Histone modifications induced by activated signalling cascades are crucial to cell-lineage decisions. Osteoblast and adipocyte differentiation from common mesenchymal stem cells is under transcriptional control by numerous factors. Although PPAR-gamma (peroxisome proliferator activated receptor-gamma) has been established as a prime inducer of adipogenesis, cellular signalling factors that determine cell lineage in bone marrow remain generally unknown. Here, we show that the non-canonical Wnt pathway through CaMKII-TAK1-TAB2-NLK transcriptionally represses PPAR-gamma transactivation and induces Runx2 expression, promoting osteoblastogenesis in preference to adipogenesis in bone marrow mesenchymal progenitors. Wnt-5a activates NLK (Nemo-like kinase), which in turn phosphorylates a histone methyltransferase, SETDB1 (SET domain bifurcated 1), leading to the formation of a co-repressor complex that inactivates PPAR-gamma function through histone H3-K9 methylation. These findings suggest that the non-canonical Wnt signalling pathway suppresses PPAR-gamma function through chromatin inactivation triggered by recruitment of a repressing histone methyltransferase, thus leading to an osteoblastic cell lineage from mesenchymal stem cells.

Janus kinases (Jaks) play an important role in signal transduction via cytokine receptors. Tyk2 is a Janus kinase, and we developed tyk2-deficient mice to study the requirement for tyk2 in vivo. Tyk2-deficient mice show no overt developmental abnormalities; however, they display a lack of responsiveness to a small amount of IFNalpha, although a high concentration of IFNalpha can fully transduce its signal even in the absence of tyk2. Furthermore, IL-12-induced T cell function is defective in these mice. In contrast, these mice respond normally to IL-6 and IL-10, both of which activate tyk2 in vitro. These observations demonstrate that tyk2 plays only a restricted role in mediating IFNalpha-dependent signaling while being required in mediating IL-12-dependent biological responses.

Core-binding factor beta (CBFbeta, also called polyomavirus enhancer binding protein 2beta (PEBP2B)) is associated with an inversion of chromosome 16 and is associated with acute myeloid leukemia in humans. CBFbeta forms a heterodimer with RUNX1 (runt-related transcription factor 1), which has a DNA binding domain homologous to the pair-rule protein runt in Drosophila melanogaster. Both RUNX1 and CBFbeta are essential for hematopoiesis. Haploinsufficiency of another runt-related protein, RUNX2 (also called CBFA1), causes cleidocranial dysplasia in humans and is essential in skeletal development by regulating osteoblast differentiation and chondrocyte maturation. Mice deficient in Cbfb (Cbfb(-/-)) die at midgestation, so the function of Cbfbeta in skeletal development has yet to be ascertained. To investigate this issue, we rescued hematopoiesis of Cbfb(-/-) mice by introducing Cbfb using the Gata1 promoter. The rescued Cbfb(-/-) mice recapitulated fetal liver hematopoiesis in erythroid and megakaryocytic lineages and survived until birth, but showed severely delayed bone formation. Although mesenchymal cells differentiated into immature osteoblasts, intramembranous bones were poorly formed. The maturation of chondrocytes into hypertrophic cells was markedly delayed, and no endochondral bones were formed. Electrophoretic mobility shift assays and reporter assays showed that Cbfbeta was necessary for the efficient DNA binding of Runx2 and for Runx2-dependent transcriptional activation. These findings indicate that Cbfbeta is required for the function of Runx2 in skeletal development.

No abstract