Different duration of parathyroid hormone exposure distinctively regulates primary response genes Nurr1 and RANKL in osteoblasts

Abstract

Parathyroid hormone (PTH) exerts dual effects, anabolic or catabolic, on bone when administrated intermittently or continuously, via mechanisms that remain largely unknown. PTH binding to cells induces PTH-responsive genes including primary response genes (PRGs). PRGs are rapidly induced without the need for de novo protein synthesis, thereby playing pivotal roles in directing subsequent molecular responses. In this study, to understand the role of PRGs in mediating osteoblastic cellular responses to PTH, we investigated whether various durations of PTH differentially induce PRGs in primary osteoblasts and MC3T3-E1. Nurr1 and RANKL, PRGs known for their anabolic and catabolic roles in bone metabolism respectively, presented distinctive transient vs. sustained induction kinetics. Corroborating their roles, maximum induction of Nurr1 was sufficiently achieved by brief PTH in as little as 30 minutes and continued beyond that, while maximum induction of RANKL was achieved only by prolonged PTH over 4 hours. Our data suggested distinctive regulatory mechanisms for Nurr1 and RANKL: PKA-mediated chromatin rearrangement for transcriptional regulation of both PRGs and ERK-mediated transcriptional regulation for RANKL but not Nurr1. Lastly, we classified PRGs into two groups based on the induction kinetics: The group that required brief PTH for maximum induction included Nur77, cox-2, and Nurr1, all of which are reported to play roles in bone formation. The other group that required prolonged PTH for maximum induction included IL-6 and RANKL, which play roles in bone resorption. Together, our data suggested the crucial role of PRG groups in mediating differential osteoblastic cellular responses to intermittent vs. continuous PTH. Continued research into the regulatory mechanisms of PKA and ERK for PRGs will help us better understand the molecular mechanisms underlying the dual effects of PTH, thereby optimizing the current therapeutic use of PTH for osteoporosis.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1. Primary response genes Nurr1 and RANKL were differentially induced by brief vs. prolonged PTH in osteoblasts.

(A) qPCR analysis of Nurr1 (left) and RANKL (right) mRNA expression in pOBs. Cells were pre-treated with 3μg/ml Cycloheximide (CHX) for 30 minutes and subsequently with PTH for 1 hour (Nurr1) or 2 hours (RANKL). Values were normalized by GAPDH and presented as a percentage of the maximum expression level. Results indicated that Nurr1 and RANKL are PTH-induced primary response genes in pOBs (n = 3, mean±SEM, *p<0.05, **p<0.01). (B) Schematic diagram showing brief vs. prolonged PTH treatment regimes for Fig 1(C). Cells were treated with PTH (indicated as gray bars) for various periods of time, washed twice with PBS, and changed into a PTH-free medium (blank bar) for brief PTH or treated with PTH throughout the time-course for prolonged PTH. Cells were collected altogether at the end of time-course. (C,D) qPCR analysis of Nurr1 (left) and RANKL (right) mRNA expression in pOBs treated with brief or prolonged PTH. In the 4-hour time course (C), cells were treated with either prolonged PTH for 4 hours or brief PTH for 0.5 to 2 hours. In the 24-hour time course (D), cells were treated with either prolonged PTH for 24 hours or brief PTH for 2 to 8 hours. The Nurr1 expression level peaked at 2 hours, and the maximum expression levels induced by all brief vs. prolonged treatment regimes were not significantly different. The RANKL expression level peaked at 4 hours, and the maximal expression level was achieved after at least 4 hours of PTH. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 5, mean±SEM, *p<0.05, **p<0.01).

Fig 2. PTH induced histone H4 acetylation and chromatin decondensation near the transcription start sites of Nurr1 and RANKL for transcriptional regulation.

