Targeting SOD3-Mediated Oxidative Stress with Betulonic Acid Protects Against Steroid-Induced Femoral Head Necrosis Authors: Qi Chen, Yangquan Hao, Wenxing Yu, Yuheng Yan, Jiadong Chen, Chao Lu Published: 2026 ABSTRACT BackgroundSteroid-induced osteonecrosis of the femoral head (SONFH) is a devastating complication of glucocorticoid therapy, fundamentally driven by oxidative stress. Although Betulonic acid (Ba), a natural pentacyclic triterpenoid, exhibits potent antioxidant properties, its therapeutic efficacy in SONFH remains unknown. This study investigates the protective effects of Ba against SONFH, specifically focusing on its capacity to restore extracellular redox homeostasis via superoxide dismutase 3 (SOD3).MethodsA rat SONFH model was induced using lipopolysaccharide and methylprednisolone, followed by Ba treatment. Femoral head architecture was evaluated via Micro-CT and histopathology. In vitro, MC3T3-E1 osteoblasts were treated with dexamethasone (Dex) and Ba. Cell viability, intracellular ROS, and superoxide anion levels were assessed. RNA sequencing identified differentially expressed genes, and the mechanism was validated using Sod3 siRNA knockdown and lentiviral overexpression assays.ResultsBa administration significantly attenuated osteonecrosis in vivo, preserving bone mineral density (p < 0.01) and preventing trabecular collapse. In vitro, Ba mitigated Dex-induced osteoblast cytotoxicity and excessive reactive oxygen species (ROS) accumulation. Transcriptomic profiling revealed that Ba predominantly restored antioxidant pathways, notably reversing the Dex-induced downregulation of Sod3. Crucially, targeted functional assays demonstrated that SOD3 is an important mediator of this protection: Sod3 silencing abolished Ba’s therapeutic efficacy, whereas Sod3 overexpression independently conferred profound resistance to oxidative injury.ConclusionBa effectively ameliorates glucocorticoid-induced osteonecrosis by reconstituting the critical extracellular antioxidant barrier, a process closely associated with SOD3 upregulation. These findings establish Ba as a promising natural therapeutic agent and highlight extracellular redox modulation as a novel strategy for SONFH management. Introduction Osteonecrosis of the femoral head (ONFH) is an orthopedic condition notoriously difficult to treat; its progression involves the necrosis of subchondral bone and collapse of the articular cartilage, ultimately leading to the destruction of the hip joint surface.1 Glucocorticoids (GCs) represent the cornerstone of treatment for a wide spectrum of autoimmune and inflammatory disorders, such as systemic lupus erythematosus, nephrotic syndrome, and rheumatoid arthritis. While GCs are also critical in managing acute respiratory crises like severe COVID-19, their clinical utility is inextricably linked to severe adverse effects. Indeed, GC therapy acts as a “double-edged sword”; it remains the leading non-traumatic cause of steroid-induced osteonecrosis of the femoral head (SONFH), placing a heavy burden on patients and healthcare systems.2 The precise mechanisms underlying SONFH remain incompletely elucidated; however, current evidence suggests that osteocyte apoptosis, vascular endothelial cell injury, and oxidative stress contribute to impaired bone repair.3,4 SONFH is a complex focal bone loss disorder; its pathogenesis involves a sophisticated interplay among osteoblast dysfunction, osteoclast overactivation, vascular endothelial impairment, and disruption of the marrow niche. No effective pharmacological intervention for SONFH is currently available in clinical practice. Therefore, the identification of predictive therapeutic targets and the development of novel pharmacological agents are urgently required. Oxidative stress, characterized by an imbalance between the generation of reactive oxygen species (ROS) and antioxidant defense mechanisms, results in cellular damage, mitochondrial dysfunction, and ultimately, tissue necrosis.5,6 Corticosteroids exacerbate oxidative stress by increasing ROS production and impairing the efficacy of the antioxidant system.5 Specifically, steroid-induced overproduction of free radicals oxidizes cellular components such as lipids, proteins, and DNA, thereby triggering inflammatory cascades and apoptosis. In bone cells, this oxidative damage leads to mitochondrial dysfunction, apoptosis, and impaired bone remodeling, all of which contribute to the pathogenesis of osteonecrosis.6 Emerging evidence indicates that oxidative stress plays a pivotal role in the development of SONFH.7 The role of antioxidants in modulating oxidative stress has garnered significant attention in the context of bone health. Superoxide dismutase 3 (SOD3), also known as extracellular superoxide dismutase, localized within the extracellular matrix, plays a crucial role in the antioxidant defense system.8 The superoxide dismutase family comprises SOD1, SOD2, and SOD3, among which only SOD3 is a secreted protein.9 While SOD1 and SOD2 are confined intracellularly, SOD3 is uniquely secreted and anchored to the extracellular matrix.10 This distinct localization is crucial in bone injury; unlike its intracellular counterparts, SOD3 specifically protects the structural integrity of the extracellular matrix and bone marrow niche from ROS-induced fragmentation, thereby maintaining the stable microenvironment essential for bone remodeling. Increased levels of SOD3 have been demonstrated to ameliorate oxidative stress across various tissues.11 This suggests its potential as a therapeutic target for the prevention of SONFH. However, the precise mechanisms by which oxidative stress regulates SOD3 expression in bone cells under the influence of corticosteroids remain incompletely investigated. As an antioxidant enzyme expressed in numerous organs and systems, SOD3 catalyzes the dismutation of superoxide radicals into oxygen and hydrogen peroxide, thereby protecting the liver, vasculature, and other tissues from oxidative damage.12–14 SOD3 effectively mitigates neurovascular injury following cerebral ischemia-reperfusion.15 In oncology, SOD3 is implicated in tumor immunomodulation and has been shown to enhance tumor infiltration by effector T cells.16 Furthermore, SOD3 modulates oxidative stress at the vitreoretinal interface, reducing the incidence of diabetic vitreoretinopathy.17 As aforementioned, SOD3 plays a significant role in the pathophysiology of numerous systems and organs. However, research concerning its role in bone tissue is relatively scarce, with limited reports available. Only a limited number of studies have thus far identified an association between SOD3 and bone metabolic homeostasis.18,19 Nevertheless, whether SOD3 is implicated in SONFH has not yet been subjected to in-depth investigation. The concept of utilizing naturally derived bioactive compounds to modulate oxidative stress for the prevention of SONFH has garnered increasing attention in recent years due to their multi-target efficacy and favorable safety profiles.20 Betulonic acid (Ba), a representative pentacyclic triterpenoid, is a key secondary metabolite enriched in various medicinal plants, including the bark of the white birch tree (Betula platyphylla Suk), which has been historically utilized in traditional remedies for inflammatory conditions.21 Ba has been reported to possess