Adiponectin Stimulates RANKL and Inhibits OPG Expression in Human Osteoblasts Through the MAPK Signaling Pathway

Authors

  • Xiang-Hang Luo,

    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    Search for more papers by this author
    • These authors contributed equally to this work

  • Li-Juan Guo,

    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    Search for more papers by this author
    • These authors contributed equally to this work

  • Hui Xie,

    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    Search for more papers by this author
  • Ling-Qing Yuan,

    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    Search for more papers by this author
  • Xian-Ping Wu,

    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    Search for more papers by this author
  • Hou-De Zhou,

    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    Search for more papers by this author
  • Er-Yuan Liao

    Corresponding author
    1. Institute of Endocrinology and Metabolism, The Second Xiangya Hospital of Central South University, Changsha, Hunan 410011, China
    • Er-Yuan Liao, MD Institute of Endocrinology and Metabolism The Second Xiangya Hospital of Central South University 139# Middle Renmin Road Changsha, Hunan 410011, China
    Search for more papers by this author

  • The authors state that they have no conflicts of interest.

  • Published online on July 17, 2006;

Abstract

Our study indicates that recombinant adiponectin induced RANKL and inhibited OPG expression in human osteoblasts through the AdipoR1/p38 MAPK pathway, and these responses contributed to the adiponectin-induced osteoclasts formation in the co-culture of osteoblast and peripheral blood monocytes systems. These findings showed that adiponectin increased osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts. It also suggests the pharmacological nature of recombinant adiponectin that indirectly induces osteoclasts formation.

Introduction: Recently, adiponectin has emerged as an element in the regulation of bone metabolism, but the mechanism remains. This study was undertaken to investigate the action of adiponectin on osteoclastogenesis through revealing RANKL and osteoprotegerin (OPG) expression in osteoblasts and osteoclast formation.

Materials and Methods: Real-time quantitative PCR and ELISA were used to detect RANKL and OPG mRNA and protein expression in cultured human osteoblasts. The involved signal pathway was studied using mitogen-activated protein kinase (MAPK) inhibitor and adiponectin receptor 1 (AdipoR1) siRNA. The effects of recombinant adiponectin on osteoclasts formation also were examined in the co-culture systems of osteoblast and peripheral blood monocytes (PBMCs) systems or purified CD14 + PBMCs cultures.

Results: Our study showed that recombinant adiponectin induced RANKL and inhibited OPG mRNA expression in human osteoblasts in a dose- and time-dependent manner. Adiponectin also increased soluble RANKL and decreased OPG secretion in osteoblasts conditioned media. Suppression of AdipoR1 with siRNA abolished the adiponectin-regulated RANKL and OPG mRNA expression in osteoblasts. Furthermore, pretreatment of osteoblasts with the MAPK inhibitor SB203580 abolished adiponectin-regulated RANKL and OPG mRNA expression. Adiponectin induced osteoclast formation in the co-culture systems of osteoblast and PBMCs systems, and OPG entirely blocked this response. However, adiponectin had no direct effect on the differentiation of osteoclast precursor purified CD14 + PBMCs.

Conclusions: These data indicate that recombinant adiponectin induced RANKL and inhibited OPG expression in human osteoblasts through the AdipoR1/p38 MAPK pathway, and these responses contributed to the adiponectin-induced osteoclast formation in the co-culture of osteoblast and PBMCs systems. These findings showed that adiponectin increased osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts. It suggests the pharmacological nature of recombinant adiponectin that indirectly induces osteoclasts formation.

INTRODUCTION

Recently, adiponectin has emerged as an element in the regulation of bone metabolism.(1–4) However, the mechanism by which adiponectin acts on the bone remains unclear.

Adiponectin, also known as apM1 (adipose most abundant gene transcript 1), Acrp30 (adipocyte complement-related protein of 30 kDa), adipoQ, and GBP28 (gelatin binding protein of 28 kDa), is highly and specifically expressed in differentiated adipocytes and is abundantly present in plasma.(5–7) Adiponectin is an ∼30-kDa polypeptide containing an N-terminal signal sequence, a variable domain, a collagen-like domain, and a C-terminal globular domain.(5–7) Adiponectin receptors (AdipoR) 1 and 2 have been identified, and the biological effects of adiponectin are mediated through the two adiponectin receptor subtypes.(8)

Our study in vitro has shown that adiponectin-induced human osteoblast proliferation and differentiation through mitogen-activated protein kinase (MAPK) pathway and suggested that osteoblasts are the direct targets of adiponectin.(1) Oshima et al.(3) reported in vitro and in vivo that adiponectin exerts an activity to increase bone mass by suppressing osteoclastogenesis and by activating osteoblastogenesis. However, Lenchik et al.(4) and Jurimae et al.(9) showed in vivo that adiponectin exert an independent negative effect on BMD in men or women and might has an unfavoring effect on bone metabolism. These data suggested that others mechanism may be involved in the action of adiponectin on bone metabolism.

