Rat BMSCs were obtained from Wuhan Servicebio Technology Co., Ltd. Since BMSCs from passages 3–6 exhibit a homogeneous population, consistent morphology and robust osteogenic differentiation functionality, all experiments were conducted using cells within this passage range. Adherent BMSCs were cultured in DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (FBS; Zhejiang Tianhang Biotechnology Co., Ltd.) in a humidified incubator maintained at 37°C with 5% CO₂.

BMSCs were seeded at a density of 1×10⁵ cells/well in 12-well plates. Following medium renewal on day 2, the BMSCs were induced to differentiate by culturing them in low-glucose DMEM supplemented with 10% FBS, 10⁻⁸ M dexamethasone (Sigma-Aldrich; Merck KGaA; cat: D4902), 50 µg/ml ascorbic acid 2-phosphate (Sigma-Aldrich, cat: 49752) and 10 mM β-glycerophosphate (Sigma-Aldrich; Merck KGaA; cat. no. G9422). The culture medium was refreshed every 3 days. Osteogenic differentiation medium was supplemented with 5 µM SB216763 (19–21) or 10 µM XAV939 (22,23) (both from Shanghai Aladdin Biochemical Technology Co., Ltd.) to activate or inhibit the Wnt/β-catenin signaling pathway during BMSC culture in 37°C, respectively. The DMSO group was supplemented with the same volume of DMSO as the groups treated with SB216763 or XAV939.

BMSCs were seeded at a density of 1×10 cells/well in 12-well plates. ALP activity was assessed following 10 days of osteogenic differentiation, in accordance with the manufacturer's protocol, using an ALP activity assay kit (Beyotime Biotechnology, cat: C3206). Total protein concentrations in the lysates (Beyotime Biotechnology, cat: P0013) were determined using the bicinchoninic acid assay (Pierce; Thermo Fisher Scientific, Inc.). Results were normalized to total protein content and expressed relative to the control condition.

The degree of mineralization was determined by ARS staining. BMSCs were seeded at a density of 1×10⁵ cells/well in 12-well plates. After osteogenic differentiation, the cells were fixed with 95% ethanol at 25°C for 30 min, followed by incubation with 0.1% ARS solution (pH 4.2; Beijing Solarbio Science & Technology Co., Ltd.) for 20 min at room temperature. To quantify mineralization, the calcium-bound dye was solubilized using 10% cetylpyridinium chloride (Sigma-Aldrich; Merck KGaA) for 1 h in 25°C, and the eluate was assayed spectrophotometrically at 562 nm. Staining intensity was captured in light microscope and normalized to total protein content and reported relative to the undifferentiated control.

Total RNA was extracted from BMSCs using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), followed by cDNA synthesis via RT with random primers and an M-MLV Reverse Transcriptase kit (Invitrogen; Thermo Fisher Scientific, Inc.) according to manufacturer's protocol. Subsequently, qPCR analyses were performed using a SYBR Green PCR kit (Takara Biotechnology Co., Ltd.). Thermocycling conditions were as follows: 94°C, 30 sec; 55°C, 30 sec. Step 3: 72°C, 1 min). GAPDH expression was used for normalization. The ΔCq values were calculated relative to GAPDH, and relative quantification of gene expression was determined using the 2^−ΔΔCq method (24). Each sample was assessed in triplicate. The primers used are shown in Table I.

Cell protein was obtained using lysis buffer (Beyotime Biotechnology, cat: P0013) and quantified by BCA method. Protein lysates (30 µg/lane) underwent electrophoretic separation on 10% SDS-polyacrylamide gels followed by wet transfer to PVDF membranes (Sigma-Aldrich; Merck KGaA). Membranes were then blocked for 1 h in 0.1% TBS-Tween (TBST) containing 5% non-fat dry milk at room temperature. Primary antibody (FOXJ3: Solarbio, Cat: K008825P. active β-catenin: Solarbio, Cat: K009589P. β-catenin: Solarbio, Cat: K008788P. AKT: Solarbio, Cat: K109232P. p-AKT: Solarbio, Cat: Cat:K000186M. ERK: Solarbio, Cat: K200062M. p-ERK: Solarbio, Cat: K009730P. GAPDH: Solarbio, cat. no. K200057M) incubation was performed overnight at 4°C (1:1,000). After washing, the blots were incubated for 1 h at room temperature with a HRP-linked goat anti-rabbit secondary antibody (1:1,000) (Solarbio, Cat. no. SE132 and SE134). Following three 5-min TBST washes, protein bands were treated by ECL kit (Beyotime Biotechnology, cat: P0018S) and detected by enhanced chemiluminescence after substrate application. GAPDH blotting served as the normalization control. ImageJ software (National Institutes of Health, V1.47) is used for protein quantification.

