MC3T3-E1 cells (a mouse preosteoblast cell line), purchased from American Type Culture Collection (ATCC, VA, United States), were cultured in α-minimum essential medium (α-MEM, 5.6 mM D-glucose, Servicebio technology, Wuhan, China) containing 10% fetal bovine serum (Gibco, Grand Island, NY, United States) and 1% antibiotics (10,000 U/ml penicillin and 10 mg/ml streptomycin, Gibco) at 37°C with 5% CO₂. To induce osteogenic differentiation, a basic culture medium, adding β-glycerophosphate (10 mM, Yuanye Biotechnology, Shanghai, China) and ascorbic acid (50 mg/ml, Solarbio Life Sciences, Beijing, China), was used to culture cells (Zhao L. et al., 2021). The osteogenic differentiation medium was replaced every 3 days.

Cells were treated with vehicle, high glucose (30 mM, Sigma-Aldrich, St. Louis, MO, United States) based on previous studies or RES (5 μM, MedChemExpress, NJ, United States) in a basic culture medium for 24 h or osteogenic differentiation medium for 14 days. When using wortmannin (MedChemExpress) as an inhibitor of phosphatidylinositol 3-kinase (PI3K)/AKT, cells were pretreated with 10 μM wortmannin for 6 h and then subjected to indicated treatments for another 24 h before conducting experiments.

Cells were transfected with negative control-short hairpin RNA (NC-shRNA) or Nrf2-shRNA (GenePharma, Shanghai, China) via cell transfection reagent (Obio Technology, Shanghai, China). Procedures were conducted by a standard protocol based on the manufacturer’s instructions.

Cell viability was performed by Cell Counting Kit 8 (CCK8, Beyotime Biotechnology, Shanghai, China). Cells were planted in 96-well plates, then treated with or without HG and RES (0.1, 1, and 5 μM) for 24 h, and finally measured. In order to ensure the effective time of RES, cells were treated with HG or HG combined with RES (5 μM) for 12, 24, or 48 h. After indicated treatments, 10 μl CCK8 reagent was added to each well and incubated at 37°C before detecting OD values at 450 nm. Cell viability was calculated using OD values and expressed by percentage.

In order to detect cell apoptosis, an In Situ Cell Death Detection Kit (Sigma-Aldrich) was used, and nuclei were counterstained with DAPI (Abcam, Cambridge, United Kingdom). For evaluating the ability of osteogenic differentiation, alkaline phosphatase (ALP) staining and alizarin red staining (ARS) were carried out by using a BCIP/NBT ALP color development kit (Beyotime Biotechnology) and Alizarin Red S (Solarbio Life Sciences), respectively. Measuring ROS levels, cells were stained with a dihydroethidium (DHE) fluorescence kit (Beyotime Biotechnology) and 2’,7’-dichlorofluorescin diacetate (DCFH-DA) fluorescence kit (Beyotime Biotechnology). For immunofluorescent staining, primary antibodies, anti-FYN antibody (1:500, Cell Signaling Technology, Danvers, MA, United States), and anti-NRF2 antibody (1:200, Abcam) were used, respectively. The secondary antibody was an anti-Rabbit IgG H and L Alexa Fluor 647 (1:200, Abcam). DAPI was applied to stain nuclei. All above staining was conducted according to the manufacturers’ instructions. Staining results were observed and photographed by light microscope (Nikon, Tokyo, Japan) or fluorescence microscope (Nikon) and calculated by ImageJ software.

Cells were cultured in an osteogenic differentiation medium for 7 days with indicated treatments, harvested, and prepared for the detection of ALP activity by the ALP activity assay kit (Beyotime Biotechnology) following the standard protocol. The absorbance was measured at 405 nm.

Cells were seeded in six-well culture plates and treated with HG or RES or both for 24 h to measure the activities of antioxidant enzymes (CAT and SOD). Then, cell lysates were collected and measured by CAT and SOD Assay Kit (Solarbio Life Sciences), respectively, following the manufacturer’s instructions.