Recent studies have identified osteoprotegerin (OPG) and RANKL as the essential effectors for osteoclastogenesis.(10–14) Thus, this study was undertaken to investigate the action of adiponectin on the osteoclastogenesis through revealing the OPG and RANKL expression in osteoblasts and osteoclast formation.

MATERIALS AND METHODS

Reagents

Recombinant human adiponectin, granulocyte-macrophage colony-stimulating factor (GM-CSF), recombinant human RANKL, and OPG were purchased from R&D systems (Minneapolis, MN, USA). OPG and sRANKL ELISA kits were purchased from Biomedica Group (Windham, NH, USA). Anti-bovine adiponectin polyclonal antibody, TRACP staining assay, and 1α,25-dihydroxyvitamin D3 (1,25 vitD) were purchased from Sigma (St Louis, MO, USA). Anti-human AdipoR1 and AdipoR2 antibody was purchased from Phoenix Pharmaceuticals (Belmont, CA, USA). SB203580 and SP600125 were purchased from Calbiochem Corp. (San Diego, CA, USA).

Adiponectin-free FBS was prepared by the passage of FBS through anti-adiponectin antibody Sepharose 4B affinity columns (Amersham Pharmacia Biotech) to remove adiponectin as described previously.(1)

Primary human osteoblast cultures

Bone samples were obtained with informed consent from donors and after approval by the Local Research Ethics Committee. Primary cultures of normal human osteoblasts were prepared from trabecular bone obtained from surgery caused by trauma as previously described.(15–17) Osteoblasts were isolated from trabecular bone as previously described.(15–17) Briefly, samples were rinsed extensively with serum-free α-MEM (Sigma Chemical) and digested with type IV collagenase (Sigma). The digested chips were cultured in phenol red-free α-MEM containing 10% FBS (Gibco-BRL Corp., Grand Island, NY, USA), 100 U/ml penicillin, 100 μg/ml streptomycin, and 50 μg/ml ascorbic acid at 37°C. After 15 days, cells migrated from within the bone particles and reached confluence after 25 days. They were passaged and subcultured in α-MEM containing 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 50 μg/ml ascorbic acid (Sigma). These cultured human osteoblasts were characterized and displayed an osteoblastic phenotype as our previously described.(15–17)

CD14 + peripheral blood mononuclear cell cultures

CD14 + peripheral blood mononuclear cells (PBMCs) were isolated and purified as osteoclast precursors as previous described.(18–20) Whole blood was obtained from healthy donors under a protocol approved by Local Research Ethics Commitee. PBMCs were isolated from buffy coats (Duo-Add blood bags; Baxter, Deerfield, IL, USA) using Ficoll-Paque (Amersham Pharmacia Biotech, Arlington Heights, IL, USA) as previous described.(18–20) Before further purification, granulocyte contamination was reduced to <1% by a second Ficoll-Paque separation step. Cell purification was achieved using CD14 antibody-coated magnetic cell sorting (MACS) MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany), and a VS+ selection column was mounted on a magnetic separator using standard techniques. Purity was assessed using flow cytometry (FACSCalibre; Becton Dickinson, Bedford, MA, USA). The CD14 + PBMCs were cultured as osteoclasts precursors in osteoclast formation experiments.

Detection of AdipoR in cultured osteoblasts and CD14 + PBMCs by RT-PCR andimmunoblot analysis

For studying the expression of AdipoR mRNA in human osteoblasts and CD14 + PBMCs, RT-PCR was performed as our previously described.(1) Total RNA from cultured osteoblasts and subcutaneous adipose tissue was isolated using Trizol reagent (Gibco) according to the manufacturer's recommended protocol. RT was performed using 1.0 μg total RNA and the reverse transcription system (Promega, Madison, WI, USA). For AdipoR1, the PCR primers were 5′-CCCTGACTGGCTAAAGGACA-3′ and 5′-CAGTACAGCCGCCTTCTAGG-3′, yielding a 812-bp fragment. For AdipoR2, the PCR primers were 5′-TTTGGAGCCCAGCTTAGAGA-3′ and 5′-TAGCCAGCCTATCTGCCCTA-3′, yielding a 796-bp fragment. PCR was performed as follows: 94°C for 1 minute, 60°C for 1 minute, and 72°C for 1.5 minutes for 35 cycles followed by a 10-minute incubation at 72°C. The identities of PCR products were confirmed by direct sequencing using an automatic DNA sequence (PE Applied Biosystems).

