indian and sonic hedgehogs regulate synchondrosis growth plate and cranial base development and...

99 (2006) 272–282www.elsevier.com/locate/ydbio

Developmental Biology 2

Indian and sonic hedgehogs regulate synchondrosis growth plate and cranialbase development and function

Blanche Young a, Nancy Minugh-Purvis b, Tsuyoshi Shimo c, Benoit St-Jacques d,Masahiro Iwamoto a, Motomi Enomoto-Iwamoto a, Eiki Koyama a, Maurizio Pacifici a,⁎

a Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, PA 19017, USAb Department of Anatomy and Cell Biology, University of Pennsylvania, Philadelphia, PA 19104, USA

c Department of Oral and Maxillofacial Surgery, Okayama University, Okayama 700-8525, Japand Shriners Hospital for Children, Genetics Unit, Montreal, Quebec, Canada H3G 1A6

Received for publication 26 May 2006; revised 21 July 2006; accepted 25 July 2006Available online 29 July 2006

Abstract

The synchondroses consist of mirror-image growth plates and are critical for cranial base elongation, but relatively little is known about theirformation and regulation. Here we show that synchondrosis development is abnormal in Indian hedgehog-null mice. The Ihh−/− cranial basesdisplayed reduced growth and chondrocyte proliferation, but chondrocyte hypertrophy was widespread. Rather than forming a typical narrowzone, Ihh−/− hypertrophic chondrocytes occupied an elongated central portion of each growth plate and were flanked by immature collagen II-expressing chondrocytes facing perichondrial tissues. Endochondral ossification was delayed in much of the Ihh−/− cranial bases but, surprisingly,was unaffected most posteriorly. Searching for an explanation, we found that notochord remnants near incipient spheno-occipital synchondroses atE13.5 expressed Sonic hedgehog and local chondrocytes expressed Patched, suggesting that Shh had sustained chondrocyte maturation andoccipital ossification. Equally unexpected, Ihh−/− growth plates stained poorly with Alcian blue and contained low aggrecan transcript levels. Acomparable difference was seen in cultured wild-type versus Ihh−/− synchondrosis chondrocytes. Treatment with exogenous Ihh did not fullyrestore normal proteoglycan levels in mutant cultures, but a combination of Ihh and BMP-2 did. In summary, Ihh is required for multiple processesduring synchondrosis and cranial base development, including growth plate zone organization, chondrocyte orientation, and proteoglycanproduction. The cranial base appears to be a skeletal structure in which growth and ossification patterns along its antero-posterior axis areorchestrated by both Ihh and Shh.© 2006 Elsevier Inc. All rights reserved.

Keywords: Indian hedgehog; Sonic hedgehog; Cranial base; Synchondrosis; Growth plate; Skeletogenesis; Extracellular matrix; Aggrecan; BMPs

Introduction

The cranial base is a morphologically complex portion of theskull that functions as a supporting platform for the developingbrain, provides special physical niches to organs such as thepituitary, and establishes distinct articulations with differentskeletal elements, including the mandible. The central region ofthe cranial base derives from the prechordal, hypophyseal, andparachordal cartilaginous plates, three pairs of precursorstructures that form between the first and second month of

⁎ Corresponding author.E-mail address: [email protected] (M. Pacifici).

0012-1606/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.ydbio.2006.07.028

gestation in humans (Larsen, 2001; Sperber, 2001; Thorogood,1988). The plates are arranged in series, lie below the developingbrain, and eventually fuse to form an uninterrupted cartilaginousstructure spanning the region from foramen magnum tointerorbital junction. With further development, primary ossifi-cation centers emerge within the unified cartilaginous structure,first posteriorly and then anteriorly (Kjaer, 1990). The majorcartilaginous segments persisting between the ossificationcenters represent the synchondroses, termed ethmoidal, intras-phenoidal, spheno-occipital, and intraoccipital according to theiranatomical location. The cranial base synchondroses display aremarkable palindromic organization, each consisting of twomirror-image growth plates arranged in opposing directions. The

mailto:[email protected]

http://dx.doi.org/10.1016/j.ydbio.2006.07.028

273B. Young et al. / Developmental Biology 299 (2006) 272–282

two growth plates share a central reserve/resting zone, displayzones of chondrocyte proliferation, pre-hypertrophy andhypertrophy and endochondral ossification, but lack an articularsynovial layer that is instead typical of growth plates indeveloping long bones. Given this orientation, the synchon-droses produce growth in opposing directions and are respon-sible for lengthening and ossification of the cranial base mainlyalong its anterior–posterior axis (Mooney et al., 1993). Recentstudies have provided some insights into the cell and molecularbasis of synchondrosis development and growth (Rice et al.,2003; Shum et al., 2003; Gakunka et al., 2000; Chen et al., 1999;Ishii-Suzuki et al., 1999), but much remains unclear, particularlyat the molecular signaling level.

Indian hedgehog (Ihh) is a critical regulator of skeletogenesis.Studies on long bone development have shown that Ihh isexpressed in the pre-hypertrophic zone of growth plates(Vortkamp et al., 1996; Koyama et al., 1996a; Bitgood andMcMahon, 1995). We found that Ihh expression coincides withossification around the diaphysis and that exogenous Ihhpromotes osteogenic cell differentiation in vitro, leading us topropose that a key role of Ihh is to initiate ossification inperichondrial tissues (Nakamura et al., 1997; Koyama et al.,1996a). Others found that Ihh limits the rates of growth platechondrocyte maturation in concert with periarticular derivedparathyroid hormone-related protein (PTHrP) (Lanske et al.,1996; Vortkamp et al., 1996). Studies with Ihh−/− mice providedclear support for both roles and indicated also that Ihh has a rolein chondrocyte proliferation (Chung et al., 2001; St-Jacqueset al., 1999). Currently, the view is that Ihh is a key paracrineregulator of distinct growth plate events in developing longbones: it stimulates chondrocyte proliferation and inhibitsmaturation together with periarticular tissue-derived PTHrPand induces intramembranous and endochondral ossification(Kronenberg, 2003). The present study was conducted todetermine whether Ihh regulates growth plate function in cranialbase synchondroses as well.

