Skeletal unloading induced by hindlimb suspension in rats reduces bone formation and induces osteopenia, but its effect on adipogenesis is unknown. We assessed the effects of unloading and transforming growth factor (TGF) ␤2 on bone marrow stromal cell adipocyte differentiation in relation with osteoblast differentiation. Skeletal unloading rapidly (4 -7 days) decreased osteoblast transcription factor Runx2, osteocalcin (OC), and type I collagen messenger RNA (mRNA) levels and reduced bone formation in the long bone metaphysis. Conversely, unloading increased expression of the adipocyte transcription factor peroxisome proliferatoractivated receptor ␥2 (PPAR␥2) at 4 days and increased expression of the adipocyte differentiation genes lipoprotein lipase (LPL) and aP2 in the bone marrow stroma at 7 days. Consistently, unloading increased the number and volume of adipocytes in the bone marrow stroma. Continuous (0 -7 days) and late (4 -7 days) treatments with TGF-␤2 corrected the abnormal expression of Cbfa1/Runx2, OC, and type I collagen mRNAs and normalized bone formation in unloaded metaphyseal bone. Moreover, both TGF-␤2 treatments decreased PPAR␥2 and C/EBP␣ mRNA levels at 4 days and normalized aP2 and LPL expression and adipocyte number and volume at 7 days. These results show that skeletal unloading increases adipocyte differentiation concomitantly with inhibition of osteoblast differentiation. These abnormalities are prevented and reversed by TGF-␤2, suggesting a role for TGF-␤ in the control of adipogenic differentiation in the bone marrow stroma. (J Bone Miner Res 2002;17:668 -677)

The effects of skeletal unloading on osteoblastic cells were evaluated in tail-suspended rats. Hindlimb elevation for 14 days induced osteopenia, decreased histomorphometric indexes of bone formation in tibial metaphysis, and reduced plasma osteocalcin and alkaline phosphatase (ALP) levels compared with controls. The in vitro proliferation of osteoblastic cells isolated from the endosteal bone surface of suspended tibias was decreased by 42 and 31% at 2 and 4 days of culture, respectively, compared with controls, as shown by [3H]thymidine labeling and cell number. The proliferation of ALP-positive marrow stromal cells was also decreased by 20-24% at 1 and 2 days of culture. However, ALP activity in bone-derived cells and marrow stromal cells was not different in unloaded and control rats, and the number of bone cells synthesizing osteocalcin, osteonectin, and type I or type III collagen was identical in the two groups. The results indicate that the inhibition of bone formation induced by skeletal unloading is related to a decreased proliferation of putative osteoblast precursor cells present along the endosteal bone surface and in the marrow stroma.

Studies on calcium nutrition in appropriate large animal models can be directly relevant to humans. We have examined the effect of dietary Ca deficiency on various bone and bone-related variables, including plasma markers, histomorphometry, mineral content and breaking strength in pigs. Three groups of eight 38-d-old female pigs were fed adequate (0.9%; control), low (0.4%; LCa) or very low (0.1%; VLCa) Ca diets for 32 d. Plasma Ca significantly decreased over time only in the VLCa-deficient pigs. The concentrations of the parathyroid hormones (PTH) and calcitriol increased as Ca deficiency developed, and the plasma PTH and calcitriol levels varied inversely with dietary Ca. The total bone ash contents, bending moments, trabecular bone volume and the mineral apposition rate all decreased as the calcium intake decreased. The osteoclast surface areas were greater than those of controls in both Ca-deficient groups, whereas the osteoblast surface areas were greater only in the VLCa group. The plasma osteoblast-related markers (alkaline phosphatase, carboxy-terminal propeptide of type I procollagen and osteocalcin) were either greater or unaffected in the Ca-deficient pigs. The results indicate that deficient bone mineralization combined with an increased bone resorption led to bone loss and fragility. The differences in the changes in bone cells (number and activity) between LCa and VLCa groups might be due to differences (time and extent) of circulating PTH and calcitriol. The defective mineralization in both Ca-depleted groups resulted mainly from the lack of Ca because their osteoblast activity was either maintained or stimulated. The results also underline the progressive sensitivity of pigs to Ca supply and the usefulness of this model.

No abstract