For studying the expression of AdipoR protein in cells, an immunoblot analysis was performed. The cells layers were homogenized with Triton lysis buffer (50 mM Tris-HCl, pH 8.0, containing 150 mM NaCl, 1% Triton X-100, 0.02% sodium azide, 10 mM EDTA, 10 μg/ml aprotinin, and 1 μg/ml aminoethylbenzenesulfonyl fluoride). The protein concentrations were determined using a Bradford protein assay. One hundred micrograms of protein from each cell layer was loaded onto a 7.5%polyacrylamide gel. After electrophoresis, the SDS-PAGE separated proteins were transferred to a nitrocellulose membrane (Amersham Pharmacia Biotech). The membrane was blocked with 2.5% nonfat milk in PBS and incubated with mouse monoclonal antibody against human AdipoR1 or AdipoR2 in PBS for 2 h. The membrane was reprobed with rabbit anti-mouse IgG conjugated with horseradish peroxidase for 1 h. Blots were processed using an ECL kit (Santa Cruz) and exposed to the film.

Real-time quantitative PCR assay for OPG and RANKL mRNA expression

Human osteoblasts were plated in 25-cm2 flasks in α-MEM containing 10% FBS and 50 μg/ml ascorbic acid. After 4 days of culture, cells were subsequently treated with vehicle (serum-free α-MEM) or adiponectin at 3, 10, or 30 μg/ml for 48 h in serum-free α-MEM. Cultures were also exposed to fresh serum-free medium with or without 30 μg/ml adiponectin for 12–48 h.

Real-time quantitative PCR analysis was done using Roche Molecular LightCycler (Roche Applied Science, Indianapolis, IN, USA) as described previously,(21,22) which is a combined thermal cycler and fluorescence detector with the ability to monitor the progress of individual PCR reactions optically during amplification. Total RNA from cultured cells was isolated using Trizol reagent (Gibco), and reverse transcription was performed using 1.0 μg total RNA and the reverse transcription system (Promega). Amplification reactions were set up in 25-μl reaction volumes containing amplification primers and SYBR Green PCR Master Mix (PE Applied Biosystems). A 1-μl volume of cDNA was used in each amplification reaction. Preliminary experiments were carried out for primer concentration optimization. Primer sequences are detailed as previous described.(19) For OPG, the PCR primers were 5′-CGTCAAGCAGGAGTGCAATC-3′ and 5′-CCAGCTTGCACCACTCCAA-3′. For RANKL, the PCR primers were 5′-TCGTTGGATCACAGCACATCA-3′ and 5′-TATGGGAACCAGATGGGATGTC-3′. For β-actin, the PCR primers were 5′-CCCAGCCATGTACGTTGCTA-3′ and 5′-AGGGCATACCCCTCGTAGATG-3′.

Amplifications were performed, and calibration curves were run in parallel in triplicates for each analysis. Each sample was analyzed six times during each experiment. The experiments were carried out at least twice. Amplification data were analyzed using the Sequence Detector System Software (PE Applied Biosystems). Relative quantification were calculated by normalizing the test crossing thresholds (Ct) with the β-actin amplified control Ct. The results were normalized to β-actin and expressed as percentage of controls.

OPG and soluble RANKL protein secretion assay

Osteoblasts were seeded onto 24-well plates (2 × 104 cells/well). After 4 days of culture, cells were subsequently treated with vehicle (serum-free α-MEM) or adiponectin at 3, 10, or 30 μg/ml for 48 h in serum-free α-MEM. Cultures were also exposed to fresh serum-free medium with or without 30 μg/ml adiponectin for 12–48 h. The conditioned media were collected for OPG and soluble RANKL (sRANKL) protein secretion assay using OPG and sRANKL ELISA (Biomedica Group, Windham, NH, USA).(23) Cells were also harvested and counting. sRANKL and OPG levels were normalized to cell numbers in each well.

RNA interference for AdipoR1

Our previous study indicated that human osteoblasts primarily expressed AdipoR1.(1) RNA interference was used to downregulate the expression of AdipoR1 in human osteoblasts as our previous study described.(1) Two pairs of small interfering RNA (siRNA) were synthesized by Genesil Biotechnology Co. (Wuhan, China). The sequences of the sense human AdipoR1 siRNA used were GGACAACGACUAUCUGCUACATT as previous described.(8) Control siRNA was synthesized by Genesil. Human AdipoR2 siRNA was also used. The sequences of the sense human AdipoR2 siRNA were GGAGUUUCGUUUCAUGAUCGGTT as previous described.(8) For gene knockdown experiments, osteoblasts were plated in 60-mm-diameter dish and cultured for 24 h in medium without antibiotics. Cells were transfected with siRNA (0.4 nmol/well) using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. After 24 h of culture, cells were retransfected with siRNA and recultured for another 48 h. Gene mRNA expression was analyzed by real-time quantitative PCR analysis.

