Secreted protein combination GAPDH/S100A8/S100A9 from human expanded potential stem cells counteracts mesenchymal stromal cell senescence

Abstract

Background

Human mesenchymal stromal cells (hMSCs) remain the most clinically advanced adult stem cell source; however, their therapeutic potential is limited by rapid replicative senescence during ex vivo expansion. Replicative senescence in hMSCs is characterized by cell cycle arrest, acquisition of senescence-associated β-galactosidase (SA-β-Gal) activity, and secretion of the senescence-associated secretory phenotype (SASP) factors.

Methods

We investigated whether conditioned medium derived from human extended pluripotent stem cells (hEPSCs), which possess both embryonic and extra-embryonic developmental potential beyond that of conventional embryonic stem cells (hESCs), could attenuate replicative senescence in human Wharton’s Jelly-derived MSCs (WJMSCs). Using sequential ultrafiltration (10 kDa and 3 kDa) followed by liquid chromatography-tandem mass spectrometry, we identified several proteins from hEPSC-conditioned medium. We then tested the combination of S100A9/GAPDH/S100A8 proteins for their effects on doxorubicin (DOXO)-induced and replicative senescence.

Result

hEPSC-conditioned medium markedly attenuated replicative senescence in hMSCs. Notably, the combination of S100A9/GAPDH/S100A8 proteins not only mitigated doxorubicin-induced senescence but also counteracted replicative senescence, as evidenced by a significant reduction in SA-β-Gal-positive cells and downregulated mRNA expression of senescence-associated genes, including p16, p21, and the SASP factor IL-6. Furthermore, EdU incorporation assays revealed significantly enhanced proliferative capacity following treatment.

Conclusions

Collectively, our findings establish a defined protein combination (S100A9/GAPDH/S100A8) that counteracts both replicative and stress-induced senescence, offering a novel, cell-free strategy to enhance the clinical utility of WJMSCs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-026-05178-z.

Introduction

Human mesenchymal stromal cells (hMSCs) exhibit remarkable immunomodulatory, anti-inflammatory properties and anti-aging, making them highly promising for a wide range of clinical applications [1–3]. hMSCs secrete a variety of anti-inflammatory cytokines and growth factors, such as interleukin-10 (IL-10), transforming growth factor-β (TGF-β), and hepatocyte growth factor (HGF) [4]. These molecules attenuate immune cell activation and proliferation, thereby alleviating inflammation [5]. IL-10 can inhibit the production of pro-inflammatory cytokines like tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ) by immune cells [6]; hMSCs can polarize macrophages from the pro-inflammatory M1 phenotype to the anti-inflammatory M2 phenotype [7, 8], further dampening the inflammatory response. Ongoing research is focused on optimizing hMSC-based therapies, including enhancing the anti-inflammatory and anti-aging potency, understanding the long-term effects of hMSC treatment.

hMSCs are prone to replicative senescence during extended in vitro culture [9]. Such aging is characterized by reduced proliferative and differentiation capacities, altered secretory profiles, attenuated paracrine activity, and compromised immunomodulatory properties, thereby limiting their therapeutic applicability [10, 11]. Senescent hMSCs can be identified by multiple established biomarkers [12]. Senescence-associated β-galactosidase (SA-β-Gal) activity, a hallmark of cellular senescence, markedly increases with progressive cellular aging. Concurrently, the tumor suppressor p53, a master regulator of senescence, and its downstream effectors CDKN2A (encoding p16) and CDKN1A (encoding p21) are significantly upregulated, collectively enforcing cell cycle arrest. Furthermore, senescent hMSCs exhibit senescence-associated secretory phenotype (SASP), characterized by elevated secretion of pro-inflammatory cytokines such as IL-6 and TNF-α, which subsequently perturbs the cellular microenvironment [10]. Collectively, these biomarkers facilitate the monitoring of MSC aging in culture systems and inform the rational design of senotherapeutic strategies [10, 13].

Recent studies have revealed that secreted proteins and exosomes from cultured human embryonic stem cells (hESCs) effectively delay bone marrow-derived MSC (BM-MSC) senescence both in vitro and in vivo [14]. Proteomic analysis has identified 4122 proteins within hESC-derived small extracellular vesicles (hESC-sEVs), which are enriched in functional networks and signaling pathways regulating cellular senescence and osteogenic differentiation. Notably, hESC-sEVs exhibit superior anti-aging efficacy compared to hMSC-sEVs in osteoarthritis models [15, 16], with the FOXO1A-mediated autophagy axis serving as a critical mechanistic pathway [16]. Human Wharton’s Jelly-derived MSCs (WJMSCs) are also widely used in clinical practice; however, whether hESC-derived secreted proteins and exosomes can delay WJMSC senescence remains unknown. While these findings collectively demonstrate that hESC-secreted factors attenuate cellular aging, the specific bioactive protein(s) responsible remain to be elucidated. Current efforts are therefore directed toward identifying key secreted factors that can effectively counteract hMSC senescence during in vitro expansion.

