Parathyroid hormone increases activating transcription factor 4 expression and activity in osteoblasts: requirement for osteocalcin gene expression

Abstract

PTH is an important peptide hormone regulator of calcium homeostasis and osteoblast function. However, its mechanism of action in osteoblasts is poorly understood. Our previous study demonstrated that PTH activates mouse osteocalcin (Ocn) gene 2 promoter through the osteoblast-specific element 1 site, a recently identified activating transcription factor-4 (ATF4) -binding element. In the present study, we examined effects of PTH on ATF4 expression and activity as well as the requirement for ATF4 in the regulation of Ocn by PTH. Results show that PTH elevated levels of ATF4 mRNA and protein in a dose- and time-dependent manner. This PTH regulation requires transcriptional activity but not de novo protein synthesis. PTH also increased binding of nuclear extracts to osteoblast-specific element 1 DNA. PTH stimulated ATF4-dependent transcriptional activity mainly through protein kinase A with a lesser requirement for protein kinase C and MAPK/ERK pathways. Lastly, PTH stimulation of Ocn expression was lost by small interfering RNA down-regulation of ATF4 in MC-4 cells and Atf4(-/-) bone marrow stromal cells. Collectively, these studies for the first time demonstrate that PTH increases ATF4 expression and activity and that ATF4 is required for PTH induction of Ocn expression in osteoblasts.

Figure 1

PTH increases levels of ATF4 expression in osteoblasts. A, Effect of PTH on Atf4 mRNA. MC-4 cells were seeded at a density of 50,000 cells/cm² in 35-mm dishes and cultured in 10% FBS medium overnight. Cells were then treated with various concentration of PTH for 6 h. For each group, total RNA (20 μg/lane) was loaded for Northern hybridization using cDNA probes for mouse Atf4 mRNA and 18S rRNAs (for normalization). B, Effect of PTH on ATF4 proteins (dose response). MC-4 cells were treated with indicated concentrations of PTH for 6 h and nuclear extracts were prepared for Western blot analysis for ATF4. C, Effect of PTH on ATF4 proteins (time course). MC-4 cells were treated with 10⁻⁷ m PTH for indicated time (h). Experiments were repeated three to four times, and qualitatively identical results were obtained.

Figure 2

Effects of CHX/ActD treatment on PTH induction of Atf4 mRNA. MC-4 cells were treated with vehicle or 10 μg/ml CHX (A) or ActD (B) in the absence or presence of PTH for 6 h. Atf4 and Gapdh mRNAs were determined by quantitative real-time RT-PCR analysis. Experiments were repeated three times, and qualitatively identical results were obtained. *, P < 0.05 [control (ctrl) vs. PTH]; #, P < 0.05 (CHX vs. CHX/PTH); †, < 0.05 (control vs. CHX).

Figure 3

PTH increases ATF4-dependent transcriptional activity in MC-4 cells. A, Target cell specificity. Cells (MC-4, UMR106–01, and primary BMSCs) were transiently transfected with p4OSE1-luc and renilla luciferase normalization plasmid and treated with 10⁻⁷ m PTH for 6 h before being harvested and assayed for dual-luciferase activity. Firefly luciferase activity was normalized to renilla luciferase activity (for transfection efficiency). B, Dose dependence. MC-4 cells were transiently transfected as in Fig. 2A and treated with indicated concentration of PTH (from 10⁻¹¹ to 10⁻⁷ m) for 6 h followed by dual-luciferase assay. C, Time course. MC-4 cells were transiently transfected as in Fig. 2A and treated with 10⁻⁷ m PTH for indicated times. Data represent mean ± sd. Experiments were repeated three to four times and qualitatively identical results were obtained. *, P < 0.05 [control (ctrl) vs. PTH].

