All animal procedures were approved by the Institutional Animal Care and Use Committee of Zhejiang University (approval nos. ZJU20230513 and ZJU20240033) and were conducted in accordance with institutional guidelines and the ARRIVE recommendations. A total of 40 mice were used in the present study. Animals were purchased from Laboratory Animal Center of Zhejiang University and housed under specific pathogen-free (SPF) conditions at 22±2°C with 50±10% relative humidity under a 12-h light/dark cycle, with free access to standard chow and water. Animals were monitored daily for general health status, including body condition, activity, grooming behavior, food and water intake and signs of pain or distress. Humane endpoints were predefined as follows: Body weight loss of 20–25% or evidence of cachexia and wasting syndrome; complete anorexia for 24 h; partial anorexia (<50% of normal caloric intake) for 3 consecutive days; and inability or extreme reluctance to stand or inability to obtain food or water, persisting for 24 h in the absence of anesthesia or analgesia. No animals were found dead prior to scheduled euthanasia or reached the predefined humane endpoints during the study. No longitudinal in vivo intervention was performed in the present study. Animals were euthanized immediately prior to tissue collection for MSC isolation, and all tissue collection procedures were completed within the same day. Animals were euthanized by gradual-fill CO₂ inhalation at a flow rate of 30% of the chamber volume per minute, followed by cervical dislocation. Death was confirmed by the absence of heartbeat and respiration together with loss of corneal and pedal reflexes before tissue collection. The mice used in the present study were male and female in equal proportions, including both young (8 weeks old, weighing 20–25 g) and aged (20 months old, weighing 35–45 g) at the time of experiments.

MSCs were isolated from the femora and tibiae of young (2-month-old) and aged (20-month-old) C57BL/6 mice via flushing and were cultured in Dulbecco's Modified Eagle Medium high glucose (4.5 g/l) (DMEM high glucose (4.5 g/l); Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc.) at 37°C within a 5% CO₂ incubator. After 7–10 days of culture, the cells were passaged using 0.25% trypsin (Gibco; Thermo Fisher Scientific, Inc.). BAY 11–7082 (1 and 2 µM), JSH-23 (20 and 50 µM) and vehicle controls were added to media as indicated. After 24 h of treatment, the culture medium was refreshed.

To knock down STC1 in MSCs, Stc1 small interfering RNAs (siRNAs; si-Stc1) or scrambled siRNAs (si-Scr) were transfected into MSCs using Lipofectamine^® RNAiMax reagent (Thermo Fisher Scientific, Inc.). Specifically, siRNAs (final concentration; 50 nM) and Lipofectamine^® RNAiMax were diluted separately in Opti-MEM medium (Gibco; Thermo Fisher Scientific, Inc.), combined and then incubated at room temperature for 10 min to induce complex formation. The transfection mixture was added directly to the cell culture medium. The transfection was performed at 37°C in a 5% CO₂ incubator for 6 h, after which the medium was replaced with fresh complete medium, and subsequent experiments were conducted 48 h after transfection. Table I lists the siRNA sequences.

Osteogenic and adipogenic differentiation of MSCs was conducted as previously described (29,30). Briefly, osteogenic differentiation was induced by culturing MSCs in osteogenic induction medium consisting of 100 µM ascorbic acid (MilliporeSigma), 2 mM β-glycerophosphate (MilliporeSigma), and 10⁻⁷ M dexamethasone (MilliporeSigma). Adipogenic differentiation was induced using adipogenic induction medium containing 1 µM dexamethasone (MilliporeSigma), 10 µg/ml insulin (MilliporeSigma), 0.5 mM 3-isobutyl-1-methylxanthine (MilliporeSigma), and 0.2 mM indomethacin (MilliporeSigma). For chondrogenic induction, the inducing media were prepared using DMEM high glucose (4.5 g/l) supplemented with 10⁻⁷ M dexamethasone (MilliporeSigma), 1 µM ascorbate-2-phosphate (MilliporeSigma) and 10 ng/ml transforming growth factor-b1 (ABclonal Biotech Co., Ltd.), along with 10% FBS and 1% penicillin/streptomycin.