Materials and methods

Mouse embryo genotyping and anatomical analysis

Heterozygous Ihh+/−mice in F1 mixed background (129/SV:C57Bl6/J) weremated to generate homozygous Ihh−/− embryos. Fragments of embryonic andneonatal tails were genotyped by PCR using the following primers: for wild-typeIhh exon 1, forward primer 5′CACCCCCAACTACAATCCCGACATCA3′ andreverse primer 5′AATGACCAGGGCTGGGCTGTGAGAA3′ generating a 464bp product; and for neomycin gene, forward primer 5′ AGGAGGCAGGGA-CATGGATAGGGTG 3′ and reverse primer 5′ TACCGGTGGATGTG-GAATGTGTGCG 3′ generating a 300 bp product. Thermocycler conditionswere 95°C for 30 s, 64°C for 1 min and 72°C for 1 min for 38 cycles. A total ofover 400 wild-type and mutant embryos and newborns were analyzed in thestudy. Skeletonswere stainedwith Alcian blue (Inouye, 1976), cranial bases werephotographed and measured microscopically. Statistical analyses were per-formed by one way ANOVAwith Tukey post hoc comparisons.

Gene expression analyses

For in situ hybridization, paraffin-embedded serial tissue sections werepretreated with 1 μg/ml proteinase K (Sigma) for 1 min at 37°C, post-fixed in 4%paraformaldehyde, washed with PBS containing 2 mg/ml glycine, and treated

with 0.25% acetic anhydride in triethanolamine buffer (Koyama et al., 1996a).Sections were hybridized with antisense or sense 35S-labeled probes (approxi-mately 1×106 DPM/section) at 50°C for 16 h. Mouse cDNA clones were:aggrecan (nucleotides 880–1733; NM_007424); histone H4C (nt. 549–799;AY158963); PTHrP (nt. 66–1386; NM_008970); PTHrP-R (nt. 671–1450;NM_011199); osteopontin (nt. 1–267; AF515708); Osterix (nt. 40–1727;NM_130458); MMP-9 (nt. 1937–2258; NM_008610); VEGF (nt. 115–539; gi/249858); Ihh (nt. 897–1954; MN_010544); Shh (full coding sequence;NM_009170); Patched-1 (nt. 81–841; NM_008957); Smoothened (nt. 450–1000; BC048091); collagen X (nt. 1302–1816; NM009925); and collagen II (nt.1095–1344; X57982). After hybridization, slides were washed with 2× SSCcontaining 50% formamide at 50°C, treated with 20 μg/ml RNase A for 30min at37°C, and washed three times with 0.1× SSC at 50°C for 10 min/wash. Sectionswere dehydrated with 70, 90, and 100% ethanol for 5 min/step, coated withKodak NTB-3 emulsion diluted 1:1 with water, and exposed for 10 to 14 days.Slides were developed with Kodak D-19 at 20°C and stained with hematoxylin.Dark and bright field images were captured using a digital camera, and dark fieldimages were pseudo-colored using Adobe Photoshop software.

For in situ hybridization using DIG labeled probes, paraffin serial sectionswere pretreated with 0.2 N HCl for 10 min, 0.25% acetic anhydride intriethanolamine buffer for 15 min, and 1 μg/ml proteinase K in 50 mM Tris–HCl, 5 mM EDTA pH 7.5 for 1 min at room temperature and then immediatelypost-fixed in 4% paraformaldehyde for 10 min. Sections were hybridized withantisense or sense DIG-probes (approximately 0.5 μg/ml) at 55°C for 16 h.Slides were washed with 2× SSC containing 50% formamide at 55°C for 30 min,three times with TN (10 mM Tris–HCl pH 7.5; 0.5 M NaCl), treated with 20 μg/ml RNase A for 30 min at 37°C, washed two times with 1× SSC containing 25%formamide at 55°C for 30 min/wash, and finally stained in NBT/BCIP colorsubstrate solution.

Chondrocyte cultures and proteoglycan analyses

Synchondroses were isolated microsurgically from E17.5 wild-type and Ihhembryos under a dissecting microscope, making sure they were devoid ofadhering tissues as much as possible. Tissue fragments were incubated for30 min at 37°C in 1× trypsin/EDTA mixture (Gibco) in Ca/Mg-free Hank'ssaline containing 60 μg/ml kanamycin and 1×Fungizone (Gibco) with gentlevortexing every 10 min. Released cells were discarded, and remaining tissuefragments were further incubated for 1.5 h in serum-free DMEM containing1.7 U/ml of type I collagenase (Worthington) at 37°C. After removal of enzymemixture, tissue fragments were dissociated into a single cell suspension by gentlepipetting and cells were plated at an initial density of 2×104/well in type Icollagen-coated 48 multi-well tissue culture plates. Medium was HG-DMEMcontaining 10% fetal bovine serum (Hyclone) and 10 μg/ml ascorbic acid. Ondays 4–5 of culture, the subconfluent cells were given fresh medium containingappropriate additives, including 100 ng/ml rhBMP-2 (Yamanouchi Phar. Co.),1 μg/ml rhIHH (R&D) and/or 0.3 μg/ml Noggin (R&D). Medium was changedevery other day, and fresh additives were provided.

To monitor proteoglycan accumulation, cultures were fixed with 3% aceticacid pH 1.0 for 5 min and stained with 0.1% Alcian blue in 3% acetic acid pH1.0 (Enomoto-Iwamoto et al., 2000). After exhaustive rinsing, cell layer-associated Alcian blue was solubilized in DMSO and quantified spectro-photometrically. DNA content was measured (Johnson-Wint and Wellis, 1982)and used for normalization.

Results

Deranged synchondrosis development in Ihh−/− mice

Skulls from wild-type and Ihh−/− E15.5 and E17.5 mouseembryos and newborn (P0) mice were stained with Alcian blueto depict cartilaginous structures. Wild-type cranial basesdisplayed their expected staining features and an elongatedshape along the antero-posterior axis. The central portion wasuniformly cartilaginous and Alcian blue-positive at E15.5

274 B. Young et al. / Developmental Biology 299 (2006) 272–282

(Figs. 1A, D, red arrow) and exhibited well-defined intrasphe-noidal (is) and spheno-occipital (so) synchondroses by E17.5and P0 flanking posterior sphenoid (ps) and basi-occipital (b-oc) ossification centers (Figs. 1B–C, E–F). In contrast, theantero-posterior length of Ihh−/− cranial bases was reduced ateach stage examined and as much as 25% at P0 (n=7; p<0.05)and was accompanied by lateral bulging of the neurocranium(Figs. 1G–I). The synchondroses were at appropriate locationsbut were markedly reduced in size and stained poorly withAlcian blue (Figs. 1H–I, K–L). The reduction in length was notsurprising given known Ihh roles in long bone elongation(St-Jacques et al., 1999), but the reduced Alcian blue stainingwas unexpected.