Human extended pluripotent stem cells (hEPSCs) exhibit broader developmental plasticity compared to conventional hESCs, with demonstrated capacity to integrate into both embryonic and extraembryonic lineages [17, 18]. Here, we report that hEPSC-conditioned medium effectively attenuates WJMSCs senescence. To identify the underlying bioactive components, we performed LC–MS/MS-based proteomic analysis of the hEPSC secretome, revealing that a combination of Protein S100-A9 (S100A9), Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and Protein S100-A8 (S100A8) proteins effectively mitigates both doxorubicin-induced and replicative senescence in WJMSCs.

Results

hEPSCs-conditioned medium attenuates replicative senescence of WJMSCs

During in vitro culture, WJMSCs undergo replicative senescence, characterized by increased expression of senescence-associated markers, elevated proportions of SA-β-Gal-positive cells, and decreased proliferative capacity. To characterize these senescence-related alterations, WJMSCs at passage 4 (P4) and passage 8 (P8) were analyzed. P8 WJMSCs exhibited a significantly increase in the proportion of SA-β-Gal-positive cells, reduced proliferation rates as assessed by CCK-8 assay, and elevated p21 mRNA levels (Additional file 1: Fig. S1). Collectively, these observations confirm the rapid onset of senescence in WJMSCs during extended in vitro culture.

Previous studies have demonstrated that certain protein components isolated from hESC-conditioned medium can ameliorate cellular aging in hMSCs and fibroblast [14, 15]. Compared to conventional hESCs, hEPSCs possess extended developmental potential, capable of chimerizing both embryonic and extraembryonic tissues [17, 18]. hESCs cultured with LCDM medium can be converted to hEPSCs. We therefore collected EPSCs-conditioned medium and supplemented WJMSC cultures with it (1:5 dilution) during passaging every three days. After 15 days, SA-β-Gal-positive cells were significantly reduced (Fig. 1A, B), EdU incorporation assay and CCK-8 analysis further demonstrated enhanced proliferation in hEPSCs-treated WJMSCs, accompanied by decreased p16 and p21 mRNA levels (Fig. 1C–F). These results indicate that hEPSCs-conditioned medium effectively alleviates WJMSCs replicative senescence.

Fig. 1.

hEPSCs-conditioned medium attenuates WJMSCs senescence. A, B Senescence-associated β-galactosidase (SA-β-gal) staining of WJMSCs treated with EPSC-conditioned medium for 15 days. Representative images (A) and quantification of SA-β-gal-positive cells (B) are shown. Data represent mean ± SEM; n = 3 biological replicates. ***, P < 0.001 versus control. C, D EdU incorporation assay assessing cell proliferation. Representative fluorescence images (C) and quantification of EdU-positive cells (D) are presented. Data represent mean ± SEM; n = 3 biological replicates. **, P < 0.01 versus control. E Cell viability was evaluated using the CCK-8 assay by measuring absorbance at 450 nm. Data represent mean ± SEM; n = 3 biological replicates. **, P < 0.01 versus control. F Relative mRNA expression levels of senescence markers CDKN2A (p16) and CDKN1A (p21) were determined by quantitative RT-PCR. Data represent mean ± SEM; n = 3 biological replicates. **, P < 0.01 versus control. G, H Secreted proteins with molecular weight below 10 kDa were enriched from hEPSC-conditioned medium and analyzed by mass spectrometry (G). This approach identified 13 distinct protein components in the 10 kDa fraction (H). I Gene Ontology (GO) enrichment analysis of mass spectrometry-identified proteins, showing significantly enriched biological processes

Proteomic identification of hEPSCs-secreted factors via mass spectrometry

Previous studies have demonstrated that low-molecular-weight (< 40 kDa) secreted proteins play crucial roles in cell growth and signal transduction, rendering them promising candidates for clinical therapeutics [19]. To identify bioactive low-molecular-weight protein components within the hEPSC secretome, conditioned medium was subjected to sequential ultrafiltration (10 kDa and 3 kDa molecular weight cutoffs), followed by trichloroacetic acid (TCA) precipitation and liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis of gel slices. This approach identified 13 distinct protein components in the 10 kDa fraction (Fig. 1G, H). Molecular weight fractionation revealed a specific protein signature comprising both low-molecular-weight (< 40 kDa) factors, including S100A9, LGALS7, CALML5, and GAPDH; and high-molecular-weight (> 40 kDa) proteins such as CLIP1, UTY, DSG1, TGM1, TGM5, IGH3G, and GSDMA. Gene ontology (GO) enrichment analysis indicated significant involvement in cornified envelope formation, mitotic processes, neutrophil degranulation, NABA matrisome-associated pathways, and microtubule cytoskeleton organization. These findings suggest that hEPSC-secreted proteins likely function in extracellular matrix remodeling and the regulation of cell proliferation and tissue repair.

