Fat mass is an important determinant of bone density, but the mechanism of this relationship is uncertain. Leptin, as a circulating peptide of adipocyte origin, is a potential contributor to this relationship. Recently it was shown that intracerebroventricular administration of leptin is associated with bone loss, suggesting that obesity should be associated with low bone mass, the opposite of what is actually found. Since leptin originates in the periphery, an examination of its direct effects on bone is necessary to address this major discrepancy. Leptin (>10(-11) m) increased proliferation of isolated fetal rat osteoblasts comparably with IGF-I, and these cells expressed the signalling form of the leptin receptor. In mouse bone marrow cultures, leptin (>or=10(-11) m) inhibited osteoclastogenesis, but it had no effect on bone resorption in two assays of mature osteoclasts. Systemic administration of leptin to adult male mice (20 injections of 43 micro g/day over 4 weeks) reduced bone fragility (increased work to fracture by 27% and displacement to fracture by 21%, P<0.001). Changes in tibial histomorphometry were not statistically significant apart from an increase in growth plate thickness in animals receiving leptin. Leptin stimulated proliferation of isolated chondrocytes, and these cells also expressed the signalling form of the leptin receptor. It is concluded that the direct bone effects of leptin tend to reduce bone fragility and could contribute to the high bone mass and low fracture rates of obesity. When administered systemically, the direct actions of leptin outweigh its centrally mediated effects on bone, the latter possibly being mediated by leptin's regulation of insulin sensitivity.