For ALP staining, osteogenic induction medium was added for 1 week (with the medium refreshed every 3 days), until >70% confluency was confirmed. Cells were fixed using 70% ethanol at room temperature for 15 min and incubated with ALP staining solution (Beyotime Biotechnology) at 37°C in the dark for 5 min. Images were acquired using a CanoScan LiDE 110 scanner (Canon, Inc.). For quantification of ALP activity, cell lysates were prepared in RIPA buffer (without protease/phosphatase inhibitors; Beyotime Biotechnology) and centrifuged at 10,000 × g at 4°C for 5 min. The resulting supernatants were analyzed using an ALP assay kit (Beyotime Biotechnology) according to the manufacturer's instructions. ALP activity was normalized to the total protein concentration of each lysate, as determined using a bicinchoninic acid (BCA) protein assay kit (Beyotime Biotechnology).

For ARS staining, osteogenic induction medium was added for 3 weeks (with the medium refreshed every 3 days), until >70% confluency was confirmed. Cells were fixed using 70% ethanol at room temperature for 1 h and stained using 1% ARS solution (MilliporeSigma) at room temperature for 30 min. For quantification, the plate was destained using 10% cetylpyridinium chloride solution (MilliporeSigma), and optical density was measured at 562 nm.

For Oil Red O staining, adipogenic induction medium was added for 1 week (with the medium refreshed every 3 days), until >90% confluency was confirmed. Cells were fixed using 70% at room temperature ethanol for 10 min and then stained using Oil Red O working solution at room temperature for 5 min. After removal of the staining solution, the cells were rinsed twice using double distilled H₂O. The plate was air-dried, and the stained lipid droplets were eluted in isopropanol. Absorbance was measured at 450 nm using a microplate reader (Omega Bio-Tek, Inc.) to quantify lipid accumulation.

For Alcian blue staining, chondrogenic induction medium (prepared as aforementioned in this section) was added for 3–4 weeks (with the medium refreshed every 3 days), until >90% confluency was confirmed. Cells were fixed using 4% paraformaldehyde at room temperature for 15 min and then stained with Alcian blue solution (MilliporeSigma) for at room temperature 30 min. After image acquisition using the aforementioned scanner, stained Alcian blue was extracted using 6 M guanidine hydrochloride for 6 h at room temperature. Absorbance was measured at 630 nm.

SA-β-Gal staining was performed following previously published procedures (29,31). Cells were fixed using a freshly prepared solution containing 2% formaldehyde and 0.2% glutaraldehyde for 5 min at room temperature. After fixation, the cells were rinsed twice using PBS and incubated with X-gal staining solution (containing 1 mg/ml X-gal (Shanghai Macklin Biochemical Co., Ltd.), 5 mM potassium ferricyanide (Shanghai Macklin Biochemical Co., Ltd.), 5 mM potassium ferrocyanide (Shanghai Macklin Biochemical Co., Ltd.), 150 mM NaCl (Shanghai Macklin Biochemical Co., Ltd.) and 2 mM MgCl₂ (Shanghai Macklin Biochemical Co., Ltd.) in a pH 6.0 buffer system) at 37°C in the dark for 6–8 h. Images were captured under a bright-field microscope (Leica Microsystems GmbH) from five randomly selected fields per sample. SA-β-Gal-positive cells were quantified as the percentage of blue stained cells among the total number of cells and statistically analyzed.

Cells were seeded in 96-well plates at a density of 1×10⁴ cells per well. On days 1, 3 and 5 after seeding, CCK-8 reagent (Beyotime Biotechnology) was added to each well according to the manufacturer's instructions. Afterward, the plates were incubated to allow the reagent to react with the cells. After 2 h of incubation at 37°C, absorbance was measured at 450 nm.

Total RNA was extracted using TRIzol™ reagent (Invitrogen; Thermo Fisher Scientific, Inc.), and 1 µg aliquots of RNA were transcribed to cDNA using reverse transcriptase (Vazyme Biotech Co., Ltd.) at 50°C for 10 min followed by 85°C for 10 sec. Afterward, qPCR was performed using an SYBR Green PCR kit (Vazyme Biotech Co., Ltd.) on a QuantStudio™ 7 Flex System (Applied Biosystems; Thermo Fisher Scientific, Inc.). PCR cycling involved initial denaturation at 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 30 sec, annealing at 60°C for 30 sec and extension at 72°C for 30 sec. Relative gene expression was normalized to β-actin and analyzed via the 2^−∆∆Cq method (32). Table I contains the primer sequences.

