Removal of matrix-bound zoledronate prevents post-extraction osteonecrosis of the jaw by rescuing osteoclast function

Abstract

Unlike other antiresorptive medications, bisphosphonate molecules accumulate in the bone matrix. Previous studies of side-effects of anti-resorptive treatment focused mainly on systemic effects. We hypothesize that matrix-bound bisphosphonate molecules contribute to the pathogenesis of bisphosphonate-related osteonecrosis of the jaw (BRONJ). In this study, we examined the effect of matrix-bound bisphosphonates on osteoclast differentiation in vitro using TRAP staining and resorption assay, with and without pretreatment with EDTA. We also tested the effect of zoledronate chelation on the healing of post-extraction defect in rats. Our results confirmed that bisphosphonates bind to, and can be chelated from, mineralized matrix in vitro in a dose-dependent manner. Matrix-bound bisphosphonates impaired the differentiation of osteoclasts, evidenced by TRAP activity and resorption assay. Zoledronate-treated rats that underwent bilateral dental extraction with unilateral EDTA treatment showed significant improvement in mucosal healing and micro-CT analysis on the chelated sides. The results suggest that matrix-bound bisphosphonates are accessible to osteoclasts and chelating agents and contribute to the pathogenesis of BRONJ. The use of topical chelating agents is a promising strategy for the prevention of BRONJ following dental procedures in bisphosphonate-treated patients.

Keywords: Alveolar bone; Bisphosphonates; Bone matrix; Chelating agents; Osteonecrosis; Osteoporosis treatment; Zoledronate.

Fig. 1. Quantification of zoledronate chelation

A) Fluorescence imaging of AF647-Zol-treated dentin discs before (upper panel) and after (lower panel) EDTA chelation. B) Quantification of fluorescent intensity in the dentin discs showed significant decrease after EDTA application at the 0.1 μM (***p<0.001). C) Quantification of endpoint fluorescence in the chelation supernatant solution showed an increase in the amount of released Zol in a dose dependent manner with statistical significance at both the 0.1 μM (*p<0.043) and 1 μM (***p<0.001). D) Quantification of zoledronate in supernatant solutions from zoledronate-treated discs using liquid chromatography coupled with tandem mass spectrometry (LC-MS) detected Zol molecules released into the supernatant solution after chelation.

Fig. 2. Localized Zoledronate within the bone matrix inhibits osteoclast differentiation and resorption pit formation in-vitro

A) Representative images of TRAP staining (upper panel) and resorption pit formation (lower panel) on OsteoAssay plate treated with PBS (control) or increasing Zol concentrations for 24 hours prior to RAW264.7 cell culture; B) Percentage of TRAP-positive cells to the total number of cells on day 5 declined as Zol concentration increased (p=0.01); C). Area of resorption pits formation at day 21 days declined with increasing concentration of Zol (p=0.02); D) Proliferation assay results showing that proliferation of RAW264.7 cells only declined when the matrix was treated with 100μM (p=0.01); E–G) RT-PCR results showing decline in mRNA expression of Cathepsin-K (p<0.0001), RANK (p=0.003) and DC-STAMP (p=0.001) in RAW 264.7 in cells pretreated with Zol 10 μM relative to the controls.

Fig. 3. EDTA-induced removal of Zol rescues osteoclast differentiation and function in vitro

A) No change in RAW264.7 cell adhesion to OsteoAssay plate after treatment of the Zol-free matrix with EDTA for 30 minutes prior to seeding; B) Percentage of TRAP positive cells increased in the Zol-treated matrix groups but not in controls when the matrix was treated with EDTA prior to seeding (p<0.001 in both doses); C) Calcium levels in the supernatant fluid of RAW264.7 cells seeded on matrix treated with PBS (control) or zoledronate alone or with subsequent EDTA prior to seeding show consistent increase in the Zol+EDTA groups, compared to Zol only (p<0.0001); D) PYD levels in the supernatants fluid of RAW264.7 cells seeded on matrix treated with PBS (control) or zoledronate alone or with subsequent EDTA prior to seeding show consistent increase in the Zol+EDTA groups, compared to Zol only (p<0.0001)

Fig. 4. Localized chelation of zoledronate after dental extraction prevents BRONJ in vivo

A) Representative images of mucosal healing on the EDTA and saline-treated sides in Zol-treated and control animals, showing exposure of necrotic bone on the saline-treated side in Zol-treated animals (Zol-Saline). B) Representative micro-CT images of the alveolar bone at the sagittal, axial, and coronal planes (left-right) on the EDTA and saline sides in a Zol-treated animal taken in 3 planes of the reconstructed images (left-right: sagittal, axial, and coronal), showing sequestered bone on the saline but not the EDTA-treaed side. C) Healing scores were significantly higher in the control versus the Zol-treated animals (**p=0.003) and on the EDTA versus saline-treated sides in Zol-treated animals (*p=0.0156). D) Representative 20x photomicrographs of H&E stained histologic sections from the extraction sites in the four groups showing prevalence of empty lacunae on the saline-treated sides in Zol-treated animals (Zol-Saline). E) 2.5 field outlining the ROI (red circle) guided by the beginning and end of the mesial root of the third molar (x). F) The percentage of empty osteocyte lacunae was higher in Zol-treated versus control animals, and on the saline versus EDTA-treated sides in Zol-treated animals (***p<0.0001). G) No difference in BMD of alveolar bone was detected by microCT between the EDTA versus saline-treated sides.