To test whether length reduction resulted, in part, from lowchondrocyte proliferation, longitudinal sections of E15.5.E17.5 and P0 wild-type and Ihh−/− synchondroses were

Fig. 1. The cranial base is abnormal in Ihh−/− mice. Bird's eye view of (A–F)wild-type and (G–L) Ihh−/− cranial bases from E15.5 and E17.5 embryos andnewborn (P0) mice stained with Alcian blue. (A, D, G, J) In E15.5 wild-type andmutant embryos, the cranial base is largely cartilaginous and stains almostuniformly with Alcian blue (red arrow). (B–C, E–F) By E17.5 and P0, wild-typecranial bases display well-defined and blue-stained intrasphenoidal (is) andspheno-occipital (so) synchondroses that flank primary gray-stained ossificationcenters, including the posterior sphenoidal (ps) and basi-occipital (b-oc) centers.(H–I, K–L) In both E17.5 and P0 Ihh−/− cranial bases, the synchondroses are illdefined, are reduced in size, and stain quite weakly with Alcian blue. Scale bars:(A–C, G–I) 1 mm; (D–F, J–L) 1.2 mm.

processed for expression of histone H4C, a universal markerof proliferating cells, by in situ hybridization. H4C-positivechondrocytes were readily detectable in the proliferative zones(pz) of wild-type growth plates in both intrasphenoidal andspheno-occipital synchondroses at each stage tested (Figs. 2A–F, arrows); there were also several proliferative chondrocytes inthe central reserve zone (rz) at E15.5 that had decreased by P0(Figs. 2A–F, bracket). In mutant synchondroses, the number ofH4C-positive chondrocytes was extremely low, regardless ofstage or location examined (Figs. 2G–L). Positive cells couldbe seen in surrounding tissues (Figs. 2G–L, arrowheads),indicating that the proliferative decay was cartilage-specific.

We then asked what could account for the unexpecteddecrease in Alcian blue staining in mutant synchondroses.Several possibilities were considered, including reducedaggrecan gene expression, increased proteoglycan degradationand reduced glycosaminoglycan sulfation. To test the first (andmore likely) possibility, wild-type and mutant sections wereanalyzed for aggrecan gene expression by in situ hybridization.In wild-type specimens, aggrecan transcripts were abundantthroughout the reserve, proliferative and pre-hypertrophiczones of growth plates (Figs. 3A–C, bracket), while expressionwas much reduced in the hypertrophic zone (Figs. 3A–C, hz),patterns typical also of long bone growth plates (Sandell et al.,1994). In Ihh−/− specimens, however, aggrecan transcriptswere severely reduced throughout the synchondroses at eachstage examined (Figs. 3G–I). To investigate mechanisms, weisolated chondrocytes from E17.5 wild-type and Ihh−/−

synchondroses, grew them in primary monolayer culture,treated them with exogenous rhIHH (1 μg/ml), rhBMP-2(100 ng/ml), rhIHH plus BMP-2, or vehicle for 48 h, andfinally measured proteoglycan content. Proteoglycan contentwas higher in wild-type than Ihh−/− cultures (Fig. 3D), in goodcorrelation with the in vivo gene expression levels. ExogenousrhIHH treatment did not change these patterns significantly,but rhBMP-2, and rhBMP-2 plus rhIHH, in particular,markedly increased proteoglycan content in both mutant andwild-type cultures (Fig. 3D). When the cells were treated withthe rhBMP antagonist Noggin (0.3 μg/ml), there was an acrossthe board reduction in proteoglycan content; this wasmeasurable biochemically (Fig. 3J) and was also illustratedby the dull phase microscopic appearance of the Noggin-treated cells (Fig. 3F) compared to the more refractile, and thusmatrix-rich, appearance of companion cultures receiving noNoggin (Fig. 3E).

Altered gene expression patterns and growth plate function inIhh−/− synchondroses

We moved on to determine the gene expression patterns,maturation levels, and organization of chondrocytes in thesynchondroses. At E15.5, wild-type intrasphenoidal andspheno-occipital synchondroses already displayed well-estab-lished growth plates (Figs. 4A, G) containing Ihh-expressingpre-hypertrophic chondrocytes (Figs. 4B, H, arrowheads) andPatched-expressing proliferating chondrocytes (Figs. 4C, I,arrowheads). Smoothened and PTHrP were expressed

Fig. 3. Aggrecan gene expression in synchondroses and cultured chondrocytes.Parasagittal sections from E15.5, E17.5, and P0 wild-type (A–C) and Ihh−/−

(G–I) synchondroses were analyzed for aggrecan gene expression (dark browncolor) by in situ hybridization. (D–E) Chondrocytes from E17.5 wild-type andIhh−/− synchondroses were maintained in primary culture and treated withrhIHH, rhBMP-2, rhIHH plus rhBMP-2, or left untreated and were thenprocessed for quantification of proteoglycan content on day 7 of culture (D) orviewed under phase microscopy (E). Companion cultures were subjected tosimilar treatments in the presence of Noggin and processed for proteoglycancontent analysis (J) or viewed by phase microscopy (F). Scale bars: (A–C) and(G–I) 75 μm; (E–F) 15 μm.

Fig. 2. Chondrocyte proliferation is reduced in Ihh−/− synchondroses.Parasagittal sections from E15.5, E17.5, and P0 wild-type (A–F) and Ihh−/−

(G–L) mice were analyzed for H4C gene expression (dark purple color) by insitu hybridization. (A–F) At each stage examined, numerous H4C-positivechondrocytes are present in the proliferative zones (pz) of intrasphenoidal (is)and spheno-occipital (so) wild-type synchondroses flanking a reserve zone (rz)that contains scattered proliferating chondrocytes. No mitotic chondrocytes areseen in the hypertrophic zone (hz), as expected. (G–L) In Ihh−/− synchondroses,there is a marked decrease in proliferating chondrocytes, but several positivecells can be seen in surrounding tissues (arrowheads). Scale bar: 150 μm.


throughout the resting and proliferative areas (Figs. 4D–E, J–K,brackets), and PTHrP-R instead characterized pre-hypertrophicand early hypertrophic chondrocytes (Figs. 4F, L, arrowheads).Ihh−/− synchondroses were histologically disorganized (Figs.4M, S) and contained barely detectable amounts of Patched andPTHrP transcripts (Figs. 4O, Q, U, W). Smoothened geneexpression persisted (Figs. 4P, V), and PTHrP-R transcripts werenow scattered throughout the cartilaginous tissue (Figs. 4R, X).