Total cellular protein was extracted using RIPA lysis buffer (Beyotime Biotechnology) on ice. Protein concentration in the lysates was determined using a BCA assay kit according to the manufacturer's instructions. Aliquots of the lysates were separated via 10% SDS-PAGE and subsequently transferred onto a PVDF membrane. A total of 20–30 µg of protein per lane was loaded. The membrane was blocked using 5% BSA (Beyotime Biotechnology) in TBST (Tris-buffered saline containing 0.1% Tween-20) for 1 h at room temperature, followed by incubation with specific primary antibodies overnight at 4°C. After washing with TBST, the membrane was incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence substrate (Beyotime Biotechnology) and recorded using a chemiluminescence imaging system (Bio-Rad Laboratories, Inc.). The following antibodies were used: Rabbit anti-NF-κB p65 (1:1,000; cat. no. A19653; ABclonal Biotech Co., Ltd.), rabbit anti-phospho-NF-κB p65 (1:1,000; cat. no. AP0124; ABclonal Biotech Co., Ltd.), rabbit anti-IκBα (1:1,000; cat. no. A19714; ABclonal Biotech Co., Ltd.), rabbit anti-phospho-IκBα (1:1,000; cat. no. AP0707; ABclonal Biotech Co., Ltd.), rabbit anti-β-Actin (1:50,000; cat. no. AC050; ABclonal Biotech Co., Ltd.) and HRP-conjugated goat anti-rabbit IgG (1:70,000; cat. no. HA1001; HUABIO).

Total RNA was isolated using TRIzol reagent. Library construction and RNA sequencing were performed at Shanghai OE Biomedical Technology Co., Ltd. RNA libraries were prepared using VAHTS Universal V10 RNA-seq Library Prep Kit (cat. no. NR616-01; Vazyme Biotech Co., Ltd.) according to the manufacturer's instructions. Subsequently, the FASTQ files of raw paired-end RNA-seq reads were mapped to the mouse reference genome NCBI38/mm10 using STAR (version 2.7.11a; http://github.com/alexdobin/STAR). The mapped reads were then quantified to generate raw counts using featureCounts (version 2.0.3) (33,34). Principal component analysis (PCA) was performed, and significantly differentially expressed genes (DEGs) (si-Stc1 versus si-Scr) were identified using the DESeq2 package (version 1.48.2; http://github.com/thelovelab/DESeq2) in R (4.5.1; RStudio, Inc.), with a threshold P<0.05 and fold changes of >1.5. Heatmaps of the relative expression of representative genes were further generated using the Complexheatmap package (version 2.24.1; http://github.com/jokergoo/ComplexHeatmap) in R.

The gene expression omnibus (GEO) dataset (accession GSE113253, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE113253, http://pubmed.ncbi.nlm.nih.gov/30833796/), originally reported by Rauch et al (35), comprises transcriptional profiles of primary MSCs at multiple time points (0 days, 4 h, 1 day, 3 days and 7 days) during osteoblast and adipocyte differentiation (36). Raw expression matrices were downloaded from GEO and processed in R (version 4.3.1). DEGs were identified using DESeq2 and subsequently subjected to K-means clustering analysis. Read counts were normalized to transcripts per million to facilitate downstream downstream analyses. Gene set enrichment analysis (GSEA) was performed using the R package clusterProfiler (version 4.10.0, http://bioconductor.org/packages/clusterProfiler/) based on Hallmark and KEGG gene sets obtained from the Molecular Signatures Database (https://www.gsea-msigdb.org/gsea/msigdb).

The ATAC-sequencing library was constructed using a Hyperactive ATAC-seq Library Prep Kit (Vazyme Biotech Co., Ltd.) according to the manufacturer's instructions. Specifically, nuclei were isolated from 50,000 cells via incubation on ice for 5 min, followed by centrifugation at 500 × g at 4°C. Tagmentation was performed at 37°C for 30 min. After purification, DNA fragments were amplified using PCR enzymes and indexed primers (Vazyme Biotech Co., Ltd.). Size selection was performed using ATAC DNA Clean Beads (Vazyme Biotech Co., Ltd.), and libraries were sequenced on the Illumina Xplus platform (Illumina, Inc.) using PE150. Raw ATAC-seq read FASTQ files were first processed using Trim Galore (version 0.6.10; http://github.com/FelixKrueger/TrimGalore) to remove the adaptors and were then aligned to NCBI38/mm10 mouse reference genome using Bowtie2 (version 2.4.4–1) (37). Uniquely mapped reads were retained for peak calling using MACS2 with a significance threshold of P<0.05 (38). To generate a consensus set of unique peaks across all samples, read counts within these peak regions were quantified using featureCounts. Differentially accessible regions (DARs) were identified using the DESeq2 package in R, with a P<0.05 and fold change of >1.5. Subsequently, the ChIPseeker package in R (version 1.44.0; http://github.com/YuLab-SMU/ChIPseeker) in R was used to identify the nearest genes, with the transcription start site region defined as −3 kb to +3 kb.