Inhibition of MAPK signal pathway

Our previous study showed that adiponectin induces activation of p38 MAPK and c-jun N-terminal kinase (JNK) in osteoblasts.(1) To study the role of MAPK signal pathway on the adiponectin-regulated OPG and RANKL expression, cells were pretreated with MAPK inhibitors SB203580 or SP600125 for 2 h before adiponectin treatment. Total RNA from cultured cells was isolated, and real-time quantitative PCR analysis was performed.

CD14 + PBMCs cultures and osteoclast formation

The CD14 + PBMCs were plated at a density of 1 × 106 cells/cm2 in 24-well plates in α-MEM containing 10% adiponectin-free FBS and cultured for 14 days in the presence of M-CSF (25 ng/ml) and RANKL (50 ng/ml) with or without recombinant human adiponectin. The cultures were stained for TRACP (a marker enzyme of osteoclasts) using a TRACP staining assay (Sigma). Cells were fixed with 1% (vol/vol) formalin/PBS for 10 minutes and stained for acid phosphatase using 10 mg/ml naphthol AS-BI phosphatase as substrate in the presence of acetate tartrate buffer at pH 5.0 (50 mM sodium acetate, 40 mM potassium-sodium tartrate), and the product was coupled with Fast garnet GBC salt. TRACP+ multinucleated cells containing more than three nuclei were observed under a microscope and counted as osteoclasts.(21–23)

Co-culture of osteoblasts and CD14 + PBMCs and osteoclast formation

Osteoblasts were seeded onto 24-well plates and cultured to subconfluence. CD14 + PBMCs (1 × 106 cells/well) were added to osteoblasts as previous described,(24) and they were co-cultured for 14 days in DMEM containing 10% adiponectin-free FBS in the presence of 10−7 M 1α,25-dihydroxyvitamin D3 (1,25 vitD). Cells were treated with 30 ng/ml adiponectin, 200 ng/ml OPG, media alone, or 50 ng/ml RANKL as a positive control. The cultures were stained for TRACP, and TRACP+ multinucleated cells containing more than three nuclei were observed under a microscope and counted as osteoclasts.

Statistical analyses

Data are presented as means ± SD. Comparisons were made using a one-way ANOVA. All experiments were repeated at least three times, and representative experiments are shown.

RESULTS

AdipoR1 and 2 expression in cultured osteoblasts and CD14 + PBMCs

Using RT-PCR, we confirmed that AdipoR1 and AdipoR2 mRNA were expressed in human osteoblasts and CD14 + PBMCs, and human subcutaneous adipose tissue as a positive control (Fig. 1A). The results of RT-PCR showed an 812-bp fragment specific to the AdipoR1 and a 796-bp fragment specific to the AdipoR2. Immunoblot analysis revealed that only AdipoR1 protein expression (42 kDa) could be detected in primary human osteoblasts but not in CD14 + PBMCs (Fig. 1B). Human subcutaneous adipose tissue was used as a positive control. Adiponectin R2 protein expression was not detected in human osteoblasts and CD14 + PBMCs. Our results showed that osteoblasts primarily expressed AdipoR1, and there was no detected AdipoR1 and 2 protein expression by immunoblot analysis in CD14 + PBMCs.

Figure Figure 1.

AdipoR1 and 2 expression in cultured osteoblasts and CD14 + PBMCs. (A) Adiponectin R1 and 2 mRNA expression in cultured human osteoblasts and CD14 + PBMCs. Total RNA was subjected to RT-PCR. The PCR products (812 bp for AdipoR 1 and 796 bp specific for AdipoR 2) were visualized in a 1.5% agarose gel stained with ethidium bromide. Lane 1, Human subcutaneous adipose tissue as a positive control; lane 2, Human osteoblasts; lane 3, Human CD14 + PBMCs; lane 4, H2O as negative control. (B) Representative results of Western blot analysis using an AdipoR1 antibody. Total cellular protein was subjected to immunoblot analysis using anti-AdipoR1 and anti-AdipoR2 antibody. The anti-AdipoR1 antibody identified a band at 42 kDa. However, adiponectin R2 protein expression was not detected in osteoblasts and CD14 + PBMCs. Left margin indicates WT standards (kDa). Lane 1, human subcutaneous adipose tissue as a positive control; lane 2, human osteoblasts lysate; lane 3, human CD14 + PBMCs lysate.