S100A9/GAPDH protein combination mitigates Doxorubicin-induced senescence in WJMSCs

To investigate the anti-senescence effects of the identified secretory proteins, we established a rapid senescence cell model using WJMSCs treated with DOXO in a dose-dependent manner [20, 21]. DOXO is a widely chemotherapeutic agent known to rapidly induce senescence in hMSCs [21]. We determined that treatment with 100 nM DOXO for 3 days represented optimal conditions for inducing senescence, as this concentration significantly increased the proportion of SA-β-Gal-positive cells without causing excessive cell death (Fig. 2A, B, Additional file 1: Fig. S2A). This treatment paradigm was also associated with elevated p21 mRNA expression (Additional file 1: Fig. S2B, C). Notably, passage 4 (P4) WJMSCs exhibited higher senescence inducibility compared to passage 8 (P8) cells and were therefore selected for subsequent experiments (Additional file 1: Fig. S2D).

Fig. 2.

S100A9/GAPDH/S100A8 protein combination mitigates doxorubicin-induced senescence in WJMSCs. A, B Establishment and validation of a rapid senescence model. WJMSCs were treated with doxorubicin (DOXO) to induce cellular senescence, and the percentage of SA-β-gal-positive cells was quantified. Representative images (A) and statistical analysis (B) are shown. Data represent mean ± SEM; n = 3 biological replicates. ****, P < 0.0001 versus untreated control. C Validation of EGFP as a negative control. Purified EGFP protein (50 nM) was added to DOXO-induced senescent WJMSCs, and relative mRNA levels of p16 and p21 were determined by quantitative RT-PCR. D Individual and combined effects of GAPDH and S100A9. Purified EGFP, GAPDH, or S100A9 proteins were added individually (50 nM each), or GAPDH and S100A9 were combined (50 nM each), to DOXO-induced WJMSCs. Relative mRNA levels of p16, p21, and IL-6 were assessed by quantitative RT-PCR. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; **, P < 0.01; ***, P < 0.001 versus EGFP control. E Synergistic effect of S100A8 and S100A9. Purified EGFP or S100A9 proteins were added individually, or S100A8 and S100A9 were combined (50 nM each), to DOXO-induced MSCs. Relative mRNA levels of p16, p21, and IL-6 were determined. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001 versus EGFP control. F Triple combination of GAPDH, S100A8, and S100A9. The three proteins were combined (50 nM each) and added to DOXO-induced WJMSCs, with EGFP treatment as control. Relative mRNA levels of p16, p21, and IL6 were assessed. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; ****, P < 0.0001 versus EGFP control. G Dose-dependent efficacy of the triple combination. GAPDH, S100A8, and S100A9 proteins were combined at 50 nM or 200 nM and added to DOXO-induced WJMSCs. Relative mRNA levels of p16, p21, and IL-6 were determined. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; **, P < 0.01; ***, P < 0.001 versus respective controls. H, I Dose–response analysis of cell viability. The triple protein combination was administered at 50 nM, 200 nM, or 500 nM to DOXO-induced WJMSCs. Cell viability was assessed by CCK-8 assay measuring absorbance at 450 nm. Representative data (H) and quantification (I) are shown. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001 versus control. J Requirement for intact protein structure. Trypsin-digested or intact (non-digested) protein mixtures were added to DOXO-induced WJMSCs. Relative mRNA levels of p16 and p21 were determined by quantitative RT-PCR. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; ***, P < 0.001 versus intact protein treatment

Among the candidate proteins identified, GAPDH and S100A9 were selected for further validation. GAPDH, traditionally recognized as a glycolytic enzyme, has been shown to localize extracellularly and participates in diverse biological processes including exosome clustering, biogenesis and secretion, apoptosis, DNA repair, and autophagy [22, 23]. S100A9, a Ca²⁺-binding protein, is secreted via non-classical pathways [24]; and has been detected at elevated levels in inflammation conditions, neoplastic cells and various human cancers [25, 26].

We purified recombinant GAPDH, S100A9, and enhanced green fluorescent protein (EGFP, as a negative control) from E coli. In the DOXO-induced senescence model, EGFP added to the culture medium at 50 nM did not affect p16 and p21 mRNA expression levels (Fig. 2C). However, when GAPDH and S100A9 proteins (each at 50 nM) were supplemented every 3 days, the combination treatment significantly reduced the mRNA levels of p16, p21, and IL-6 compared to EGFP control (Fig. 2D, Additional file 1: Fig. S2D, E). These findings suggest GAPDH and S100A9 may function as potential mediators of the anti-senescence effects mediated by EPSCs.

