Fig. 4: The proinflammatory and chondrocyte catabolic effects of RANKL on FLSs.

A Design of scRNA-seq experiment. Created in BioRender. Hu, Y. (2025) https://BioRender.com/esd3e6h. B Visualization of UMAP plot pooled from 14 cell clusters isolated from knee joints of Sham, DMM and DMM + denosumab mouse single-cell transcriptomes. C Percentage of each cell type within knee joints among Sham, DMM and DMM + denosumab mice based on the UMAP distribution. D Number and strength of intercellular communication among the FLS, chondrocyte and macrophage cell populations in the indicated groups. The edge width is proportional to the indicated number or strength of intercellular communication. E Western blot of MMP-13 and TNF-α in vehicle-, RANKL- and RANKL + denosumab-treated FLSs (n = 3, biological replicates). F Western blotting of Col2a1 and Col10a1 in chondrocytes incubated with conditioned medium collected from FLSs treated with RANKL or RANKL + denosumab (n = 3, biological replicates). G Western blotting of Arg1 and CD86 in macrophages incubated with conditioned medium collected from FLSs treated with RANKL or RANKL + denosumab (n = 3, biological replicates). H Safranin O/Fast green staining of knee joints in all groups at 4 weeks after RANKL or RANKL + denosumab injection. Scale bar = 200 μm. I The OA severity of the knee joints was evaluated using the OARSI score (n = 6, biological replicates). J Representative H&E staining images showing the infiltration of inflammatory cells and thickness of the synovium (dashed lines) in the indicated group at week 4. Magnified images of the boxed areas are shown in the panel below. Scale bar = 500 μm (upper image); 100 μm (lower image). K Quantification of synovitis score (n = 6, biological replicates). Source data and P values are provided in the Source Data file. Data are presented as median with interquartile range. Statistical analysis was performed using one-way ANOVA with Tukey’s post hoc test (E–G, I, K). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.