Effects of adiponectin on RANKL and OPG mRNA expression

Figure 2A shows the dose response and time-course of effects of adiponectin on the RANKL mRNA expression in cultured osteoblasts. After 48 h of culture, the RANKL mRNA expression at 3 μg/ml adiponectin concentration was greater than that of controls (p < 0.05). At 10 and 30 μg/ml adiponectin concentrations, the RANKL mRNA expression increased greatly (p < 0.05). After being in culture for 12 h, the RANKL mRNA expression increased slightly compared with control (180 ± 12% of control, p < 0.05). After 24 or 48 h in culture, the RANKL mRNA expression increased more obviously (220 ± 19% and 316 ± 14% of control, respectively, p < 0.001).

Figure Figure 2.

Effects of adiponectin on RANKL and OPG mRNA expression. Cells were exposed to 3–30 μg/ml adiponectin for 48 h and to 30 μg/ml adiponectin for 12–48 h. RANKL and OPG mRNA expression was determined by real-time quantitative PCR. Results are expressed as percent of control. (A) The dose and time response of adiponectin on RANKL mRNA expression in cultured osteoblasts. Bar represents mean ± SD (n = 3; *p < 0.05 vs. control). (B) The dose and time response of adiponectin on OPG mRNA expression in cultured osteoblasts. Dots represent percent of control expression level at various time-points (*p < 0.05 vs. control).

Figure 2B shows that adiponectin inhibited the OPG mRNA expression in a dose- and time-dependent manner in osteoblasts. Treatment with 3–30 μg/ml adiponectin caused a significant decrease in OPG mRNA expression. After 12–48 h in culture with 30 μg/ml adiponectin, the OPG mRNA expression decreased several-fold above that seen in controls (p < 0.05).

Effects of adiponectin on sRANKL and OPG secretion

Figure 3A shows that adiponectin induced sRANKL production in media in a dose- and time-dependent manner in osteoblasts. Treatment with 3–30 μg/ml adiponectin for 48 h caused a significant increase in sRANKL production compared with controls (p < 0.05). After being in culture for 12 h, the sRANKL production increased slightly compared with controls (p < 0.05). After 24 or 48 h in culture, the sRANKL production increased more significantly (p < 0.05).

Figure Figure 3.

Effects of adiponectin on sRANKL and OPG secretion. Cells were exposed to 3–30 μg/ml adiponectin for 48 h and to 30 μg/ml adiponectin for 12–48 h. sRANKL and OPG production was determined by ELISA. sRANKL and OPG levels were normalized to cell numbers in each well (ng/5 × 104 cells). (A) The dose and time response of adiponectin on sRANKL production in cultured osteoblasts. Bar represents mean ± SD (n = 5; *p < 0.05 vs. control). Dots represent the percent of control expression level at various time-points (n = 5; *p < 0.05 vs. control). (B) The dose and time response of adiponectin on OPG production in cultured osteoblasts. Bar represents mean ± SD (n = 5; *p < 0.05 vs. control). Dots represent percent of control expression level at various time-points (n = 5; *p < 0.05 vs. control).

Figure 3B shows that adiponectin inhibited the OPG production in a dose- and time-dependent manner in osteoblasts. Treatment with 3–30 μg/ml adiponectin caused the decrease in OPG production. After 12–48 h in culture with 30 μg/ml adiponectin, the OPG production decreased significantly above that seen in controls (p < 0.05).

AdipoR 1 and p38 signaling pathway mediated adiponectin-regulated RANKL and mRNA expression

Figure 4A shows that treatment with siRNA-AdipoR1, but not siRNA control and siRNA-AdipoR2, blocked the expression of AdipoR1 protein in osteoblasts. It showed the adipoR1 knockdown efficiency in osteoblasts by RNA interference. Figure 4B shows that pretreatment of cells with the p38 inhibitor SB203580, but not the JNK inhibitor SP600125, blocked the increasing RANKL mRNA expression by adiponectin. To exclude the nonspecific effect of SB20358, as the adequate control, both 10−7 M 1,25 vitD and 1,25 vitD + SB203580 induced RANKL mRNA expression. SB203580 did not blocked the increasing RANKL mRNA expression by 1,25 vitD. Suppression of AdipoR1 with siRNA-AdipoR1, but not siRNA control and siRNA-AdipoR2, inhibited the increasing RANKL mRNA expression by adiponectin. These data indicated that adiponectin-induced RANKL mRNA expression is mediated by the AdipoR1/p38 pathway.

Figure Figure 4.