Synergistic anti-senescence effects of the GAPDH/S100A8/S100A9 protein combination in doxorubicin-treated WJMSCs

S100A9 and S100A8 are prominent members of the S100 protein family that primarily exist as a heterodimer (S100A8/A9), also known as calprotectin. This heterodimer plays crucial regulatory roles in inflammation and immune responses [25]. Under high calcium conditions, S100A8 and S100A9 can form homodimers, heterodimers, and other multimeric configurations; however, the heterodimer represents the most stable and physiologically relevant structure [27, 28]. As multi-functional proteins, S100A8 and S100A9 are implicated in cancer cell growth, inflammation response, and various signal pathway [26, 29, 30].

In our experiments system, combined treatment with purified S100A9 and S100A8 proteins significantly reduced mRNA levels of the senescence markers p16 and p21, as well as the SASP factor IL-6 (Fig. 2E, Additional file 1: Fig. S2F). We further evaluated a triple combination of GAPDH, S100A8 and S100A9 proteins (each at 50 nM), which similarly suppressed the expression of these senescence-associated genes (p16, p21, and IL-6) (Fig. 2F). Notably, increasing the protein concentration to 200 nM each resulted in superior efficacy, demonstrated clear dose-dependent response (Fig. 2G, Additional file 1: Fig. S2G).

Cell proliferation assays using the CCK-8 method confirmed that this protein combination enhanced cell proliferation at effective concentrations. However, at 500 nM each, the combination induced cytotoxicity, indicating a narrow therapeutic window (Fig. 2H, I). To determine whether the biological activity resided in the intact proteins or their proteolytic fragments, we digested the protein mixture with trypsin. The proteins were completely digested into peptides and then added to the WJMSC culture medium. The results suggested that the intact protein mixture was more effective than its corresponding peptide fragments (Fig. 2J).

Collectively, these findings demonstrate that the combination of GAPDH, S100A8, and S100A9 effectively counteracts DOXO-induced senescence in WJMSCs. These data support a model wherein low-dose protein treatment confers protective anti-senescence effects, whereas excessive exposure compromises cell viability.

Protein combination GAPDH/S100A8/S100A9 counteracts replicative senescence in long-term culture

To evaluate the long-term anti-senescence effects of the protein combination (hereafter referred to as “Pros,” comprising GAPDH, S100A8, and S100A9), WJMSCs were continuously cultured in the presence of Pros at concentrations of 50 nM and 200 nM for 20 days. Immunocytochemical analysis revealed that the proportion of SA-β-Gal-positive cells decreased significantly in a dose-dependent manner following Pros treatment (Fig. 3A, B). Conversely, EdU incorporation assays demonstrated a significant increase in proliferating cells (Fig. 3C, D), and CCK-8 proliferation assays confirmed enhanced cellular proliferation (Fig. 3E), collectively indicating delayed replicative senescence. Molecular characterization further supported these findings: the relative expression levels of key senescence markers, including p16, p21, p53 and the SASP factor IL-6, were markedly reduced at both mRNA and protein levels following 20 days of treatment, with higher concentrations showing greater efficacy (Fig. 3F, G). Comparison analysis of short-term (10-day) versus long-term (20-day) treatment revealed that prolonged Pros exposure significantly enhances anti-senescence efficacy, suggesting time-dependent cumulative benefits (Fig. 3G).

Fig. 3.

GAPDH/S100A8/S100A9 protein combination attenuates replicative senescence in WJMSCs. (A-B) Long-term treatment with the protein combination (Pros) reduces senescence markers. WJMSCs were cultured for 20 days with EGFP (control) or Pros at 50 nM and 200 nM. (A) Representative SA-β-gal staining images. (B) Quantification of SA-β-gal-positive cells. Data represent mean ± SEM; n = 3 biological replicates. ***, P < 0.001; ****, P < 0.0001 versus EGFP control. (C-D) Pros treatment enhances cell proliferation. (C) Representative EdU incorporation images. (D) Quantification of EdU-positive cells. Data represent mean ± SEM; n = 3 biological replicates. ***, P < 0.001; ****, P < 0.0001 versus EGFP control. (E) Cell viability assessment. WJMSCs treated with EGFP or Pros (50 nM and 200 nM) for 20 days were analyzed by CCK-8 assay measuring absorbance at 450 nm. Data represent mean ± SEM; n = 3 biological replicates. ***, P < 0.001 versus EGFP control. (F) Downregulation of senescence-associated proteins. Western blot analysis of p21 and p53 expression in WJMSCs treated with EGFP, 50 nM Pros, or 200 nM Pros for 20 days. β-actin served as loading control. Quantification of relative protein levels is shown below. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 versus EGFP control. (G) Time-dependent suppression of senescence markers. Relative mRNA levels of p16, p21, TP53 (p53), and IL6 were measured by quantitative RT-PCR at day 10 and day 20 following treatment with EGFP, 50 nM Pros, or 200 nM Pros. Data represent mean ± SEM; n = 3 biological replicates. ***, P < 0.001; ****, P < 0.0001 versus respective EGFP controls. (H) GAPDH C152S mutant is required for optimal anti-senescence efficacy. Relative mRNA levels of p16, p21, and IL6 were measured at day 20 in WJMSCs treated with EGFP, wild-type GAPDH-containing Pros, or Pros containing GAPDH mutant (C152S) in various combinations. Data represent mean ± SEM; n = 3 biological replicates. *, P < 0.05; **, P < 0.01; ***, P < 0.001 versus respective controls

The protective effects of Pros were not restricted to WJMSCs. Similar anti-senescence activity was observed in HFF-1 human fibroblasts, where 200 nM Pros treatment for 12 days significantly reduced mRNA levels of IL-6, p16, and p21 while improving cell proliferation (Additional file 1: Fig. S3), demonstrating broad applicability across distinct cell types and senescence models.

