Osteoprotegerin Induces Apoptosis of Osteoclasts and Osteoclast Precursor Cells via the Fas/Fas Ligand Pathway

Abstract

Osteoprotegerin (OPG) is known to inhibit differentiation and activation of osteoclasts (OCs) by functioning as a decoy receptor blocking interactions between RANK and RANKL. However, the exact role of OPG in the survival/apoptosis of OCs remains unclear. OPG caused increased rates of apoptosis of both OCs and osteoclast precursor cells (OPCs). The expression of Fas and activated caspase-8 was increased by both 20 ng/mL and 40 ng/mL of OPG, but was markedly decreased at 80 ng/mL. Interestingly, we noted that while levels of Fas ligand (FasL) increased with increasing doses of OPG, the soluble form of FasL in the supernatant decreased. The results of a co-immunoprecipitation assay suggested that the decrease of sFasL might be caused by the binding of OPG. This would block the inhibition of the apoptosis of OCs and OPCs. Furthermore, changes in expression levels of Bax/Bcl-2, cleaved-caspase-9, cleaved-caspased-3 and the translocation of cytochrome c, illustrated that OPG induced apoptosis of OCs and OPCs via the classic Fas/FasL apoptosis pathway, and was mediated by mitochondria. Altogether, our results demonstrate that OPG induces OCs and OPCs apoptosis partly by the Fas/FasL signaling pathway.

Fig 1. OPG induces apoptosis in primary osteoclasts (OCs) and osteoclast precursor cells (OPCs) in a dose-dependent manner.

Cells deprived of RANKL had added to them various doses of OPG (20, 40, 80 ng/mL). (A) Apoptosis of OCs was evaluated after 24 h using a TUNEL-derived method. Results are expressed as the percentage of apoptotic OCs. Data are expressed as mean ± SD (n = 3) relative to control. *P < 0.05, **P < 0.01 in comparison to the control by one-way ANOVA. (B) Apoptosis of OPCs collected by trypsinization was determined by flow cytometry for Annexin-V-FITC and propidium iodide (PI) dual labeling. Cells in the Q2 and Q4 quadrants represent apoptotic cells. The mean is present in the Q2 and Q4 quadrant. (C) We calculated the population of apoptotic cell. Data are expressed as mean ± SD (n = 3) relative to control. ^a P < 0.05 and ^b P < 0.01 represent the data of Q2 in comparison to the control, ^c P < 0.01 represents the data of Q4 the in comparison to the control by one-way ANOVA.

Fig 2. OPG activates the Fas/FasL apoptotic pathway in OCs and OPCs.

At the end of culture, OCs and OPCs were harvested by corresponding methods. We evaluated Fas, FasL and cleaved-caspase-8 expression by western blotting using the relevant antibodies in OCs (A) and OPCs (E). Blots for Fas, FasL and cleaved-caspase-8 in OCs (B-D) and OPCs (F-H) were semi-quantified using Image LabTM software. Data are expressed as mean ± SD (n = 3) relative to control. **P < 0.01 in comparison to the control by one-way ANOVA.

Fig 3. Western blotting shows that co-treatment with the caspase-8 inhibitor Z-IETD-FMK reverses OPG-induced activation of caspase-8 and caspase-3 in OCs and OPCs.

Cells were pretreated with Z-IETD-FMK (40 μM) for 4 h prior to treatment with OPG (40ng/mL) for 24 h. Cleaved-caspase-8 and cleaved-caspase-3 expression in OCs (A) and OPCs (D) was detected by Western blotting using the relevant antibodies. All experiments were performed in triplicate. Blots for cleaved-caspase-8 and cleaved-caspase-3 in OCs (B and C) and OPCs (E and F) were semi-quantified using Image LabTM software. Data are expressed as mean ± SD (n = 3) relative to the control. **P < 0.01 in comparison to the control by one-way ANOVA. ^## P < 0.01 in comparison to the OPG treatment by one-way ANOVA.

Fig 4. OPG induces apoptosis of OCs by binding the soluble form of FasL in the supernatant.

At the end of the culture, OPG was added at different doses, after removal of RANKL. (A) sFasL concentrations were measured by ELISA in the culture media 24 h later (n = 3) (**P < 0.01) (B) OCs and OPCs were treated with 0 and 40 ng/mL OPG for 24 h. Supernatants and OCs/OPCs were subjected to immunoprecipitation with polyclonal anti-FasL (IP FasL) followed by immunoblotting with OPG antibodies. The lysate of osteoblasts was used as the positive control for OPG expression. (C) In the three experiments, the apoptotic rate of OCs and OPCs was evaluated by corresponding methods. A correlation analysis between sFasL concentrations in the culture media and the percentage of apoptotic OCs and OPCs detected in the same cultures, was performed with Spearman test and linear regression.

Fig 5. OPG induces apoptosis via the classic Fas/FasL apoptosis pathway in OCs and OPCs.

At the end of the cultures, OCs and OPCs were harvested by corresponding methods. Bax, Bcl-2, cleaved-caspase-3 and cleaved-caspase-9 expression was evaluated by western blotting using relevant antibodies in OCs (A) and OPCs (B). Blots for Bax, Bcl-2, cleaved-caspase-3 and cleaved-caspase-9 in OCs C) and OPCs (D) were semi-quantified using Image LabTM software. The levels of cytochrome c compared to β-actin were determined by Western blotting prior to immunostaining and fluorescence microscopy of the mitochondria and cytosol of OCs (E) and OPCs (F). Blots for cytochrome c in OCs (G) and OPCs (H) were semi-quantified using Image LabTM software. Data are expressed as mean ± SD (n = 3) relative to control. *P < 0.05, **P < 0.01 in comparison to the control by one-way ANOVA.

Fig 6. Schematic representation of the proposed OPG-induced pathway in OCs and OPCs.

OPG induces Fas-mediated apoptotic signal transduction pathways by binding to sFasL, or by downregulating the shedding of membrane-bound FasL. FasL induces the clustering of intracellular death domains that recruit FADD. FADD, in turn, recruits the upstream procaspase-8 to the receptor complexes via a DED–DED interaction. As procaspase-8 has substantial enzymatic activity, recruited procaspases can be cleaved. Activated caspase-8 can directly cleave procaspase-3 (pathway 1) or can activate the mitochondria-dependent pathway of apoptosis (pathway 2).