In both WT mice and TLR2/4 KO mice, a significantly higher number of inflammatory cells were found infiltrating around the peri-implant tissues in the ligation group compared with the non-ligation group (Fig. 4a–c). However, the number of inflammatory cells in tissues of the ligation group was not significantly changed when treated with anti-RANKL antibody alone in both WT and TLR2/4 KO mice compared with the ligation group (Fig. 4b, c). MiR-146a treatment additional to anti-RANKL antibody significantly decreased the number of inflammatory cells in WT mice but not in TLR2/4 KO mice when compared with the anti-RANKL antibody alone group. Taken together, anti-RANKL antibody alone did not affect inflammatory cell infiltration in peri-implantitis, and miR-146a showed anti-inflammatory effects only on WT mice but not on TLR2/4-deficient mice.

Gingival TNF-α mRNA showed a significant upregulation in the ligation group compared with the non-ligation group in both WT and TLR2/4 KO mice (Fig. 5a, b). Moreover, TNF-α showed no significant decrease when treated with anti-RANKL antibody alone in both WT and TLR2/4 KO mice compared with the ligation group. However, additional miR-146a treatment significantly decreased TNF-α expression in WT mice but not in TLR2/4 KO mice (Fig. 5a, b), suggesting the consistent results of anti-inflammation effects characterized by quantification of infiltrating inflammation cells (Fig. 4a–c). Meanwhile, gingival RANKL mRNA expression was significantly decreased with anti-RANKL antibody alone in both WT and TLR2/4 KO mice, and additional miR-146a treatment did not show significant difference compared with the anti-RANKL antibody treatment group (Fig. 5c, d). Taken together, anti-RANKL antibody showed the inhibition effects on RANKL expression in both WT and TLR2/4 KOTLR2/4 KO mice, and miR-146a showed anti-inflammation effect through downregulation of TNF-α mRNA only in WT mice.