AdipoR 1 and p38 signaling pathways mediate adiponectin-regulated RANKL and OPG mRNA expression. (A) Treatment with siRNA-AdipoR1 blocked the expression of AdipoR1 protein in osteoblasts. Cells were treated with siRNA control, siRNA-AdipoR1, or siRNA-AdipoR2. Total cellular protein was subjected to immunoblot analysis using anti-AdipoR1 antibody. Lane 1, human osteoblasts lysate; lane 2, lysate from osteoblasts treated with siRNA control; lane 3, lysate from osteoblasts treated with siRNA-AdipoR1; lane 4, lysate from osteoblasts treated with siRNA-AdipoR2. (B) AdipoR 1 and p38 signaling pathways mediate adiponectin-induced RANKL mRNA expression. Cells were incubated with SB203580 (10 μM) or SP600125 (10 μM) for 2 h before treatment with 30 μg/ml adiponectin for 30 minutes. Cells were treated with 10−7 M 1α,25-dihydroxyvitamin D3 (1,25 vitD), or 1,25 vitD + 10 μM SB203580 as another control. Cells were also treated with siRNA control, siRNA-AdipoR1, or siRNA-AdipoR2 in the presence of 30 μg/ml adiponectin. RANKL mRNA expression was determined by real-time quantitative PCR. Results are expressed as percent of control. Bars represent mean ± SD (n = 3; *p < 0.05 vs. adiponectin-treated control). (C) AdipoR1 and p38 signaling pathways mediate adiponectin-inhibited OPG mRNA expression. Cells were treated as above described. OPG mRNA expression was determined by real-time quantitative PCR. Results are expressed as percent of control. Bars represent mean ± SD (n = 3; *p < 0.05 vs. adiponectin-treated control).

Figure 4C shows that pretreatment of cells with the p38 inhibitor SB203580, but not the JNK inhibitor SP600125, blocked the decreasing OPG mRNA expression by adiponectin. To exclude the nonspecific effect of SB20358, as the adequate control, both 10−7 M 1,25 vitD and 1,25 vitD + SB203580 inhibited OPG mRNA expression. SB203580 did not blocked the decreasing OPG mRNA expression by 1,25 vitD. Suppression of AdipoR1 with siRNA-AdipoR1, but not siRNA control and siRNA-AdipoR2, inhibited the decreasing OPG mRNA expression by adiponectin. These data indicated that adiponectin-inhibited OPG mRNA expression is mediated by the AdipoR1/p38 pathway.

Effects of adiponectin on osteoclast formation in CD14 + PBMCs cultures

Table 1 shows that 3–30 μg/ml adiponectin can not induce osteoclast formation in osteoclast precursor purified CD14 + PBMCs. As a positive control, 25 ng/ml M-CSF and 50 ng/ml RANKL induced osteoclast formation compared with the control and adiponectin treatment (p < 0.05). There were no significant difference between 25 ng/ml M-CSF and 50 ng/ml RANKL treatment and 30 μg/ml adiponectin + (25 ng/ml M-CSF + 50 ng/ml RANKL) treatment (p > 0.05). These data suggest that adiponectin have no direct effects on the differentiation of osteoclast precursor purified CD14 + PBMCs.

Table Table 1.. Effects of Adiponectin on Osteoclast Formation in CD14 + PBMCs Cultures
original image

Effects of adiponectin on osteoclast formation in co-culture of osteoblasts and CD14 + PBMCs

Figure 5A shows a microscopic view of TRACP staining. TRACP+ multinucleated cells containing more than three nuclei were observed under a microscope and counted as osteoclasts. Adiponectin (30 μg/ml) increased TRACP+ multinucleated osteoclasts formation in co-culture of osteoblasts and CD14 + PBMCs for 14 days compared with controls. As a positive control, 50 ng/ml RANKL induced osteoclast formation more obviously.

Figure Figure 5.

Effects of adiponectin (Ad) on osteoclast formation in co-culture of osteoblasts and CD14 + PBMCs. Osteoblasts were seeded onto 24-well plates and cultured to subconfluence. CD14 + PBMCs were added to osteoblasts and were co-cultured in DMEM containing 10% adiponectin-free FBS in the presence of 10−7 M 1,25 vitD. Cells were treated with 30 μg/ml adiponectin, 200 ng/ml OPG, media alone, or 50 ng/ml RANKL as a positive control. The cultures were stained for TRACP. TRACP+ red multinucleated cells were counted as osteoclasts. (A) Representative microscopic view of effects of adiponectin on osteoclast formation in co-culture of osteoblasts and CD14 + PBMCs. Osteoblasts and CD14 + PBMCs were co-cultured for 14 days. Shown are representative a microscopic view at a magnification of ×200. Osteoclasts were indicated with blue arrows. (a) Control. (b) Treatment with 30 μg/ml adiponectin. (c) Treatment with 50 ng/ml RANKL. (B) Time-course of effects of adiponectin on osteoclast formation in co-culture of osteoblasts and CD14 + PBMCs. Cells were treated with control (media alone), 30 μg/ml adiponectin, or 50 ng/ml RANKL for 2, 5, 7, and 14 days. Dots represent mean value (n = 3). (C) Inhibition effect of OPG on osteoclasts formation by adiponectin. Cells were treated with control (media alone), 30 μg/ml adiponectin, 200 ng/ml OPG, 50 ng/ml RANKL, or 30 μg/ml adiponectin + 50 ng/ml RANKL for 14 days. Bar represents mean ± SD (n = 3; *p < 0.05 vs. control. #p < 0.05 vs. adiponectin treatment).

