Coordinated production and remodeling of the extracellular matrix is essential during development. It is of particular importance for skeletogenesis, as the ability of cartilage and bone to provide structural support is determined by the composition and organization of the extracellular matrix. Connective tissue growth factor (CTGF, CCN2) is a secreted protein containing several domains that mediate interactions with growth factors, integrins and extracellular matrix components. A role for CTGF in extracellular matrix production is suggested by its ability to mediate collagen deposition during wound healing. CTGF also induces neovascularization in vitro, suggesting a role in angiogenesis in vivo. To test whether CTGF is required for extracellular matrix remodeling and/or angiogenesis during development, we examined the pattern of Ctgf expression and generated Ctgf-deficient mice. Ctgf is expressed in a variety of tissues in midgestation embryos, with highest levels in vascular tissues and maturing chondrocytes. We confirmed that CTGF is a crucial regulator of cartilage extracellular matrix remodeling by generating Ctgf^-/- mice. Ctgf deficiency leads to skeletal dysmorphisms as a result of impaired chondrocyte proliferation and extracellular matrix composition within the hypertrophic zone. Decreased expression of specific extracellular matrix components and matrix metalloproteinases suggests that matrix remodeling within the hypertrophic zones in Ctgf mutants is defective. The mutant phenotype also revealed a role for Ctgf in growth plate angiogenesis. Hypertrophic zones of Ctgf mutant growth plates are expanded, and endochondral ossification is impaired. These defects are linked to decreased expression of vascular endothelial growth factor (VEGF) in the hypertrophic zones of Ctgf mutants. These results demonstrate that CTGF is important for cell proliferation and matrix remodeling during chondrogenesis, and is a key regulator coupling extracellular matrix remodeling to angiogenesis at the growth plate.

Previous studies have demonstrated the ability of bone morphogenetic proteins (BMPs) to promote chondrogenic differentiation in vitro . However, the in vivo role of BMP signaling during chondrogenesis has been unclear. We report here that BMP signaling is essential for multiple aspects of early chondrogenesis. Whereas mice deficient in type 1 receptors Bmpr1a or Bmpr1b in cartilage are able to form intact cartilaginous elements, double mutants develop a severe generalized chondrodysplasia. The majority of skeletal elements that form through endochondral ossification are absent, and the ones that form are rudimentary. The few cartilage condensations that form in double mutants are delayed in the prechondrocytic state and never form an organized growth plate. The reduced size of mutant condensations results from increased apoptosis and decreased proliferation. Moreover, the expression of cartilage-specific extracellular matrix proteins is severely reduced in mutant elements. We demonstrate that this defect in chondrocytic differentiation can be attributed to lack of Sox9 , L-Sox5 , and Sox6 expression in precartilaginous condensations in double mutants. In summary, our study demonstrates that BMPR1A and BMPR1B are functionally redundant during early chondrogenesis and that BMP signaling is required for chondrocyte proliferation, survival, and differentiation in vivo .

Bone morphogenetic protein (BMP) signaling is required for endochondral bone formation. However, whether or not the effects of BMPs are mediated via canonical Smad pathways or through noncanonical pathways is unknown. In this study we have determined the role of receptor Smads 1, 5 and 8 in chondrogenesis. Deletion of individual Smads results in viable and fertile mice. Combined loss of Smads 1, 5 and 8, however, results in severe chondrodysplasia. Smad1/5^CKO (cartilage-specific knockout) mutant mice are nearly identical to Smad1/5^CKO;Smad8^-/- mutants, indicating that Smads 1 and 5 have overlapping functions and are more important than Smad8 in cartilage. The Smad1/5^CKO phenotype is more severe than that of Smad4^CKO mice, challenging the dogma, at least in chondrocytes, that Smad4 is required to mediate Smad signaling through BMP pathways. The chondrodysplasia in Smad1/5^CKO mice is accompanied by imbalances in cross-talk between the BMP, FGF and Ihh/PTHrP pathways. We show that Ihh is a direct target of BMP pathways in chondrocytes, and that FGF exerts antagonistic effects on Ihh expression. Finally, we tested whether FGF exerts its antagonistic effects directly through Smad linker phosphorylation. The results support the alternative conclusion that the effects of FGFs on BMP signaling are indirect in vivo.

Mice carrying a targeted disruption of BmprIB were generated by homologous recombination in embryonic stem cells. BmprIB−/− mice are viable and, in spite of the widespread expression of BMPRIB throughout the developing skeleton, exhibit defects that are largely restricted to the appendicular skeleton. Using molecular markers, we show that the initial formation of the digital rays occurs normally in null mutants, but proliferation of prechondrogenic cells and chondrocyte differentiation in the phalangeal region are markedly reduced. Our results suggest that BMPRIB-mediated signaling is required for cell proliferation after commitment to the chondrogenic lineage. Analyses of BmprIB and Gdf5 single mutants, as well as BmprIB; Gdf5 double mutants suggests that GDF5 is a ligand for BMPRIB in vivo. BmprIB; Bmp7 double mutants were constructed in order to examine whether BMPRIB has overlapping functions with other type I BMP receptors. BmprIB; Bmp7 double mutants exhibit severe appendicular skeletal defects, suggesting that BMPRIB and BMP7 act in distinct, but overlapping pathways. These results also demonstrate that in the absence of BMPRIB, BMP7 plays an essential role in appendicular skeletal development. Therefore, rather than having a unique role, BMPRIB has broadly overlapping functions with other BMP receptors during skeletal development.