As suggested by these observations, the chondrocytematuration and ossification programs were also severelyaffected (Fig. 5). Wild-type E15.5 and E17.5 synchondrosesdisplayed mature collagen X- and osteopontin-positive hyper-trophic chondrocytes at the growth plate end facing endochon-dral bone (Figs. 5A–C, E–G, arrows). The cells flankedposterior sphenoidal and basi-occipital ossification centers thatstained positively with the mineral stain alizarin red andexpressed also osteopontin (Figs. 5C–D, G–H). Chondrocytesin Ihh−/− synchondroses were not arranged in characteristicgrowth plates (Figs. 5I, M). The presumptive posterior

sphenoidal ossification center was ill defined and wasactually occupied by collagen X-expressing chondrocytes atE15.5 (ps, Figs. 5J–L, arrowhead) and by collagen X-, alizarinred- and osteopontin-positive hypertrophic mineralizing chon-drocytes by E17.5 (Figs. 5N–P, arrowhead). Mineralizingchondrocytes were also scattered throughout much of theremaining cartilaginous tissue (Figs. 5N–P). When weexamined the basi-occipital ossification center, we weresurprised to find that it appeared quite normal in mutantspecimens in terms of histology, collagen X and osteopontingene expression and alizarin red staining (b-oc, Figs. 5I–L) and

Fig. 4. Expression of Ihh receptors and related factors is abnormal in Ihh−/− synchondroses. (A–L) In E15.5 embryos, the wild-type intrasphenoidal (is) and spheno-occipital (so) synchondroses (A, G, respectively) exhibit: Ihh gene expression in the pre-hypertrophic zone (arrowheads, B, H); Patched expression in proliferatingzone (arrowheads, C, I); Smoothened and PTHrP expression throughout the reserve/proliferative area (bracket, D–E, J–K); and PTHrP-R expression in pre-hypertrophic and hypertrophic zones (arrowheads, F, L). In contrast, the Ihh−/− synchondroses have: abnormal growth plates (M, S); barely detectable expression ofPatched (O, U) and PTHrP (Q, W); fairly normal Smoothened expression (P, V); and widespread PTHrP-R expression (R, X). Scale bar: (A–X) 150 μm.


was in fact virtually indistinguishable from that in wild-typespecimens (b-oc, Figs. 5A–D).

Abnormal distribution of hypertrophic chondrocytes anddelayed expression of terminal gene markers in Ihh−/−

synchondroses

In order to permit and sustain a seamless transition frommature mineralized cartilage to endochondral bone, hyper-trophic chondrocytes normally abut the perichondrial tissue thatis the source of bone and vascular progenitor cells. Closerinspection of the synchondroses shown above revealed,however, that there were obvious differences in the topogra-phical distribution of hypertrophic chondrocytes in wild-typeversus mutant cartilage. The wild-type collagen X-, osteopon-tin- and mineralizing hypertrophic chondrocytes occupied theentire zone, and the more peripheral among them were in closecontact with neighboring perichondrial tissues (Figs. 6A–D,arrows); in contrast, the Ihh−/− hypertrophic chondrocytestended to occupy a more central location throughout thesynchondroses (Figs. 6G–J) and were actually flanked by less

mature collagen II-expressing chondrocytes facing and abuttingthe perichondrial tissues (Fig. 6L, arrowheads) absent in wild-type tissue (Fig. 6F, arrowheads). These topographical andphenotypic differences were re-affirmed by the expressionpatterns of Osterix (Nakashima et al., 2002) whose transcriptswere present throughout wild-type hypertrophic zone andadjacent perichondrial tissues and bone (Fig. 6E) but wereseen only in the centrally located chondrocytes in mutantspecimens (Fig. 6K).

In addition to an appropriate distribution, hypertrophicchondrocytes need to express genes particularly important forthe cartilage-to-bone transition, including metalloproteasefamily members and vascular endothelial growth factor(VEGF) (Zelzer et al., 2002; Engsig et al., 2000). In situhybridization showed that wild-type hypertrophic chondrocytesexpressed significant levels of MMP-9 and VEGF throughoutthe late hypertrophic zone facing endochondral bone andneighboring perichondrial tissues (Figs. 7A–F, arrowheads).Instead, the distribution of MMP-9- and VEGF-expressinghypertrophic chondrocytes was clearly altered in mutantsynchondroses and particularly so in the intrasphenoidal

Fig. 5. Derangement of chondrocyte maturation and ossification patterns in Ihh−/− synchondroses. E15.5 and E17.5 wild-type synchondroses (A, E) contain maturehypertrophic chondrocytes exhibiting expression of collagen X (Col X) and osteopontin (OP) (arrows, B–C, F–G) and flanking the posterior sphenoidal (ps) and basi-occipital (b-oc) ossification centers that are positive with alizarin red (AR) (D, H). In mutant E15.5 and E17.5 synchondroses, the growth plates are mal-organized (I,M), and the posterior sphenoidal ossification center is occupied by collagen X-positive and osteopontin/mineral-negative chondrocytes at E15.5 (arrowhead, J–L) andby collagen X-, osteopontin- and mineral-positive hypertrophic chondrocytes at E17.5 (arrowhead, N–P). Note, however, that the basi-occipital ossification center inthe mutants appears quite normal at both stages (I–P). Scale bars: (A–D and I–L) 250 μm; (E–H and M–P) 150 μm.


synchondrosis (Figs. 7G–L, arrowheads) and the number ofthese cells expressing VEGF actually seemed higher.