A lentiviral vector encoding FOXJ3 (lentiviral vector backbone: pCDH-EF1a-MCS-IRES-puro; OE-FOXJ3) was generated by Guangzhou iGene Biotechnology Co., Ltd. In short, 293T cells (Guangzhou iGene Biotechnology Co.) were co-transfected using a third-generation lentiviral system, with plasmid ratios of 4 µg (target plasmid): 3 µg (psPAX2): 1 µg (pMD2.G). Virus supernatants were collected in batches at 48 and 72 h after transfection and filtered through a 0.45 µm filter membrane. The virus particles were concentrated by ultracentrifugation (~70,000-100,000 × g, 2 h) in 4°C, and the precipitate was resuspended in a 500 µl of buffer. Finally, the samples were aliquoted and stored at −80°C. When cells reached 80–90% confluency, lentiviral transduction was performed for 24 h at 37°C. The viral supernatant was diluted in complete medium to achieve a multiplicity of infection of 10 and was applied to cells supplemented with polybrene (8 µg/ml; Sigma-Aldrich; Merck KGaA) in 24 h. The negative control was prepared by transducing the cells with the lentiviral vector backbone lacking the target gene. Transduction efficiency was assessed by RT-qPCR analysis of FOXJ3 mRNA levels. Subsequent experiments were performed in 24 h later.

Gene silencing was performed using FOXJ3-targeting siRNAs (Shanghai GenePharma Co., Ltd.), with a non-targeting scrambled siRNA (Shanghai GenePharma Co., Ltd.) used as a negative control. Transfection of 1×10⁵ BMSCs was carried out using 50 nM siRNA with Lipofectamine^® RNAiMAX (Invitrogen; Thermo Fisher Scientific, Inc.) according to standard procedures in 37°C in 6 h. Subsequent experiments were performed in 6 h later. The siRNA sequences are shown in Table II.

All data were obtained from experiments repeated at least three times. Results are presented as the mean ± standard deviation. All statistical analyses were conducted using SPSS 18.0 (IBM Corp.). Paired Student's t-test was used for two-group comparisons, whereas one-way ANOVA with Tukey's HSD post hoc multiple comparisons test applied for multi-group analyses. The Pearson correlation test was performed for correlation analyses. P<0.05 was considered to indicate a statistically significant difference.

Firstly, osteogenic differentiation of BMSCs was induced and dynamic changes in FOXJ3 expression were detected during this differentiation process. The results of ARS staining indicated that BMSCs were effectively differentiated into osteoblasts after a 14-day culture in osteogenic induction medium (Fig. 1A). RT-qPCR results demonstrated that FOXJ3 expression progressively increased with prolonged osteogenic induction time, reaching peak levels on day 10 of induction with a ~2-fold increase compared with that in the non-induced group (Fig. 1B). Furthermore, RT-qPCR analysis revealed a positive correlation between FOXJ3 expression and the expression of osteogenesis-related genes Runt-related transcription factor 2 (RUNX2) (Fig. 1C) and osteocalcin (OCN) (Fig. 1D). Additionally, western blotting demonstrated a progressive elevation in FOXJ3 protein expression with extended osteogenic induction, reaching a maximum on day 10 of induction (Fig. 1E and F). These findings collectively suggested that FOXJ3 may be positively associated with osteogenic differentiation of BMSCs and could serve a regulatory role in BMSCs osteogenic differentiation processes.

To investigate the regulatory role of FOXJ3 in BMSCs osteogenic differentiation, BMSCs were transfected with siRNA to knock down FOXJ3 expression. The results of RT-qPCR demonstrated that siFOXJ3-1 and siFOXJ3-2 exhibited significant knockdown efficiencies, whereas siFOXJ3-3 showed no substantial effect compared to the siControl group (Fig. 2A). Western blotting further confirmed that siFOXJ3-1 and siFOXJ3-2 effectively reduced FOXJ3 protein expression in BMSCs (Fig. 2B and C). Therefore, siFOXJ3-1 and siFOXJ3-2 for the subsequent experiments. Following FOXJ3 knockdown, osteogenic differentiation was induced in BMSCs. Quantitative ALP analysis revealed decreased ALP activity in both siFOXJ3-1 and siFOXJ3-2 groups compared with that in the control, indicating that FOXJ3 knockdown suppressed ALP activity in BMSCs (Fig. 2D). ARS staining showed that the numbers of osteogenic nodules in the siFOXJ3-1 and siFOXJ3-2 groups were markedly reduced compared with those in the control group (Fig. 2E). Quantitative analysis of ARS staining further confirmed that knockdown of FOXJ3 expression inhibited osteogenic differentiation of BMSCs (Fig. 2F). RT-qPCR demonstrated that the expression of osteogenic differentiation-related genes RUNX2 and OCN was suppressed following FOXJ3 knockdown (Fig. 2G). These findings collectively demonstrated that FOXJ3 knockdown may impair the osteogenic differentiation potential in BMSCs.