Total RNA was extracted from cells using TRIzol reagent (Cwbio, Valencia, CA, United States). Then, 1 μg of total RNA was used for reverse transcription into cDNA by HiFiScript cDNA Synthesis Kit (Cwbio). Finally, cDNA was amplified by a real-time PCR detection system (Bio-Rad, Hercules, CA, United States) with UltraSYBR Mixture (Cwbio) using the following gene-specific primers (Sangon Biotech, Shanghai, China): heme oxygenase 1 (Ho1), NAD(P)H quinone dehydrogenase 1 (Nqo1), Cat, Sod, osterix (Osx), collagen 1 (Col1), osteopontin (Opn), runt-related transcription factor 2 (Runx2), and glyceraldehyde-3-phosphate dehydrogenase (Gapdh). PCR was performed according to the manufacturer’s instructions.

Proteins were extracted from cells via radioimmunoprecipitation assay buffer (Beyotime Biotechnology) after different treatments, then equally separated by 10% SDS-PAGE gel, and ultimately transferred into nitrocellulose membranes (GE Healthcare Life Sciences, Beijing, China). Membranes were blocked in 5% nonfat milk for 1 h, and incubated overnight at 4°C with anti-phosphorylated-AKT (P-AKT, 1:1,000, Cell Signaling Technology), anti-AKT (1:1,000, Cell Signaling Technology), anti-P-GSK3β (1:1,000, Cell Signaling Technology), anti-GSK3β (1:1,000, Cell Signaling Technology), anti-FYN (1:1,000), anti-NRF2 (1:1,000), anti-BAX (1:1,000, Cell Signaling Technology), anti-BCL-2 (1:1,000, Abcam), anti-CLEAVED CASPASE-3 (C-CASPASE 3, 1:1,000, Cell Signaling Technology), anti-GAPDH (1:1,000, Servicebio technology), and anti-β-ACTIN (1:2000, Servicebio technology). The catalogs of antibodies are shown in Supplementary Table S1.

Data were analyzed by GraphPad Prism using one-way analysis of variance (ANOVA) or two-way ANOVA with Tukey’s or Sidak’s post hoc test as appropriate and presented as means ± standard deviation (SD). p < 0.05 was accepted significant difference among compared groups.

To ascertain the effect of RES on MC3T3-E1 cells under a diabetic condition, we first assessed the cell viability by CCK8 assay with different concentrations of RES. As shown in Figure 1A, cell viability was decreased after HG treatment, but RES ameliorated HG-inhibited cell proliferation in a dose-dependent manner. Figure 1B also indicated that this damage was significantly ameliorated by 5 μM RES for 24 h. Hence, 5 μM RES for 24 h was selected in the following experimental design. Given that apoptosis is a major contributor to HG-triggered osteoblast dysfunction, apoptotic cells were determined by TUNEL staining. Results showed that TUNEL-positive cells induced by HG could be reduced following RES treatment (Figure 1C). The inhibitory effects of RES on cell apoptosis were further confirmed by detecting apoptosis-related proteins, including BAX, BCL-2, and C-CASPASE3. Results of Western blot analysis showed that RES decreased BAX/BCL-2 ratio and C-CASPASE3 expression in HG-treated MC3T3-E1 cells (Figure 1D). The above findings implied that RES possessed an effective anti-apoptotic property under the HG condition.