(A) qPCR analysis of PTH-induced Nurr1 (left) and RANKL (right) mRNA and heteronuclear RNA (hnRNA) levels in pOBs. mRNA and hnRNA levels were determined by qPCR using primer sets amplifying exon-exon and exon-intron junction, respectively. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 5, mean±SEM). (B,C) PTH-stimulated changes in histone H4 acetylation near transcription start sites of Nurr1 (left panels) and RANKL (right panels). For the ChIP assay, cross-linked chromatins from pOBs treated with PTH for the indicated number of hours were immunoprecipitated with anti-acetylated Histone H4 antibody or without the antibody. Immunoprecipitated chromatins were reverse-crosslinked and subjected to PCR (B) or qPCR (C) analysis with primers amplifying near the transcription start site of Nurr1 and RANKL (primer sequences indicated in S1 Table). Representative acrylamide gel images of PCR analysis (B) from three independent experiments with similar results are shown. For qPCR, values are shown as a percentage of the input (n = 3, mean±SEM). Note that the kinetics of Histone H4 acetylation and hnRNA levels are similar for both Nurr1 and RANKL. (D) The restriction endonuclease assay (REA) to monitor PTH-induced nucleosome remodeling near the transcription start site of Nurr1 and RANKL. The schematic diagram shows the locations of the restriction enzyme HindIII and KpnI recognition sites used for Nurr1 and RANKL, respectively (top panels). For REA, nuclear lysates from POBs treated with PTH for the indicated number of hours were digested with indicated restriction enzymes and subjected to southern blot analysis with a gene-specific 32P-labled probe for -100 bp to +50bp regions of Nurr1 and RANKL. Representative images from three independent experiments with similar results are shown. Note that the kinetics between Histone H4 acetylation, hnRNA level, and PTH-induced nucleosome remodeling are similar for both Nurr1 and RANKL.

Fig 3. Inhibition of PKA affected transcription and histone H4 acetylation near the transcription start sites of both Nurr1 and RANKL.

(A,B) qPCR analysis of the mRNA level (A) and hnRNA level (B) of Nurr1 (left) and RANKL (right) in POBs pre-treated with PKA inhibitor H89 15 minutes prior to PTH treatment. hnRNA were determined by qPCR using primer sets amplifying exon-intron junction. Note that the maximum mRNA and hnRNA levels of both Nurr1 and RANKL were significantly down-regulated by H89 pre-treatment. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 3, mean±SEM, *p<0.05, **p<0.01). (C) PTH-stimulated changes in histone H4 acetylation near transcription start sites of Nurr1 (left panels) and RANKL (right panels). For the ChIP assay, cross-linked chromatins from pOBs treated with PTH for indicated number of hours were immunoprecipitated with the anti-acetylated Histone H4 antibody or without the antibody. Immunoprecipitated chromatins were reverse-cross-linked and subjected to qPCR analysis with primers amplifying near the transcription start sites of Nurr1 and RANKL (primer sequences indicated in S1 Table). Values were shown as a percentage of the input. Note that the Histone H4 acetylation level was significantly down-regulated by H89 pre-treatment for both Nurr1 and RANKL (n = 3, mean±SEM, *p<0.05, **p<0.01). (D,E) qPCR analysis of mRNA level (D) and hnRNA level (E) of Nurr1 (left) and RANKL (right) in POBs received PKA inhibitor H89 post-treatment 1 or 2 hours after PTH treatment. Note that the maximum mRNA and hnRNA levels of Nurr1 were not affected by H89 post-treatment. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 5, mean±SEM, *p<0.05, **p<0.01).

Fig 4. Inhibition of ERK affected RANKL transcription, but not Nurr1.

(A-C) qPCR analysis of mRNA level (A,B) and hnRNA level (B) of Nurr1 (left) and RANKL (right) in pOBs received MEK inhibitor U0126 pre-treatment (A) 15 minutes prior to PTH or post-treatment (B,C) 1 or 2 hours after PTH. The level of hnRNA was determined by qPCR using primer sets amplifying exon-intron junction. Note that the maximum mRNA and hnRNA levels of only RANKL, but not Nurr1, were significantly down-regulated by U0126 pre-and post-treatment. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 5, mean±SEM, *p<0.05, **p<0.01). (D,E) No significant effects of MEK inhibitor U0126 treatment on PTH-induced histone H4 acetylation level near transcription start sites of Nurr1 and RANKL. Cross-linked chromatins from pOBs, treated with PTH for the indicated number of hours with either U0126 pre-treatment (A) or post-treatment (B,C) were immunoprecipitated with the anti-acetylated Histone H4 antibody or without the antibody. Immunoprecipitated chromatins were reverse-cross-linked and subjected to qPCR analysis with primers amplifying near the transcription start sites of Nurr1 and RANKL (primer sequences indicated in S1 Table). Values are shown as a percentage of the input (n = 5, mean±SEM, not significant (n.s.)).