Sonic hedgehog expression

We were very intrigued by our findings above that theoccipital primary ossification center appeared to be able to formand exhibit a fairly normal phenotype in the Ihh−/− cranial bases(Fig. 5) and that mutant hypertrophic chondrocytes in thespheno-occipital synchondroses seemed less affected phenoty-pically than those in the intrasphenoidal synchondrosis (Fig. 7).How could these observations be explained? There were two

obvious possibilities: (a) the occipital portion of the cranial basecould be less dependent on Ihh signaling than more anteriorregions; or (b) the occipital portion may express anotherhedgehog protein such as Sonic hedgehog (Shh) or be adjacentto a hedgehog-expressing tissue (Eberhard et al., 2006; Nie etal., 2005). To test these possibilities, we performed in situhybridization on serial sections of E13.5 cranial base using Ihhand Shh riboprobes. Indeed, we found that Shh was stronglyexpressed by notochord remnants adjacent to incipient spheno-occipital synchondrosis both in wild-type (Figs. 8A–B) andIhh−/− specimens (Figs. 8F–G), in good correlation withprevious findings (Nie et al., 2005). No Shh transcripts were

Fig. 6. Topographical distribution of hypertrophic chondrocytes is abnormal inIhh−/− synchondroses. (A–F) In wild-type E17.5 intrasphenoidal synchondrosis,the hypertrophic chondrocytes characterized by expression of collagen X (ColX), osteopontin (OP), and osterix (Osx) and by mineralization (AR) (B–E) forma well defined growth plate zone and abut the perichondrial tissues (arrows,B–D). As expected, the cells lack collagen II transcripts (Col II) (arrowheads, F)that are instead typical of resting and proliferating zones (F, left side). (G–L) InIhh−/− synchondroses, the mature collagen X-, osteopontin-mineral- andOsterix-positive chondrocytes occupy the central portion of the growth plate(G–K) and are flanked by immature collagen II-expressing chondrocytes facingand abutting the perichondrial tissues (arrowheads, L). Scale bar: (A–L) 60 μm.


seen near the more anterior intrasphenoidal synchondrosis(Figs. 8B, G). A higher magnification view showed moreclearly that the Shh-expressing tissue was located very close tothe developing synchondrosis (Figs. 8C–D and H–I). Indeed,Patched was expressed not only near the Shh-expressing tissue,but also in the underlying chondrocytes in both wild-type andIhh−/− specimens (Figs. 8E, J, asterisk).

Discussion

Our data demonstrate that Ihh is required for synchondrosisorganization and function and for cranial base elongation andoverall configuration. The Ihh−/− synchondrosis growth plates

are disorganized, chondrocyte proliferation is markedlydepressed, and chondrocyte maturation is initially delayedbut then becomes widespread, as exemplified by the presenceof collagen X-positive chondrocytes occupying the prospec-tive posterior sphenoidal ossification center site at E15.5.These growth plate defects are as serious as, and reminiscentof, those seen in Ihh−/− long bone anlagen (St-Jacques et al.,1999), indicating that Ihh is a universal regulator of growthplate function, regardless of the skeletal structure in which it isexpressed and operates. Synchondroses, however, differ fromdeveloping long bones in one crucial respect, namelysynchondrosis growth plates are not overlaid by a periarti-cular/articular layer. This layer is important for long bonedevelopment where it regulates chondrocyte proliferation andmaturation rates via its Ihh-dependent production of PTHrP(Kronenberg, 2003). In the synchondroses then, the Ihh-PTHrP relay could instead function within the growth plateitself and specifically between Ihh-producing pre-hypertrophicchondrocytes and PTHrP-expressing reserve/proliferatingchondrocytes (Fig. 4). Notably, since the growth plates indeveloping long bones eventually become separated from thearticular layer by a secondary ossification center, it followsthat even the Ihh-PTHrP loop operating in long bones wouldeventually have to wean itself of periarticular influences andinterdependencies as previously suggested (van der Eerdenet al., 2000).

An unexpected finding in our study is that Ihh−/−

synchondroses stain much less with Alcian blue and containmarkedly lower amounts of aggrecan transcripts. Though Ihhregulates multiple aspects of growth plate biology, it had notbeen linked previously to the regulation of aggrecan geneexpression and matrix accumulation. Aggrecan gene expressionis normally down-regulated in the hypertrophic zone of growthplates as our data (Fig. 3) and previous studies on long bonesdemonstrate (Sandell et al., 1994). Thus, aggrecan down-regulation in Ihh−/− synchondroses could reflect widespreadchondrocyte maturation. This is an unlikely possibility since thedecrease in gene expression characterizes even E15.5 synchon-droses when chondrocyte maturation has not become wide-spread yet. An alternative possibility suggested by our in vitrodata is that the Ihh−/− chondrocytes lack, or fail to receive,factors needed to keep aggrecan gene expression at physiologiclevels. We find that, while exogenous Ihh by itself does notsignificantly boost proteoglycan levels in wild-type and Ihh−/−

chondrocyte cultures, it markedly enhances the responses of thecells to exogenous BMP-2. Ihh thus appears to sensitize thecells to the beneficial effects of BMP signaling. There isevidence that BMPs act as positive enhancers of thechondrocyte phenotype and genetic interference with BMPaction is incompatible with normal phenotype and function andthat hedgehog proteins are positive regulators of BMP geneexpression (Grimsrud et al., 2001; Minina et al., 2001;Enomoto-Iwamoto et al., 1998; Zou et al., 1997). In summary,it may be that, in the absence of Ihh, the growth plates containsuboptimal amounts/types of BMPs or abnormal amounts ofrelated factors, resulting in substandard matrix gene expressionand reduced matrix content. Given the fundamental role of

Fig. 7. Terminal markers of chondrocyte maturation and hypertrophy are expressed suboptimally in Ihh−/− synchondroses. (A–F) In E17.5 wild-type intrasphenoidaland spheno-occipital synchondroses, MMP-9 and VEGF transcripts are abundant and characterize hypertrophic and post-hypertrophic chondrocytes and ossification-associated cells (arrowheads, B–C, E–F). (G–L) In mutant E17.5 specimens. both genes are expressed in an altered pattern (arrowheads, H–I, K–L). Note that thesedeficiencies appear to be more severe in intrasphenoidal (G–I) than spheno-occipital (J–L) synchondrosis. Scale bar: (A–L) 75 μm.


aggrecan in cartilage biology, a depressed aggrecan geneexpression could conceivably have serious and broad con-sequences on cartilage structure and physiology. In addition,cartilage matrix is not only composed of aggrecan, but alsocontains several other proteoglycans (in addition to severalcollagens and other macromolecules). It is plausible then thatIhh−/− synchondrosis growth plates may lack, or expressabnormal levels of, other proteoglycans as well. In a recentstudy on developing Ihh−/− long bones, we showed that mutantgrowth plates lack syndecan-4 transcripts and display a severelyabnormal distribution of syndecan-3 transcripts, while bothtranscripts are largely restricted to the proliferative zone incompanion wild-type specimens (Pacifici et al., 2005). Ourprevious work with chick embryos suggested that syndecans,and syndecan-3 in particular, have important roles in restrictingand regulating Ihh action in growth plate and proliferative zone(Shimo et al., 2004; Shimazu et al., 1996). Lastly, perlecan-nullmice have markedly abnormal growth plates with markedlyabnormal Ihh gene expression patterns (Arikawa-Hirasawaet al., 1999). Together, the above data and considerations lead tothe conclusion that Ihh−/− growth plates, regardless of theiranatomical location, are characterized by deficient and/orabnormal expression of cell surface- and matrix-associatedproteoglycans, and these aberrations could have important rolesin establishing phenotypic abnormalities in both Ihh−/− longbones and cranial base.

