Osteoporosis is one of the most common bone pathologies, which are characterized by a decrease in bone mass. It is well established that bone mass, which results from a balanced bone formation and bone resorption, is regulated by many hormonal, environmental and genetic factors. Here we report that the immune semaphorin 4D (Sema4D) is a novel factor controlling bone resorption. Sema4D-deficient primary osteoclasts showed impaired spreading, adhesion, migration and resorption due to altered ß3 integrin sub-unit downstream signaling. In apparent accordance with these in vitro results, Sema4D deletion in sexually mature female mice led to a high bone mass phenotype due to defective bone resorption by osteoclasts. Mutant males, however, displayed normal bone mass and the female osteopetrotic phenotype was only detected at the onset of sexual maturity, indicating that, in vivo, this intrinsic osteoclast defect might be overcome in these mice. Using bone marrow cross transplantation, we confirmed that Sema4D controls bone resorption through an indirect mechanism. In addition, we show that Sema4D −/− mice were less fertile than their WT littermates. A decrease in Gnrh1 hypothalamic expression and a reduced number of ovarian follicles can explain this attenuated fertility. Interestingly, ovariectomy abrogated the bone resorption phenotype in Sema4D −/− mice, providing the evidence that the observed high bone mass phenotype is strictly dependent on ovarian function. Altogether, this study reveals that, in vivo, Sema4D is an indirect regulator of bone resorption, which acts via its effect on reproductive function.

Recent evidence indicates that microglial cells may not derive from blood circulating mature monocytes as they express features of myeloid progenitors. Here, we observed that a subpopulation of microglial cells expressed CD34 and B220 antigens during brain development. We thus hypothesized that microglia, or a subset of microglial cells, originate from blood circulating CD34+/B220+ myeloid progenitors, which could target the brain under developmental or neuroinflammatory conditions. Using experimental allergic encephalomyelitis (EAE) as a model of chronic neuroinflammation, we found that a discrete population of CD34+/B220+ cells expands in both blood and brain of diseased animals. In EAE mice, intravenous transfer experiments showed that macrophage-colony stimulating factor (M-CSF) -expanded CD34+ myeloid progenitors target the inflamed central nervous system (CNS) while keeping their immature phenotype. Based on these results, we then assessed whether CD34+/B220+ cells display in vitro differentiation potential toward microglia. For this purpose, CD34+/B220+ cells were sorted from M-CSF-stimulated bone marrow (BM) cultures and exposed to a glial cell conditioned medium. Under these experimental conditions, CD34+/B220+ cells were able to differentiate into microglial-like cells showing the morphological and phenotypic features of native microglia. Overall, our data suggest that under developmental or neuroinflammatory conditions, a subpopulation of microglial cells derive from CNS-invading CD34+/B220+ myeloid progenitors.