antitumor, anti-inflammatory, antiviral, and antioxidant properties.22,23 Mechanistically, Ba shares a conserved skeleton with other pentacyclic triterpenoids that function as free radical scavengers, contributing to the protection of cells against oxidative damage.24 Given these shared structural features, it is noteworthy that a recent study demonstrated that the analog betulinic acid reduces intracellular ROS levels,25 indicating that Ba may hold similar potential as a therapeutic agent for preventing oxidative stress-induced bone injury. This study aimed to investigate the role of Ba in the process of steroid-induced osteonecrosis of the femoral head and its underlying mechanisms through both in vitro and in vivo approaches. We focused on the mechanism by which Ba modulates oxidative stress and its potential to halt the progression of osteonecrosis. Using a glucocorticoid-induced rat model of ONFH, we found that Ba mitigates GCs-induced oxidative stress damage in bone-forming cells by upregulating SOD3 expression. Crucially, we moved beyond phenotypic observation to mechanistic validation using gene silencing and overexpression techniques. These rigorous gain- and loss-of-function analyses established SOD3 as an important functional mediator required for Ba’s protective efficacy against steroid-induced osteonecrosis. Materials and Methods Animals and Grouping All animal experiments complied with the ethical guidelines outlined in the Guide for the Care and Use of Laboratory Animals and received approval from the Animal Ethics Committee of Shaanxi University of Chinese Medicine. Thirty-two male Sprague-Dawley rats, aged eight weeks and weighing 200–230 g, were sourced from the Animal Center of Xi’an Jiaotong University. These animals were randomly assigned to four experimental groups using a random number table One group underwent sham surgery and functioned as the control; the other three groups were utilized to generate a glucocorticoid-induced model of SONFH.26 The groups were designated as follows: Model group (MPS + PBS), Low-dose Ba treatment group (MPS + Ba Low), and High-dose Ba treatment group (MPS + Ba High). Model Setup and Drug Treatment The rat model of SONFH was generated following a standardized protocol. All animals were maintained in a SPF environment, with controlled temperature (23 ± 2 °C) and a 12-hour light/dark cycle. They had continuous access to food and water and were allowed to move without restriction. Prior to the initiation of pharmacological interventions, body weights of the rats were recorded. The induction regimen commenced with daily intraperitoneal injections of lipopolysaccharide (LPS; 20 μg/kg, Sigma, USA) over three consecutive days. This was followed by daily intramuscular administration of methylprednisolone (MPS; 60 mg/kg, Pfizer, USA) for the subsequent four-day period. LPS was administered to induce a systemic inflammatory and hypercoagulable state, thereby enhancing the susceptibility of bone tissue to corticosteroids and increasing the incidence of osteonecrosis.26 MPS injections were alternated between the left and right gluteal muscles. Sodium penicillin (80,000 U) was administered intraperitoneally twice weekly to prevent infection. The low-dose Ba treatment group received 15 mg/kg Ba (MCE, China) via oral gavage five times per week, while the high-dose Ba treatment group received 30 mg/kg Ba (MCE, China) via oral gavage five times per week.27 The control group received equivalent volumes of normal saline. At the 5-week endpoint, euthanasia was performed via an intraperitoneal overdose of Pentobarbital Sodium (150 mg/kg). Sample collection proceeded only after the absence of respiration and heartbeat was verified. Micro-CT Analysis To evaluate imaging changes, high-resolution Micro-CT (Bruker AXS GmbH, Karlsruhe, Germany) was employed to assess bone mineral density, bone mass, and trabecular microstructure at the femoral head region.28 Parameters derived from these data included trabecular bone mineral density (BMD), trabecular number (Tb.N, 1/mm), trabecular thickness (Tb.Th, mm), trabecular separation (Tb.Sp, mm), and bone volume to tissue volume ratio (BV/TV, %). Scanning parameters were set as follows: current 400 µA, voltage 60 kV, 360-degree rotation scanning with an angular increment of 0.9°, and a scanning time of approximately 20–30 minutes per sample. The entire femur was scanned, and three-dimensional images were reconstructed at an isotropic resolution of 12 µm, with a minimum of 80 slices analyzed in the selected region of the subchondral bone in the femoral head. These tomographic images were integrated into three-dimensional reconstructions using MicroView V.2.1 software. Histological and H&E Staining Femoral head specimens were fixed in 4% paraformaldehyde for 48 hours and decalcified in 10% EDTA (Servicebio, China) for 4 weeks. The samples were subsequently embedded in paraffin and sectioned into 6 μm-thick slices. H&E staining was performed to examine the overall architecture and evaluate the trabecular bone structure. Quantitative assessments were conducted by two independent observers. The empty lacunae rate was calculated as the ratio of empty lacunae to the total number of osteocytes within the trabecular bone across at least three randomly selected fields per sample. IHC Staining, TUNEL Staining and IF Staining IHC staining was performed to determine the expression levels of the bone marker osteocalcin (OCN). Briefly, the paraffin-embedded sections were deparaffinized in xylene and rehydrated through a graded ethanol series. Antigen retrieval was performed in citrate buffer using a pressure cooker. After quenching endogenous peroxidase activity with 3% hydrogen peroxide and blocking with 5% bovine serum albumin for 1 hour at room temperature, the sections were incubated with a primary antibody against OCN (1:400 dilution, Cat. No. 23418-1-AP, Proteintech, China) at 4°C overnight. Subsequently, the sections were incubated with a biotin-conjugated secondary antibody, followed by incubation with HRP-labeled streptavidin (Maixin Biotechnologies, China). Chromogenic development was induced using a DAB chromogen kit (ZSGB-BIO, China), and the nuclei were counterstained with hematoxylin. Quantitative assessments of the stained sections were performed by two independent observers. Cellular apoptosis was assessed quantitatively using a TUNEL assay kit, which labels DNA fragmentation. Following permeabilization and damage repair, the prepared reagents were applied to the samples. Cell nuclei were then counterstained with DAPI during a 10-minute incubation in the dark. Finally, sections were coverslipped using an anti-fade mounting medium. Tissue sections were examined under a fluorescence microscope, and the fluorescence intensity of TUNEL staining was quantified using ImageJ software (NIH, USA). IF staining was employed to elucidate the expression levels and localization of SOD3. Tissue sections were incubated with a primary antibody against SOD3 (1:150 dilution, Cat. No. DF7753, Affinity Biosciences, China) at 4°C overnight. Following PBS washes, the sections were incubated with a corresponding fluorescent secondary antibody, Anti-Rabbit IgG (Alexa Fluor® 594 Conjugate) (1:1000 dilution, Cat. No. #8889, Cell Signaling Technology, USA), for 1 hour at room temperature in the dark. Tissue