Tri bone system

Mice were sacri ced and their hind limbs were removed, xed in 70% ethanol, and analyzed for bone histomorphometry. Microstructural changes in femoral bone were evaluated using a micro-computed tomography (micro-CT) system (CosmoScan GX; Rigaku corporation, Tokyo, Japan). The distal and diaphyseal femur were scanned at 90kV, 160μA, and a voxel size of 20 × 20 × 20μm. Three-dimensional reconstruction using the TRI-BONE system (Ratoc system Engineering, Co., Ltd., Tokyo, Japan) was performed. For bone morphometric analysis we evaluated tissue mineral density of cortical bone (TMD, mg/cm 3 ), cortical thickness (Ct.Th, μm), cortical bone area (Ct.Ar, mm 2 ), and BMC (mg) as parameters.
TMD was evaluated in cortical bone of the metaphyseal femur, and Ct.Th, Ct.Ar and BMC were evaluated in cortical bone of the diaphyseal femur. The metaphysis was measured from 200μm below the growth plate and to 4000μm distal, and the diaphyseal was 500μm wide.
Histopathological and uorescent immunohistochemical analysis

Bone morphology and microarchitecture were assessed by high-resolution micro-computed tomography (µCT) (inspeXio SMX-90CT; Shimadzu, Tokyo, Japan) as previously described (Harada et al. 2011) . The proximal tibia and diaphysial tibia were scanned at an X-ray energy of 70 keV, with an integration time of 0.12 s and voxel size of 0.025 mm/pix. Threedimensional (3D) reconstruction of mineralized tissue was performed using the TRI-BONE system (Ratoc System Engineering, Co., Ltd., Tokyo, Japan).
For the trabecular bone region of the proximal tibia, we determined trabecular bone volume (BV/TV, %), trabecular thickness (Tb.Th, μm), trabecular number (Tb.N, 1/mm), trabecular separation (Tb.Sp, mm), and structure model index (SMI) (Hildebrand and Rüegsegger 1997) . For the cortical bone region in the tibial diaphysis, we assessed medullary volume/total bone volume (MV/TV, %), cortical bone volume/total bone volume (CV/TV), and cortical thickness (Ct.Th, μm).