Colony stimulating factor-1 (CSF1) is required for osteoclast progenitor proliferation and differentiation, the latter dependent in part on activation of the transcription factor, Mitf. Recent evidence indicates an important role for CSF1 in states of accelerated bone loss including estrogen-deficiency, inflammatory arthritis and malignancy-induced bone resorption. Alternative splicing of the CSF-1 gene results in production of soluble (sCSF1) and cell-surface (mCSF1) isoforms. We have found that restricted over-expression of either mCSF1 or sCSF1 in osteoblasts causes bone loss in vivo. We then crossed op/op mice with these transgenic mice to create animals with selective over expression of each isoform in bone, in the absence of endogenous CSF1. Interestingly, while mCSF1 completely rescued the osteopetrotic phenotype of the op/p mouse, the bone density in sCSF1-op/op mice did not completely normalize. Surprisingly, following ovariectomy, sCSF-op/op mice lost significantly more bone than the mCSF1-op/op mice, suggesting that sCSF1 plays a more important role in mediating estrogen-deficiency bone loss. Consistent with this finding, isoform-specific real-time PCR documented increased expression of sCSF1 in the bone RNA of ovariectomized mice while levels of mCSF1 did not change. We are now beginning studies in mCSF1 and sCSF1 knock out mice. Preliminary results from mCSF1 knock out mice indicate that they have increased bone mass and reduced number of osteoclasts. To identify CSF-1 target genes in osteoclasts, we have used a gene profiling approach and found 7 genes induced by CSF1 in an Mitf-dependent manner. To pursue these findings and to test the hypothesis that the two CSF1 isoforms serve non-redundant roles in bone, we will determine if isoform-selective deletion of CSF1 alters the skeletal response to resorptive stimuli including ovariectomy and continuous PTH infusion and conversely, if the absence of each isoform differentially affects the anabolic response to PTH. We will determine the isoform of CSF1 preferentially up regulated in osteoblasts following estrogen withdrawal and assay the importance of this change in vivo by ovariectomizing mice with isoform-selective deletion of each CSF1 isoform in osteoblasts. We will determine the effect of sCSF1 and mCSF1 on the extent and time course of induction of 2 the 7 candidate genes identified in the CSFl-""""""""Mitf pathway, one, Jun-dimerization protein-2, is required for osteoclastogenesis and one, pi 10a, is known to have a role in differentiated osteoclast function. We will use these two genes as probes to further test the hypothesis that the two CSF1 isoforms act differently in osteoclasts. We will also use EMSA and mutational promoter analyses to determine the role of Mitf in CSF1 isoform-induced transcription of these two genes. In the aggregate, these studies will clarify the roles of the two CSF1 isoforms in bone as well as begin to define CSF-1 regulated genes in osteoclasts. Since CSF1 is emerging as an important target for drug discovery in treating low bone mass, these data will provide critical information that will inform and refine our approach to this target.
Showing the most recent 10 out of 24 publications
