Gene targeting by the vitamin D response element binding protein reveals a role for vitamin D in osteoblast mTOR signaling

Abstract

Transcriptional regulation by hormonal 1,25-dihydroxyvitamin D(3) [1,25(OH)(2)D(3)] involves occupancy of vitamin D response elements (VDREs) by the VDRE binding protein (VDRE-BP) or 1,25(OH)(2)D(3)-bound vitamin D receptor (VDR). This relationship is disrupted by elevated VDRE-BP, causing a form of hereditary vitamin D-resistant rickets (HVDRR). DNA array analysis showed that of 114 genes regulated by 1,25(OH)(2)D(3) in control cells, almost all (113) were rendered insensitive to the hormone in VDRE-BP-overexpressing HVDRR cells. Among these was the gene for DNA-damage-inducible transcript 4 (DDIT4), an inhibitor of mammalian target of rapamycin (mTOR) signaling. Chromatin immunoprecipitation PCR using 1,25(OH)(2)D(3)-treated osteoblasts confirmed that VDR and VDRE-BP compete for binding to the DDIT4 gene promoter. Expression of DDIT4 mRNA in these cells was induced (1.6-6 fold) by 1,25(OH)(2)D(3) (10-100 nM), and Western blot and flow cytometry analysis showed that this response involved suppression of phosphorylated S6K1(T389) (a downstream target of mTOR) similar to rapamycin treatment. siRNA knockdown of DDIT4 completely abrogated antiproliferative responses to 1,25(OH)(2)D(3), whereas overexpression of VDRE-BP exerted a dominant-negative effect on transcription of 1,25(OH)(2)D(3)-target genes. DDIT4, an inhibitor of mTOR signaling, is a direct target for 1,25(OH)(2)D(3) and VDRE-BP, and functions to suppress cell proliferation in response to vitamin D.

Figure 1.

Increased expression of VDRE-BP in a patient with HVDRR causes cellular resistance to 1,25(OH)₂D₃. A) SDS-PAGE analysis utilizing B cells from a patient with HVDRR or an age/sex-matched wild-type (WT) control subject. Cells were treated with vehicle or 1,25(OH)₂D₃ (10 nM, 24 h). B) Dose-dependent effect of 1,25(OH)₂D₃ (6 h) on expression of mRNA for 24-hydroxylase (CYP24A1) in HVDRR and WT cells. Data are shown as a fold-induction of mRNA levels relative to vehicle-treated control cells (n=3, means±sd). *P ≤ 0.001, **P ≤ 0.05; 2-factor ANOVA with Bonferroni post hoc test.

Figure 2.

Identification of VDRE-BP-dependent gene signatures in a patient with HVDRR. A) DNA array analysis of genes induced or suppressed by 1,25(OH)₂D₃ in control or HVDRR cells (1.5-fold difference; P≤0.05 cutoff). B) Heat map of differentially regulated genes [10 nM 1,25(OH)₂D₃, 6 h]. C) RT-qPCR validation of selected genes in control and HVDRR cells (n=3, means±sd). KCNN4, potassium intermediate/small conductance calcium-activated channel subfamily N, member 4; FAM46C, family with sequence similarity 46, member C; ZFP14, zinc finger protein 14 homologue. *P ≤ 0.001, **P ≤ 0.05; 2-way ANOVA with posttest.

Figure 3.

1,25(OH)₂D₃-induced regulation of VDRE-BP target genes in human osteoblastic cells. A–C) Effect of 1,25(OH)₂D₃ (0.1–10 nM, 24 h) on mRNA expression in primary hOBs at early proliferation (A) or quiescent (B) or late (C) stages of differentiation. Long-term hOBs were cultured in normal (unstimulated) or osteogenic (stimulation) medium (n=3, means±se). *P ≤ 0.05; 2-way ANOVA. D) Effect of 1,25(OH)₂D₃ (24 h) on mRNA expression in osteoblastic MG-63 cells. E) Effect of transfected VDRE-BP cDNA on 1,25(OH)₂D₃-induced gene expression in MG-63 cells and hOBs, respectively (n=3, means±se). DIV, days in vitro. *P ≤ 0.05 vs. empty-vector control; 1-way ANOVA (P≤0.001) with Bonferroni multiple-comparison test.

