Although many regulators of skeletogenesis have been functionally characterized, one current challenge is to integrate this information into regulatory networks. Here, we discuss how the canonical Wnt and Smad-dependent BMP pathways interact together and play antagonistic or cooperative roles at different steps of osteogenesis, in the context of the developing vertebrate embryo. Early on, BMP signaling specifies multipotent mesenchymal cells into osteochondroprogenitors. In turn, the function of Wnt signaling is to drive these osteochondroprogenitors towards an osteoblastic fate. Subsequently, both pathways promote osteoblast differentiation, albeit with notable mechanistic differences. In osteocytes, the ultimate stage of osteogenic differentiation, the Wnt and BMP pathways exert opposite effects on the control of bone resorption by osteoclasts. We describe how the dynamic molecular wiring of the canonical Wnt and Smad-dependent BMP signaling into the skeletal cell genetic programme is critical for the generation of bone-specific cell types during development.

Osteogenesis is the fundamental process by which bones are formed, maintained and regenerated. The osteoblasts deposit the bone mineralized matrix by secreting large amounts of extracellular proteins and by allowing the biochemical conditions for the nucleation of hydroxyapatite crystals. Normal bone formation requires a tight control of osteoblastic activity, and therefore, osteoblasts represent a major focus of interest in biomedical research. Several crucial features of osteogenesis can be readily recapitulated using murine, avian and fish primary and immortalized osteoblastic cultures. Here, we describe a novel and straightforward in vitro culture of primary osteoblasts from the amphibian Xenopus tropicalis, a major vertebrate model organism. X. tropicalis osteoblasts can readily be extracted from the frontoparietal bone of pre-metamorphosing tadpole skulls by series of gentle protease treatments. Such primary cultures efficiently proliferate and can conveniently be grown at room temperature, in the absence of CO2, on a variety of substrates. X. tropicalis primary osteoblasts express well-characterized genes known to be active during osteogenesis of teleost fish, chick, mouse and human. Upon differentiation, such cultures mineralize and activate DMP1, an osteocyte-specific gene. Importantly, X. tropicalis primary osteoblasts can be efficiently transfected and respond to the forced activation of the bone morphogenetic protein pathway by increasing their nuclear levels of phospho-Smad. Therefore, this novel primary culture is amenable to experimental manipulations and represents a valuable tool for improving our understanding of the complex network of molecular interactions that govern vertebrate bone formation.

The formation of the complex osteocytic network relies on the emission of long cellular processes involved in communication, mechanical strain sensing, and bone turnover control. Newly deposited osteocytic processes rapidly become trapped within the calcifying matrix, and, therefore, they must adopt their definitive conformation and contact their targets in a single morphogenetic event. However, the cellular mechanisms ensuring the robustness of this unique mode of morphogenesis remain unknown. To address this issue, we examined the developing calvaria of the amphibian Xenopus tropicalis by confocal, two-photon, and super-resolution imaging, and described flattened osteocytes lying within a woven bone structured in lamellae of randomly oriented collagen fibers. While most cells emit peripheral and perpendicular processes, we report two osteocytes morphotypes, located at different depth within the bone matrix and exhibiting distinct number and orientation of perpendicular cell processes. We show that this pattern is conserved with the chick Gallus gallus and suggest that the cellular microenvironment, and more particularly cell-cell contact, plays a fundamental role in the induction and stabilization of osteocytic processes. We propose that this intrinsic property might have been evolutionarily selected for its ability to robustly generate self-organizing osteocytic networks harbored by the wide variety of bone shapes and architectures found in extant and extinct vertebrates.