Ogle, R.C. (1996). Dura mater secretes soluble heparin-binding factors required for cranial suture morphogenesis. In vitro Cell Dev. Biol. Animal. 32, 627-632. Opperman, L.A., Passarelli, R., Reintjes, M., Walford, T.P., and Ogle, R.C. (1993a). Tissue interactions during calvarial ogenesis require heparin-binding factors. Mol. Biol. Cell 4 Suppl., 145a. Opperman, L.A., Persing, J.A., Sheen, R., and Ogle, R.C. (1994). In the absence of periosteum, transplanted fetal and neonatal rat coronal sutures resist osseous obliteration. J. Craniofac. Surg. 5, 327-332. Opperman, L.A., Sweeney, T.M., Redmon, J., Persing, J.A., and Ogle, R.C. (1993b). Tissue interactions with underlying dura mater inhibit osseous obliteration of developing cranial sutures. Dev. Dyn. 198, 312- 322. Orr-Urtreger, A., Givol, D., Yayon, A., Lonai, P., and Yarden, Y. (1991). Developmental expression of two murine fibroblast growth factor receptors flg and bek. Development 113, 1419-1434. Ortega, S., Ittmann, M., Tsang, S.H., Ehrlich, M., and Basilico, C. (1998). Neuronal defects and delayed wound healing in mice lacking fibroblast growth factor 2. Proc. Natl. Acad. Sci. USA 95, 5672-5677. Otto, F., Thornell, A.P., Crompton, T., Denzel, A., Gilmour, K.C., Rosewell, I.R., Stamp, G.W., Beddington, R.S., Mundlos, S., Olsen, B.R., Selby, P.B., and Owen, M.J. (1997). Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 89, 765-771. Park, W.J., Theda, C., Maestri, N.E., Meyers, G.A., Fryburg, J.S., Dufresne, C., Cohen, M.M.J., and Jabs, E.W. (1995). Analysis of phenotypic features and FGFR2 mutations in Apert syndrome. Am. J. Hum. Genet. 57, 321-328. Parr, B.A., and McMahon, A.P. (1994). Wnt genes and vertebrate development. Curr. Opin. Genet. Dev. 4, 523- 528. Partanen, J., Makela, T.P., Eerola, E., Korhonen, J., Hirvonen, H., Claesson-Welsh, L., and Alitalo, K. (1991). FGFR-4, a novel acidic fibroblast growth factor receptor with a distinct expression pattern. EMBO J. 10, 1347-1354. Partanen, J., Schwartz, L., and Rossant, J. (1998). Opposite phenotypes of hypomorphic and Y766 phosphorylation site mutations reveal a function for Fgfr1 in anteroposterior patterning of mouse embryos. Genes Dev. 12, 2332-2344. Patarca, R., Freeman, G.J., Singh, R.P., Wei, F.Y., Durfee, T., Blattner, F., Regnier, D.C., Kozak, C.A., Mock, B.A., and Morse, H.C. (1989). Structural and functional studies of the early T lymphocyte activation 1

87 (Eta-1) gene. Definition of a novel T cell-dependent response associated with genetic resistance to bacterial infection. J. Exp. Med. 170, 145-161. Paznekas, W.A., Cunningham, M.L., Howard, T.D., Korf, B.R., Lipson, M.H., Grix, A.W., Feingold, M., Goldberg, R., Borochowitz, Z., Aleck, K., Mulliken, J., Yin, M., and Jabs, E.W. (1998). Genetic heterogeneity of Saethre-Chotzen syndrome, due to TWIST and FGFR mutations. Am. J. Hum. Genet. 62, 1370-1380. Pelton, R.W., Saxena, B., Jones, M., Moses, H.L., and Gold, L.I. (1991). Immunohistochemical localization of TGF beta 1, TGF beta 2, and TGF beta 3 in the mouse embryo: expression patterns suggest multiple roles during embryonic development. J. Cell Biol. 115, 1091-1105. Persson, K.M., Roy, W.A., Persing, J.A., Rodeheaver, G.T., and Winn, H.R. (1979). Craniofacial growth following experimental craniosynostosis and craniectomy in rabbits. J. Neurosurg. 50, 187-197. Persson, M. (1995). The role of sutures in normal and abnormal craniofacial growth. Acta Odontol. Scand. 