Figure 1.
Mechanical load represses transactivation by Smad2/3 in tibial osteocytes.
Images of luciferase activity in three individual SBE-luciferase mice 5h after in vivo loading reveals less Smad2/3-mediated transactivation in the loaded tibiae compared to nonloaded tibiae. Bioluminescence imaging and quantitation (A, B) show consistent reductions in luminescence in the loaded tibiae as opposed to the nonloaded tibiae. Radiance measurements from the nonloaded and loaded limb of the same animal are denoted by a line connecting the dots; black bars indicate the average radiance from nonloaded and loaded tibiae of 6 mice. Immunostaining for luciferase (C) in sections of tibial cortical bone and quantitation of luciferase-expressing osteocytes (D) reveal a decrease in luciferase-positive osteocytes in the loaded tibiae compared to the nonloaded control. UMR-106 whole cell lysates (E, upper panel) and tibial cortical bone lysates (E, lower panel) were evaluated for the level of phosphorylated Smad3, total Smad3, and β-actin by Western analysis. UMR-106 cells were harvested 2 h after treatment with vehicle (DMSO), TGFβ1 (5 ng/ml), or an inhibitor of the TGFβ type 1 receptor, Alk5 (TβRI-I, SB431542). Tibiae were harvested 3h after loading (n = 4 mice). Loaded tibiae (L) were compared to the nonloaded tibiae (NL) from the same mouse. (* p<0.05).
Figure 2.
TGFβ signaling is required for load-induced bone formation.
Micro-computed tomography images of loaded and nonloaded WT and DNTβRII tibiae (A) show that loading increases cortical bone thickness of WT bone more than DNTβRII bone. The overall fluorochrome intake was reduced in the DNTβRII mice despite a slightly increased basal mineral apposition rate in DNTβRII mice relative to WT (B). Results of dynamic bone histomorphometry are consistent with micro-CT, showing that the relative load-mediated increase in bone mineral apposition rate is significantly lower in DNTβRII tibiae than in WT (C) (* p<0.05).
Table 1.
Quantitative measures of micro-computed tomography analyses.
Figure 3.
Load-sensitive Sclerostin regulation requires TGFβ signaling.
Immunohistochemistry shows reduced Sclerostin expression between the loaded and nonloaded tibiae of WT mice, but no difference in Sclerostin expression in tibiae of DNTβRII mice (A). Control IgG used in the primary step of immunohistochemistry verifies the specificity of the SOST staining. Quantitation of Sclerostin-positive osteocytes in the tibiae confirm an 8% reduction in Sclerostin expression in loaded WT, but not in DNTβRII tibiae (B). (* p<0.05).
Figure 4.
TGFβ induces SOST expression through Smad3.
Treating UMR-106 cells with TGFβ1 (5 ng/ml) for 2, 8, or 24 h results in an increase in SOST mRNA, while inhibiting TGFβ signaling with SB431542 results in a decrease in SOST expression (A). Smad3 overexpression with pRK5-Smad3 induces the SOST promoter-reporter construct, 3XECR-hSOSTpLuc, in a TGFβ dose-dependent manner within 24 h of TGFβ treatment (B). Blocking translation with cycloheximide (CHX) or transcription with actinomycin-D (ActD) for 2 h prevents TGFβ (5 ng/ml) induction of SOST expression (C). siRNA mediated knockdown of Runx2 did not prevent TGFβ (5 ng/ml) induction of SOST expression (D). (For panel C, * represents p<0.05 computed by comparing samples with added TGFβ to samples without added TGFβ in each treatment group; for all other panels, * p<0.05 computed by comparing samples to untreated cells).
Figure 5.
The role of TGFβ signaling in load-induced bone formation.
Load represses TGFβ signaling through Smad2/3 in osteocytes, which is required for the activation of SOST expression through an indirect, Runx2-independent mechanism. Loss of function mutations in the TGFβ type II receptor impair load-mediated repression of SOST and new bone formation.