One intriguing finding in the present study is thetopographical distribution of hypertrophic chondrocyteswithin the Ihh−/− synchondroses. Whereas the wild-typecells form a characteristic narrow zone that is well defined tri-dimensionally and abuts the perichondrial tissues, the mutantcells are apparently unable to form a well-defined zone andoccupy a central linear region of the tissue. Indeed, they aresurrounded by non-hypertrophic and collagen II-expressingchondrocytes that take their place in facing and confrontingthe perichondrial tissues. An abnormal distribution ofhypertrophic chondrocytes was recently reported in Ihh−/−

long bone anlagen also (Colnot et al., 2005). The data suggestthat Ihh directly or indirectly dictates the topographicaldistribution of chondrocytes within developing skeletalelements and enables the cells to organize themselves intozones of maturation. What may be the nature of thesenavigational devices and why are mutant synchondrosischondrocytes located centrally? We do not know whetherthe former may involve action by primary cilia. These cellsurface protrusions are present in most cells (Wheatley, 1995),have been suggested to have roles in chondrocyte orientationor polarity within cartilage (Olsen et al., 2005; Poole et al.,1985), and have recently been found to also have a criticalrole in hedgehog signaling (Huangfu and Anderson, 2005;Huangfu et al., 2003). Thus, without Ihh, mutant chondro-cytes would lack both cilia-dependent hedgehog signaling andnavigational functions. With regard to the central location ofmutant hypertrophic chondrocytes, we can surmise that, bybeing a non-random distribution, it must reflect additionalbiological processes. One possibility is that factors emanatefrom perichondrial tissues that are negative regulators ofchondrocyte maturation and hypertrophy. In the absence ofIhh signaling, the only microenvironment suitable forchondrocyte maturation would be the central portion of thegrowth plate. This is not a far-fetched possibility since studieshave pointed to negative influences of perichondrium onchondrocyte maturation and function (Alvarez et al., 2001;Serra et al., 1999), a phenomenon that Holtzer and co-workers termed “interference” nearly four decades ago(Chacko et al., 1969).

One could argue, however, that if these negative perichon-drial influences were to remain unchanged and unchallenged,chondrocytes would never mature in an appropriate manner andendochondral ossification would never occur. Thus, an alter-native explanation, and one we favor, is that perichondrialtissues are not uniform and those abutting the reserve andproliferative zones of growth plate may be structurally andfunctionally different from those abutting early and late

Fig. 8. Shh and Ihh are expressed in the occipital regions of the early cranialbase. (A) In E13.5 wild-type cranial base, there is already obvious geneexpression of Ihh in incipient intrasphenoidal (is) and spheno-occipital (so)synchondroses. (B) Abundant Shh transcripts are present in notochord remnantsnear nascent spheno-occipital synchondrosis, but not near the intrasphenoidalsynchondrosis. As expected, Shh is also expressed in oral ectoderm (oe). (C–E)Higher magnification view of boxed area. (E) Patched transcripts are present inboth chondrocytes (asterisk) and surrounding tissues. (F–J) Identical analysesperformed on cranial base of E13.5 Ihh−/− littermates. Note the persistence ofShh expression in notochord remnants in G–I and expression of Patched inchondrocytes (asterisk) and adjacent tissues in J. Scale bars: (A–B, F–G)150 μm; (C–E, H–J) 50 μm.


hypertrophic zones. In previous studies, we monitored expres-sion of extracellular matrix genes along the perichondrial tissuesand indeed found that the patterns expressed by perichondriumsurrounding the resting/proliferative zones were markedlydifferent from those expressed around the hypertrophic/minera-lizing zones (Koyama et al., 1996b). In studies on retinoidsignaling during chick long bone development, we found thatretinoid-rich perichondrial tissues flanking growth plate matura-tion zones are rich in retinoids that could trigger or enhancechondrocyte hypertrophy (Koyama et al., 1999), a possibilitysuggested also by studies on endogenous retinoid distributionusing transgenic reporter mice (von Schroeder and Heersche,1998). In summary, it is possible that the character and functionof perichondrial tissues change along the growth plate. Ihhwould be crucial in enacting this change; in its absence,perichondrial tissues would remain hostile to chondrocytematuration and hypertrophy, the hypertrophic cells wouldremain located centrally, and the normal transition from

mineralized hypertrophic cartilage to bone would be hamperedif not completely prevented. It was recently shown that Ihh−/−

long bone anlagen contain a poorly defined and thin perichon-drium that lacks a normal complement of blood vesselscompared to wild-type (Colnot et al., 2005). We see a similarsubstandard distribution of blood vessels in Ihh−/− synchon-drosis perichondrium (unpublished observations), strengtheningthe argument that perichondrial tissues in mutant skeletalelements are structurally and functionally deficient and couldcontribute to cartilage and bone defects.

In closing, we need to address our finding that Shh isexpressed by notochord remnants in the vicinity of nascentspheno-occipital synchondroses. This is actually not surprisinggiven that notochord is initially in close proximity withparachordal cartilages that contribute the occipital portion ofthe developing cranial base. Our data show that notochordremnant-associated Shh expression is invariant in both E13.5wild-type and Ihh−/− embryos and that adjacent chondrocytesstrongly express Patched. The data indicate that the spheno-occipital synchondrosis and in particular its most posteriorportion are recipients of Shh signaling and were thus able todevelop and ossify equally well in both wild-type and Ihh−/−

embryos. This is in keeping with our observation that the moreanterior intrasphenoidal synchondrosis was fully defective inIhh−/− mice, likely beyond reach of Shh. Obviously, thesescenarios need further testing but are reasonable and are alsotantalizing for the following reasons. First, primary ossificationcenters along the developing cranial base have long beenknown to form first posteriorly and then anteriorly (Kjaer,1990); Shh could be the reason for this asynchrony. Second,the synchondroses do not behave in unison and one notableand important difference is that they undergo involution andclose at different postnatal ages, suggesting that they are underdistinct mechanisms (Ingervall and Thilander, 1972). Insummary, the cranial base not only has a complex embryonicorigin but also a complex spatio-temporal physiology andappears to represent a skeletal structure in which Shh and Ihhhave roles in pacing and orchestrating development, behavior,and function of its various portions along the anterior–posterior axis.