Figure 5B shows the time-course effects of adiponectin on osteoclastogenesis. Treatment with 30 μg/ml adiponectin or 50 ng/ml RANKL for 2 days had no effects on osteoclast formation. After 5, 7, or 14 days in culture with 30 μg/ml adiponectin, TRACP+ multinucleated osteoclast formation was increased compared with controls (p < 0.05). As a positive control, RANKL induced osteoclasts formation dramatically compared with controls and adiponectin treatment (all p < 0.05).

Figure 5C shows that 30 μg/ml adiponectin increased TRACP+ multinucleated osteoclast formation compared with controls (p < 0.05). Treatment of cells with 200 ng/ml OPG blocked the increasing osteoclast formation by adiponectin or RANKL. A combination of adiponectin–RANKL achieved a synergistic rise in osteoclast formation. These data indirectly show that adiponectin increased osteoclast formation indirectly through regulating RANKL/OPG production.

DISCUSSION

Adipocytes can highly and specifically express adiponectin protein, and recently, our studies showed that adiponectin induced human osteoblast proliferation and differentiation.(1) However, the action of adiponectin on osteoclasts remains unclear. Our data showed that adiponectin induced RANKL and inhibited OPG expression in human osteoblasts through the adiponectin receptor (AdipoR1)/p38 MAPK pathway, and these responses contributed to the adiponectin-induced osteoclasts formation in co-culture of osteoblasts and CD14 + PBMCs. However, adiponectin had no direct effects on the differentiation of osteoclast precursor purified CD14 + PBMCs. These findings suggest that adiponectin increase osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts.

The key cytokines that regulate osteoclastogenesis are RANKL, a stimulator of osteoclast differentiation, activity, and survival, and OPG, an inhibitor of osteoclastogenesis. Osteoblast/stromal cells express these two genes.(10–14) RANKL exists in two forms: membrane bound form and soluble form derived by cleavage of the full-length protein. RANKL seems to be both necessary and sufficient for the complete differentiation of osteoclast precursor cells into mature osteoclasts.(10–14) Injection of exogenous RANKL in mice induces an osteoporotic phenotype, and RANKL knockout mice have severe osteopetrosis,(25) whereas OPG attaches to RANKL, thereby blocking osteoclastogenesis. OPG is cleaved from the membrane and is secreted in a soluble form. Targeted ablation of OPG in mice leads to early onset osteoporosis.(26,27)

We examined the effects of adiponectin on RANKL and OPG expression in osteoblasts. In this study, adiponectin induced RANKL and inhibited OPG mRNA expression in human osteoblasts in a dose- and time-dependent manner. Adiponectin also increased soluble RANKL and OPG secretion in osteoblasts conditioned media. Adiponectin induced the osteoclasts formation in co-culture of osteoblasts and CD14 + PBMCs, and OPG entirely blocked this response. A combination of adiponectin–RANKL achieved a synergistic rise in osteoclasts formation in co-culture of osteoblasts and CD14 + PBMCs. Furthermore, our results showed that adiponectin had no effects on the differentiation of osteoclast precursor purified CD14 + PBMCs. These data suggest that adiponectin induce osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts.

Our study revealed that adiponectin had no directly effects on the differentiation of osteoclast precursor CD14 + PBMCs. This is in conflict with the reported by Oshima et al.,(3) who showed that adiponectin inhibited differentiation of osteoclast precursor cells into osteoclasts and also suppressed the bone resorption activity of osteoclast. The grounds of this disparity are not clear but could be because of using of adiponectin-free FBS in osteoclast precursor cells cultures or another unknown variable. Recent studies have shown that the levels of adiponectin in 10%FBS are ∼6 μg/ml.(1) Oshima et al.(3) cultured mouse marrow macrophages and PBMCs in α-MEM containing 10%FBS and RANKL for 7 days with or without 1–30 μg/ml recombinant adiponectin to observe the osteoclast differentiation. Therefore, the presence of adiponectin in 10% FBS may interfere with the process of observing the osteoclast differentiation in the study of Oshima et al.(3) However, in our experiment, the use of adiponectin-free FBS in osteoclast precursor cell cultures may avoid this interfering. Our results showed that adiponectin had no direct effects on the differentiation of osteoclast precursor purified CD14 + PBMCs, but induced osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts.