Figure 4

PTH increases binding of ATF4 to OSE1 DNA. A, PTH increases binding of osteoblast nuclear extracts (NE) to OSE1. Nuclear extracts were prepared from MC-4 cells with (P) (lanes 3–7) or without (C) (lane 2) PTH treatment for 6 h. One microgram of each nuclear extract was incubated with end-labeled double-stranded OSE1 (TGC TTA CAT CAG AGA GCA) and analyzed by electrophoresis on 4% polyacrylamide gels. DNA binding to labeled wild-type OSE1 probe was analyzed in the presence of 25- to 50-fold molar excesses of cold wt (lanes 6 and 7) or mt (lanes 4 and 5) OSE1 (TGC TTA gta CAG AGA GCA) by GMSA using 1 μg of nuclear extracts from PTH-treated MC-4 cells. B, Binding site specificity. Labeled wt (lanes 1–3) and mt (lanes 4–6) OSE1 probes were incubated with 1 μg nuclear extracts from MC-4 cells with and without PTH treatment. C, The nuclear complex binding OSE1 contains ATF4. Labeled wild-type OSE1 probe was incubated with 1 μg nuclear extracts from PTH-treated MC-4 cells in the presence of normal control IgG (lane 3), ATF4 antibody (lane 4), Runx2 antibody (lane 5), CREB antibody (lane 6), ATF1 antibody (lane 7), and Fra-1 antibody (lane 8). Experiments were repeated three to four times, and qualitatively identical results were obtained.

Figure 5

PKA is the major signaling pathway mediating the PTH response. A, Effects of inhibitors/activators on PTH-induced ATF4 transcriptional activity. MC-4 cells were transiently transfected with p4OSE1-luc and renilla luciferase normalization plasmid. After 42 h, cells were treated with 10 μm inhibitors/activators in the absence or presence of 10⁻⁷ m PTH for 6 h followed by dual-luciferase assay. Compounds used were: H89, a PKA inhibitor; FSK, a PKA activator; GF109203X, a PKC inhibitor; PMA, a PKC activator; U0126, a MAPK inhibitor; and U0124, an inactive analog of U0126. B and C, Dose-response of FSK (B) and PMA (C) on PTH stimulation of ATF4 transcriptional activity. MC-4 cells were transiently transfected as in Fig. 5A and treated with indicated concentration of respective activator for 6 h in the absence and presence of 10⁻⁷ m PTH followed by dual-luciferase assay. Data represent mean ± sd. Experiments were repeated three times and qualitatively identical results were obtained. *, P < 0.05 [control (ctrl) vs. PTH].

Figure 6

ATF4 siRNA blocks PTH stimulation of Ocn expression. A and B, MC-4 cells were transiently transfected with Atf4 siRNA (A) or negative control (Ctrl) siRNA (B). After 48 h, total RNA was prepared for quantitative real-time RT-PCR analyses for Atf4 mRNA, which was normalized to Gapdh mRNA. C and D, MC-4 cells were transiently transfected with 40 nm Atf4 siRNA or negative control siRNAs. After 42 h, cells were treated with and without 10⁻⁷ m PTH for 6 h followed by RNA preparation and quantitative real-time RT-PCR analyses for Ocn and Col1(I) mRNAs, which were normalized to the Gapdh mRNAs. *, P < 0.05 (ctrl vs. PTH); #, P < 0.05 (ctrl siRNA vs. ATF4 siRNA). Data represent mean ± sd. Experiments were repeated three times, and qualitatively identical results were obtained.

Figure 7

PTH stimulation of Ocn expression is lost in Atf4^−/− BMSCs. A, PCR genotyping was performed on tail DNA using a cocktail of three primers (see Materials and Methods). A 700-bp DNA PCR product is amplified from Atf4^−/− mouse tail DNA and a 900-bp product from wild-type mice. B–E, Effects of ATF4 deficiency on PTH stimulation of Atf4 (B), Ocn (C), Opn (D), and Pth1r (E) expression in BMSCs. Primary BMSCs were seeded at a density of 50,000 cells/cm² in 35-mm dishes and cultured in 10% FBS medium overnight. Cells were then treated with 10⁻⁷ m PTH for 6 h followed by RNA preparation and quantitative real-time RT/PCR for Atf4 (B), Ocn (C), Opn (D), and Pth1r (E) mRNA, which were normalized to the Gapdh mRNAs. *, P < 0.05 (ctrl vs. PTH); #, P < 0.05 (wt vs. mt). Data represent mean ± sd. Experiments were repeated three times, and qualitatively identical results were obtained.