53, 152-161. Pesce, S., and Benezra, R. (1993). The loop region of the helix-loop-helix protein Id1 is critical for its dominant negative activity. Mol. Cell. Biol. 13, 7874-7880. Peters, K., Ornitz, D., Werner, S., and Williams, L. (1993). Unique expression pattern of the FGF receptor 3 gene during mouse organogenesis. Dev. Biol. 155, 423-430. Peters, K.G., Werner, S., Chen, G., and Williams, L.T. (1992). Two FGF receptor genes are differentially expressed in epithelial and mesenchymal tissues during limb formation and organogenesis in the mouse. Development 114, 233-243. Piccolo, S., Sasai, Y., Lu, B., and 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. Polakowska, R.R., Piacentini, M., Bartlett, R., Goldsmith, L.A., and Haake, A.R. (1994). Apoptosis in human skin development: morphogenesis, periderm, and stem cells. Dev. Dyn. 199, 176-188. Pritchard, J.J., Scott, J.H., and Girgis, F.G. (1956). The structure and development of cranial and facial sutures. J. Anat. 90, 73-89. Prockop, D.J., Kuivaniemi, H., and Tromp, G. (1994). Molecular basis of osteogenesis imperfecta and related disorders of bone. Clin. Plast. Surg. 21,407-413. Proetzel, G., Pawlowski, S.A., Wiles, M.V., Yin, M., Boivin, G.P., Howles, P.N., Ding, J., Ferguson, M.W., and Doetschman, T. (1995). Transforming growth factor-beta 3 is required for secondary palate fusion. Nat. Genet. 11, 409-414. Przylepa, K.A., Paznekas, W., Zhang, M., Golabi, M., Bias, W., Bamshad, M.J., Carey, J.C., Hall, B.D., Stevenson, R., Orlow, S., Cohen, M.M.J., and Jabs, E.W. (1996). Fibroblast growth factor receptor 2 mutations in Beare-Stevenson cutis gyrata syndrome. Nat. Genet. 13, 492-494. Pyke, C., Ralfkiaer, E., Tryggvason, K., and Dano, K. (1993). Messenger RNA for two type IV collagenases is located in stromal cells in human colon cancer. Am. J. Pathol. 142, 359-365. Raff, M.C. (1992). Social controls on cell survival and cell death. Nature 356, 397-400. Raff, M.C., Barres, B.A., Burne, J.F., Coles, H.S., Ishizaki, Y., and Jacobson, M.D. (1993). Programmed cell death and the control of cell survival: lessons from the nervous system. Science 262, 695-700. Raynal, C., Delmas, P.D., and Chenu, C. (1996). Bone sialoprotein stimulates in vitro bone resorption. Endocrinology 137, 2347-2354. Reardon, W., Winter, R.M., Rutland, P., Pulleyn, L.J., Jones, B.M., and Malcolm, S. (1994). Mutations in the fibroblast growth factor receptor 2 gene cause Crouzon syndrome. Nat. Genet. 8, 98-103. Reardon, W., Wilkes, D., Rutland, P., Pulleyn, L.J., Malcolm, S., Dean, J.C., Evans, R.D., Jones, B.M., Hayward, R., Hall, C.M., Nevin, N.C., Baraister, M., and Winter, R.M. (1997). Craniosynostosis associated with FGFR3 pro250arg mutation results in a range of clinical presentations including unisutural sporadic craniosynostosis. J. Med. Genet. 34, 632-636. Reddi, A.H. (1994). Bone and cartilage differentiation. Curr. Opin. Genet. Dev. 4, 737-744. Reinholt, F.P., Hultenby, K., Oldberg, A., and Heinegard, D. (1990a). Osteopontin—a possible anchor of osteoclasts to bone. Proc. Natl. Acad. Sci. USA 87, 4473-4475. Reinholt, F.P., Widholm, S.M., Ek-Rylander, B., and Andersson, G. (1990b). Ultrastructural localization of a tartrate-resistant acid ATPase in bone. J. Bone Miner. Res. 5, 1055-1061. Renier, D., Sainte-Rose, C., Marchac, D., and Hirsch, J.F. (1982). Intracranial pressure in craniostenosis. J. Neurosurg. 57, 370-377. Reponen, P., Sahlberg, C., Munaut, C., Thesleff, I., and Tryggvason, K. (1994). High expression of 92-kD type IV collagenase (gelatinase B) in the osteoclast lineage during mouse development. J. Cell Biol. 124, 1091-1102. Rijli, F.M., Mark, M., Lakkaraju, S., Dierich, A., Dolle, P., and Chambon, P. (1993). A homeotic transformation is generated in the rostral branchial region of the head by disruption of Hoxa-2, which acts as a selector gene. Cell 75, 1333-1349. Roach, H.I., Erenpreisa, J., and Aigner, T. (1995). Osteogenic differentiation of hypertrophic chondrocytes involves asymmetric cell divisions and apoptosis. J. Cell Biol. 131, 483-494. Roberts, D.J., Johnson, R.L., Burke, A.C., Nelson, C.E., Morgan, B.A., and Tabin, C. (1995). Sonic hedgehog is

