| Abstract: | Colony-stimulating factor-1 (CSF-1) released by stromal cells in the bone microenvironment is essential for the proliferation of osteoclast progenitors. In op/op mutant mice, a thymidine insertion in the coding sequence of the CSF-1 gene results in CSF-1 deficiency that in turn leads to decreased osteoclast production and osteopetrosis. Because the osteopetrotic defect is due to the failure of stromal cells to produce CSF-1, we determined if retroviral-mediated gene transfer of the wild-type CSF-1 cDNA into op/op stromal cells would restore their ability to support osteoclast formation in vitro. A retroviral vector, L-CSF-1-SN, was constructed by inserting 1,867 bp of the wild-type CSF-1 cDNA into pLXSN. After transduction with L-CSF-1-SN or LXSN constructs, a stable PA317 packaging cell line that produced a high viral titre was isolated. Viral supernatant from this line was used to infect op/op bone marrow stromal cells. Stable L-CSF-1-SN op/op stromal clones overexpressed CSF-1 mRNA and released CSF-1 into conditioned medium, compared with no CSF-1 released by LXSN op/op stroma. The amount of CSF-1 produced by two clones was similar to the physiologic level released by normal littermate stroma. Southern blot analysis confirmed the presence of intact proviral sequences in transduced cells. In coculture assays, L-CSF-1-SN, but not LXSN, op/op stromal cells supported the formation of TRAP-positive multinucleated cells in the absence of exogenous CSF-1. These findings indicate that genetically engineered stromal cells may be used to improve defective osteoclastogenesis and suggest that targeting stromal cells to bone is a potentially useful therapeutic modality for treating bone disorders. J. Cell. Physiol. 176:323–331, 1998. © 1998 Wiley-Liss, Inc. |