Acknowledgments

We express our gratitude to the Yamanouchi PharmaceuticalCo. for kindly providing recombinant rhBMP-2. This study wassupported by NIH grants AR46000 and AR47543. B. Youngwas supported by a postdoctoral position on the NIEHS traininggrant T32EESO7282-12.

References

Alvarez, J., Horton, J., Sohn, P., Serra, R., 2001. The perichondrium plays animportant role in mediating the effects of TGF-β1 on endochondral boneformation. Dev. Dyn. 221, 311–321.

Arikawa-Hirasawa, E., Watanabe, H., Takami, H., Hassell, J.R., Yamada, Y.,1999. Perlecan is essential for cartilage and cephalic development. Nat.Genet. 23, 354–358.


Bitgood, M.J., McMahon, A.P., 1995. Hedgehog and Bmp genes arecoexpressed at many diverse sites of cell–cell interaction in the mouseembryo. Dev. Biol. 172, 126–138.

Chacko, S., Holtzer, S., Holtzer, H., 1969. Suppression of chondrogenicexpression in mixtures of normal chondrocytes and BUDR-altered chon-drocytes growing in vitro. Biochem. Biophys. Res. Commun. 34, 183–189.

Chen, L., Adar, R., Yang, X., Monsonego, E.O., Cuiling, L., Hauschka, P.V.,Yayon, A., Deng, C.-X., 1999. Gly369Cys mutation in mouse FGFR3causes achrondroplasia by affecting both chondrogenesis and osteogenesis.J. Clin. Invest. 104, 1517–1525.

Chung, U.-I., Schipani, E., McMahon, A.P., Kronenberg, H.M., 2001. Indianhedgehog couples chondrogenesis to osteogenesis in endochondral bonedevelopment. J. Clin. Invest. 107, 295–304.

Colnot, C., de la Fuente, L., Huang, S., Hu, D., Lu, C., St-Jacques, B., Helms,J.A., 2005. Indian hedgehog synchronizes skeletal angiogenesis andperichondrial maturation with cartilage development. Development 132,1057–1067.

Eberhard, J.K., Swartz, M.E., Crump, J.G., Kimmel, C.B., 2006. Earlyhedgehog signaling from neural to oral epithelium organizes anteriorcraniofacial development. Development 133, 1069–1077.

Engsig,M.T., Chen, Q.-J., Vu, T.H., Pedersen, A.-C., Therkidsen, B., Lund, L.R.,Henriksen, K., Lenhard, T., Foged, N.T., Werb, Z., Delaisse', J.-M., 2000.Matrix metalloproteinase 9 and vascular endothelial growth factor areessential for osteoclast recruitment into developing long bones. J. Cell Biol.151, 879–889.

Enomoto-Iwamoto, M., Iwamoto, M., Mukudai, Y., Kawakami, Y., Nohno, T.,Higuchi, Y., Takemoto, S., Ohuchi, H., Noji, S., Kurisu, K., 1998. Bonemorphogenetic protein signaling is required for maintenance of differen-tiated phenotype, control of proliferation, and hypertrophy in chondrocytes.J. Cell Biol. 140, 409–418.

Enomoto-Iwamoto, M., Nakamura, T., Aikawa, T., Higuchi, Y., Yuasa, T.,Yamaguchi, A., Nohno, T., Noji, S., Matsuya, T., Kurisu, K., Koyama, E.,Pacifici, M., Iwamoto, M., 2000. Hedgehog proteins stimulate chondrogeniccell differentiation and cartilage formation. J. Bone Min. Res. 15,1659–1668.

Gakunka, P.T., Koboki, Y., Opperman, L.A., 2000. Hyaluronan is essential forthe expansion of the cranial base growth plates. J. Craniofac. Genet. Dev.Biol. 20, 53–63.

Grimsrud, C.D., Romano, P.R., D'Souza, M., Puzas, J.E., Schwarz, E.M.,Reynolds, P.R., Rosier, R.N., O'Keefe, R.J., 2001. BMP signaling stimulateschondrocyte maturation and the expression of Indian hedgehog. J. Orthop.Res. 19, 18–25.

Huangfu, D., Anderson, K.V., 2005. Cilia and Hedgehog responsiveness in themouse. Proc. Natl. Acad. Sci. U.S.A. 102, 11325–11330.

Huangfu, D., Liu, A., Rakeman, A.S., Murcia, N.S., Niswander, L., Anderson,K.V., 2003. Hedgehog signaling in the mouse requires intraflagellartransport proteins. Nature 426, 83–87.

Ingervall, B., Thilander, B., 1972. The human spheno-occipital synchondrosis: I.The time of closure appraised macroscopically. Acta Odontol. 30.

Inouye, M., 1976. Differential stain for cartilage and bone in fetal mammalianskeletons. Cong. Anom. 16, 171–173.

Ishii-Suzuki, M., Suda, N., Yamazaki, K., Kuroda, T., Senior, P.V., Beck, F.,Hammond, V.E., 1999. Differential responses to parathyroid hormone-related protein (PTHrP) deficiency in the various craniofacial cartilages.Anat. Rec. 255, 452–457.

Johnson-Wint, B., Wellis, S., 1982. A rapid in situ deoxyribonucleic acid assayfor determining cell number in culture and tissue. Anal. Biochem. 122,338–344.

Kjaer, I., 1990. Radiographic determination of prenatal basicranial ossification.J. Craniofac. Genet. Dev. Biol. 10, 113–123.

Koyama, E., Leatherman, J.L., Noji, S., Pacifici, M., 1996a. Early chick limbcartilaginous elements possess polarizing activity and express Hedgehog-related morphogenetic factors. Dev. Dyn. 207, 344–354.

Koyama, E., Shimazu, A., Leatherman, J.L., Golden, E.B., Nah, H.-D., Pacifici,M., 1996b. Expression of syndecan-3 and tenascin-C: possible involvementin periosteum development. J. Orthop. Res. 14, 403–412.

Koyama, E., Golden, E.B., Kirsch, T., Adams, S.L., Chandraratna, R.A.S.,Michaille, J.-J., Pacifici, M., 1999. Retinoid signaling is required for

chondrocyte maturation and endochondral bone formation during limbskeletogenesis. Dev. Biol. 208, 375–391.

Kronenberg, H.M., 2003. Developmental regulation of the growth plate. Science423, 332–336.