88 an endodermal signal inducing Bmp-4 and Hox genes during induction and regionalization of the chick hindgut. Development 121, 3163-3174. Rodan, S.B., Wesolowski, G., Yoon, K., and Rodan, G.A. (1989). Opposing effects of fibroblast growth factor and pertussis toxin on alkaline phosphatase, osteopontin, osteocalcin, and type I collagen mRNA levels in ROS 17/2.8 cells. J. Biol. Chem. 264, 19934-19941. Roodman, G.D. (1996). Advances in bone biology: the osteoclast. Endocr. Rev. 17, 308-332. Rossert, J., and de Crombrugghe, B. (1996). Type I collagen: structure, synthesis, and regulation. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, 1996; Chapter 10. Roth, D.A., Longaker, M.T., McCarthy, J.G., Rosen, D.M., McMullen, H.F., Levine, J.P., Sung, J., and Gold, L.I. (1997). Studies in cranial suture biology: Part I. Increased immunoreactivity for TGF-beta isoforms (beta 1, beta 2, and beta 3) during rat cranial suture fusion. J. Bone Miner. Res. 12, 311-321. Roy, W.A., Iorio, R.J., and Meyer, G.A. (1981). Craniosynostosis in vitamin D-resistant rickets. A mouse model. J. Neurosurg. 55, 265-271. Rushton, M.A. (1937). The dental condition in cleidocranial dysostosis. Guys Hosp. Rep. 87, 345-361. Rushton, M.A. (1956). An anomaly of cementum in cleidocranial dysostosis. Br. Dent. J. 100, 81-83. Rutland, P., Pulleyn, L.J., Reardon, W., Baraitser, M., Hayward, R., Jones, B., Malcolm, S., Winter, R.M., Oldridge, M., and Slaney, S.F. (1995). Identical mutations in the FGFR2 gene cause both Pfeiffer and Crouzon syndrome phenotypes. Nat. Genet. 9, 173-176. Sabbieti, M.G., Marchetti, L., Abreu, C., Montero, A., Hand, A.R., Raisz, L.G., and Hurley, M.M. (1999). Prostaglandins regulate the expression of fibroblast growth factor-2 in bone. Endocrinology 140, 434- 444. Salo, J., Lehenkari, P., Mulari, M., Metsikko, K., and Väänänen, H.K. (1997). Removal of osteoclast bone resorption products by transcytosis. Science 276, 270-273. Salo, T., Lyons, J.G., Rahemtulla, F., Birkedal-Hansen, H., and Larjava, H. (1991). Transforming growth factor- beta 1 up-regulates type IV collagenase expression in cultured human keratinocytes. J. Biol. Chem. 266, 11436-11441. Sandberg, M., Autio-Harmainen, H., and Vuorio, E. (1988). Localization of the expression of types I, III, and IV collagen, TGF-beta 1 and c-fos genes in developing human calvarial bones. Dev. Biol. 130, 324-334. Sanford, L.P., Ormsby, I., Gittenberger-de Groot, A.C., Sariola, H., Friedman, R., Boivin, G.P., Cardell, E.L., and Doetschman, T. (1997). TGFbeta2 knockout mice have multiple developmental defects that are non- overlapping with other TGFbeta knockout phenotypes. Development 124, 2659-2670. Satokata, I., and Maas, R. (1994). Msx1 deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development. Nat. Genet. 6, 348-356. Schaffer, J. (1930). Die Stützgewebe. In: “Handbuch der mikroskopischen Anatome des Menschen”, von Möllendorf, W ed. vol2, 338-350. Julius Springer, Berlin. Cited in Vinkka, H. (1982). Secondary cartilages in the facial skeleton of the rat. Proc. Finn. Dent. Soc. 78, suppl. 