To investigate the mechanistic basis of GAPDH-mediated protection, we substituted wild-type GAPDH with a catalytically inactive mutant (C152S) in the Pros combination. Notably, the mutant-containing combination failed to suppress senescence markers to the same extent as the wild-type combination (Fig. 3H), indicating that the catalytic cysteine residue of GAPDH (C152) is essential for anti-senescence activity; however, this may reflect either a requirement for enzymatic catalysis or a structural contribution to protein folding, stability, and non-catalytic protein–protein interactions. Collectively, these results demonstrate that the GAPDH/S100A8/S100A9 protein combination effectively counteracts replicative senescence during long-term culture, with efficacy dependent on treatment duration.

Protein combination GAPDH/S100A8/S100A9 remodels the transcriptomic landscape of WJMSCs

To elucidate the transcriptional mechanisms underlying Pros-mediated anti-senescence effects, we performed RNA-seq analysis on WJMSCs from three experimental groups: untreated control (Blank), EGFP-treated control (EGFP), and cells treated with 200 nM GAPDH/S100A8/S100A9 combination (Pros). Comparative analysis revealed that Pros treatment significantly altered global gene expression patterns. Relative to the blank control, Pros upregulated 305 transcripts and downregulated 159 transcripts (Additional file 1: Fig. S4A). When compared against the EGFP control, Pros induced upregulation of 304 transcripts and downregulation of 202 transcripts (Fig. 4A). Notably, approximately two-thirds of upregulated genes overlapped between these two comparisons, with remarkable concordance (90% overlap) observed among the top 100 upregulated genes. These findings confirm that long-term EGFP exposure minimally impacts cellular transcriptional profiles (Additional file 1: Fig. S4B), whereas the protein combination serves as the primary driver of transcriptomic reprogramming in WJMSCs.

Fig. 4.

GAPDH/S100A8/S100A9 protein combination remodels the transcriptomic landscape of WJMSCs. A Volcano plot visualization of differentially expressed genes (DEGs) in WJMSCs treated with the protein combination (Pros: GAPDH/S100A8/S100A9, 200 nM) versus EGFP control (200 nM) for 20 days. Significantly upregulated genes (log₂ fold change > 1, adjusted P < 0.05) are shown in red; significantly downregulated genes (log₂ fold change <  − 1, adjusted P < 0.05) are shown in blue. B RNA sequencing read coverage tracks for representative senescence-associated genes. Genome browser views showing normalized read coverage for IL6, TP53, CDKN2A (encoding p16), and CDKN1A (encoding p21) in EGFP-treated (blue) versus Pros-treated (red) WJMSCs. Reduced read depth in Pros-treated cells indicates transcriptional downregulation of these markers. C Gene Ontology (GO) enrichment analysis of differentially expressed genes. Significantly enriched biological processes for upregulated (up) and downregulated (down) genes are displayed. D Validation of RNA-seq findings by quantitative RT-PCR. Relative mRNA expression levels of selected upregulated (WNT2B, ANO3, H19, PPP1R12B, PPP1R14A) and downregulated (GPNMB, AC007938.2, SGCG, PSAT1, EPHA3) genes were measured in WJMSCs treated with Pros or EGFP for 20 days. Data represent mean ± SEM; n = 3 biological replicates. **, P < 0.01; ***, P < 0.001 versus EGFP control

RNA-seq read coverage tracks for key senescence markers, including IL6, TP53, CDKN2A (encoding p16), and CDKN1A (encoding p21), demonstrated markedly reduced expression in Pros-treated cells (red) compared to untreated controls (blue) (Fig. 4B). Principal component analysis and Pearson correlation coefficients revealed that Pros-treated WJMSCs formed a transcriptionally distinct cluster, clearly segregated from both passage 12 (P12) senescent cells and EGFP-treated controls (Additional file 1: Fig. S4C), indicating substantial transcriptomic divergence. Gene ontology (GO) enrichment analysis of differentially expressed genes revealed significant functional reprogramming. Upregulated genes were enriched in biological processes related to extracellular matrix organization, including “NABA core matrisome,” O-glycosylation, actin cytoskeleton organization, tube morphogenesis, and system process regulation (Fig. 4C). Conversely, downregulated genes were associated with corticotropin-releasing hormone response, muscle organ development, amino acid metabolism, epithelial cell proliferation regulation, metal binding (metallothioneins), and inflammatory response (Fig. 4C). Notably, the suppression of inflammation-related pathways aligns with our previous phenotypic observations.

