Fig. 4

Competitive binding of HDAC6 and AR on Runx2 promoter to modulate Runx2 expression in BMSCs. Runx2 promoter (− 1280 bp ~  + 270 bp) deletion mutant-driven luciferase vectors were established (A). Luciferase activity of every vector was measured by transient reporter assay in BMSCs from young mice (B). Runx2 promoter (− 347 bp ~ − 1 bp) deletion mutant-driven luciferase vectors were established (C). Luciferase activity of every vector was measured in BMSCs young mice (D). Luciferase activity of Runx2 promoter (− 327 bp ~ − 1 bp)-driven reporter vector was measured in response to AR-WT or AR-DN co-transfection (E). ChIP was employed to evaluate AR and HDAC6 occupancy of the Runx2 proximal promoter in BMSCs from young and aged mice with and without osteogenic induction (F). ChIP was employed to evaluate AR and HDAC6 occupancy of the Runx2 proximal promoter in response to AR-WT or AR-DN overexpression (G). Lentivirus vector loading siRNA against HDAC6 was employed to knockdown HDAC6 expression (H). ChIP was employed to evaluate AR and HDAC6 occupancy of the Runx2 proximal promoter in response to HDAC6 knockdown (I). Luc: luciferase, AR: androgen receptor, AR-WT: wild type AR, AR-DN: dominant negative AR, PC: positive control, NC: negative control. In transient reporter assay, the results were expressed as fold changes in relative luciferase unit (RLU) of every vector relative to empty luciferase vector. Data are shown as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001