Co-authors Andy McMahon (left) and Stephen Rodda
The mammalian skeleton is important for many physiological functions; as a scaffold, it supports the body, enabling motion and mechanical loading, and it provides protection for the internal organs. The bones also provide a niche for postnatal hematopoiesis and are the central store for calcium. As such, any modulation in skeletal strength or integrity, such as that experienced through injury (e.g., fracture) or disease state (e.g., osteoporosis) can severely compromise the quality of life and survival of all vertebrates.
While there are several cell lineages contributing to the skeleton during development, each gives rise to a common bone-generating cell type, the osteoblast. Despite the commonality of this cell type to all bone formation, bone can be produced by two distinct mechanisms: direct differentiation from the mesechymal progenitor (occurring in the skull and face, called intramembranous ossification) or the formation of bone on a cartilage scaffold (called endochondral ossification, used by the rest of the skeleton).
Endochondral ossification, the focus of our current study (Rodda and McMahon, Development 133, 3231-3244), is initiated by the condensation of multipotent mesenchymal progenitor cells into structures that anticipate elements of the adult skeleton. Chondrocytes are the first cell type to form, starting out as immature proliferative cells, which initiate the formation of the cartilage matrix, then differentiating to postmitotic hypertrophic chondrocytes. Osteoblast progenitor cells are first evident in the inner layer of perichondral cells, which lie adjacent to the zone of hypertrophic chondrocytes.
Several lines of evidence implicate Hedgehog (Hh) and canonical Wnt signaling in the regulation of endochondral ossification. Indian hedgehog (Ihh) is expressed by prehypertrophic chondrocytes and acts directly on perichondrial mesenchyme to initiate an osteogenic program. Several recent developmental studies have suggested that canonical Wnts play a role in specification of osteoblasts. Specifically, Cre recombinase expressed under the control of promoters that are active broadly during skeletal development have been used to conditionally inactivate b-catenin function, resulting in an overall failure to develop bone. Despite a common result, these studies disagreed with respect to where along the osteoblast lineage canonical Wnt signaling functions and whether it acts directly.
In this study, we used a broad acting cre mouse line, Col2cre, to remove b-catenin function in both the chondrocyte lineage and osteoblast lineage, permitting the detailed characterization of where along the osteoblast lineage these cells arrest. While mutant embryos failed to produce terminally differentiated osteoblasts, they were able to produce osteoblast precursor cells that express Osx1, suggesting a role for canonical Wnt downstream of Osx1. To test this hypothesis we generated a novel Osx1-GFP::Cre mouse strain, where cre expression is restricted to the osteoblast lineage. Using this mouse line to remove b-catenin function resulted in a phenotype consistent with the Col2cre removal of b-catenin function, demonstrating that canonical Wnt functions directly on the osteoblast lineage and downstream of Osx1 expression. Interestingly, cell fate analysis demonstrates that on removal of b-catenin activity, Osx1 expressing osteoblast precursors gives rise to ectopic chondrocytes instead of osteoblasts. This suggests that, along with earlier reports, there is an extended role for canonical Wnt signaling in suppression of an alternate chondrocytic fate within osteoblast precursors. In contrast to loss-of-function studies, use of the Osx1-GFP::Cre line to enhance b-catenin activity rapidly accelerated this program, leading to a dramatic expansion of osteoblast precursors and the premature synthesis of a mineralized bone matrix in the long bones.
Finally, using the Osx1-GFP::Cre line, we were able to conditionally block the ability of osteoblast precursors to respond to a Hh signal, demonstrating that once osteoblast progenitor cells reach an Osx1 state, Hh signaling is dispensable for the rest of terminal differentiation.
Overall, this study has defined the precise states of cell differentiation along the osteoblast lineage that Hh and canonical Wnt function, and importantly identified canonical Wnt as a key mediator in the regulation and onset of mineralized matrix production, both key aspects of bone homeostasis and fracture repair.