To validate the RNA-seq findings, we conducted quantitative RT-PCR on selected differentially expressed genes. Among upregulated candidates, WNT2B, a secreted activator of Wnt/β-catenin signaling crucial for MSC proliferation and differentiation [31], showed increased expression. Similarly, ANO3 (a calcium-responsive transmembrane protein) [32, 33], and H19 (a proliferation-associated long noncoding RNA) [34], and PPP1R14A/PPP1R12B (protein phosphatase 1 regulatory subunits implicated in cell cycle and metabolic regulation) [35] were all significantly upregulated in Pros-treated cells (Fig. 4D). Conversely, GPNMB, AC007938.2, SGCG, PSAT1 and EPHA3 exhibited downregulated expression patterns consistent with the transcriptomic data (Fig. 4D). These results suggest that the GAPDH/S100A8/S100A9 combination functions as an upstream regulatory signal that modulates gene expression networks involved in extracellular matrix remodeling, cell proliferation, and inflammation suppression, thereby coordinately antagonizing cellular senescence.

Discussion

MSC aging is characterized by several interwined hallmarks: (1) elevated senescence-associated β-galactosidase (SA-β-Gal) activity, the most widely used biomarker for detecting cellular senescence [36]; (2) the development of a pro-inflammatory senescence-associated secretory phenotype (SASP) that reinforces paracrine senescence, and (3) cell-cycle arrest in G0/G1 phase accompanied by accumulation of p16 and p21 [37, 38]. Notably, SA-β-Gal lacks absolute specificity, as it can be elevated under non-senescent conditions (e.g., high cell density, serum starvation, lysosomal stress) [39]; thus, definitive classification ideally employs a multi-parameter framework [40]. Human expanded Potential Stem Cells (hEPSCs) possess the unique dual capacity to differentiate into both embryonic and trophoblast lineages in vitro and in chimera models [17]. Here, we demonstrate that hEPSC-conditioned medium and specific secreted proteins combinations (GAPDH/S100A9/S100A8) counteract WJMSC senescence by decreasing SA-β-Gal activity, suppressing p16/p21 expression and restoring cell-cycle progression. These secreted protein combinations appear poised to enhance WJMSC engraftment and reparative capacity.

Calcium ions (Ca²⁺), functioning as ubiquitous second messengers, collaborate with their binding proteins to form a highly conserved, multi-layered regulatory network that governs both cell-cycle progression and immune response [41–43]. S100A8 and S100A9, constituting the heterodimer calprotectin, undergo Ca²⁺-mediated heterodimerization and conformational switching, thereby regulating innate immunity and cell growth [25]. Recent studies have reported that S100A9/S100A8 dimers activate Toll-like receptor-4 (TLR4); however, high extracellular calcium concentrations induce the formation of S100A8/S100A9 tetramers, which prevent TLR4 binding and limit inflammatory activity [44–46]. In our study, we speculate that supplementation with S100A9/S100A8 may have induced extracellular tetramer formation, thereby attenuating TLR4 signaling and consequently suppressing downstream NF-κB activation. Notably, NF-κB activation is both necessary and sufficient for the establishment and maintenance of SASP; its genetic or pharmacologic inhibition uncouples inflammatory paracrine signaling from permanent cell-cycle arrest [47–49].

The combination of S100A8/S100A9/GAPDH significantly reduced mRNA levels of p16, p21, and IL-6 in a dose-dependent manner. Traditionally regarded as a housekeeping gene, GAPDH has recently been revealed to possess multiple functions beyond its canonical role in glycolysis [23, 50–52]. Notably, when a glycolytic-inactive GAPDH mutant (C152S) replaced wild-type GAPDH in the combination, the anti-senescence efficacy was diminished, indicating that GAPDH’s protective activity may partially dependent on its enzymatic function. Furthermore, continuous supplementation of S100A8/S100A9/GAPDH protein combinations was required to sustain anti-aging effects, suggesting that these proteins function not merely as transient signaling molecules but as persistent effectors that coordinate multiple pathways to maintain cellular rejuvenation. Here, we demonstrate for the first time that S100A8/S100A9/GAPDH secreted by hEPSCs durably counteracts WJMSCs senescence. This protein-based approach offers a safer strategy to delay MSC aging in vitro while providing a higher-quality culture system for clinical-scale MSC expansion. Although the triple protein combination (Pros) exerts anti-senescence effects on WJMSCs, its clinical application requires careful assessment of tumorigenic risk.

Conclusions

Collectively, our findings establish a defined protein combination (S100A9/GAPDH/S100A8) that counteracts both replicative and stress-induced senescence in WJMSCs, offering a novel, cell-free strategy to enhance their clinical utility.