sections were examined under a fluorescence microscope, and the SOD3 fluorescence intensity was quantitatively analyzed using ImageJ software. Cell Culture and Cell Viability Assays Mouse osteoblastic MC3T3-E1 cells (Shanghai ZhongQiaoXinZhou Biotechnology, China) were cultured in MEM-α medium (Gibco, USA) supplemented with 10% fetal bovine serum (Gibco, USA). Cells were maintained in a humidified incubator at 37°C with 5% CO2. Cell viability of dexamethasone (Dex)-treated MC3T3-E1 cells was assessed using the CCK-8 assay (Beyotime, China). MC3T3-E1 cells were seeded into 96-well plates at a density of 2,500 cells/well. To evaluate the protective effect of Ba, the cells were divided into three experimental groups: (1) Control group: cultured in standard medium without any treatment; (2) Dex group: treated with 100 μM Dex (MCE, China) for 24 hours; (3) Dex + Ba group: pre-treated with varying concentrations of Ba (5 or 10 μM) for 1 hour, followed by co-incubation with 100 μM Dex for 24 hours. ROS The culture medium from each group of cells was aspirated, and the cells were washed twice with PBS. After aspirating the PBS, an appropriate volume of DCFH-DA working solution was added, and the cells were incubated at 37°C with 5% CO2 in a light-protected environment for 20 minutes. Following three additional washes with PBS, the cells were observed under a fluorescence microscope. Fluorescence intensity was quantified using ImageJ software (NIH, USA) by calculating the mean fluorescence density across at least three randomly selected fields. Superoxide Anion Active Oxygen Detection The culture medium was aspirated from each group of cells, followed by a single wash with PBS. After aspirating the PBS, an appropriate volume of DHE staining solution was added, and the cells were incubated at 37°C with 5% CO2 in a light-protected environment for 30 minutes. The cells were then washed twice with PBS and subsequently observed under a fluorescence microscope. Fluorescence intensity was quantified using ImageJ software (NIH, USA) by calculating the mean fluorescence density across at least three randomly selected fields. RNA Extraction The culture medium was completely aspirated, and Trizol reagent was added, followed by incubation at room temperature for 5 minutes to ensure complete sample lysis. Chloroform was added, and after vigorous shaking, the mixture was incubated at room temperature for 2–3 minutes. Centrifugation was performed at 12,000 × g and 4°C for 15 minutes, resulting in visible phase separation. The upper aqueous phase containing total RNA was carefully transferred to a new centrifuge tube, and anhydrous isopropanol was added, mixed thoroughly, and precipitated at room temperature for 10 minutes. The sample was centrifuged at 12,000 × g and 4°C for 10 minutes, revealing an RNA pellet at the bottom of the tube; the supernatant was discarded, and wash buffer was added and mixed, followed by centrifugation at 7,500 × g and 4°C for 5 minutes, after which the supernatant was again discarded. A brief centrifugation step was performed to pellet any remaining liquid, which was then carefully aspirated. DEPC-treated water was added to dissolve the RNA. RNA-Seq Analysis Three experimental groups were established: a normal control group (NC), a Dex-treated group (DEX), and a Dex plus Ba co-treatment group (DEX+BA). Each group was set up with three biological replicates. Total RNA was extracted from the cells of each group using TRIzol reagent, followed by rigorous quality control assessment of RNA integrity, purity, and concentration. RNA samples passing the quality control were used for subsequent library construction. High-throughput sequencing of multiple samples was performed using the paired-end sequencing mode on the Illumina NovaSeq 6000 platform. Identification of DEGs and Functional Enrichment Analysis Differentially expressed genes (DEGs) were identified using the DESeq2 package in R software. We compared the gene expression profiles between the DEX group and the NC group, as well as between the DEX_BA group and the DEX group. Genes meeting the thresholds of |log2Fold Change| ≥ 0.5 and an adjusted P-value ≤ 0.05, based on the Benjamini-Hochberg method for multiple testing correction, were defined as statistically significant DEGs. Concurrently, to systematically interpret the biological functions of the DEGs at the pathway level, we employed Gene Set Variation Analysis (GSVA). Enrichment scores for specific pathways within the KEGG gene sets were calculated to evaluate the activity changes of these pathways across different samples. Molecular Docking Analysis To investigate the direct interaction between Ba and the SOD3 protein, molecular docking was performed. The 2D structure of the Ba ligand was obtained from the PubChem database (http://pubchem.ncbi.nlm.nih.gov/). This 2D structure was then imported into ChemOffice 20.0 software to generate its 3D conformation and saved as a mol2 file. The target protein sequence of SOD3 (UniProtKB: O09164) was retrieved from the UniProt database (https://www.uniprot.org/), and its 3D protein crystal model was constructed using the AlphaFold Protein Structure Database (https://alphafold.ebi.ac.uk) and saved in PDB format. Molecular Operating Environment (MOE) 2019 software was utilized to perform energy minimization for the compound, pre-process the target protein, and identify the active binding pockets. Finally, molecular docking was executed using MOE 2019, with the number of computational runs set to 50. The binding affinity was evaluated based on the binding energy score. Visualization of the docking results, including the spatial conformation and specific interactions, was processed using PyMOL 2.6.0 and Discovery Studio 2019 software. siRNA Transfection Specific small interfering RNAs against Sod3 and a scrambled negative control siRNA were sourced from GenePharma (GenePharma, China). Cells cultured in 6-well plates were transfected with the respective siRNAs at an optimal concentration of 50 nM using Lipofectamine 2000 in Opti-MEM medium. Cells were harvested 48 hours post-transfection for subsequent experiments. The target sequences were as follows: Sod3_#1 Sense (5’-3’): CCAGACAAAGGAGCGCAAGTT, Antisense (5’-3’): CUUGCGCUCCUUUGUCUGGTT; Sod3_#2 Sense (5’-3’): GCUUGAGGCCUAUUUCAGUTT, Antisense (5’-3’): ACUGAAAUAGGCCUCAAGCTT. Lentiviral Transfection A stable MC3T3-E1 cell line overexpressing Sod3 was established using lentiviral transduction. The mouse Sod3 gene was cloned into a Lenti-SFFV-3×Flag-IRES-PURO-WPRE vector. Lentiviral particles were packaged in HEK293T cells using the psPAX2 and pMD2.G system. MC3T3-E1 cells were transduced with the purified lentivirus and selected with puromycin (2 μg/mL) to obtain stable pools. Quantitative Real-Time PCR cDNA was synthesized from total RNA using the PrimeScript RT Master Mix (TaKaRa, Japan). Quantitative PCR was performed on a LightCycler 96 system (Bio-Rad, USA) using TB Green Premix Ex Taq II (TaKaRa). Each 20 μL reaction contained 10 μL of Premix, 0.8 μL each of forward and reverse primers (10 μM), 0.4 μL of ROX Reference Dye, 2 μL of cDNA, and nuclease-free water. The following primers were used: Mouse Sod3: Forward: 5’-TGCTGCTCGCTCACATAACA-3’, Reverse: 5’-GAGGTTCTCTGCACCTGTCA-3’ Mouse Gapdh: Forward: 5’-AGGTCGGTGTGAACGGATTTG-3’, Reverse: 5’-GGGGTCGTTGATGGCAACA-3’ Western Blot Analysis The protein expression level of SOD3 was detected using Western blot analysis. Proteins (20 μg per lane) were separated on 10% SDS-PAGE gels and subsequently transferred onto PVDF membranes. After blocking with 5% skim milk for 1 hour at room temperature, the membranes were incubated with specific primary antibodies at 4°C overnight. The primary antibodies used were rabbit monoclonal anti-SOD3 (1:1000 dilution, Cat. No. #24000, Cell Signaling Technology, USA) and rabbit polyclonal anti-β-actin (1:5000 dilution, Cat. No. 20536-1-AP, Proteintech, China), which served as the internal loading control. Subsequently, the membranes were incubated with an HRP-conjugated Goat Anti-Rabbit IgG (H+L) secondary antibody (1:5000 dilution, Cat. No. SA00001-2, Proteintech, China) at room temperature for 1 hour. Finally, protein bands were visualized using a Bio-Rad gel imaging system (Chemidoc XRS+, USA). The gray values of the protein bands were quantified using ImageJ software (NIH, USA) and normalized to β-actin. Western blot analyses were performed in triplicate independently. Statistical Analysis Statistical analysis was performed using GraphPad Prism 7.0 software (GraphPad, USA), and data are presented as mean ± standard deviation (SD). The independent sample size “n” explicitly denotes the number of true independent biological replicates, representing independent animal subjects for in vivo assays or independent cell experiments performed on separate days for in vitro assays. For specific in vitro and imaging analyses utilizing a limited sample size (n = 3), experiments were rigorously controlled to minimize intra-group variance; nevertheless, these particular datasets should be interpreted as supportive and exploratory rather than definitive. For all histological and immunofluorescence imaging quantifications, multiple microscopic fields were captured per sample and averaged to produce a single value per biological replicate. Before conducting parametric analyses, the assumption of normality was rigorously verified using the Shapiro–Wilk test, and the homogeneity of variance was assessed using the Brown-Forsythe test. For datasets meeting these criteria, statistical significance among multiple groups was determined using ordinary one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test for multiple comparisons. Exact p-values were computed by the software, with p < 0.05 considered statistically significant. Results Ba Attenuates MPS-Induced ONFH in Rats The administration of Ba markedly mitigated MPS-induced osteonecrosis in the femoral head of rats. In this study, an established rat model of ONFH was subjected to micro-CT scanning, enabling visualization and subsequent morphological quantification of both cortical and trabecular bone structures. The control group exhibited a smooth and uniformly colored cartilage surface on the femoral head, whereas the model group displayed discoloration and surface irregularities in the weight-bearing region. However, Ba treatment reduced lesion severity (Figure 1a), with a pronounced alleviation observed particularly at high doses. Three-dimensional reconstructions from sagittal, coronal, and transverse views, along with 3D models (Figure 1a), revealed severe trabecular bone loss and collapse in the weight-bearing area of the MPS + PBS group. Ba treatment significantly attenuated the aforementioned femoral head damage. Micro-CT analysis of the region of interest (ROI) in the weight-bearing area demonstrated that the mean bone mineral density (BMD) in the MPS group was significantly lower than that in the control group. Further analysis indicated that Ba treatment enhanced BMD, bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular number (Tb.N). Compared to the model group, trabecular separation (Tb.Sp) and bone surface to bone volume ratio (BS/BV) were significantly decreased (Figure 1b–g). Figure 1Ba reduced SONFH in a rat model. (a) Micro-CT images. The red highlighted area represents the ROI within the subchondral bone selected for quantitative analysis. (b–g) Quantitative analysis of BMD, BV/TV, BS/BV, Tb.Th, Tb.Sp, and Tb.N. Ba: Betulonic acid; MPS: Methylprednisolone; PBS: Phosphate-Buffered Saline; BMD: Bone Mineral Density; BV/TV: bone volume/tissue volume; BS/BV: bone surface/bone volume; Tb.Th: trabecular thickness (mm); Tb.Sp: trabecular separation (mm); Tb.N: trabecular number (1/mm). (Mean ± SD, n = 8 biological replicates. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01, *p < 0.05).The image A shows micro-CT scans of rat femoral heads under different treatments: Control, MPS+PBS, MPS+Ba Low, and MPS+Ba High. Views include coronal (COR), transverse (TRA), sagittal (SAG) and 3D reconstructions. The image B shows a bar graph of Bone Mineral Density (BMD) in mg per cm cubed, comparing Control, MPS plus PBS, MPS plus Ba Low and MPS plus Ba High groups. The image C shows a bar graph of bone volume/tissue volume (BV/TV) in percent, comparing the same groups. The image D shows a bar graph of bone surface/bone volume (BS/BV) in mm superscript -1. The image E shows a bar graph of trabecular thickness (Tb. Th) in mm. The image F shows a bar graph of trabecular separation (Tb. Sp) in mm. The image G shows a bar graph of trabecular number (Tb. N) in mm superscript -1. Statistical significance is indicated with asterisks: p < 0.001, p < 0.01, p < 0.05.Micro-CT images and graphs show Ba′s effect on SONFH in rats, analyzing BMD, BV/TV, BS/BV, Tb. Th, Tb. Sp and Tb. N. Ba reduced SONFH in a rat model. (a) Micro-CT images. The red highlighted area represents the ROI within the subchondral bone selected for quantitative analysis. (b–g) Quantitative analysis of BMD, BV/TV, BS/BV, Tb.Th, Tb.Sp, and Tb.N. Ba: Betulonic acid; MPS: Methylprednisolone; PBS: Phosphate-Buffered Saline; BMD: Bone Mineral Density; BV/TV: bone volume/tissue volume; BS/BV: bone surface/bone volume; Tb.Th: trabecular thickness (mm); Tb.Sp: trabecular separation (mm); Tb.N: trabecular number (1/mm). (Mean ± SD, n = 8 biological replicates. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01, *p < 0.05). Histological analysis further confirmed the protective effect of Ba against MPS-induced osteocyte impairment. In H&E staining, the MPS + PBS group exhibited sparse and collapsed trabeculae along with a higher number of empty lacunae. In contrast, the Ba-treated groups showed minimal bone necrosis, less replacement of trabeculae by necrotic tissue, and a significantly reduced empty lacunae rate. The H&E staining results were consistent with the micro-CT findings, indicating that Ba confers protection against MPS-induced trabecular bone loss in ONFH (Figure 2a and b). Decalcification can induce a certain number of empty lacunae, albeit at a low incidence rate. Typically, decalcification of normal bone tissue results in no more than 10% empty lacunae.29 TUNEL staining for detecting apoptosis in femoral head tissue revealed the lowest TUNEL fluorescence intensity in the control group. The TUNEL intensity was significantly higher in the MPS group compared to the control, whereas it was markedly lower in the Ba-treated groups relative to the MPS group. These results demonstrate that Ba attenuates MPS-induced apoptosis in rat femoral head cells (Figure 2c and d). Bone histomorphometric analysis indicated that Ba administration helped maintain bone mass compared to the MPS + PBS group. Immunohistochemical staining for the osteoblast biomarker OCN yielded consistent results (Figure 2e and f), aligning with the micro-CT data. Figure 2Histological analysis of Ba alleviating SONFH in rats. (a and b) H&E-stained sections from the respective experimental groups. (c and d) TUNEL staining and quantitative analysis of TUNEL intensity. (e and f) Immunohistochemical assay of OCN in the femoral head of the indicated group. (Mean ± SD, n = 3 biological replicates. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01). Note: A baseline of up to 10% empty lacunae was considered an acceptable artifact secondary to the decalcification process, which was accounted for during histological evaluation.Image A displays H&E sections for the Control, MPS+PBS, MPS+Ba Low, and MPS+Ba High groups. Each group includes a low magnification view and two higher magnification views. Image B is a bar chart titled ′H&E′ showing the empty lacunae ratio: Control (~2%), MPS+PBS (~45-50%), MPS+Ba Low (~20%), MPS+Ba High (~8-10%). A bracket with three asterisks highlights differences from MPS+PBS to MPS+Ba High. Image C presents TUNEL microscopy images for the same groups. Image D is a bar chart showing TUNEL intensity: Control (~1.1), MPS+PBS (~3.5), MPS+Ba Low (~3.0), MPS+Ba High (~2.3). A bracket with three asterisks spans Control to MPS+PBS and two asterisks span MPS+Ba Low to MPS+Ba High. Image E shows OCN immunohistochemical assay images for the groups. Image F is a bar chart showing OCN area occupied: Control (~48-50%), MPS+PBS (~22-24%), MPS+Ba Low (~35-37%), MPS+Ba High (~38-40%). Two brackets with three asterisks compare MPS+PBS with MPS+Ba Low and MPS+Ba High.Infographic: Ba reduces empty lacunae and TUNEL intensity, and increases OCN expression in rat femoral head histology. Histological analysis of Ba alleviating SONFH in rats. (a and b) H&E-stained sections from the respective experimental groups. (c and d) TUNEL staining and quantitative analysis of TUNEL intensity. (e and f) Immunohistochemical assay of OCN in the femoral head of the indicated group. (Mean ± SD, n = 3 biological replicates. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01). Note: A baseline of up to 10% empty lacunae was considered an acceptable artifact secondary to the decalcification process, which was accounted for during histological evaluation. Protective Effect of Ba Against Dex-Induced Osteoblast Injury To investigate the therapeutic effect of Ba (Figure 3a), we employed the CCK-8 assay to evaluate its efficacy in mitigating Dex-induced osteoblast injury. Results showed that multiple concentrations of Ba (5 μM, 10 μM) alleviated the damage caused by dexamethasone pretreatment in osteoblasts (Figure 3b). Compared to the model group, the 10 μM Ba treatment significantly enhanced osteoblast viability. Consequently, the concentration of 10 μM was identified as the optimal dosage for Ba. These findings indicate that Ba can effectively reduce Dex-induced damage in osteoblasts. Figure 3Ba alleviates Dex-induced osteoblast injury and RNA-seq analysis. (a) The chemical structure of Ba. (b) CCK-8 assay results showed that various Ba concentrations inhibited the dexamethasone-induced decrease in osteoblast viability. (c) RNA-seq workflow diagram. (d) Volcano plot showing DEGs between the control and Dex groups. (e) Volcano plot showing DEGs between the Dex and Dex+Ba groups. (f) GSVA heatmap illustrating the pathways significantly altered by Dex and reversed by Ba treatment, including oxidative stress-related pathways. (g) Molecular docking simulation illustrating the binding mode and interaction between Ba and the SOD3 protein. (Mean ± SD; n = 6 biological replicates for CCK-8, n = 3 for others. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001).Image A displays Ba′s chemical structure in 2D and 3D models. Image B features a bar chart titled ′Cell Counting Kit-8′ with cell viability data: Control (~100), Dex (~50), Dex + Ba 5 microM (~60), Dex + Ba 10 microM (~70), with asterisks indicating comparisons. Image C illustrates an RNA-seq workflow with a culture dish, pipette, tube, circular RNA icon and sequencing machine. Image D shows a volcano plot ′DEX vs NC′ with axes for Log2 Fold Change and -Log10 P-value. Image E presents a volcano plot ′DEX+BA vs DEX′ with a labeled point, Sod3. Image F displays a heatmap ′GSVA Enrichment Variable Pathways′ with groups NC, DEX, DEX_BA and pathways: Necroptosis, Ferroptosis, Pentose phosphate pathway, Peroxisome, Glutathione metabolism. Image G shows molecular docking of Ba with SOD3, highlighting residues ARG-174, LEU-177, THR-178, ARG-106.Multi-panel infographic on Ba: CCK-8 viability, RNA-seq, DEGs, GSVA pathways and SOD3 docking. Ba alleviates Dex-induced osteoblast injury and RNA-seq analysis. (a) The chemical structure of Ba. (b) CCK-8 assay results showed that various Ba concentrations inhibited the dexamethasone-induced decrease in osteoblast viability. (c) RNA-seq workflow diagram. (d) Volcano plot showing DEGs between the control and Dex groups. (e) Volcano plot showing DEGs between the Dex and Dex+Ba groups. (f) GSVA heatmap illustrating the pathways significantly altered by Dex and reversed by Ba treatment, including oxidative stress-related pathways. (g) Molecular docking simulation illustrating the binding mode and interaction between Ba and the SOD3 protein. (Mean ± SD; n = 6 biological replicates for CCK-8, n = 3 for others. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001). RNA-Seq Analysis of the Protective Effects of Ba To comprehensively analyze the mechanism of Ba in treating SONFH, we performed RNA-seq (Figure 3c). RNA-seq data revealed 7,743 DEGs in the Dex group compared to the control group, comprising 3,398 significantly upregulated and 4,345 significantly downregulated genes (Figure 3d). Comparison between the Dex+Ba group and the Dex group identified 679 genes with significantly altered expression, including 286 significantly upregulated and 393 significantly downregulated genes (Figure 3e). To gain deeper insight into the biological pathways underlying these changes, we performed GSVA. This pathway-centric analysis revealed that Dex significantly disrupted multiple metabolic and stress response pathways compared to the control group. Notably, Ba treatment significantly reversed the activity of several key pathways related to oxidative stress homeostasis, including the pentose phosphate pathway, peroxisome, and glutathione metabolism (Figure 3f). This pathway-level reversal strongly suggests that the therapeutic effect of Ba is closely associated with the restoration of redox homeostasis. Our analysis of individual DEGs confirmed that Ba effectively reversed the aberrant expression of core antioxidant genes, such as Sod3, Aldh3a1, Gsta4, and Cyb5r3, induced by Dex. This demonstrates the potential efficacy of Ba in treating steroid-induced osteonecrosis by suppressing oxidative stress. It is noteworthy that Sod3 gene expression was significantly downregulated in the Dex group but was upregulated following Ba treatment. This key finding indicates that the induction of SOD3 expression may be a central event in the therapeutic mechanism of Ba. To further validate this hypothesis and explore the direct interaction between Ba and the SOD3 