Figure 4.

DDIT4 is a target for VDR/VDRE-BP-chromatin interaction in osteoblasts. ChIP-qPCR analysis of MG-63 cells treated with vehicle or 1,25(OH)₂D₃ for 15 min. Transcriptional start site (−1 kb) primers were designed for both CYP24A1 and DDIT4, while a nonintergenic VDRE-negative primer set for CYP24A1 (−2 kb) was used as a control. Data are presented as fold enrichment of chromatin normalized for non-specific (IgG) antibody (n=3, means±sd; P≤0.05, 1-way ANOVA).

Figure 5.

1,25(OH)₂D₃-mediated regulation of mTOR signaling in osteoblasts. A) Effect of 1,25(OH)₂D₃ (0, 1, and 10 nM; 18 h) on VDRE-BP-related protein expression in MG-63 cells. Expression of S6K1 and p-S6K1^Thr389 (nuclear and cytoplasmic isoforms p85/p70, respectively) was assessed under normal conditions and without growth factor (−GF). B) Immunofluorescent detection of DDIT4 (panels 1–3) and p-S6K^Thr389 (panels 4–6) within MG-63 cells treated with 5% FBS (panels 1, 4), serum starvation (panels 2, 5), or 10 nM 1,25(OH)₂D₃ (panels 3, 6). Cells/samples were analyzed at 18 h after treatment. Merge images show transmission overlay; see text for explanation of arrows.

Figure 6.

Regulation of osteoblast proliferation and cell size. A) 1,25(OH)₂D₃ (10 nM) and rapamycin (25 nM) decrease MG-63 cell proliferation (n=6, means±se). *P ≤ 0.01; 2-way ANOVA. B) Cell cycle analyses at 48 h. G₀/G₁ population assessed using the Student's t test (n=3). C) Representative flow cytometry profiles showing cell size distribution (forward scattering) of G₁-gated MG-63 cells treated for 48 h (blue) or untreated (red). D) Quantification of cell size change (n=3; means±se). *P < 0.05; Student's t test.

Figure 7.

DDIT4 is essential for the regulation of osteoblast proliferation by 1,25(OH)₂D₃. A) Expression of DDIT4 and CYP24A1 mRNA in MG-63 cells transfected with control (nonspecific) or DDIT4 siRNA (48 h) in the presence or absence of 1,25(OH)₂D₃ (10 nM, 24 h; n=3, means± sd). *P ≤ 0.001; 1-way ANOVA with Tukey's multiple-comparison test. B) Western blot analysis of MG-63 cells treated with control or DDIT4 siRNA (48 h). C) 1,25(OH)₂D₃ [10 nM] effect on MG-63 cell proliferation treated with control or DDIT4 siRNA (n=6, means±se). *P ≤ 0.001; Student's t test. D) Effect of 1,25(OH)₂D₃ (48 h) on MG-63 cell proliferation under transfection with control or hnRNPC1 (VDRE-BP) cDNA (n=6, mean±se). *P ≤ 0.05 vs. corresponding control; 1-way ANOVA (P≤0.001) with post hoc test.

Figure 8.

Regulation of osteoblastic mTOR signaling by 1,25(OH)₂D₃ and VDRE-BP. DDIT4, a stress-induced negative regulator of mTOR signaling, is activated in bone cells by 1,25(OH)₂D₃, DDIT4 up-regulation results in activation of TSC1/2 leading to suppression of mTOR activity through reduced phosphorylation of p70 S6 kinase 1 for decreased cell proliferation. Induction of DDIT by 1,25(OH)₂D₃ can be blocked by overexpression of VDRE-BP. Our findings highlight the importance of vitamin D-mTOR singling in controlling cellular metabolism within osteoblastic bone cells.