IL-17-mediated M1/M2 macrophage alteration contributes to pathogenesis of bisphosphonate-related osteonecrosis of the jaws

Abstract

Purpose: Osteonecrosis of the jaw (ONJ) is emerging as one of the important complications in cancer patients treated with antiresorptive agents. This study explored the potential role of interleukin (IL)-17-mediated M1/M2 macrophage alterations in the pathogenesis of bisphosphonate-related osteonecrosis of the jaw (BRONJ).

Experimental design: The expression of IL-17 and M1 and M2 macrophage markers at the local mucosal site of human BRONJ lesions was examined by immunofluorescence studies. BRONJ-like disease was induced in C57BL/6 mice and multiple myeloma-burdened mice by intravenous injection of zoledronate to evaluate the correlation of elevated IL-17 levels with changes in M1 and M2 macrophage phenotypes and the therapeutic effects of blocking IL-17 on pathogenesis of BRONJ-like disease.

Results: Increased T-helper (TH)17 cells and IL-17 cytokine correlate with an increase in M1/M2 macrophages ratio at the local mucosal site of both murine and human BRONJ lesion. Convincingly, in mice burdened with multiple myeloma, a combination of elevated suprabasal level and drug-induced IL-17 activity augmented the incidence of BRONJ; both systemic increase of IL-17 and disease severity could be reversed by adoptive transfer of ex vivo expanded M2 macrophages. Targeting IL-17 via specific neutralizing antibodies or a small inhibitory molecule, laquinimod, significantly decreased M1/M2 ratio and concomitantly suppressed BRONJ-like condition in mice. Mechanistically, IL-17 enhanced IFN-γ-induced M1 polarization through augmenting STAT-1 phosphorylation while suppressing IL-4-mediated M2 conversion via inhibiting STAT-6 activation.

Conclusions: These findings have established a compelling linkage between activated IL-17-mediated polarization of M1 macrophages and the development of BRONJ-like conditions in both human disease and murine models.

Figure 1

Increased expression of IL-17 in oral mucosal tissues bordering the non-healing extraction socket of BRONJ patients. A and B, Immunofluorescence studies showed increased expression of IL-17 in both CD4⁺ T cells and CD68⁺ macrophages in mucosal tissues bordering the extraction sockets of patients manifested BRONJ lesions (BRONJ) or patients with history of zoledronate treatment without clinical BRONJ (Non-BRONJ) (n=5), whereby oral mucosal tissues from healthy patients who underwent routine dental extraction for other non-inflammatory mucosa conditions (Normal) or healthy patients with diagnosis of inflammatory gum disease, specifically, active periodontitis with resultant tooth loss (Periodontitis) were used as controls (n=5). Scale bars, 50μm. C, IL-17 levels in the tissue lysates were examined by ELISA. Data are mean ± SEM of multiple fields in n=5 per group. *P < 0.05; **P<0.01.

Figure 2

Altered M1 and M2 macrophage infiltrations in oral mucosal tissues bordering the non-healing extraction socket of BRONJ patients. A–D, Immunofluorescence studies showed an increased infiltration of CD68⁺ iNOS⁺ M1 macrophages and a decreased infiltration of CD206⁺ M2 macrophages in mucosal tissues bordering the extraction sockets of patients manifested BRONJ lesions (BRONJ) or patients with history of zoledronate treatment without clinical BRONJ (Non-BRONJ) (n=5), whereby oral mucosal tissues from healthy patients who underwent routine dental extraction for other non-inflammatory mucosa conditions (Normal) or patients with diagnosis of inflammatory gum disease, specifically, active periodontitis with resultant tooth loss (Periodontitis) were used as controls (n=5). Scale bars, 50μm. E, Quantification of total CD68⁺ macrophages in 6 randomly selected high-power fields (HPFs). F and G, Quantification of M1 and M2 macrophages in 6 randomly selected high-power fields (HPFs). Data are mean ± SEM of multiple fields in n=5 per group. *P < 0.05; **P<0.01; ns, no significant statistical differences.