Fig 5. PTH-induced RANKL expression level was down-regulated by inhibition of actin polymerization inhibitor Cytochalasin D and ROCK inhibitor Y-27632 treatment.

(A) PTH-induced changes in filamentous actin (f-actin) cytoskeleton arrangement in pOBs. For visualization of f-actin, cells were treated with either prolonged PTH (PTH) or brief PTH, stained with rhodamin-phalloidin and fixed with 4% formaldehyde to be examined under fluorescence microscopy. For brief PTH, cells were treated with PTH for one hour, changed into a PTH-free medium, and prepared for staining after 30 minutes (bottom left image) or one hour (bottom right image). (B) PTH-induced changes in the amount of active Rho in pOBs. Cells treated with PTH for the indicated period of time were prepared for Rho activation assay pulling-down active Rho, which was subsequently detected by immunoblot analysis using the anti-Rho antibody. Note that activation of Rho was rapidly induced by 15 minutes of PTH treatment. A representative immunoblot image from three independent experiments with similar results is shown. (C) qPCR analysis of Nurr1 (left) and RANKL (right) mRNA level in pOBs treated with 2 μM actin polymerization inhibitor Cytochalasin D (CD) or 2 μM ROCK inhibitor Y-27632 15 minutes prior to PTH treatment for indicated hours. Note that the maximum mRNA level of only RANKL, not Nurr1, was significantly down-regulated by CD as well as Y-27632 pre-treatment. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 5, mean±SEM, *p<0.05, **p<0.01). (D) The effect of ROCK inhibitor Y-27632 treatment on PTH-induced histone H4 acetylation near the transcription start sites of Nurr1 (left panels) and RANKL (right panels). pOBs pre-treated with 2 μM ROCK inhibitor Y-27632 for 15 minutes and treated with 10 nM PTH for indicated hours. Cross-linked chromatins from POBs, treated with PTH for indicated hours, were immunoprecipitated with anti-acetylated Histone H4 antibody or without the antibody. Immunoprecipitated chromatins were reverse-cross-linked and subjected to PCR analysis using primers amplifying near the transcription start site of Nurr1 and RANKL (primer sequences indicated in S1 Table). PCR products were separated on 8% acrylamide gel. A representative image from three independent experiments with similar results is shown. Note that, for Nurr1 as well as RANKL, no significant changes in Histone H4 acetylation near the transcription start sites were observed with Y27632 pre-treatment. (E) The effect of MEK inhibitor U0126 treatment on PTH-induced Rho activation. pOBs, pre-treated with 10 μM MEK inhibitor U0126 15 minutes prior to PTH, were prepared for Rho activation assay and subsequent immunoblot analysis with anti-Rho. No significant increase in PTH-induced Rho activation was noted in cells pre-treated with U0126. Representative immunoblot images from three independent experiments with similar results are shown.

Fig 6. Primary response gene Nur77, cox-2 and IL-6 were differentially induced by brief vs. prolonged PTH, similarly to Nurr1 or RANKL.

(A) qPCR analysis of Nur77, cox-2 and IL-6 mRNA expression in pOBs. Cells were pre-treated with 3μg/ml Cycloheximide for 30 minutes and subsequently with PTH for 2 hours. Results indicate that all examined genes were PTH-induced primary response genes in pOBs. Values were normalized by GAPDH and presented as a percentage of the maximum expression level (n = 5, mean±SEM). (B) qPCR analysis of Nur77, cox-2 and IL-6 mRNA expression in pOBs treated with brief or prolonged PTH for 4 hours. For brief PTH, cells were treated with PTH for 0.5 hour to 2 hours over a 4-hour time course or for 2 to 8 hours over a 24-hour time course, then prepared for qPCR. For prolonged PTH, cells were treated with PTH for 2 hour, washed twice with PBS, and changed into a PTH-containing medium. Note that Nur77 and COX-2 maximal induction was sufficiently achieved by all treatment regimes, while IL-6 maximal induction required at least 2 hours of PTH treatment. Values were normalized by GAPDH and are presented as a percentage of the maximum expression level (n = 5, mean±SEM).

Fig 7. A schematic figure illustrating differential regulatory mechanisms of brief vs. prolonged PTH for Nurr1 and RANKL induction in osteoblasts.