Materials and methods

Cell culture

Human Wharton’s Jelly-derived MSCs (WJMSCs) were purchased from Guangzhou Celera Stem Cell Technology Co., Ltd. (Guangzhou, China). WJMSCs were routinely maintained in F12 medium supplemented with 15% fetal calf serum (FBS), 100 U/ml penicillin–streptomycin. Culture vessels were pre-coated with 0.1% (w/v) gelatin solution prior to seeding. Cells were subcultured and fresh medium was replenished every three days. HFF-1 cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Human Expanded Potential Stem Cells (hEPSCs) were established as previously described [17] and maintained on mitomycin C-inactivated mouse embryonic fibroblast (MEF) feeder cells in LCDM medium. hEPSCs were separated from feeder cells based on differential adherence properties and subsequently collected for experiments. All cell lines were routinely tested for mycoplasma contamination and confirmed negative throughout this study.

RNA isolation and quantitative real-time PCR (RT-qPCR)

Total RNA was extracted using TRIzol reagent (TAKARA, 9109) according to the manufacturer’s protocol and reverse-transcribed using reverse transcriptase (Vazyme, R223). Real-time quantitative PCR was performed using SYBR qPCR Master Mix (Vazyme, Q321) on an ABI Prism 7300 Sequence Detection System. Relative gene expression was calculated using the delta-delta CT method and normalized to GAPDH expression levels. Primer sequences are listed in Additional file 3: Table S1.

Senescence-associated β-galactosidase (SA-β-gal) assay

Cellular senescence was assessed using SA-β-gal staining. Cells are fixed with a solution containing formaldehyde and glutaraldehyde. then incubated overnight at 37 °C (withnot CO₂) in X-gal staining solution. Senescent cells were identified by blue staining under bright-field microscopy and quantified by counting positive cells in randomly selected fields.

EdU incorporation assay

Cell proliferation was measured using 5-ethynyl-2′-deoxyuridine (EdU) incorporation. Cells are incubated with EdU, which is incorporated into nascent DNA during the S phase of the cell cycle. Following incubation, EdU was detected via click chemistry using an azide-conjugated fluorescent dye. Cells were subsequently fixed, permeabilized, and counterstained for nuclei visualization under a fluorescence microscope.

Cell viability assay (CCK-8)

Cell viability and proliferation were assessed using the cell counting kit-8 (CCK-8). Cells were seeded in 96-well plates and cultured under experimental conditions. CCK-8 reagent was added to each well, and plates were incubated at 37 °C for 1–4 h. Absorbance was measured at 450 nm using a microplate reader, with optical density (OD) values proportional to viable cell numbers.

Secret proteins isolation and proteomic analysis

Secreted proteins were isolated from conditioned medium using sequential ultrafiltration. Briefly, 30 mL of cell-culture supernatant was centrifuged at 4000 rpm for 10 min at 4 °C to remove cellular debris. The supernatant was transferred to a 10 kDa molecular weight cut-off (MWCO) ultrafiltration tube and centrifuged at 4000 rpm at 4 °C for 30 min; the filtrate (flow-through) was collected from the lower chamber. This filtrate was subsequently transferred to a 3 kDa MWCO ultrafiltration tube and centrifuged at 4000 rpm at 4 °C for 35 min; the resulting filtrate was collected. Proteins were precipitated by adding one-ninth volume of 100% trichloroacetic acid (TCA), mixed thoroughly, and incubated at − 20 °C for 10 min. Following centrifugation at 15,000 ×g at 4 °C for 15 min, the supernatant was discarded and the protein pellet was washed with ice-cold acetone. The pellet was centrifuged again at 15,000 ×g at 4 °C for 5 min, air-dried briefly, and resuspended in SDS loading buffer. After heating at 95 °C for 5 min, samples were separated by SDS-PAGE, and protein bands were excised for mass spectrometry identification.

Recombinant protein expression and purification

Target plasmids were transformed into E. coli BL21 (DE3) competent cells. A single colony was inoculated into 5 mL Luria–Bertani (LB) medium containing appropriate antibiotics and cultured overnight at 37 °C with shaking. The overnight culture was diluted 1:50 into fresh LB medium and grown at 37 °C until reaching an optical density (OD₆₀₀) of 0.6–0.8. Protein expression was induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG) to a final concentration of 0.5 mM, followed by incubation at 16 °C with shaking at 180 rpm for 16–24 h. Cells were harvested by centrifugation at 4000 rpm for 10 min at 4 °C and resuspended in lysis buffer (8 mL per gram wet weight). Cell lysis was performed on ice by sonication (15 s pulses with 30 s intervals, 30% amplitude, total duration ~ 45 min). The lysate was clarified by centrifugation at 15,000 rpm for 15 min at 4 °C, and the supernatant was incubated with pre-equilibrated Ni–NTA resin for 1 h at 4 °C. The mixture was loaded onto a column, washed with binding buffer, and bound protein was eluted with elution buffer. The protein was subjected to further purification by size-exclusion chromatography (SEC). Purified protein was concentrated using a 10 kDa centrifugal filter unit, quantified by BCA assay, and stored at − 80 °C until use.