Data are presented as the mean ± standard deviation. Data analysis was conducted using GraphPad Prism 10.0 software (Dotmatics). Differences between groups were assessed using unpaired t-tests, one-way ANOVA (with Dunnett's post hoc test) or two-way ANOVA (with Tukey's post hoc test). P<0.05 was considered to indicate a statistically significantly difference.

Considering that STC1 has been reported to be upregulated in multiple aged tissues, its expression levels during MSC senescence were first examined (17), and MSCs were isolated from 2- and 20-month-old mice. As hypothesized, MSCs from 20-month-old mice exhibited significantly higher Stc1 expression than those from 2-month-old mice (Fig. S1A). Next, publicly available sequencing data (GEO: GSE113253) on osteogenic and adipogenic differentiation of MSCs was analyzed. Gene clustering analysis revealed multiple modules with distinct temporal expression patterns during MSC differentiation into osteoblasts and adipocytes (Fig. 1A). Specifically, the C1 and C2 modules captured gene expression changes associated with osteogenic differentiation. Functional enrichment analysis revealed that these modules were significantly associated with osteogenesis-related biological processes, including ‘ossification’, the ‘canonical Wnt signaling pathway’ and ‘bone development’. By contrast, the C3 and C4 modules represented gene expression programs associated with adipogenic differentiation and were enriched in ‘fatty acid metabolic process’, reflecting commitment to the adipocyte lineage. Consistent with these findings, GSEA demonstrated that osteoblast-related signaling pathways were significantly activated during osteogenic differentiation (Fig. 1B), whereas adipogenesis-related pathways were robustly enriched during adipogenic differentiation (Fig. 1C). Collectively, these results highlighted the stage-specific transcriptional programs underlying MSC lineage specification into osteoblasts and adipocytes. Furthermore, the temporal expression dynamics of Stc1 during MSC differentiation were examined. During osteogenic induction, Stc1 expression exhibited marked temporal variation, with notably higher levels observed at days 3 and 14 (Fig. 1D). By contrast, Stc1 expression exhibited a continuous upward trend throughout adipogenic differentiation (Fig. 1E). These observations indicated that Stc1 was dynamically regulated during MSC lineage commitment and might participate in both osteogenic and adipogenic differentiation processes. To further validate the upregulation of Stc1, mRNA was collected from mouse MSCs treated with osteogenic and adipogenic stimuli for 3 and 7 days, respectively. Consistently, Stc1 mRNA levels increased following osteogenic and adipogenic stimulation, with substantially higher expression observed under adipogenic conditions at day 3 (Fig. 1F and G). In addition, Stc1 expression was examined during chondrogenic differentiation of MSCs, and it was observed that Stc1 expression remained largely unchanged upon chondrogenic stimulation (Fig. S2A).

Furthermore, it was examined whether Stc1 depletion affects MSC senescence and lineage specification. Specifically, two siRNAs targeting Stc1 (si-Stc1−1 and si-Stc1−2) were used to knock down its expression (Fig. 2A). To minimize off-target effects, the two siRNAs were mixed at a ratio of 1:1 and used for transfection (si-Stc1-mix, hereafter referred to as si-Stc1). Stc1 depletion significantly reduced cellular senescence in MSCs, as evidenced by decreased SA-β-Gal staining and downregulated p21 expression, while p16 expression slightly decreased (Fig. S1B-D). Consistently, MSCs exhibited an increased proliferation rate following Stc1 depletion (Fig. S1E).

ALP staining was performed to assess the effects of Stc1 knockdown on the osteogenic commitment of MSCs. Notably, ALP activity, an indicator of early osteogenesis, was inhibited by Stc1 knockdown (Fig. 2B and C). Extracellular matrix mineralization also decreased following Stc1 knockdown, as determined via ARS staining (Fig. 2D and E). Consistently, the expression levels of osteogenic-related genes were downregulated by Stc1 knockdown (Fig. 2F). Notably, although Stc1 was notably upregulated during adipogenesis (39), significant differences were not observed between the Stc1 knockdown groups and controls in lipid droplet formation or the expression of adipogenic-related genes in MSCs (Fig. 2G-I). In addition, to assess whether STC1 is involved in chondrogenic differentiation, MSCs with STC1 knockdown were subjected to chondrogenic induction and evaluated by multiple readouts, including Alcian blue staining, quantitative glycosaminoglycan (GAG) measurement and expression of chondrogenic marker genes (Sox9 and Col2a1). As shown in Fig. S2B-E, STC1 depletion did not significantly alter matrix deposition, GAG content, or chondrogenic gene expression, indicating that STC1 is dispensable for MSC chondrogenesis under the tested conditions.