The biological effects of adiponectin are mediated through the two adiponectin receptor subtypes, AdipoR1 and 2.(8) Our data showed that osteoblasts primarily express AdipoR1 and are a target for adiponectin action.(1) We also showed that there was no detected AdipoR1 and 2 protein expression by immunoblot analysis in human CD14 + PBMCs, which may contribute to the deficient direct effects of adiponectin on CD14 + PBMC differentiation. In this study, we investigated whether adiponectin affects osteoblasts through AdipoR1. Results revealed that suppression of AdipoR1 with siRNA abolished the adiponectin-regulated RANKL and OPG expression in osteoblasts. This suggests that adiponectins regulate RANKL and OPG expression in osteoblasts through AdipoR1.

To gain further insight into the mechanisms by which adiponectin regulate RANKL and OPG expression, we evaluated the signaling events. MAPKs are well known to play an essential role in controlling cell proliferation, differentiation, and gene expression.(28–30) Recent studies have shown that adiponectin induces activation of p38 MAPK and JNK but not ERK1/2 in human osteoblasts.(1) These experiments showed that pretreatment of cells with the p38 inhibitor SB203580, but not the JNK inhibitor, abolished the adiponectin-induced RANKL expression in human osteoblasts. Results also indicated that inhibition of p38 activation by SB203580 blocked adiponectin-inhibited OPG expression, whereas not the JNK inhibitor. These data indicate that adiponectin-regulated RANKL and OPG expression are mediated by the p38 pathway. MAPKs are involved in gene expression in bone cells.(31,32) Ishida et al.(33) showed that TGF-β induced RANKL expression in vascular endothelial cells through the p38 MAPK pathway. Kusumi et al.(34) reported that cyclic tensile strain inhibited RANKL and stimulated OPG synthesis in osteoblasts through the p38 MAPK pathway. Li et al.(35) showed that annexin II stimulates RANKL expression through the MAPK pathway. In this study, our study showed that activation of p38 contributed to the adiponectin-induced RANKL expression and adiponectin-inhibited OPG expression in human osteoblasts. These findings suggest that adiponectin induces RANKL and inhibits OPG expression in human osteoblasts through the AdipoR1/p38 MAPK pathway.

Recently, some studies have implied that adiponectin may emerge as an element in the regulation of bone metabolism.(1–4) Jurimae et al.(9) reported that circulating adiponectin seems to exert an independent negative effect on BMD in perimenopausal women. Lenchik et al.(4) showed that adiponectin was a novel determinant of BMD and was inversely associated with BMD. This suggests that adiponectin has an unfavoring effect on bone metabolism. However, our recent study has shown that adiponectin induced human osteoblast proliferation and differentiation through the MAPK pathway.(1) Oshima et al.(3) reported that adiponectin exerted an activity to increase bone mass by suppressing osteoclastogenesis and activating osteoblastogenesis. However, Shinoda et al.(36) showed that adiponectin-null mice and mice overexpressing adiponectin in the liver have normal bones, and their results suggested three distinct adiponectin actions on bone formation: a positive action through the autocrine/paracrine pathway by locally produced adiponectin, a negative action through the direct pathway by circulating adiponectin, and a positive action through the indirect pathway by circulating adiponectin through enhancement of the insulin signaling. Therefore, the mechanism mediating the adiponectin effect remains controversial and others mechanism may also be involved. The current observations show that adiponectin increase osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts, which may unfavorably affect bone metabolism, and contribute to the negative effect of adiponectin on BMD. In our study, the dose of recombinant adiponectin in our study was very large (3–30 μg/ml), and Shinoda et al.(36) showed that adiponectin-null mice have normal bones. This implies the pharmacological nature of recombinant adiponectin that indirectly induces osteoclast formation.

In conclusion, this study provided evidence that adiponectin induced RANKL and inhibited OPG expression in human osteoblasts through the AdipoR1/p38 MAPK pathway, and these responses contributed to the adiponectin-induced osteoclast formation in the co-culture of osteoblast and CD14 + PBMCs systems. However, adiponectin had no direct effects on the differentiation of osteoclast precursor purified CD14 + PBMCs. These findings suggest that adiponectin increases osteoclast formation indirectly through stimulating RANKL and inhibiting OPG production in osteoblasts, and the relationship between adiponectin and bone metabolism should be further studied. It also suggests the pharmacological nature of recombinant adiponectin that indirectly induces osteoclast formation.

Acknowledgements

This work was supported by Grant 30200322 from the China National Natural Scientific Foundation, A Foundation for the Author of National Excellent Doctoral Dissertation of PR China (200259), and Hunan Provincial Outstanding Younger Foundation of P.R. China (03JJY1005).

Ancillary