Immunofluorescence staining

Cells were seeded on 15 mm glass coverslips in 12-well plates pre-coated with 0.1% gelatin (37 °C, 1 h). Following treatment, cells were rinsed twice with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde (PFA) at 4 °C for 15 min, permeabilized with 0.2% Triton X-100 at room temperature for 10 min, and blocked with 3% goat serum at room temperature for 1 h. Samples were incubated with primary antibody overnight at 4 °C, washed three times with PBST (PBS containing 0.05% Tween-20), and incubated with Alexa Fluor-conjugated secondary antibody at room temperature for 1 h in the dark. After three additional washes with PBST, coverslips were mounted cell-side down onto glass slides using mounting medium containing DAPI and sealed with nail polish. Fluorescence images were acquired using a fluorescence microscope, and samples were stored at 4 °C until imaging.

Western blot analysis

Cells were harvested, washed once with ice-cold PBS, and pelleted by centrifugation at 13,000 ×g for 30 s. Cell pellets were lysed in 0.5–1 mL RIPA buffer on ice for 15 min, and lysates were clarified by centrifugation at 15,000 ×g at 4 °C for 15 min. Protein samples (50 μg lysate) were mixed with 5 × SDS loading buffer, denatured at 95 °C for 5 min, and separated by 10% SDS-PAGE. Proteins were transferred to PVDF membranes at 0.25 A for 1 h, blocked with 5% non-fat milk in TBST for 1 h, and incubated with primary antibody (1:5000 dilution) for 1 h at room temperature or overnight at 4 °C. Following three 5-min washes with TBST, membranes were incubated with IRDye-conjugated secondary antibody (1:5000 dilution) for 1 h at room temperature in the dark. After three additional washes with TBST, protein bands were visualized using an Odyssey infrared imaging system.

RNA-sequencing analysis

RNA-seq sequencing was performed by Berry Genomics (Beijing, China). RNA quality was assessed using an Agilent 2100 Bioanalyzer, and libraries were quantified using an ABI StepOnePlus Real-Time PCR System. Sequencing was conducted on an Illumina HiSeq 2500 platform generating 150-bp paired-end reads. Raw reads in FASTQ format were filtered using Cutadapt and quality-controlled using FastQC. Clean reads were aligned to the human reference genome (hg38) using TopHat (version 1.3.2) with clean mapping ratios exceeding 94.89%. Gene expression levels were quantified using Cufflinks (version 1.1.0), normalized to fragments per kilobase of transcript per million mapped reads (FPKM) using the trimmed mean of M values (TMM) method. Differentially expressed genes are listed in Additional file 4: Table S2.

Statistical analysis

All data are presented as mean ± standard error of the mean (SEM). Statistical significance between two groups was determined using Student’s unpaired t-test, while comparisons among multiple groups were performed using one-way ANOVA unless otherwise specified. A P value < 0.05 was considered statistically significant.

Abbreviations

hMSCs

Human mesenchymal stromal cells

hESCs

Human embryonic stem cells

BM-MSC

Bone marrow-derived MSC

WJMSCs

Human Wharton’s Jelly-derived MSCs

hEPSCs

Human extended pluripotent stem cells

hESC-sEVs

hESC-derived small extracellular vesicles

DOXO

Doxorubicin

SA-β-Gal

Senescence-associated β-galactosidase

SASP

Senescence-associated secretory phenotype

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

Pros

GAPDH, S100A8, and S100A9

Author contributions

QC. F and QW. P performed most experiments, collected data, and conducted statistical analyses. HY. L, ZJ. L and ZC. L participated in data collected and analysis; HY. L additionally performed bioinformatics analyses. F.L and G.S designed experiments and wrote the manuscript with input from all authors.

Funding

This work was supported by the National Key Research and Development Program of China [2025YFA1805300 to Z.S.], the National Natural Science Foundation of China [32170802].

Availability of data and materials

The authors declare that all other data supporting the findings of this study are within the manuscript and its supplementary files are available from the corresponding authors upon request.

Declarations

Ethics approval and consent to participate

Human Wharton’s Jelly-derived MSCs (WJMSCs) were purchased from Guangzhou Celera Stem Cell Technology Co., Ltd. (Guangzhou, China). These cells were isolated from human tissue by the supplier, not by the authors of this study. The supplier confirms that all tissues were collected with written informed consent from donors and with approval from their institutional review board or independent ethics committee. The cells were used in accordance with the supplier’s terms and conditions.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Declaration of AI usage

The authors confirm that AI-assisted tools was used to improve grammar and readability in the preparation of this manuscript. This work was completed entirely by the authors without the use of artificial intelligence for research design, data analysis, or figure generation.