Gfi1 expressed in bone marrow stromal cells is a novel osteoblast suppressor in patients with multiple myeloma bone disease.

Protracted inhibition of osteoblast (OB) differentiation characterizes multiple myeloma (MM) bone disease and persists even when patients are in long-term remission. However, the underlying pathophysiology for this prolonged OB suppression is unknown. Therefore, we developed a mouse MM model in which the bone marrow stromal cells (BMSCs) remained unresponsive to OB differentiation signals after removal of MM cells. We found that BMSCs from both MM-bearing mice and MM patients had increased levels of the transcriptional repressor Gfi1 compared with controls and that Gfi1 was a novel transcriptional repressor of the critical OB transcription factor Runx2. Trichostatin-A blocked the effects of Gfi1, suggesting that it induces epigenetic changes in the Runx2 promoter. MM-BMSC cell-cell contact was not required for MM cells to increase Gfi1 and repress Runx2 levels in MC-4 before OBs or naive primary BMSCs, and Gfi1 induction was blocked by anti-TNF-α and anti-IL-7 antibodies. Importantly, BMSCs isolated from Gfi1(-/-) mice were significantly resistant to MM-induced OB suppression. Strikingly, siRNA knockdown of Gfi1 in BMSCs from MM patients significantly restored expression of Runx2 and OB differentiation markers. Thus, Gfi1 may have an important role in prolonged MM-induced OB suppression and provide a new therapeutic target for MM bone disease.

Osteoclasts are important for bone angiogenesis.

Increased osteoclastogenesis and angiogenesis occur in physiologic and pathologic conditions. However, it is unclear if or how these processes are linked. To test the hypothesis that osteoclasts stimulate angiogenesis, we modulated osteoclast formation in fetal mouse metatarsal explants or in adult mice and determined the effect on angiogenesis. Suppression of osteoclast formation with osteoprotegerin dose-dependently inhibited angiogenesis and osteoclastogenesis in metatarsal explants. Conversely, treatment with parathyroid hormone related protein (PTHrP) increased explant angiogenesis, which was completely blocked by osteoprotegerin. Further, treatment of mice with receptor activator of nuclear factor-kappaB ligand (RANKL) or PTHrP in vivo increased calvarial vessel density and osteoclast number. We next determined whether matrix metalloproteinase-9 (MMP-9), an angiogenic factor predominantly produced by osteoclasts in bone, was important for osteoclast-stimulated angiogenesis. The pro-angiogenic effects of PTHrP or RANKL were absent in metatarsal explants or calvaria in vivo, respectively, from Mmp9(-/-) mice, demonstrating the importance of MMP-9 for osteoclast-stimulated angiogenesis. Lack of MMP-9 decreased osteoclast numbers and abrogated angiogenesis in response to PTHrP or RANKL in explants and in vivo but did not decrease osteoclast differentiation in vitro. Thus, MMP-9 modulates osteoclast-stimulated angiogenesis primarily by affecting osteoclasts, most probably by previously reported migratory effects on osteoclasts. These results clearly demonstrate that osteoclasts stimulate angiogenesis in vivo through MMP-9.

Fetal liver cells transplanted in utero rescue the osteopetrotic phenotype in the oc/oc mouse.

Autosomal recessive osteopetrosis (ARO) is a group of genetic disorders that involve defects that preclude the normal function of osteoclasts, which differentiate from hematopoietic precursors. In half of human cases, ARO is the result of mutations in the TCIRG1 gene, which codes for a subunit of the vacuolar proton pump that plays a fundamental role in the acidification of the cell-bone interface. Functional mutations of this pump severely impair the resorption of bone mineral. Although postnatal hematopoietic stem cell transplantation can partially rescue the hematological phenotype of ARO, other stigmata of the disease, such as secondary neurological and growth defects, are not reversed. For this reason, ARO is a paradigm for genetic diseases that would benefit from effective prenatal treatment. Using the oc/oc mutant mouse, a murine model whose osteopetrotic phenotype closely recapitulates human TCIRG1-dependent ARO, we report that in utero transplantation of adult bone marrow hematopoietic stem cells can correct the ARO phenotype in a limited number of mice. Here we report that in utero injection of allogeneic fetal liver cells, which include hematopoietic stem cells, into oc/oc mouse fetuses at 13.5 days post coitum produces a high level of engraftment, and the oc/oc phenotype is completely rescued in a high percentage of these mice. Therefore, oc/oc pathology appears to be particularly sensitive to this form of early treatment of the ARO genetic disorder.