7, 1-137. Schell, U., Hehr, A., Feldman, G.J., Robin, N.H., Zackai, E.H., de Die-Smulders, C., Viskochil, D.H., Stewart, J.M., Wolff, G., and Ohashi, H. (1995). Mutations in FGFR1 and FGFR2 cause familial and sporadic Pfeiffer syndrome. Hum. Mol. Genet. 4, 323-328. Schowing, J. (1968). Demonstration of the inductive role of the brain in osteogenesis of the embryonic skull of the chicken. J. Embryol. Exp. Morphol. 19, 83-94. Schwartz, L.M., and Osborne, B.A. (1993). Programmed cell death, apoptosis and killer genes. Immunol. Today 14, 582-590. Selander, K.S., Monkkonen, J., Karhukorpi, E.K., Harkonen, P., Hannuniemi, R., and Väänänen, H.K. (1996). Characteristics of clodronate-induced apoptosis in osteoclasts and macrophages. Mol. Pharmacol. 50, 1127-1138. Selby, P.B., and Selby, P.R. (1978). Gamma-ray-induced dominant mutations that cause skeletal abnormalities in mice. III. Description of presumed mutations. Mut. Res. 50, 341-351. Shah, N.M., Groves, A.K., and Anderson, D.J. (1996). Alternative neural crest cell fates are instructively promoted by TGFbeta superfamily members. Cell 85, 331-343. Shishido, E., Higashijima, S., Emori, Y., and Saigo, K. (1993). Two FGF-receptor homologues of Drosophila: one is expressed in mesodermal primordium in early embryos. Development 117, 751-761. Sillence, D.O., Ritchie, H.E., and Selby, P.B. (1987). Animal model: skeletal anomalies in mice with cleidocranial dysplasia. Am. J. Med. Genet. 27, 75-85. Simonet, W.S., Lacey, D.L., Dunstan, C.R., Kelley, M., Chang, M.S., Luthy, R., Nguyen, H.Q., Wooden, S., Bennett, L., Boone, T., Shimamoto, G., DeRose, M., Elliott, R., Colombero, A., Tan, H.L., Trail, G., Sullivan, J., Davy, E., Bucay, Renshaw-Gegg, L., Hughes, T.M., Hill, D., Pattison, W., Campbell, P., and Boyle, W.J. (1997). Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89, 309-319. Slack, J.M. (1994). Inducing factors in Xenopus early embryos. Curr. Biol. 4, 116-126. Smith, D.W., and Töndury, G. (1978). Origin of the calvaria and its sutures. Am. J. Dis. Child. 132, 662-666.

89 Somerman, M.J., Fisher, L.W., Foster, R.A., and Sauk, J.J. (1988). Human bone sialoprotein I and II enhance fibroblast attachment in vitro. Calcif. Tissue Int. 43, 50-53. Somerman, M.J., Prince, C.W., Sauk, J.J., Foster, R.A., and Butler, W.T. (1987). Mechanism of fibroblast attachment to bone extracellular matrix: role of a 44 kilodalton bone phosphoprotein. J. Bone Miner. Res. 2, 259-265. Sood, S., Eldadah, Z.A., Krause, W.L., McIntosh, I., and Dietz, H.C. (1996). Mutation in fibrillin-1 and the Marfanoid-craniosynostosis (Shprintzen-Goldberg) syndrome. Nat. Genet. 12, 209-211. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. (1991). Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64, 693-702. Spicer, D.B., Rhee, J., Cheung, W.L., and Lassar, A.B. (1996). Inhibition of myogenic bHLH and MEF2 transcription factors by the bHLH protein Twist. Science 272, 1476-1480. Steinberger, D., Muller, U., Junger, T.H., Howaldt, H.P., and Christophis, P. (1999). Mutation of FGFR2 (cys278phe) in craniolacunia and pansynostosis. J. Med. Genet. 36, 499-500. Stelnicki, E.J., Vanderwall, K., Hoffman, W.Y., Harrison, M.R., Glowacki, J., and Longaker, M.T. (1998). A new in utero sheep model for unilateral coronal craniosynostosis. Plast. Reconstr. Surg. 101, 278-286. Storm, E.E., Huynh, T.V., Copeland, N.G., Jenkins, N.A., Kingsley, D.M., and Lee, S.J. (1994). Limb alterations in brachypodism mice due to mutations in a new member of the TGF beta-superfamily. Nature 368, 639- 643. Sung, V., Stubbs, J.T., Fisher, L., Aaron, A.D., and Thompson, E.W. (1998). Bone sialoprotein supports breast cancer cell adhesion proliferation and migration through differential usage of the alpha(v)beta3 and alpha(v)beta5 integrins. J. Cell Physiol. 