Role of diabetes- and obesity-related protein in the regulation of osteoblast differentiation

Abstract

Although thyroid hormone (TH) is known to exert important effects on the skeleton, the nuclear factors constituting the TH receptor coactivator complex and the molecular pathways by which TH mediates its effects on target gene expression in osteoblasts remain poorly understood. A recent study demonstrated that the actions of TH on myoblast differentiation are dependent on diabetes- and obesity-related protein (DOR). However, the role of DOR in osteoblast differentiation is unknown. We found DOR expression increased during in vitro differentiation of bone marrow stromal cells into osteoblasts and also in MC3T3-E1 cells treated with TH. However, DOR expression decreased during cellular proliferation. To determine whether DOR acts as a modulator of TH action during osteoblast differentiation, we examined whether overexpression or knockdown of DOR in MC3T3-E1 cells affects the ability of TH to induce osteoblast differentiation by evaluating alkaline phosphatase (ALP) activity. ALP activity was markedly increased in DOR-overexpressing cells treated with TH. In contrast, loss of DOR dramatically reduced TH stimulation of ALP activity in MC3T3-E1 cells and primary calvaria osteoblasts transduced with lentiviral DOR shRNA. Consistent with reduced ALP activity, mRNA levels of osteocalcin, ALP, and Runx2 were decreased significantly in DOR shRNA cells. In addition, a common single nucleotide polymorphism (SNP), DOR1 found on the promoter of human DOR gene, was associated with circulating osteocalcin levels in nondiabetic subjects. Based on these data, we conclude that DOR plays an important role in TH-mediated osteoblast differentiation, and a DOR SNP associates with plasma osteocalcin in men.

Fig. 1.

Diabetes- and obesity-related protein (DOR) expression during osteoblast proliferation and differentiation. A: DOR expression decreases during osteoblast proliferation. MC3T3-E1 cells were treated with or without fibroblast growth factor (FGF; 1 ng/ml) or 10% calf serum (CS) for 24 h. DOR expression was determined by real-time RT-PCR analysis. DOR expression in vehicle-treated control (ΔC_T = 9.74 ± 0.07). Values are presented as %vehicle-treated control ± SE (n = 3). ^AP < 0.01 vs. vehicle. B: DOR expression increases during bone marrow stromal cell (BMSC) differentiation, as measured by real time RT-PCR. BMSCs were cultured with α-MEM containing 10% CS in the presence or absence of ascorbic acid (100 μg/ml) and β-glycerophosphate (10 mM) for 24 days to induce osteoblast differentiation. Alkaline phosphatase (ALP) gene expression was also assessed to confirm BMSC differentiation. DOR expression in undifferentiated BMSCs (ΔC_T = 11.24 ± 0.27). Values are presented as fold change relative to day 0 ± SE (n = 6). ^AP < 0.001 vs. undifferentiated BMSCs. C: thyroid hormone (TH) stimulates DOR gene expression. MC3T3-E1 cells were incubated in the presence or absence of 3,3′,5-triiodothyronine (T₃; 10 ng/ml) or l-thyroxine (T₄; 100 ng/ml) for 6 or 24 h. DOR expression was determined by real-time RT-PCR analysis. Values are presented as fold change from corresponding vehicle control ± SE (n = 3–5). Data with different letters indicate significant difference at P < 0.05. Treatment groups with same letters are not statistically different. D and E: TH induction of DOR is acute and is minimally affected by de novo protein synthesis. MC3T3-E1 cells were pretreated with cycloheximide (1 μM) for 1 h prior to the addition of T₃ (10 ng/ml) or T₄ (100 ng/ml) for 6 (D) or 24 h (E). DOR expression was determined by real-time RT-PCR analysis. DOR expression was 60% lower in cycloheximide vehicle group compared with vehicle alone at 24 h. ΔC_T = 9.39 ± 0.11, DOR expression in vehicle-treated control. Values are presented as fold change from corresponding vehicle control ± SE (n = 3–5). ^AP < 0.05 vs. vehicle control; ^BP < 0.05 vs. T₃ alone.

Fig. 2.

