Zoledronic acid modulates osteoclast apoptosis through activation of the NF-κB signaling pathway in ovariectomized rats

Abstract

Bone mass loss (osteoporosis) seen in postmenopausal women is an adverse factor for implant denture. Using an ovariectomized rat model, we studied the mechanism of estrogen-deficiency-caused bone loss and the therapeutic effect of Zoledronic acid. We observed that ovariectomized-caused resorption of bone tissue in the mandible was evident at four weeks and had not fully recovered by 12 weeks post-ovariectomized compared with the sham-operated controls. Further evaluation with a TUNEL assay showed ovariectomized enhanced apoptosis of osteoblasts but inhibited apoptosis of osteoclasts in the mandible. Zoledronic acid given subcutaneously as a single low dose was shown to counteract both of these ovariectomized effects. Immunohistochemical staining showed that ovariectomized induced the protein levels of RANKL and the 65-kD subunit of the NF-κB complex mainly in osteoclasts, as confirmed by staining for TRAP, a marker for osteoclasts, whereas zoledronic acid inhibited these inductions. Western blotting showed that the levels of RANKL, p65, as well as the phosphorylated form of p65, and IκB-α were all higher in the ovariectomized group than in the sham and ovariectomized + zoledronic acid groups at both the 4th- and 12th-week time points in the mandible. These data collectively suggest that ovariectomized causes bone mass loss by enhancing apoptosis of osteoblasts and inhibiting apoptosis of osteoclasts. In osteoclasts, these cellular effects may be achieved by activating RANKL-NF-κB signalling. Moreover, zoledronic acid elicits its therapeutic effects in the mandible by counteracting these cellular and molecular consequences of ovariectomized.

Keywords: NF-κB signaling pathway; RANKL; Zoledronic acid; osteoblast; osteoclast; osteoporosis.

Figure 1.

Involvement of RANK-NF-κB signalling in the differentiation of osteoclasts in osteoporosis. RANKL secreted by mature osteoblasts binds to and activates RANK on osteoclast precursors, activating TRAF6 that in turn activates the NF-κB signalling pathway. As a downstream event, NFATc1 is activated and enters into the nucleus to regulate expression of genes related to osteoclast differentiation and maturation. (A color version of this figure is available in the online journal.)

Figure 2.

X-ray imaging of rat mandibles in the Sham, OVX, and ZOL groups at 4 and 12 weeks after tooth extraction. Note that the bone mineral density (red arrows) at the bottom of the alveolar fossa was higher in the ZOL group, was higher than in the OVX group, especially at 12 weeks, but was similar between the ZOL and Sham groups. On the other hand, alveolar ridge absorption (blue arrows) was obvious in the OVX group, especially at the fourth week, but was less evident in the sham group and not obvious in the ZOL group. The bone healing of the two groups was already complete at 12 weeks. (A color version of this figure is available in the online journal.)

Figure 3.

Histology of the mandible. Although the bone in each group is still more loose and less dense at the 4th week than the 12th week time point, a larger number of bone resorption lacunae and osteoclasts (red arrow) are discerned in the OVX group than in the sham and ZOL groups, and more osteocytes (black arrow) are found in the sham and ZOL groups. Neovascularization (blue arrow) can be discerned 12 weeks post OVX, especially in the sham and ZOL group. (A color version of this figure is available in the online journal.)

Figure 4.

Masson staining of collagen fibres in the mandible, in which mature bone appears red, whereas immature braided bone that is fraught with collagen fibres appears blue (a). Quantification of stained collagen (b) shows that at the fourth-week time point, the OVX group has an abundance of collagen when compared with the sham group, whereas ZOL decreases the abundance of collagen; the collagen abundance was still more than the sham group. The collagen abundance shows little difference among the three groups 12 weeks post OVX. (Data are presented as mean ± SD. * and *** indicate a P < 0.05 and <0.001, respectively, when compared with the Sham group; # and ## indicate a P < 0.05 and 0.01, respectively, when compared with OVX group.) (A color version of this figure is available in the online journal.)

Figure 5.

TUNEL staining detecting apoptotic osteoblasts in the osteoblast-abundant areas of the mandible. Note that the number of apoptotic osteoblasts (arrows) is more abundant in the OVX group than in the sham and ZOL groups, especially at the fourth-week time point (a), with the average staining intensity of five randomly selected areas per animal quantified and statistically compared (b). (Data are presented as mean ± SD. * and *** indicate a P < 0.05 and 0.001, respectively, when compared with the Sham group; ## and ### indicate a P < 0.01 and 0.001, respectively, when compared the OVX group with the ZOL group.). (A color version of this figure is available in the online journal.)

Figure 6.

Detection of apoptotic osteoclasts with a TUNEL assay and with TRAP staining in the mandible. Note that the number of apoptotic osteoclasts (arrows) is less abundant in the OVX group than in the Sham and ZOL groups (a). The osteoclasts are distinctive by their multiple nuclei and very large cellular sizes, which were confirmed by staining for TRAP (b), a marker for osteoclasts. The number of TUNEL-positive osteoclasts and the staining intensity of individual osteoclasts are both more in the OVX group than the sham and ZOL groups at both time points (a), which is also confirmed by the quantitation of the average TUNEL staining intensity of five randomly selected areas per animal (c). (Data are presented as mean ± SD. ** indicate a P< 0.01, respectively, when compared with the Sham group; # and ## indicate a P < 0.05 and 0.01, respectively, when compared with the OVX group.). (A color version of this figure is available in the online journal.)

Figure 7.

IHC staining for RANKL in the mandible. RANKL positive cells are mainly osteoclasts (a). The number of positive osteoclasts and the average staining intensity of individual osteoclasts are both higher in the OVX group than in the sham and ZOL groups at both time points (a); quantitation of average staining intensity, which is mainly contributed by osteoclasts, of five randomly selected areas per animal shows similar results (b). (Data are presented as mean ± SD. * and **** indicate a P < 0.05 and 0.0001, respectively, when compared with the sham group; ## and ### indicate a P < 0.01 and 0.001, respectively, when compared with the OVX group.). (A color version of this figure is available in the online journal.)

Figure 8.

IHC staining for the p65 component of the NF-κB complex in the mandible. Positive cells are mainly osteoclasts (a). The number of positive osteoclasts and the average staining intensity of individual osteoclasts are both higher in the OVX group than in the Sham and ZOL groups at both time points (a); quantitation of average staining intensity, which is mainly contributed by osteoclasts, of five randomly selected areas per animal shows similar results (b). (Data are presented as mean ± SD. **** indicate a P < 0.0001, respectively, when compared with the Sham group; #### indicate a P < 0.0001, respectively, when compared with the OVX group.). (A color version of this figure is available in the online journal.)

Figure 9.

WB assay of RANKL and certain NF-κB components in the mandibular tissue. (a) OVX induces RANKL as well as the total protein and the phosphorylated form of not only p65 but also IκB-α four weeks post OVX, while ZOL prevents these effects of OVX. These changes are still discerned but much less evident at the later time point. (b) Quantitation of the band intensity shows that these differences, after correction with the signal of β-Actin included for the protein loading control, are still statistically significant. Data are presented as mean ± SD. *, ** and *** indicate a P < 0.05, 0.01 and 0.001, respectively, when compared with the Sham group; ## and ### indicate a P < 0.01 and 0.001, respectively, when compared with the OVX group.