176, 482-494. Szebenyi, G., Savage, M.P., Olwin, B.B., and Fallon, J.F. (1995). Changes in the expression of fibroblast growth factor receptors mark distinct stages of chondrogenesis in vitro and during chick limb skeletal patterning. Dev. Dyn. 204, 446-456. Takahashi, Y., and Le Douarin, N. (1990). cDNA cloning of a quail homeobox gene and its expression in neural crest-derived mesenchyme and lateral plate mesoderm. Proc. Natl. Acad. Sci. USA 87, 7482-7486. Takahashi, Y., Bontoux, M., and Le Douarin, N.M. (1991). Epithelio-mesenchymal interactions are critical for Quox 7 expression and membrane bone differentiation in the neural crest derived mandibular mesenchyme. EMBO J. 10, 2387-2393. Takeuchi, Y., Nakayama, K., and Matsumoto, T. (1996). Differentiation and cell surface expression of transforming growth factor-beta receptors are regulated by interaction with matrix collagen in murine osteoblastic cells. J. Biol. Chem. 271, 3938-3944. Tan, K.L., and Tan, L.K. (1981). Cleidocranial dysostosis in infancy. Pediatr. Radiol. 11, 114-116. Tan, S.S., and Morriss-Kay, G. (1985). The development and distribution of the cranial neural crest in the rat embryo. Cell Tissue Res. 240, 403-416. Tang, K.T., Capparelli, C., Stein, J.L., Stein, G.S., Lian, J.B., Huber, A.C., Braverman, L.E., and DeVito, W.J. (1996). Acidic fibroblast growth factor inhibits osteoblast differentiation in vitro: altered expression of collagenase, cell growth-related, and mineralization-associated genes. J. Cell Biochem. 61, 152-166. Tavella, S., Raffo, P., Tacchetti, C., Cancedda, R., and Castagnola, P. (1994). N-CAM and N-cadherin expression during in vitro chondrogenesis. Exp. Cell Res. 215, 354-362. Ten Cate, A.R., Freeman, E., and Dickinson, J.B. (1977). Sutural development: structure and its response to rapid expansion. Am. J. Orthod. 71, 622-636. Tezuka, K., Nemoto, K., Tezuka, Y., Sato, T., Ikeda, Y., Kobori, M., Kawashima, Eguchi, H., Hakeda, Y., and Kumegawa, M. (1994). Identification of matrix metalloproteinase 9 in rabbit osteoclasts. J. Biol. Chem. 269, 15006-15009. Tezuka, K. Sato, T., Kamioka, H., Nijweide, P.J., Tanaka, K., Matsuo, T., Ohta, M., Kurihara, N., Hakeda, Y., and Kumegawa, M. (1992). Identification of osteopontin in isolated rabbit osteoclasts. Biochem. Biophys. Res. Commun. 186, 911-917. Ting, K., Vastardis, H., Mulliken, J.B., Soo, C., Tieu, A., Do, H., Kwong, E., Bertolami, C.N., Kawamoto, H., Kuroda, S., and Longaker, M.T. (1999). Human NELL-1 expressed in unilateral coronal synostosis. J. Bone Miner. Res. 14, 80-89. Thaller, S.R., Hoyt, J., Tesluk, H., and Holmes, R. (1993). The effect of insulin growth factor-1 on calvarial sutures in a Sprague-Dawley rat. J. Craniofac. Surg. 4, 35-39. Thesingh, C.W., Groot, C.G., and Wassenaar, A.M. (1991). Transdifferentiation of hypertrophic chondrocytes into osteoblasts in murine fetal metatarsal bones, induced by co-cultured cerebrum. Bone Miner. 12, 25- 40. Thesleff, I., and Sharpe, P. (1997). Signalling networks regulating dental development. Mech. Dev. 67, 111-123. Thompson, S.W. (1966). Selected histological and histopathological methods. Charles C. Thomas Publishers, Springfield, Illinois. p 644. Thorogood, P. (1993). Differentiation and morphogenesis of cranial skeletal tissues. In: The Skull. Hanken, J., and Hall, B.K. eds. Chicago University Press. Chapter 4. Thorogood, P.V., and Hinchliffe, J.R. (1975). An analysis of the condensation process during chondrogenesis in