MC3T3-E1 cells were cultured in the osteogenic differentiation medium for 14 days to test the effect of RES on osteogenic differentiation. The osteogenic differentiation ability was measured by ARS, which detected the mineralizing matrix. The HG-inhibited capability of osteogenic differentiation was significantly rescued by RES, as shown in Figures 2A,C. ALP, another biomarker of osteogenic differentiation, was used to evaluate the differentiation potential of preosteoblasts in the HG environment. Figures 2B,D demonstrated that the impaired process of osteogenic differentiation following the HG treatment was improved by RES. Consistent with the above data, mRNA expression of osteoblast-related genes (Osx, Col1, Opn, and Runx2) were upregulated by RES in HG-treated MC3T3-E1 cells (Figures 2E–H). Collectively, these findings suggested that RES could promote osteogenesis in MC3T3-E1 cells under the HG condition.

Oxidative stress is a major factor responsible for osteoblast dysfunction. Herein, we confirmed the effect of RES on HG-induced oxidative damage by DHE staining. As shown in Figures 3A,C, excessive ROS production induced by HG was largely reduced by RES treatment. Consistent with the DHE data, RES significantly reduced oxidative stress levels increased by HG as measured by DCFH-DA staining (Figures 3B,D). Moreover, relative mRNA levels of antioxidant genes (Ho1 and Nqo1) were detected by RT-qRCR. Significant increases in their expression were shown in the HG combined with RES group compared with the HG group (Figures 3E,F). Although the mRNA levels of Cat and Sod had no apparent change in the HG group, HG treatment could significantly decrease the enzyme activities of CAT and SOD, which were upregulated following RES treatment (Figures 3G–J). Our findings indicated that RES acted as a strong candidate to reverse HG-induced oxidative stress.

NRF2 plays a critical role in the cellular response to oxidative stress. When treated with HG, total and nuclear NRF2 expression significantly declined in MC3T3-E1 cells, supported by Western blot and nuclear immunofluorescent localization, which were obviously reversed by RES (Figures 4A,E,G). Moreover, several mRNA levels of NRF2-targeted genes, Ho1 and Nqo1, were reduced in the HG condition, but they were completely recovered after RES treatment (Figures 3E,F), which suggested that RES could activate NRF2 transcriptional function in MC3T3-E1 cells under HG condition. To further investigate the effect of NRF2 in the protection of RES against HG-induced osteoblast damage, the Nrf2 gene was knocked down by Nrf2-shRNA. Results of fluorescent probe DHE and Western blot indicated that knockdown of Nrf2 blocked the protective effect of RES on oxidative stress and apoptosis (Figures 5A,B,D).

Previous studies indicated that the reduction of oxidative damage by RES was mainly associated with AKT-regulated NRF2 activation. Furthermore, the protective effect of NRF2 can be modulated by AKT/GSK3β/FYN. We found that exposing MC3T3-E1 cells to HG condition significantly decreased the expression of P-AKT and P-GSK3β and increased FYN expression in nuclei. RES reversed HG-induced changes in P-AKT, P-GSK3β, and nuclear FYN protein expression (Figures 4A–D,F). In order to figure out whether the PI3K/AKT pathway was critical to all protective effects of RES under a diabetic condition, wortmannin, a specific pharmacological inhibitor of the PI3K/AKT pathway, was applied to test parameters in the following experiments. From Figures 6A–E, the protein expression of P-AKT, P-GSK3β, and NRF2 significantly decreased, while that of FYN increased following wortmannin treatment. Most importantly, decreased protein expression of P-AKT, P-GSK3β, and NRF2 and increased protein expression of FYN in wortmannin groups following HG treatment could not be reversed by RES (Figures 6A–E). Moreover, DHE staining showed that the anti-oxidative ability of RES was largely abolished by wortmannin (Figures 6F,G). The mechanism of RES in regulating NRF2 was further verified by the knockdown of the Nrf2 gene. As shown in Figure 5C, E–H, downregulation of the NRF2 expression did not affect the expression of P-AKT, P-GSK3β, and FYN in the Nrf2-shRNA groups compared with NC-shRNA groups. These findings proved that RES largely restrained HG-increased oxidative stress levels and prevented osteoblast dysfunction under the HG condition by activating NRF2 via AKT/GSK3β/FYN pathway.