DOR overexpression exacerbates T₃-induced increase in ALP activity. A: expression of DOR in MC3T3-E1 cells transduced with murine leukemia virus-β-galactosidase (MLV-β-Gal) or MLV-DOR as determined by Western blot using whole cell lysates. B: MLV-β-Gal- and MLV-DOR-overexpressing MC3T3-E1 cells were treated with T₃ (0.1, 1, 3, and 10 ng/ml), and ALP activity was determined 72 h after treatment. Values are presented as %vehicle-treated MLV-β-Gal ± SE (n = 5–8). Basal ALP activity values of MLV-β-Gal (1.48 mU/mg protein ± 0.11) and DOR-overexpressing cells (2.52 mU/mg protein ± 0.18). Data were analyzed by 2-way ANOVA, and a significant interaction was observed between gene and treatment. ^AP < 0.05 vs. vehicle-treated control; ^BP < 0.05 vs. MLV-β-Gal at corresponding dose.

Fig. 3.

Inhibition of DOR reduces TH stimulation of ALP activity in MC3T3-E1 cells. A: expression of DOR in MC3T3-E1 cells transduced with scramble control short-hairpin RNA (shRNA) or DOR shRNA as determined by real-time RT-PCR analysis. Values are presented as means ± SE (n = 4). ^AP < 0.01 vs. scramble control. B: DOR protein level as determined by Western blot using whole cell lysates. Scramble control and DOR shRNA MC3T3-E1 cells were treated with T₃ (0.1, 1, 3, and 10 ng/ml; C) or T₄ (10, 100, and 1,000 ng/ml; D), and ALP activity was determined 72 h after treatment. Values are presented as %vehicle-treated scramble control ± SE (n = 6–8). Basal ALP activity values of scramble control (0.35 mU/mg protein ± 0.01) and DOR shRNA (0.23 mU/mg protein ± 0.01). Data were analyzed by 2-way ANOVA, and a significant interaction was observed between gene and treatment. ^AP < 0.05 vs. vehicle-treated control; ^BP < 0.05 vs. scramble control at corresponding dose.

Fig. 4.

TH-induced increase in ALP activity is severely reduced in primary calvaria osteoblasts transduced with DOR shRNA. A: transduction efficiency of primary calvaria osteoblasts transduced with scramble control or DOR shRNA. Pictures are representative of scramble control-bright field (I), scramble control-GFP fluorescence (II), DOR shRNA-bright field (III), and DOR shRNA-GFP fluorescence (IV). Images were taken at ×40 magnification with an Olympus IX70. B: scramble control and DOR shRNA primary calvaria osteoblasts were treated with T₃ (0.1, 1, and 10 ng/ml) or T₄ (100 ng/ml), and ALP activity was determined 72 h after treatment. Values are presented as %vehicle-treated scramble control ± SE (n = 6). Basal ALP activity values of scramble control (43.1 mU/mg protein ± 0.67) and DOR shRNA (21.2 mU/mg protein ± 1.5). Data were analyzed by 2-way ANOVA, and a significant interaction was observed between gene and treatment. ^AP < 0.05 vs. vehicle treated control; ^BP < 0.001 vs. scramble control at corresponding dose.

Fig. 5.

DOR deficiency decreases the gene expression of osteoblast-specific markers. Scramble control and DOR shRNA MC3T3-E1 cells were incubated in the presence or absence of T₄ (100 ng/ml) for 24 (osteocalcin) or 72 h (ALP and Runx2). All treatments contained ascorbic acid (100 μg/ml) and β-glycerophosphate (10 mM). Expression of osteocalcin (A), ALP (B), and Runx2 (C) was determined by real-time RT-PCR. Values are presented as fold change ± SE (n = 4). ^AP < 0.05 vs. vehicle control; ^BP < 0.05 vs. scramble control. D: DOR is localized to the nucleus. DOR expression was evaluated by Western blot in MC3T3-E1 cells incubated in the presence or absence of T₃ (10 ng/ml) or T₄ (100 ng/ml) for 1 h. Cell lysates from cytosolic and nuclear fractions were analyzed, and histone H3, β-actin, and β-tubulin served as internal controls. Data are representative of 4 independent experiments (n = 4).

Fig. 6.

Ninety-five percent confidence interval (95% CI) of circulating osteocalcin levels according to DOR/Tp53inp2 gene polymorphism. Osteocalcin levels in G/G vs. C/C carriers were significantly different.