90 the embryonic chick hind limb. J. Embryol. Exp. Morphol. 33, 581-606. Tondravi, M.M., McKercher, S.R., Anderson, K., Erdmann, J.M., Quiroz, M., Maki, R., and Teitelbaum, S.L. (1997). Osteopetrosis in mice lacking haematopoietic transcription factor PU.1. Nature 386, 81-84. Towler, D.A., Rutledge, S.J., and Rodan, G.A. (1994). Msx-2/Hox 8.1: a transcriptional regulator of the rat osteocalcin promoter. Mol. Endocrinol. 8, 1484-1493. Triffitt, J.T. (1996). The stem cell of the osteoblast. In: Bilezikian, J.P., Raisz, L.G., and Rodan, G.A. eds. Principles of Bone Biology. San Diego, CA, USA: Academic Press, Chapter 4. Tucker, A.S., Yamada, G., Grigoriou, M., Pachnis, V., and Sharpe, P.T. (1999). Fgf-8 determines rostral-caudal polarity in the first branchial arch. Development 126, 51-61. Tyler, M.S., and Hall, B.K. (1977). Epithelial influences on skeletogenesis in the mandible of the embryonic chick. Anat. Rec. 188, 229-239. Vaahtokari, A., Åberg, T., and Thesleff, I. (1996). Apoptosis in the developing tooth: association with an embryonic signaling center and suppression by EGF and FGF-4. Development 122, 121-129. Väänänen, H.K., and Horton, M. (1995). The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J. Cell Sci. 108, 2729-2732. Väänänen, H.K., Karhukorpi, E.K., Sundquist, K., Wallmark, B., Roininen, I., Hentunen, T., Tuukkanen, J., and Lakkakorpi, P. (1990). Evidence for the presence of a proton pump of the vacuolar H(+)-ATPase type in the ruffled borders of osteoclasts. J. Cell Biol. 111, 1305-1311. Vainio, S., Jalkanen, M., Vaahtokari, A., Sahlberg, C., Mali, M., Bernfield, M., and Thesleff, I. (1991). Expression of syndecan gene is induced early, is transient, and correlates with changes in mesenchymal cell proliferation during tooth organogenesis. Dev. Biol. 147, 322-333. Vainio, S., Karavanova, I., Jowett, A., and Thesleff, I. (1993). Identification of BMP-4 as a signal mediating secondary induction between epithelial and mesenchymal tissues during early tooth development. Cell 75, 45-58. Vincent, J.B., and Averill, B.A. (1990). An enzyme with a double identity: purple acid phosphatase and tartrate- resistant acid phosphatase. FASEB J. 4, 3009-3014. Virchow, R. (1851). Über den Cretinismus, namentlich in Franken, und über pathologische Schädelformen. Verh Phys Med Ges Würzburg 2:230-271. Cited in Kreiborg, S. (1986). Postnatal growth and development of the craniofacial complex in premature craniosynostosis. In: Cohen, M.M.Jr. ed. (1986) Craniosynostosis: Diagnosis, Evaluation, and Management. Raven Press, New York. Vortkamp, A., Gessler, M., and Grzeschik, K.H. (1991). GLI3 zinc-finger gene interrupted by translocations in Greig syndrome families. Nature 352, 539-540. Vortkamp, A., Lee, K., Lanske, B., Segre, G.V., Kronenberg, H.M., and Tabin, C.J. (1996). Regulation of rate of cartilage differentiation by Indian hedgehog and PTH-related protein. Science 273, 613-622. Vu, T.H., Shipley, J.M., Bergers, G., Berger, J.E., Helms, J.A., Hanahan, D., Shapiro, S.D., Senior, R.M., and Werb, Z. (1998). MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes. Cell 93, 411-422. Walker, D.G. (1975). Bone resorption restored in osteopetrotic mice by transplants of normal bone marrow and spleen cells. Science 190, 784-785. Wang, Z.Q., Ovitt, C., Grigoriadis, A.E., Mohle-Steinlein, U., Ruther, U., and Wagner, E.F (1992). Bone and haematopoietic defects in mice lacking c-fos. Nature 360, 741-745. Waymire, K.G., Mahuren, J.D., Jaje, J.M., Guilarte, T.R., Coburn, S.P., and MacGregor, G.R. (1995). Mice lacking tissue non-specific alkaline phosphatase die from seizures due to defective metabolism of vitamin B-6. Nat. Genet. 