protein, molecular docking analysis was conducted (Figure 3g). The simulation results revealed that Ba successfully docked into the active binding domain of the SOD3 protein with an estimated binding energy of −5.5570 kcal/mol. Generally, a binding energy lower than −5.0 kcal/mol is considered to indicate good binding affinity. Specifically, structural analysis of the binding complex demonstrated that Ba formed stable hydrogen bonds with the amino acid residues Arg106, Thr178, and Arg174 of the SOD3 protein. Furthermore, the residue Leu177 formed carbon-hydrogen bonds and hydrophobic interactions with the Ba molecule. These robust computational findings provide crucial structural insights, demonstrating that Ba exhibits a strong and stable binding affinity with SOD3, thereby supporting its role as a key functional target. Based on this transcriptomic evidence, we subsequently detected SOD3 protein expression in rat femoral head tissue using IF staining (Figure 4a–d). The results corroborated the RNA-seq findings, showing the lowest SOD3 fluorescence intensity in the Dex group, while the Ba group exhibited a significant increase, confirming the upregulation of SOD3 at the protein level. Figure 4Ba upregulates SOD3 expression to alleviate Dex-induced oxidative stress injury. (a) In situ IF staining of SOD3 in rat femoral heads. (b) In situ ROS staining. (c) In situ superoxide anion staining using DHE. (d) Quantitative analysis of SOD3 intensity. (e) Quantitative analysis of ROS mean density. (f) Quantitative analysis of DHE red density. (Mean ± SD, n = 3 biological replicates. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01, *p < 0.05).The figure consists of six panels labeled a to f. Panel a shows in situ IF staining of SOD3 in rat femoral heads under Control, MPS+PBS, MPS+Ba Low, and MPS+Ba High conditions, with blue indicating nuclei and red indicating SOD3. Panel b displays ROS staining with bright field, ROS and Merge views for Control, Dex and Dex plus Ba, where green indicates ROS presence. Panel c presents DAPI and DHE staining with a Merge view for Control, Dex and Dex plus Ba, with blue for nuclei and red for DHE. Panel d is a bar graph showing SOD3 intensity, with significant increases in MPS+Ba High compared to MPS+PBS. Panel e shows ROS mean density, significantly higher in Dex compared to Control, reduced with Dex plus Ba. Panel f illustrates DHE Red density, higher in Dex, reduced with Dex plus Ba. Ba treatment mitigates Dex-induced oxidative stress, increasing SOD3 and reducing ROS and DHE signals.Multi-panel figure showing SOD3, ROS and DHE changes in rat femoral heads with Dex and Ba treatments. Ba upregulates SOD3 expression to alleviate Dex-induced oxidative stress injury. (a) In situ IF staining of SOD3 in rat femoral heads. (b) In situ ROS staining. (c) In situ superoxide anion staining using DHE. (d) Quantitative analysis of SOD3 intensity. (e) Quantitative analysis of ROS mean density. (f) Quantitative analysis of DHE red density. (Mean ± SD, n = 3 biological replicates. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01, *p < 0.05). Protective Effect of Ba Against Dex-Induced Oxidative Stress Injury in Osteoblasts Glucocorticoid-induced osteoblast injury is recognized as a critical factor in ONFH. Dex is known to trigger oxidative stress, which subsequently leads to osteoblast dysfunction and the development of osteoporosis.30 The regulatory effect of Ba on oxidative stress was assessed using an intracellular ROS staining assay. Since excessive intracellular ROS levels can induce severe cytotoxicity, we measured total ROS to evaluate intracellular ROS generation (Figure 4b–e). Cells treated with Dex exhibited stronger fluorescence intensity, reflecting excessive ROS production induced by Dex. The results showed a significant increase in ROS levels in osteoblasts from the Dex group, indicating that Dex induced a pronounced oxidative stress response. However, ROS levels were markedly reduced in the Dex + Ba treatment group, demonstrating that Ba significantly alleviates the oxidative damage caused by dexamethasone. Concurrently, a superoxide anion assay was performed. The results indicated that while Dex enhanced ethidium (EDH) fluorescence intensity, Ba significantly attenuated the Dex-induced monomer aggregation in the Dex+Ba group compared to the Dex group (Figure 4c–f). This suggests that Ba can partially reverse the intracellular superoxide anion levels elevated by Dex. These results collectively indicate that Ba possesses antioxidant properties and can effectively mitigate Dex-induced oxidative stress injury in osteoblasts. Ba Protects Against Dex-Induced Oxidative Stress Injury in Osteoblasts by Regulating SOD3 Sod3 gene expression was successfully knocked down in osteoblasts using siRNA technology (Figure 5a–c). SOD3 is a crucial antioxidant enzyme that scavenges superoxide anions, thereby protecting cells from oxidative stress damage. Results showed that compared to the Dex+Ba+siNC group, the reversal effect of Ba on the Dex-induced decrease in cell viability was attenuated following Sod3 knockdown (Figure 5d). Concurrently, the inhibitory effect of Ba on Dex-induced ROS levels was significantly weakened (Figure 5e and f). This finding further underscores the critical role of SOD3 in the mechanism by which Ba mitigates dexamethasone-induced cellular damage via the regulation of oxidative stress pathways. Furthermore, superoxide anion detection assays across different treatment groups revealed that the Dex+Ba+siSod3 group exhibited significantly stronger red fluorescence intensity than the Dex+Ba+siNC group. This indicates that the efficacy of Ba in suppressing the Dex-induced increase in intracellular superoxide anion levels was impaired (Figure 5g and h). These results demonstrate that Ba mitigates oxidative damage by regulating SOD3 to enhance the clearance of intracellular superoxide anions (O2·−). Figure 5Sod3 silencing attenuates the protective effects of Ba on osteoblasts. (a) Protein expression levels of SOD3. (b) Sod3 mRNA levels detected by qPCR. (c) Relative protein levels of SOD3. (d) Cell viability assessed by CCK-8 assay. (e) In situ ROS staining. (f) Quantitative analysis of ROS mean density. (g) In situ DHE staining. (h) Quantitative analysis of DHE red density. (Mean ± SD; n = 6 biological replicates for CCK-8, n = 3 for others. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01, *p < 0.05).Image A shows protein expression levels of SOD3 and beta-actin under control, siSod3 #1 and siSod3 #2 conditions. Image B depicts a bar graph of Sod3 mRNA levels, showing fold change compared to control, with significant reduction in siSod3 #1 and #2. Image C presents relative protein levels of SOD3, indicating a decrease in siSod3 #1 and #2. Image D displays a bar graph of MC3T3-E1 cell viability assessed by Cell Counting Kit-8, showing reduced viability in Dex and siSod3 conditions. Image E shows bright field (BF), ROS staining and merged images for Dex+siNC, Dex+Ba+siNC, Dex+Ba+siSod3 #1 and Dex+Ba+siSod3 #2. Image F illustrates a bar graph of ROS mean density, showing increased levels in Dex+siNC compared to other groups. Image G shows DAPI, DHE staining and merged images for different conditions. Image H presents a bar graph of DHE red density, indicating higher levels in siSod3 groups compared to Dex+Ba+siNC.Sod3 silencing impacts osteoblasts: protein, mRNA, cell viability, ROS, DHE analysis. Sod3 silencing attenuates the protective effects of Ba on osteoblasts. (a) Protein expression levels of SOD3. (b) Sod3 mRNA levels detected by qPCR. (c) Relative protein levels of SOD3. (d) Cell viability assessed by CCK-8 assay. (e) In situ ROS staining. (f) Quantitative analysis of ROS mean density. (g) In situ DHE staining. (h) Quantitative analysis of DHE red density. (Mean ± SD; n = 6 biological replicates for CCK-8, n = 3 for others. