Fig. 3. miR-125b in bone matrix inhibits osteoclastogenesis by targeting Prdm1.

a Representative images of osteogenic activities in Tg and WT bone marrow cell cultures, assessed by ALP/von Kossa staining (scale bars, 4.5 mm) and the percentage of ALP/von Kossa⁺ areas (n = 3). b miR-125b levels in AGO2 immunoprecipitates from the cell and ECM fractions of WT vs. Tg bones (n = 3). c Representative images of TRAP⁺ multinucleated cells (MNCs) in WT and Tg bone marrow macrophage (BMM) cultures stimulated with RANKL and M-CSF (scale bars, 100 µm) and the number of TRAP⁺ MNCs/well (n = 6). d FACS analysis of Tg and WT osteoclast precursor (CD117⁺/CD115⁺/CD11b^low) cells; cells were gated on CD117 (n = 8). e The number of TRAP⁺ MNCs/well in RAW-D cell cultures with or without WT MVs vs. Tg MVs (n = 6). Cells were stimulated with RANKL. f Representative images of TRAP staining of either Tg or WT BMMs on Tg vs. WT calvarial bone chips (scale bars, 100 μm) and the percentage of TRAP⁺ areas (n = 4). αMEM shows bone chips without BMMs. g Prdm1 levels in RAW-D cells with MVs (1 μg protein/mL) or PBS; Rpl32 was used as internal control (n = 3). h The reporter constructs contained the miRNA response element (MRE) (Wild) of Prdm1 or its mutant, and reporter activities in RAW-D cells cotransfected with reporter plasmids and miR-125b mimic (Mimic) (n = 4). i mRNA levels of Prdm1 and associated downstream target genes in RAW-D cells transfected with Mimic and negative control miRNA (NC); Rpl32 was used as internal control (n = 3). BMMs and RAW-D cells were stimulated by RANKL plus M-CSF (c, f) and RANKL (e, g–i), respectively. *P < 0.05, **, ^##P < 0.01, and ***P < 0.001 vs. matched control by Student’s t-test (b, f, g, i) and by Tukey’s multiple comparison (e, h).