MC3T3-E1 cells was cultured in α-MEM (A1049001, Gibco) supplemented with 10% (v/v) fetal bovine serum (FBS) (10099-141, Gibco) solution and were incubated at 37°C in 5% CO₂ humidified air.

Kaempferol was dissolved in dimethyl sulfoxide (DMSO) at 100 mM and kept away from light at −20°C for use, which was diluted with α-MEM to a suitable concentration upon experiment.

Cell counting kit-8 (CCK-8) assays were employed to assess the cell viability, according to the instructions of manufacture (Dojindo laboratories, Japan). Specifically, MC3T3-E1 cells were grown in 96-well plates (3 × 10³ cells/well). To assess the effect of kaempferol on the cell viability of osteoblasts, cells were co-cultured with 1, 5, 10, 25, 50, 75, and 100 μM kaempferol, respectively. To evaluate the protective activity of kaempferol in DEX-induced osteoblast viability, cells in experimental groups were co-cultured with 5, 10, and 25 μM kaempferol, respectively, and 1 μM DEX for 3 days. CCK-8 (10 μl) with complete medium (90 μl) was then added, and cells were cultured in carbon dioxide (CO₂) incubator for an additional 2 h. The wavelength 450 nm was set for the absorbance, and the assays were conducted on a microplate reader (Thermo Fisher Scientific Inc., United States). MC3T3-E1 cells were seeded in 24-well plates for 48 h and cell proliferation was evaluated using the BeyoClick™ EdU-594 (C0078S, Beyotime). Briefly, EdU (10 µM) was added into the cell culture medium and incubated for 2 h. The cells were then treated with 4% paraformaldehyde for 15 min and 0.5% Triton X-100 for 10 min. After washing with PBS, the cells were stained at room temperature for 30 min with the Click Additive Solution. Next, the cells were incubated with Hoechst 33342 stain for 10 min. Finally, a fluorescence microscope (Olympus Corporation) was used to observe and count the percentage of EdU+ cells from 4 optical fields/well.

Cell cycle was measured by flow cytometry. Briefly, digested cells were harvested. Cell pellet was suspended with 70% ethanol at 4°C overnight. The stationary cells were co-incubated with RNase-A and propidium iodide (PI) for half an hour. Cell cycle was analyzed by FACSCalibur (BD Biosciences Pharmingen™).

The cells were fixed with 4% paraformaldehyde, then stained with Hoechst 33342 (C1026, Beyotime) for 20 min and washed with PBS for 3 times. Apoptosis of the cells was observed under fluorescence microscope (Olympus Corporation). Cells were stained with PI and annexin V-FITC to evaluate apoptosis by flow cytometry according to the manufacturer’s instructions (BD Biosciences PharMingen). Cells were washed twice with phosphate-buffered saline (PBS) and stained with 5 μl of Annexin V-FITC and 5 μl of PI for 15 min in the dark. Quantification of apoptotic cells was performed by flow cytometry using a FACScan cytofluorometer (BD Biosciences).

To explore the roles of kaempferol in osteogenic differentiation, cells in experimental groups were treated with 5, 10, and 25 μM kaempferol, respectively, and 1 μM dexamethasone. The ALP activity was measured by an ALP Assay Kit after 5 days. The wavelength of 405 nm was set for detection in microplate reader. ALP staining was conducted by BCIP/NBT Kit (C3206, Beyotime).

Mineralization activity was investigated by alizarin red staining. MC3T3-E1 cells (2×10⁵ cells/well) were induced by osteogenic induction medium (OIM, including 50 μg/ml ascorbic acid, 10 mM β-glycerophosphate). Cells in experimental groups were treated with 5, 10, and 25 μM kaempferol, respectively, and 1 μM DEX for 14 days. Then, MC3T3-E1 cells were stained using alizarin red S for half an hour, and the images of mineralization nodes were photographed. 10% cetylpyridinium chloride (CPC, Sigma) was used to extract the alizarin red, and the wavelength 562 nm was selected for detection.