To investigate the effect of Stc1 knockdown on the transcriptional profile of MSCs, RNA sequencing was performed 3 days after transfection with si-Stc1 (n=3). PCA revealed a notable separation between control and Stc1-depleted MSCs, indicating that Stc1 knockdown induced a marked global change in gene expression (Fig. 3A). Differential expression analysis identified a total of 3,146 DEGs based on a cutoff of P<0.05 and fold change >1.5, including 1,373 downregulated and 1,773 upregulated genes following Stc1 knockdown (Fig. 3B). Gene Ontology (GO) enrichment analysis revealed that the downregulated genes were significantly enriched in biological processes related to ‘osteogenic differentiation’ and ‘bone mineralization’ (Fig. 3C). By contrast, the upregulated genes were predominantly associated with inflammatory responses, including pathways related to ‘interleukin-1 production’, ‘positive regulation of interleukin-6 production’, as well as ‘canonical NF-κB signal transduction’ (Fig. 3D). GSEA further demonstrated that gene sets associated with osteogenesis and bone mineralization were significantly negatively enriched in Stc1-depleted MSCs, whereas gene sets related to inflammatory and immune responses were positively enriched following Stc1 knockdown (Fig. 3E-H). A heatmap of representative genes further confirmed enhanced inflammatory responses and impaired osteogenic capacity following Stc1 knockdown (Fig. 3I).

Furthermore, an ATAC-sequencing was performed to examine the impact of Stc1 depletion on chromatin accessibility in MSCs. Overall, Stc1 knockdown resulted in only modest changes in global chromatin accessibility (Fig. 4A and B). Differential accessibility analysis identified 4,753 DARs with decreased accessibility and 6,544 DARs with increased accessibility following Stc1 knockdown, based on a cutoff of P<0.05 (Fig. 4C). Functional annotation of DARs revealed a notable functional divergence between regions gaining or losing accessibility following Stc1 knockdown. Regions with reduced accessibility were preferentially associated with genes involved in ‘collagen-activated signaling pathway’, ‘regulation of osteoblast differentiation’ and ‘bone mineralization’, indicating selective chromatin closure at osteogenesis-related loci (Fig. 4D). By contrast, regions exhibiting increased accessibility were associated with genes involved in inflammatory processes, including ‘interleukin-2 production’, ‘regulation of canonical NF-κB signal transduction’ and ‘cell chemotaxis’ (Fig. 4E). Genome track visualization further illustrated these chromatin accessibility changes at representative loci. The promoter regions of osteogenic genes, including Alp and Hoxa3, exhibited notably reduced accessibility following Stc1 knockdown (Fig. 4F and G). By contrast, increased chromatin accessibility was observed at the promoters of inflammation-related genes, such as Cxcl12 and Il11 (Fig. 4H and I). Collectively, transcriptomic and chromatin accessibility analysis indicated that STC1 knockdown selectively repressed osteogenic programs while concomitantly activating inflammatory pathways.