Figure 3

An elevated IL-17 level was correlated with altered M1 and M2 macrophage phenotypes in mucosal tissues bordering extraction socket of C57BL/6 mice with BRONJ-like lesions. Mice were intravenously given one dose of zoledronate (Zol) (125μg/kg) one week before tooth extraction followed by intravenous injection of Zol twice a week for two weeks. A, Dynamic changes in serum IL-17 levels of C57BL/6 mice within one week after tooth extraction. Data represent mean ± SEM (n=5 per group). **P<0.01. B, An elevated serum IL17 level was correlated with BRONJ-like lesion development. Ext (−), normal mice without tooth extraction; Ext control, mice with tooth extraction but without Zol treatment; Zol(+)BRONJ(−), mice with tooth extraction and Zol treatment but without BRONJ-like lesion development; Zol(+)BRONJ(+), mice with tooth extraction and Zol treatment developed BRONJ-like lesions. Data represent mean ± SEM (n=5 per group). *P<0.05, **P<0.01 as compared with Ext control. ^$$P < 0.01 as compared with Zol (+) BRONJ (−). C–F, Real-time PCR analysis showed increased expression of IL-17, iNOS mRNA but decreased expression of arginase-1 mRNA in soft socket tissues of Zol(+)BRONJ(+) mice as compared with those of Zol(+)BRONJ(−) mice. Data represent mean ± SEM (n=5 per group). *P<0.05, **P<0.01 as compared with Ext (+) control. ^$P < 0.05, ^$$P < 0.01 as compared with Zol (+) BRONJ (−). G and H, Immunofluorescence studies showed an increased number of CD11b⁺ iNOS⁺ M1 macrophages and a decreased number of CD11b⁺ Arginase-1⁺ (Arg-1) M2 macrophages (an increased M1/M2 ratio) in mucosal tissues bordering socket tissues of Zol(+)BRONJ(+) mice as compared with those of Zol(+)BRONJ(−) mice. Scale bars, 100 μm. Data represent mean ± SEM quantification of M1/M2 macrophages (multiple images, n=3 per group). **P<0.01 as compared with Ext control. ^$P < 0.05, ^$$P < 0.01 as compared with Zol (+) BRONJ (−).

Figure 4

An elevated IL-17 level was correlated with altered M1 and M2 macrophage phenotypes in mucosal tissues bordering extraction sockets of multiple myeloma (MM)-burdened mice with BRONJ-like lesions. A, Treatment regimen. Mice were intravenously given one dose of zoledronate (Zol) (125μg/kg) one week before tooth extraction followed by intravenous injection of Zol twice a week for two weeks. B and C, Induction of BRONJ-like lesions in MM mice. Data represent mean ± SD (n=10 per group). **P<0.01 as compared with Ext control (C). D, An elevated serum IL-17 level in Zol(+) BRONJ(+) mice as compared with that of Zol(+)BRONJ(−) mice. Data represent mean ± SEM (n=5 per group). *P<0.05 as compared with Ext (+) control; ^$P < 0.05 as compared with Zol (+) BRONJ (−). E and F, An increased number of CD11b⁺ iNOS⁺ M1 macrophages and a decreased number of CD11b⁺ Arginase-1⁺ M2 macrophages in mucosal tissues bordering extract sockets of Zol(+) BRONJ(+) MM mice as compared with that of Zol(+) BRONJ(−) MM mice as determined by immunofluorescence studies. Data represent mean ± SEM quantification of M1/M2 macrophages in mucosal tissues of BRONJ-like lesions of mice (multiple images n=3 per group). ^#P<0.01 as compared with C57BL/6 mice (B6); *P<0.05, **P<0.01 as compared with Ext control; ^$P < 0.05, ^$P < 0.01 as compared with Zol (+) BRONJ (−). Ext (−), normal mice without tooth extraction; Ext control, mice with tooth extraction but without Zol treatment; Zol(+)BRONJ(−), mice with tooth extraction and Zol treatment but without manifestation of BRONJ-like lesions; Zol(+)BRONJ(+), mice with tooth extraction and Zol treatment developed BRONJ-like lesions.