TRIZOL reagent was employed for extracting total RNA, which was reversely transcribed into the cDNA, according the instructions of a DNA Reverse Transcription Kit. Reverse transcription reaction conditions: Ⅰ, 37°C, 15 min; Ⅱ, 85°C, 5 s; Ⅲ, 4°C storage. The cDNA was synthesized using a Prime Script TMRT reagent Kit with gDNA Eraser (TaKaRa). The qRT-PCR was analyzed by SYBR Premix Ex TaqII Reverse Transcriptase (TaKaRa). All primers used in this study were listed as below: Runx2 forward: 5′-gaa​tgc​act​acc​cag​cca​c-3′, reverse: 5′-tgg​cag​gta​cgt​gtg​gta​g-3′; Osx forward: 5′-agg​agg​cac​aaa​gaa​gcc​ata​c-3′, reverse: 5′-agg​gaa​ggg​tgg​gta​gtc​att-3′; β-actin forward: 5′-gcc​aac​cgt​gaa​aag​atg​ac-3′, reverse: 5′-acc​aga​ggc​ata​cag​gga​cag-3′. Amplification conditions: Ⅰ. Pre-denaturation: 95°C, 30 s; Ⅱ. Amplification: 95°C, 5 s, 60°C, 34 s, 40 cycles; Ⅲ. Dissolution curve: 95°C, 30 s, 65°C, 30 s, 97°C, 30 s; Ⅳ. Store at 4°C for gene amplification. In this study, 7,500 real-time quantitative fluorescence PCR system was used for real-time quantitative PCR analysis, and all the sample were normalized to β-actin.

The extracts of total protein were prepared by lysing MC3T3-E1 cells in a RIPA buffer containing protease inhibitors. BCA assays were employed to determine the concentrations of protein. 30 μg proteins of each sample were added to 10% SDS-PAGE and then electrotransferred to the PVDF for further immunoblotting. Antibodies were used: anti-Runx2 (1:1000, ab236639, Abcam), anti-Osx (1:1000, ab209484, Abcam), anti-p53 (1:1000, 60283-2-Ig, Proteintech), anti-CyclinD1 (1:1000, 26939-1-AP, Proteintech), anti-Bcl-2 (1:1000, 26593-1-AP, Proteintech), anti-Bax (1:1000, 60267-1-Ig, Proteintech), anti-p38 (1:1000, #8690, Cell Signal Technology), anti-p-p38 (1:1000, #4511, Cell Signal Technology), anti-JNK (1:1000, #9255, Cell Signal Technology) anti-p-JNK (1:1000, #9255, Cell Signal Technology), and anti-β-actin antibody (1:1000, AF0003, Beyotime). PVDF were incubated with diluted antibodies. Goat anti-rabbit IgG H&L (HRP) (1:5000, ab6721, Abcam) or goat anti-mouse IgG H&L (HRP) (1:5000,ab6789, Abcam) was used to co-incubate with PVDF at 37°C for 1 h. Proteins were detected by ECL, and quantitatively analyzed by scanning densitometry (Bio-Rad, United States).

Wild type zebrafish larvae (AB strain) were purchased from Shanghai FishBio Co., Ltd. Larvae of AB strain were cultured in medium (0.16 mmol/L MgSO₄, 0.33 mmol/L CaCl₂, 0.17 mmol/L KCl, 5 mmol/L NaCl, and 10 ppm methylene blue) under isothermal conditions at 28.5°C. On 9 dpf (days post-fertilization), the zebrafish larvae were stained with alizarin red. First, zebrafish larvae were immobilized in 4% polyformaldehyde for 2.5 h and dehydrated in 50% ethanol for 20 min. Then, zebrafish larvae were bleached with 1.5% H₂O₂ in 1% KOH for 25 min to remove the pigment and washed 3 times with PBST. Next, zebrafish larvae were stained with 0.01% alizarin red staining in 0.7% KOH for 4 h. Finally, the zebrafish larvae were decolorized in different proportions of 0.5% KOH and glycerin (3:1, 1:1, 1:3) for 6–8 h. The experimental protocols for quantifying larval skull mineralization were similar to those previously published (Wu et al., 2021). Zebrafish were placed on a slide covered with glycerin. Stereomicroscope was used to photograph ventral view and lateral view of zebrafish. The mineralization area and integrated optical density (IOD) of skull alizarin red staining were analyzed using Image-Pro Plus (IPP, Media Cybernetics, United States).

The data were expressed as mean ± standard deviations (SD). Statistical analysis was conducted using SPSS 19.0. A one-way analysis of variance (ANOVA) was used for multiple comparisons in the statistical analysis and a value of p < 0.05 was considered statistically significant.