To further elucidate the regulatory role of FOXJ3 in osteogenic differentiation, lentiviral infection was used to overexpress FOXJ3 in BMSCs. The results of RT-qPCR showed that the expression levels of FOXJ3 in BMSCs were significantly increased after lentiviral infection (Fig. 3A). Western blotting also revealed that FOXJ3 protein expression was elevated in the OE-FOXJ3 group compared with that in the control group (Fig. 3B). Protein semi-quantification demonstrated a ~2-fold increase in protein expression in the OE-FOXJ3 group relative to the control group (Fig. 2C). Following FOXJ3 OE, osteogenic differentiation was further induced in BMSCs. Quantitative detection of ALP revealed that ALP activity in the OE-FOXJ3 group was significantly increased compared with that in the control group (Fig. 3D). ARS staining results demonstrated a marked increase in osteogenic nodules within the OE-FOXJ3 group relative to the control group (Fig. 3E), and quantitative analysis of ARS staining further confirmed that FOXJ3 overexpression enhanced osteogenic differentiation of BMSCs (Fig. 3F). Additionally, RT-qPCR revealed that the expression levels of the osteogenic differentiation-related genes RUNX2 and OCN were upregulated following FOXJ3 overexpression (Fig. 3G). These findings collectively demonstrated that FOXJ3 gain-of-function may promote osteogenic differentiation in BMSCs.

To investigate the mechanism by which FOXJ3 regulates BMSC osteogenic differentiation, osteogenic differentiation was induced after knocking down FOXJ3 expression, and the expression levels of proteins in common osteogenic differentiation pathways, including the Wnt/β-catenin, PI3K/AKT and MAPK/ERK pathways, were examined via western blotting. The results revealed that the expression levels of active β-catenin were decreased in the siFOXJ3-1 and siFOXJ3-2 groups compared with those in the control group, whereas p-AKT and p-ERK expression showed no significant differences (Fig. 4A and B). These results indicated that FOXJ3 may primarily promote BMSCs osteogenic differentiation by regulating the Wnt/β-catenin pathway.

To further elucidate the role of the Wnt/β-catenin pathway in FOXJ3-mediated regulation of BMSCs osteogenic differentiation, rescue experiments were performed using pathway-specific inhibitors or agonists. Western blotting initially confirmed alterations in the expression levels of proteins associated with the Wnt/β-catenin pathway following combined FOXJ3 knockdown and treatment with SB216763, a Wnt/β-catenin pathway agonist. The results showed that the expression of active β-catenin in the siFOXJ3 group was decreased compared with that in the siControl group; however, after the addition of the pathway agonist SB216763, levels of active β-catenin were increased to levels comparable with the control (Fig. S1A and B). Quantitative ALP analysis revealed that ALP activity was reduced in the siFOXJ3 group compared with that in the siControl group, whereas the addition of the pathway agonist SB216763 significantly enhanced ALP activity (Fig. 5A). Furthermore, ARS staining and quantification demonstrated fewer osteogenic nodules in the siFOXJ3 group compared with that in the siControl group, whereas the addition of the pathway agonist SB216763 restored the osteogenic differentiation capacity of BMSCs (Fig. 5B and C).

Furthermore, after overexpressing FOXJ3, the findings were further validated using the Wnt/β-catenin pathway inhibitor XAV939. Western blotting demonstrated that active β-catenin expression was elevated in the OE-FOXJ3 group compared with that in the vector group; however, this enhancement was reversed following treatment with the pathway inhibitor XAV939, restoring active β-catenin expression to control levels (Fig. S1C and D). ALP activity was significantly enhanced in the OE-FOXJ3 group relative to the vector group, whereas this effect was attenuated upon pathway inhibitor treatment (Fig. 5D). ARS staining and quantification showed increased osteogenic nodule formation in the OE-FOXJ3 group compared with that in the vector group, whereas this pro-osteogenic effect was abolished in the OE-FOXJ3 + XAV939 group, with nodule formation returning to baseline control levels (Fig. 5E and F). Therefore, these results indicated that FOXJ3 could regulate the osteogenic differentiation of BMSCs in a Wnt/β-catenin pathway-dependent manner.