Lanske, B., Karaplis, A.C., Lee, K., Lutz, A., Vortkamp, A., Pirro, A.,Karperien, M., Defize, L.H.K., Ho, C., Mulligan, R., Abou-Samra, A.B.,Juppner, H., Segre, G.V., Kronenberg, H.M., 1996. PTH/PTHrp receptor inearly development and Indian hedgehog-regulated bone growth. Science273, 663–666.

Larsen, W.J., 2001. Development of the head and neck. Human Embryology.Churchill Livingstone, New York, pp. 351–378.

Minina, E., Wenzel, H.M., Kreschel, C., Karp, S., Gaffield, W., McMahon,A.P., Vortkamp, A., 2001. BMP and Ihh/PTHrP signaling interact tocoordinate chondrocyte proliferation and differentiation. Development 128,4523–4534.

Mooney, M.P., Losken, H.W., Tschakaloff, A., Siegel, M.I., Losken, A., Lalikos,J.F., 1993. Congenital bilateral coronal suture synostosis in a rabbit andcraniofacial growth comparisons with experimental models. Cleft PalateCraniofac. J. 30, 121–128.

Nakamura, T., Aikawa, T., Enomoto-Iwamoto, M., Iwamoto, M., Higuchi, Y.,Pacifici, M., Kinto, N., Yamaguchi, A., Noji, S., Kurisu, K., Matsuya, T.,1997. Induction of osteogenic differentiation by hedgehog proteins.Biochem. Biophys. Res. Commun. 237, 465–469.

Nakashima, K., Zhou, X., Kunkel, G., Zhang, Z., Deng, J.M., Behringer, R.R.,de Crombrugghe, B., 2002. The novel zinc finger-containing transcriptionfactor osterix is required for osteoblast differentiation and bone formation.Cell 108, 17–29.

Nie, X., Luukko, K., kvinnsland, I.-H., Kettunen, P., 2005. Developmentallyregulated expression of Shh and Ihh in the developing mouse cranial base:comparison with Sox9 expression. Anat. Rec. 286A, 891–898.

Olsen, B.R., Kolpakova, E., McBratney-Owen, B., Li, X., Zhou, J., Fukai, N.,2005. Genetic and epigenetic determinants of skeletal morphogenesis—Role of cellular polarity and ciliary function in skeletal development andgrowth. Oral Biosci. Med. 213, 57–65.

Pacifici, M., Shimo, T., Gentili, C., Kirsch, T., Freeman, T.A., Enomoto-Iwamoto, M., Iwamoto, M., Koyama, E., 2005. Syndecan-3: a cellsurface heparan sulfate proteoglycan important for chondrocyteproliferation and function during limb skeletogenesis. J. Bone Min.Metab. 23, 191–199.

Poole, C.A., Flint, M.H., Beaumont, B.W., 1985. Analysis of the morphologyand function of primary cilia in connective tissues: a cellular cyberneticprobe? Cell Motil. 5, 175–193.

Rice, D.P.C., Rice, R., Thesleff, I., 2003. Fgfr mRNA isoforms in craniofacialdevelopment. Bone 33, 14–27.

Sandell, L.J., Sugai, J.V., Trippel, S.B., 1994. Expression of collagens I, II, Xand XI and aggrecan mRNAs by bovine growth plate chondrocytes in situ.J. Orthop. Res. 12, 1–14.

Serra, R., Karaplis, A.C., Sohn, P., 1999. Parathyroid hormone-related peptide(PTHrP)-dependent and -independent effects of transforming growth factorβ(TGFβ) on endochondral bone formation. J. Cell Biol. 145, 783–794.

Shimazu, A., Nah, H.D., Kirsch, T., Koyama, E., Leatherman, J.L., Golden, E.B.,Kosher, R.A., Pacifici, M., 1996. Syndecan-3 and the control of chondrocyteproliferation during endochondral ossification. Exp. Cell Res. 229, 126–136.

Shimo, T., Gentili, C., Iwamoto, M., Wu, C., Koyama, E., Pacifici, M., 2004.Indian hedgehog and syndecan-3 coregulate chondrocyte proliferation andfunction during chick limb skeletogenesis. Dev. Dyn. 229, 607–617.

Shum, L., Kane, A.A., Nuckolls, G.H., 2003. BMP4 promotes chondrocyteproliferation and hypertrophy in the endochondral cranial base. Int. J. Dev.Biol. 47, 423–431.

Sperber, G.H., 2001. Cranial base. Craniofacial development. B.C. Decker,Hamilton, Canada, pp. 89–101.

St-Jacques, B., Hammerschmidt, M., McMahon, A.P., 1999. Indian hedgehogsignaling regulates proliferation and differentiation of chondrocytes and isessential for bone formation. Genes Dev. 13, 2076–2086.

Thorogood, P., 1988. The developmental specification of the vertebrate skull.Development 103 (Suppl.), 141–153.

van der Eerden, B.C., Karperien, M., Gevers, E.F., Lowik, C.W., Wit, J.M.,2000. Expression of Indian hedgehog, parathyroid hormone-related protein,


and their receptors in the postnatal growth plate of the rat: evidence for alocally acting growth restraining feedback loop after birth. J. Bone Miner.Res. 15, 1045–1055.

von Schroeder, H.P., Heersche, J.N.M., 1998. Retinoic acid responsiveness ofcells and tissues in developing fetal limbs evaluated in a RAREhsplacZtransgenic mouse model. J. Orthop. Res. 16, 355–364.

Vortkamp, A., Lee, K., Lanske, B., Segre, G.V., Kronenberg, H.M., Tabin, C.J.,1996. Regulation of rate of cartilage differentiation by Indian hedgehog andPTH-related protein. Science 273, 613–622.

Wheatley, D.N., 1995. Primary cilia in normal and pathological tissues.Pathobiology 63, 222–238.

Zelzer, E., McLean, W., Ng, Y.S., Fukai, N., Reginato, A.M., Lovejoy, S.,D'Amore, P.A., Olsen, B.R., 2002. Skeletal defects in VEGF(120/120) micereveal multiple roles of VEGf in skeletogenesis. Development 129,1893–1904.

Zou, H., Wieser, R., Massague, J., Niswander, L., 1997. Distinct roles of type Ibone morphogenetic protein receptors in the formation and differentiation ofcartilage. Genes Dev. 11, 2191–2203.

indian and sonic hedgehogs regulate synchondrosis growth plate and cranial base development and...

Documents