11, 45-51. Weeks, B.S., Schnaper, H.W., Handy, M., Holloway, E., and Kleinman, H.K. (1993). Human T lymphocytes synthesize the 92 kDa type IV collagenase (gelatinase B). J. Cell Physiol. 157, 644-649. Weinreb, M., Shinar, D., and Rodan, G.A. (1990). Different pattern of alkaline phosphatase, osteopontin, and osteocalcin expression in developing rat bone visualized by in situ hybridization. J. Bone Miner. Res. 5, 831-842. Weiss, M.J., Cole, D.E., Ray, K., Whyte, M.P., Lafferty, M.A., Mulivor, R.A., and Harris (1988). A missense mutation in the human liver/bone/kidney alkaline phosphatase gene causing a lethal form of hypophosphatasia. Proc. Natl. Acad. Sci. USA 85, 7666-7669. White, C.D., (1996). Sutural effects of fronto-occipital cranial modification. Am. J. Phys. Anthrop. 100, 397- 410. Wiktor-Jedrzejczak, W., Bartocci, A., Ferrante, A.W.J., Ahmed-Ansari, A., Sell, KW, Pollard, J.W., and Stanley, E.R. (1990). Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. Proc. Natl. Acad. Sci. USA 87, 4828-4832. Wilkes, D., Rutland, P., Pulleyn, L.J., Reardon, W., Moss, C., Ellis, J.P., Winter, RM, and Malcolm, S. (1996). A recurrent mutation, ala391glu, in the transmembrane region of FGFR3 causes Crouzon syndrome and acanthosis nigricans. J. Med. Genet. 33, 744-748. Wilkie, A.O. (1997). Craniosynostosis: genes and mechanisms. Hum. Mol. Genet. 6, 1647-1656. Wilkie, A.O., Slaney, S.F., Oldridge, M., Poole, M.D., Ashworth, G.J., Hockley, A.D., Hayward, R.D., David, D.J.,

91 Pulleyn, L.J., and Rutland, P. (1995). Apert syndrome results from localized mutations of FGFR2 and is allelic with Crouzon syndrome. Nat. Genet. 9, 165-172. Wilkinson, D.G., and Green, J. (1990). In situ hybridisation and the three-dimensional reconstruction of serial sections. In: Copp, A. J. and Cockroft, D.L. eds. Postimplantation Mammalian Embryos: A Practical Approach. IRL press, Oxford. pp. 155-171. Wilkinson, D.G., Peters, G., Dickson, C., and McMahon, A.P. (1988). Expression of the FGF-related proto- oncogene int-2 during gastrulation and neurulation in the mouse. EMBO J. 7, 691-695. Wilkinson, D.G., Bhatt, S., and McMahon, A.P. (1989). Expression pattern of the FGF-related proto-oncogene int-2 suggests multiple roles in fetal development. Development. 105, 131-136. Williams, E.J., Furness, J., Walsh, F.S., and Doherty, P. (1994). Activation of the FGF receptor underlies neurite outgrowth stimulated by L1, N-CAM, and N-cadherin. Neuron 13, 583-594. Willing, M.C., Deschenes, S.P., Scott, D.A., Byers, P.H., Slayton, R.L., Pitts, S.H., Arikat, H., and Roberts, E.J. (1994). Osteogenesis imperfecta type I: molecular heterogeneity for COL1A1 null alleles of type I collagen. Am. J.Hum. Genet. 55, 638-647. Winograd, J., Reilly, M.P., Roe, R., Lutz, J., Laughner, E., Xu, X., Hu, L., Asakura, T., Vander Kolk, C., Strandberg, J.D., and Semenza, G.L. (1997a). Perinatal lethality and multiple craniofacial malformations in MSX2 transgenic mice. Hum. Mol. Genet. 6, 369-379. Winograd, J.M., Im, M.J., Vander Kolk, C.A. (1997b). Osteoblastic and osteoclastic activation in coronal sutures undergoing fusion ex vivo. Plast. Reconstr. Surg. 100, 1103-1112. Winter, R.M., and Baraitser, M, (1996). The London Dysmorphology Database. Oxford University Press, Oxford. Wozney, J.M. (1998). The bone morphogenetic protein family: multifunctional cellular regulators in the embryo and adult. Eur. J. Oral Sci. 106 Suppl. 