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, **p < 0.01, *p < 0.05). SOD3 Overexpression Alleviates Dex-Induced Oxidative Stress Injury in Osteoblasts To further investigate the role of SOD3 in the protective mechanism of Ba against Dex-induced osteoblast injury via oxidative stress regulation, we first successfully overexpressed the Sod3 gene using lentiviral transfection (Figure 6a-c). Following successful transfection, osteoblasts were subjected to Dex treatment and subsequently evaluated using intracellular ROS staining and superoxide detection assays to evaluate the impact of Sod3 overexpression on the oxidative stress response. Sod3 overexpression effectively mitigated the Dex-induced impairment of osteoblast viability (Figure 6d). Intracellular ROS staining revealed a significant increase in ROS levels in Dex-treated control osteoblasts, indicating that Dex strongly induced an oxidative stress response. In contrast, the Sod3 overexpression group exhibited a marked decrease in ROS levels, demonstrating that Sod3 overexpression effectively alleviates Dex-induced oxidative stress damage in osteoblasts (Figure 6e and f). This finding further supports the critical role of SOD3 in counteracting oxidative stress. Using a superoxide detection assay, we also observed a significant reduction in superoxide accumulation in the Sod3 overexpression group (Figure 6g and h), further confirming its antioxidant capacity. Collectively, these experimental results demonstrate that Sod3 overexpression significantly alleviates Dex-induced oxidative stress injury, reduces ROS levels, diminishes superoxide accumulation, and restores the levels of antioxidant markers to normal. This discovery suggests that SOD3, as a key antioxidant enzyme, holds potential therapeutic value for the prevention and treatment of steroid-induced osteonecrosis of the femoral head. Figure 6Sod3 overexpression enhances the protective effect against dexamethasone-induced osteoblast injury. (a) Protein expression levels of SOD3. (b) Sod3 mRNA levels detected by qPCR. (c) Relative protein levels of SOD3. (d) Cell viability assessed by CCK-8 assay. (e) In situ ROS staining. (f) Quantitative analysis of ROS mean density. (g) In situ DHE staining. (h) Quantitative analysis of DHE red density. (Mean ± SD; n = 6 biological replicates for CCK-8, n = 3 for others. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, *p < 0.05).Image A: Immunoblot shows stronger SOD3 band in Sod3 OE compared to Control. Image B: Bar chart of Sod3 mRNA levels shows higher levels in Sod3 OE than Control, marked with ***. Image C: Bar chart of SOD3 protein levels shows higher levels in Sod3 OE than Control, marked with ***. Image D: Cell viability chart shows Dex lower than Control; Sod3 OE plus Dex higher than Dex but lower than Sod3 OE, marked with ***. Image E: Microscopy shows more ROS puncta in Dex than Control, fewer in Sod3 OE and Sod3 OE plus Dex. Image F: ROS mean density chart shows Dex highest; others lower, marked with ***. Image G: Microscopy shows more DHE puncta in Dex than Control, fewer in Sod3 OE and Sod3 OE plus Dex. Image H: DHE red density chart shows Dex highest, marked with *.Composite infographic: Sod3 OE raises SOD3 and reduces Dex-linked viability loss and ROS and DHE signals. Sod3 overexpression enhances the protective effect against dexamethasone-induced osteoblast injury. (a) Protein expression levels of SOD3. (b) Sod3 mRNA levels detected by qPCR. (c) Relative protein levels of SOD3. (d) Cell viability assessed by CCK-8 assay. (e) In situ ROS staining. (f) Quantitative analysis of ROS mean density. (g) In situ DHE staining. (h) Quantitative analysis of DHE red density. (Mean ± SD; n = 6 biological replicates for CCK-8, n = 3 for others. One-way ANOVA with Tukey’s post-hoc test. ***p < 0.001, *p < 0.05). Discussion It is well established that SONFH leads to hip dysfunction and severe disability, imposing a significant burden on patients and society. Reports indicate that 24.1% of ONFH cases are associated with GC therapy.31 Accumulating evidence suggests that GCs can reduce bone mineral density and are directly or indirectly linked to trabecular abnormalities and replacement by necrotic tissue, which lacks mechanical strength. The accumulation of bone metabolic disorders due to mechanical stress overload may ultimately lead to femoral head collapse.3 Given the irreversible nature of joint collapse, investigating early-stage therapeutic interventions appears to hold greater clinical value for patients with early-stage SONFH. Against this backdrop, natural medicines have garnered widespread attention due to their multi-target effects and high biocompatibility. Plant extracts and natural compounds have been confirmed to possess diverse biological activities, including antioxidant, anti-inflammatory, and anti-apoptotic properties, demonstrating significant potential in disease prevention and treatment.32,33 Following this research direction, this study is the first to systematically investigate the protective effects and underlying mechanisms of the natural compound Ba in SONFH. Through in vivo and in vitro experiments, this study confirms that Ba significantly alleviates SONFH by upregulating SOD3 expression, with its core mechanism being the inhibition of oxidative stress damage. Our findings not only reveal the potential of Ba, a natural triterpenoid, in the prevention and treatment of SONFH but also, for the first time, establish SOD3 as the key molecular target through which Ba exerts its bone-protective effects. This provides new experimental evidence for the clinical development of targeted antioxidant therapeutic strategies against SONFH. To date, the MPS-induced rat model is considered an ideal preclinical model for simulating GC-induced SONFH.26,34 In this study, an end-stage SONFH pathological model was established. Ba intervention was found to significantly mitigate SONFH in rats, notably reducing empty lacunae, preventing bone collapse, and lessening structural damage in the femoral head. SONFH patients exhibit substantially lowered bone mineral density, bone volume, and trabecular thickness in areas of collapse. The impairment of osteoblast function induced by glucocorticoids, which directly affects bone quality, represents a major target in SONFH treatment research. GCs can promote apoptosis and inhibit osteoblast metabolism, which are common features in SONFH.34 Our TUNEL staining results also showed that Ba significantly reduced osteocyte apoptosis, further suggesting its potential protective role in steroid-induced ONFH. Excessive use of glucocorticoids induces osteocyte death and degenerative changes in bone structure. To further validate the effect of Ba on oxidative stress in osteoblasts, we conducted cytological experiments using MC3T3-E1 osteoblastic