To explore the roles of kaempferol on the proliferation of MC3T3-E1 cells, CCK-8 assay and EdU assay were employed. Results indicated that kaempferol (1–10 μM) significantly promoted cell proliferation. However, kaempferol at the concentrations of 50–100 μM inhibited the proliferation of MC3T3-E1 cells (p < 0.01) (Figure 1A). Co-incubation of MC3T3-E1 cells with DEX for 3 days greatly inhibited the proliferation (p < 0.01). However, kaempferol (5 and 10 μM) ameliorated the inhibitory effect induced by DEX (p < 0.01) (Figure 1B). The EdU assay demonstrated that DEX significantly inhibited cell proliferation (p < 0.01), and the kaempferol (5 and 10 μM) ameliorated the inhibition of cell proliferation caused by DEX (p < 0.05) (Figures 1C,D).

Cells were co-cultured with kaempferol (1, 5, and 10 μM) and DEX (1 μM) for 3 days. The results indicated that DEX caused cell cycle arrest in the phase of G0/G1 (p < 0.01). However, kaempferol (5 and 10 μM) eased cell cycle arrest induced by DEX in MC3T3-E1 cells (Figure 2).

Cells were co-cultured with kaempferol (5, 10, and 25 μM) and DEX (1 μM) for 3 days. The results indicated that DEX caused cell apoptosis (p < 0.01). However, kaempferol (5, 10, and 25 μM) eased cell apoptosis induced by DEX in MC3T3-E1 cells (Figure 3).

MC3T3-E1 cells were co-cultured with DEX and kaempferol (5, 10, and 25 μM) for 3 days. The results showed that DEX decreased the protein expression of cyclin D1 and Bcl-2 and increased that of p53 and Bax. However, kaempferol reversed the effect of DEX, as indicated by the increased protein expression of cyclin D1 and Bcl-2 and the decreased expression of p53 and Bax in DEX-treated MC3T3-E1 cells (Figure 4).

Next, we evaluated whether kaempferol would increase the ALP activity in DEX-treated MC3T3-E1 cells. MC3T3-E1 cells were cultured in OIM and incubated with DEX (1 μM) and kaempferol (5, 10, and 25 μM) for 5 days. As a result, DEX significantly decreased the ALP activity in MC3T3-E1 cells. In contrast, kaempferol (5, 10, and 25 μM) significantly enhanced the ALP activity in DEX-treated MC3T3-E1 cells (p < 0.01) (Figure 5).

MC3T3-E1 cells were cultured in OIM and incubated with DEX (1 μM) and kaempferol (5, 10 and 25 μM) for 2 weeks. Alizarin red staining was used to visualize the calcified nodules. DEX decayed the calcified nodules formation in MC3T3-E1 cells. In contrast, kaempferol (5, 10, and 25 μM) significantly promoted the calcified nodules formation, which was indicted by the quantitative data (Figure 6).

MC3T3-E1 cells were co-incubated with DEX (1 μM) and kaempferol (5, 10, and 25 μM) for 3 days. The results indicated that the mRNA levels of Runx2 and Osx in DEX + kaempferol group were significantly increased compared with DEX group (Figures 7A,B). MC3T3-E1 cells were co-incubated with DEX (1 μM) and kaempferol (5, 10, and 25 μM) for 5 days. The results from western blot analysis showed that expressions of Runx2 and Osx were significantly decreased in DEX (1 μM) group, which were reverse by co-incubation with kaemoferol (Figures 7C,D).

The phosphorylation levels of p38 and JNK in DEX-treated MC3T3-E1 cells were measured. The results indicated that DEX decreased p38 phosphorylation, and kaempferol restored the level of phosphorylated p38 (p-p38). DEX did not significantly change the level of phosphorylated JNK (p-JNK). In contrast, kaempferol promoted JNK phosphorylation in DEX-treated MC3T3-E1 cells (Figure 8).

To further elucidate the role of p38-MAPK and JNK MAPK pathways in the protective effect of kaempferol, cells were treated with SB203580 (p38-MAPK inhibitor), SP600125 (JNK inhibitor), DEX and kaempferol. As shown in Figure 9, kaempferol (10 μM) significantly enhanced the ALP activity in DEX-treated MC3T3-E1 cells (p < 0.01). However, the addition of SB203580 or SP600125 significantly inhibited the ALP activity enhanced by kaempferol in DEX-treated MC3T3-E1 cells (p < 0.05).

Bone mineralization is a crucial indicator of bone formation. We observed the bone formation of zebrafish by staining with alizarin red. According to alizarin red staining area and IOD, the degree of skull mineralization can be determined. The results showed that bone mineralization area and IOD of skull in DEX group were obviously inhibited. However, the addition of kaempferol (10, 25, and 50 μM) attenuated the effect of DEX on decrease of bone mineralization area and IOD of skull in zebrafish (Figure 10).