Figure 5

Systemic infusion of M2 macrophages and blocking IL-17 activity suppress BRONJ-like lesion formation in MM mice. A, Treatment regimen. Mice were intravenously given one dose of zoledronate (Zol) (125μg/kg) one week before tooth extraction followed by intravenous injection of Zol twice a week for two weeks. One day after tooth extraction, mice were adoptively transferred with ex vivo expanded M2 macrophages (1x10⁶) or administered with laquinimod (Laq; 5mg/kg) or IL-17 neutralizing antibody (@IL-17, 0.3mg/mice). B–D, Amelioration of BRONJ-like lesion formation by adoptive transfer of ex vivo expanded M2 macrophages, daily oral administration of Laq or intraperitoneal (i.p.) administration of murine neutralizing @IL-17 one day after tooth extraction. Data represent mean ± SEM (n=10 per group). **P<0.01 as compared with Ext control; ^&P < 0.05 and ^&&P<0.01 as compared with Zol (+). E, A reduction in serum IL-17 level after adoptive transfer of ex vivo expanded M2 macrophages, daily oral administration of Laq or intraperitoneal (i.p.) administration of murine neutralizing @IL-17 one day after tooth extraction. Data represent mean ± SEM (n=5 per group). ^##P<0.01 as compared with C57BL/6 mice (B6); **P<0.01 as compared with Ext control; ^&P < 0.05 and ^&&P<0.01 as compared with Zol (+). F and G, A decrease in M1/M2 ratio in mucosal tissues bordering extraction sockets of mice after adoptive transfer of ex vivo expanded M2 macrophages, daily oral administration of Laq or intraperitoneal (i. p.) administration of murine neutralizing @IL-17 one day after tooth extraction. Data represent mean ± SEM of quantification of M1/M2 macrophages in BRONJ-like lesions of mice (multiple images, n=3 per group). ^#P<0.05 and ^##P<0.01 as compared with C57BL/6 mice (B6); ^$P<0.05 as compared with Ext control; *P<0.05 and **P<0.01 as compared with Zol (+) BRONJ (−); ^&P < 0.05 and ^&&P<0.01 as compared with Zol(+)BRONJ(+).

Figure 6

Mechanisms underlying IL-17-mediated alterations of M1 and M2 macrophage phenotypes. Mice bone marrow-derived monocytes were cultured in the presence of M-CSF (10ng/mL) for 6 days to induce differentiation of mice bone marrow-derived macrophages (BMDMs), while human leukemic monocytes (THP-1) were differentiated into macrophages in the presence of PMA (200 nmol/L) for 24 hours. A and B, Mice BMDMs or human macrophages were continuously cultured under M1 macrophage polarization condition (IFN-γ, 20ng/mL) in the presence or absence of murine or human IL-17 (10ng/mL) for another 24 hours. The production of murine or human TNF-α in the supernatants was determined by ELISA. C, Mice BMDMs were continuously cultured under M1 macrophage polarization condition (IFN-γ, 20ng/mL) in the presence or absence of murine IL-17 (10ng/mL) for different time periods, and the phosphorylated levels of STAT-1 were determined by Western blot. D and E, Mice BMDMs or human macrophages were continuously cultured under M2 macrophage polarization condition (IL-4 10ng/mL) in the presence or absence of murine or human IL-17 (10ng/mL) for another 24 hours, and the production of murine or human IL-10 in the supernatants were determined by ELISA. F, Mice BMDMs were continuously cultured under M2 macrophage polarization condition (IL-4 10ng/mL) in the presence or absence of murine IL-17 (10ng/mL) for another 24 hours, and the expression of arginase-1 (Arg-1) was determined by Western blot analysis. G, Mice BMDMs were continuously cultured under M2 macrophage polarization condition (IL-4 10ng/mL) in the presence or absence of murine IL-17 (10ng/mL) for another 2 hours, and the phosphorylated levels of STAT-6 were determined by Western blot analysis. Data represent mean ± SEM of three independent experiments. *P < 0.05; **P<0.01.