cells. CCK-8 viability assays demonstrated that Ba alleviated the Dex-induced reduction in osteoblast viability. Furthermore, Ba possesses antioxidant properties that help maintain cellular homeostasis, potentially underpinning its ability to improve osteoblast viability during SONFH. ROS staining and superoxide anion (O2·−) detection assays indicated that Ba significantly mitigated Dex-induced oxidative stress injury and enhanced cell proliferative activity. These results suggest that Ba protects osteoblasts from glucocorticoid damage by modulating the oxidative stress response. Studies have shown6 that oxidative stress injury triggered by glucocorticoids is a significant cause of osteoblast dysfunction and osteoporosis. In hormone-induced ONFH, ROS levels are markedly elevated, accompanied by downregulated expression of antioxidant enzymes. It is plausible that antioxidants, by inhibiting oxidative stress, could represent a potential therapeutic avenue for preventing or slowing the progression of steroid-induced ONFH. This perspective is supported by several studies,6,35 underscoring the central role of oxidative stress in the pathogenesis of SONFH. To further elucidate the mechanism of Ba in treating SONFH, RNA-seq analysis revealed that Ba could reverse a vast number of Dex-induced gene expression disturbances, with the upregulation of Sod3 being particularly prominent. GSVA further confirmed that Ba could specifically restore the activity of key antioxidant pathways, including the pentose phosphate pathway, peroxisome, and glutathione metabolism. This evidence, from both transcriptomic and pathway levels, collectively indicates that the restoration of oxidative stress homeostasis is central to the efficacy of Ba. SOD3, an antioxidant enzyme primarily located extracellularly, efficiently scavenges superoxide radicals, thereby alleviating tissue damage induced by oxidative stress. Therefore, to validate the regulatory role of the differentially expressed gene Sod3, we employed gene transfection for knockdown and overexpression. Using siRNA to silence Sod3, we confirmed that Ba effectively reversed the downregulation of Sod3 and, by regulating Sod3 expression, counteracted the oxidative stress injury caused by Dex in MC3T3-E1 osteoblasts, further demonstrating its potential to protect bone tissue. Subsequently, using lentiviral transfection to overexpress Sod3, we found that Sod3 overexpression alleviated Dex-induced damage in osteoblasts. This indicates that Sod3 could serve as a potential genetic target for treating SONFH, making the modulation of Sod3 expression an important strategy for antioxidant stress therapy in bone diseases. SOD3 plays a crucial protective role in various oxidative stress-related diseases by regulating the extracellular redox state and maintaining normal cellular function.36,37 To further elucidate the mechanism of Ba in treating SONFH, RNA-seq analysis revealed that Ba could reverse a vast number of Dex-induced gene expression disturbances. Notably, Ba treatment not only upregulated Sod3 but also significantly restored the expression of Gsta4 and Aldh3a1, both of which are key components of cellular antioxidant defense. GSVA further confirmed that Ba could specifically restore the activity of metabolic pathways critical for redox homeostasis, including the pentose phosphate pathway and glutathione metabolism. Although these transcriptomic signatures hint at the potential involvement of upstream regulators such as the Nrf2 signaling pathway, we purposefully focused our mechanistic investigation on SOD3 rather than explicitly mapping the entire upstream transcriptional axis. Unlike the ubiquitously expressed intracellular SOD1 and SOD2, SOD3 is uniquely anchored to the extracellular matrix (ECM) via its heparin-binding domain. The extracellular microenvironment is highly vulnerable to GC-induced oxidative damage, which directly contributes to vascular compromise and subsequent osteonecrosis. By demonstrating that Ba specifically restores this critical extracellular antioxidant barrier, our study highlights an innovative mechanism of action that goes beyond conventional intracellular ROS scavenging. Thus, establishing SOD3 as a key functional mediator provides a highly relevant mechanistic explanation for mitigating SONFH, although elucidating the precise upstream transcriptional sensors remains a valuable direction for future research. Furthermore, studies have shown19 that SOD3 also plays a significant role in bone metabolism by modulating the PI3K/Akt pathway, promoting osteogenesis, and inhibiting adipogenesis. This aligns with our observation that Ba restores bone homeostasis, highlighting SOD3 as a robust therapeutic target for steroid-induced ONFH. Although the findings of this study are encouraging, further investigation is needed to elucidate the molecular mechanisms by which Ba regulates SOD3 expression and how this regulation translates into clinical benefits. Future research should delve deeper into the specific signaling pathways through which Ba influences SOD3 expression and evaluate its long-term effects in animal models. Concurrently, our study primarily focused on osteoblasts. The potential effects of Ba on osteoclasts and vascular endothelial cells, which also contribute to SONFH, remain to be explored. Additionally, our study design has certain methodological limitations. Given the limited sample size in specific in vitro and imaging assays, these datasets should be interpreted as supportive and exploratory evidence. Furthermore, although our current findings highly indicate SOD3 as the critical mediator in Ba-mediated protection, our in vivo experimental design lacks targeted causal validation. The current in vivo data primarily demonstrate the systemic efficacy of Ba and the corresponding upregulation of SOD3 expression in the femoral head. However, this study did not utilize local intra-articular injections of specific SOD3 inhibitors or conditional Sod3-knockout mouse models (such as osteoblast-specific Sod3 ablation) to definitively prove that Ba’s anti-necrotic effects in vivo are exclusively dependent on local SOD3 signaling. Future investigations employing these targeted genetic or pharmacological approaches are warranted to uncouple this localized mechanism in vivo fully. In summary, our study reveals a novel pharmacological strategy that leverages the antioxidant capacity of Ba to treat steroid-associated ONFH. Mechanistic analysis identified the upregulation of SOD3 as the central event in this protective pathway. This work not only highlights Ba as a potential natural therapeutic agent but also elevates SOD3 as a robust therapeutic target for preventing and treating steroid-associated osteonecrosis. Conclusion In summary, this study reveals that Ba counteracts glucocorticoid-induced oxidative stress damage in osteonecrosis of the femoral head by modulating SOD3 expression. By identifying SOD3 as a therapeutic target, our work provides a novel direction for the prevention and treatment of this disease. As a potential protective agent, Ba may offer a new therapeutic strategy for patients clinically affected by glucocorticoid therapy-related bone disorders. However, further in vivo causal validation of SOD3 remains necessary, for example, using local SOD3 inhibition or conditional knockout models.