1, 160-166. Wuechner, C., Nordqvist, A.C., Winterpacht, A., Zabel, B., and Schalling, M. (1996). Developmental expression of splicing variants of fibroblast growth factor receptor 3 (FGFR3) in mouse. Int. J. Dev. Biol. 40, 1185- 1188. Xiao, G., Wang, D., Benson, M.D., Karsenty, G., and Franceschi, R.T. (1998). Role of the alpha2-integrin in osteoblast-specific gene expression and activation of the Osf2 transcription factor. J. Biol. Chem. 273, 32988-32994. Xu, X., Weinstein, M., Li, C., Naski, M., Cohen, R.I., Ornitz, D.M., Leder, P., and Deng C. (1998). Fibroblast growth factor receptor 2 (FGFR2)-mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development. 125, 753-765. Yamaguchi, T.P., Conlon, R.A., and Rossant, J. (1992). Expression of the fibroblast growth factor receptor FGFR-1/flg during gastrulation and segmentation in the mouse embryo. Dev. Biol. 152, 75-88. Yamaguchi, T.P., Harpal, K., Henkemeyer, M., and Rossant, J. (1994). fgfr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev. 8, 3032-3044. Yamaguchi, T.P., and Rossant, J. (1995). Fibroblast growth factors in mammalian development. Curr. Opin. Genet. Dev. 5, 485-491. Yasuda, H., Shima, N., Nakagawa, N., Yamaguchi, K., Kinosaki, M., Mochizuki, S., Tomoyasu, A., Yano, K., Goto, M., Murakami, A., Tsuda, E., Morinaga, T., Higashio, K., Udagawa, N., Takahashi, N., and Suda, T. (1998). Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc. Natl. Acad. Sci. USA 95, 3597-3602. Yaziji, H., Janckila, A.J., Lear, S.C., Martin, A.W., and Yam, L.T. (1995). Immunohistochemical detection of tartrate-resistant acid phosphatase in non-hematopoietic human tissues. Am. J. Clin. Pathol. 104, 397- 402. Yen, E.H., Yue, C.S., Suga, D.M. (1989). Effect of force level on synthesis of type III and type I collagen in mouse interparietal suture. J. Dent. Res. 68, 1746-1751. Yip, J.E., Kokich, V.G., and Shepard, T.H. (1980). The effect of high doses of retinoic acid on prenatal craniofacial development in Macaca nemestrina. Teratology 21, 29-38. Yoshida, H., Hayashi, S., Kunisada, T., Ogawa, M., Nishikawa, S., Okamura, H., Sudo, T., and Shultz, L.D. (1990). The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442-444. Zeichner-David, M., Vo, H., Tan, H., Diekwisch, T., Berman, B., Thiemann, F., Alcocer, M.D., Hsu, P., Wang, T., Eyna, J., Caton, J., Slavkin, H.C., and MacDougall, M. (1997). Timing of the expression of enamel gene products during mouse tooth development. Int. J. Dev. Biol. 41, 27-38. Zhang, R., Ducy, P., and Karsenty, G. (1997). 1,25-dihydroxyvitamin D3 inhibits Osteocalcin expression in mouse through an indirect mechanism. J. Biol. Chem. 272, 110-116. Zhao, Q., Behringer, R.R., and de Crombrugghe, B. (1996). Prenatal folic acid treatment suppresses acrania and meroanencephaly in mice mutant for the Cart1 homeobox gene. Nat. Genet. 13, 275-283. Zhou, M., Sutliff, R.L., Paul, R.J., Lorenz, J.N., Hoying, J.B., Haudenschild, C.C., Yin, M., Coffin, J.D., Kong, L., Kranias, E.G., Luo, W., Boivin, G.P., Duffy, J.J., Pawlowski, S.A., and Doetschman, T. (1998). Fibroblast growth factor 2 control of vascular tone. Nature Medicine 4 , 201-207. Zhu, X., Komiya, H., Chirino, A., Faham, S., Fox, G.M., Arakawa, T., Hsu, B.T., and Rees, D.C. (1991). Three- dimensional structures of acidic and basic fibroblast growth factors. Science 251, 90-93. 92