To further elucidate the coordinated transcriptional and chromatin accessibility changes induced by Stc1 knockdown, an integrative analysis of the RNA-sequencing and ATAC-sequencing datasets was performed. A total of 261 overlapping genes exhibited both decreased transcript abundance in RNA-sequencing and reduced chromatin accessibility in ATAC-sequencing (Fig. 5A). Functional enrichment analysis of this gene set revealed an association with osteogenesis-related processes, including ‘bone trabecular formation’, ‘osteoblast differentiation’ and ‘skeletal system morphogenesis’ (Fig. 5B). Conversely, 374 overlapping genes were identified that exhibited upregulation at the transcript level and increased chromatin accessibility (Fig. 5C). Enrichment analysis of these genes indicated predominant involvement in inflammatory and immune-related pathways, such as ‘regulation of interleukin-6 production’, ‘canonical inflammasome complex assembly’ and ‘regulation of canonical NF-κB signal transduction’ (Fig. 5D). Consistent with the Venn diagram analysis, the integrative heatmap revealed that inflammatory and NF-κB-related genes (including Tlr4, Il6ra, Pycard and Cxcl12) were upregulated and exhibited increased chromatin accessibility, whereas osteogenesis-related genes (including Col1a1, Bglap, Alpl, Smad5 and Dlx2) were downregulated and exhibited reduced accessibility (Fig. 5E). These findings indicated that Stc1 knockdown was associated with coordinated transcriptional and chromatin accessibility changes related to inflammatory activation and impaired osteogenic programs. NF-κB signaling has been well established as a central regulator of MSC osteogenic differentiation and inflammatory responses (40,41). Given the consistent enrichment of inflammatory and NF-κB signaling pathways, it was examined whether Stc1 knockdown activates NF-κB signaling in MSCs. To further confirm this pathway activation, GSEA was performed on the RNA-sequencing dataset, which revealed significant enrichment of the canonical NF-κB signaling gene set in the Stc1 knockdown group (Fig. 5F). Western blot analysis revealed a notable increase in p65 phosphorylation, accompanied by a slight increase in phosphorylated IκBα, indicating early activation of canonical NF-κB signaling (Fig. 5G). Consistently, the expression levels of several well-established NF-κB downstream target genes, including Birc3, Icam1, Vcam1, Tnf and Il1α (42), were significantly upregulated following Stc1 knockdown (Fig. 5H). Collectively, these results provided molecular evidence that STC1 depletion activates the NF-κB signaling pathway in MSCs.

To determine whether inhibition of NF-κB signaling could mitigate the impaired osteogenic differentiation induced by Stc1 knockdown, MSCs were treated with the NF-κB inhibitor BAY 11-7082 (43,44). To exclude potential cytotoxic effects, cell viability was first assessed via a CCK-8 assay. Treatment with BAY 11-7082 at the concentrations used in the present study did not significantly affect MSC viability (Fig. 6A). Consequently, the inhibitory efficiency of BAY 11-7082 was verified via western blot analysis (Fig. 6B). As shown, treatment with BAY 11-7082 markedly reduced the phosphorylation levels of p65 and IκBα. Densitometric analysis revealed that p-p65 and p-IκBα levels were decreased by approximately 60 and 36%, respectively, compared with the control group. By contrast, total p65 and IκBα protein levels remained largely unchanged, indicating effective suppression of NF-κB pathway activation rather than alterations in protein expression. ALP activity assays revealed that treatment with BAY 11-7082 at concentrations of 1 and 2 µM increased ALP activity during osteogenic differentiation of MSCs (Fig. 6C and D). Consistent with this observation, ARS staining revealed that BAY 11-7082 treatment effectively mitigated the reduction in extracellular matrix mineralization induced by Stc1 depletion (Fig. 6E and F). Moreover, the expression levels of osteogenic marker genes, Alp and Ocn, exhibited a similar increasing trend following BAY 11-7082 treatment (Fig. 6G and H).

To further validate these findings, MSCs were treated with JSH-23, an alternative NF-κB inhibitor, at concentrations of 20 and 50 µM during osteogenic induction (44). To exclude potential cytotoxic effects, cell viability was first assessed via a CCK-8 assay, which demonstrated that JSH-23 at 20 and 50 µM did not significantly affect MSC viability (Fig. S3A). Similar to BAY 11-7082, JSH-23 treatment enhanced osteogenic differentiation and restored osteogenic capacity in Stc1-knockdown MSCs (Fig. S3B-G). Overall, following treatment with either BAY 11-7082 or JSH-23, osteogenic differentiation in Stc1-depleted MSCs reached or even exceeded that observed in scrambled siRNA control cells treated with vehicle, indicating a robust mitigating effect.

Given the essential role of NF-κB in the regulation of inflammation, BAY 11-7082 treatment was expected to inhibit the expression of inflammatory factors, Il1a and Tnf, induced by Stc1 depletion. Notably, treatment with 2 µM BAY 11-7082 reduced their expression (Fig. 6I and J). Inhibition of NF-κB signaling by BAY 11-7082 further enhanced the effect of Stc1 depletion on cellular senescence, as evidenced by the SA-β-gal assay (Fig. 6K and L). Similarly, treatment with JSH-23 also inhibited the expression of Il1a and Tnfa and further mitigated senescence in STC1-depleted MSCs (Fig. S3H-K).

Collectively, these results identify STC1 as a key regulator of osteogenic differentiation and inflammation in MSCs. These findings provide mechanistic insight into the role of STC1 in MSC biology and highlight that modulation of STC1 and NF-κB signaling requires further investigation in more physiologically relevant and in vivo models of age-related bone loss.