Mechanical loading primes MSC-derived exosomes to promote cartilage repair

Abstract

Cartilage defects remain a major clinical challenge due to the limited efficacy of current therapies and the intrinsically low regenerative capacity of chondrocytes. Mechanical loading has emerged as a promising strategy to enhance stem cell-based cartilage repair; however, the underlying molecular mechanisms remain poorly understood. Here, we show that cyclic tensile strain primes mesenchymal stem cells (MSCs) to secrete exosomes enriched in microRNA-330-3p (miR-330-3p), which markedly enhances cartilage regeneration. Mechanistically, miR-330-3p restores mitochondrial quality control in chondrocytes by engaging an FKBP4-FoxO3a-dependent mitophagy program, leading to activation of PINK1/Parkin-mediated mitochondrial clearance. The regenerative efficacy of miR-330-3p-enriched exosomes was validated in a Sprague-Dawley rat model of cartilage defects. In vitro, miR-330-3p promotes chondrocyte proliferation and migration while suppressing apoptosis, senescence, and extracellular matrix degradation. Together, these findings identify mechanically primed MSC-derived exosomes as a mechanistically informed therapeutic strategy for cartilage repair.

Graphical abstract

Highlights

• •Cyclic tensile strain (CTS) primes MSCs to secrete exosomes with enhanced cartilage-regenerative capacity.
• •CTS enriches miR-330-3p in MSC-derived exosomes, enabling mechanically regulated intercellular signaling for cartilage repair.
• •Mechanically primed exosomes restore mitochondrial quality control in chondrocytes via FKBP4–FoxO3a–dependent mitophagy.
• •CTS-derived exosomes improve chondrocyte survival, matrix preservation, and osteochondral regeneration in vitro and in vivo.

Mechanically primed MSC exosomes improve cartilage repair by delivering miR-330-3p to restore mitochondrial homeostasis in chondrocytes.

1. Introduction

Osteochondral defects represent a persistent clinical challenge [1], driven by the limited intrinsic regenerative capacity of cartilage and the inability of current interventions to achieve durable structural and mechanical integration [2,3]. Despite advances in cell-based and tissue-engineering approaches, long-term functional restoration remains difficult, particularly in load-bearing joints [4]. Thus, there is a pressing need for therapeutic strategies that can enhance cartilage regeneration while maintaining mechanical integrity.

Mechanical loading has emerged as a clinically relevant regulator of cartilage homeostasis and regeneration and is increasingly incorporated into rehabilitation and regenerative strategies [5]. Mesenchymal stem cells (MSCs) based therapies [6], in particular, exhibit enhanced reparative efficacy when combined with mechanical stimulation [7,8]. However, how mechanical cues are translated into sustained pro-regenerative signals that can be therapeutically harnessed remains poorly understood [[9], [10], [11], [12]].

Exosomes (Exos), nanoscale extracellular vesicles that mediate intercellular communication, have emerged as important mediators of tissue regeneration and cellular adaptation to environmental stimuli [13,14]. MSC-derived Exos carry diverse bioactive cargos, including microRNAs, proteins, and lipids, which can regulate chondrocyte survival, metabolism, inflammatory responses, and extracellular matrix (ECM) maintenance [15,16]. Previous studies have reported that mechanical stimulation can modulate EV/Exo secretion, cargo composition, and biological activity, suggesting that physical cues may be used to engineer the therapeutic properties of MSC-derived Exos [17,18]. However, the key mechano-responsive exosomal microRNAs and downstream mechanisms linking mechanically conditioned MSC-derived Exos to mitochondrial quality control and cartilage repair remain insufficiently understood [19,20].

Here, we show that cyclic tensile strain (CTS) primes MSCs to secrete Exos with enhanced regenerative capacity. CTS selectively enriches miR-330-3p in MSC-derived Exos, which promotes cartilage repair by restoring mitochondrial quality control in chondrocytes through an FK506-binding protein 4 (FKBP4)–forkhead box O3a (FoxO3a)-dependent mitophagy program. Using complementary in vitro assays and a rat osteochondral defect model, we demonstrate that mechanically primed Exos enhance chondrocyte survival, matrix preservation, and functional cartilage regeneration. Together, these findings identify Exos-mediated signal propagation as a key mechanism linking mechanical loading to cartilage repair and establish a translational framework for mechanically guided Exos-based therapies.

2. Results

2.1. CTS enriches miR-330-3p in MSC-Exos

To investigate the influence of mechanical stimulation on exosomal miRNA profiles, human umbilical cord-derived mesenchymal stem cells (UMSCs) were cultured under static and CTS conditions using the Flexcell FX-5000™ tension system (Fig. S1). UMSCs displayed typical fibroblast-like morphology, expressed CD29, CD90, and CD105, and possessed trilineage differentiation potential (Fig. S2). Exos were isolated from conditioned media by differential centrifugation (Fig. S3). Both CTS-derived (CTS-Exos) and static-derived (STA-Exos) Exos exhibited a cup-shaped morphology with comparable size distributions (∼80 nm) (Fig. S4a–b). Western blot analysis confirmed the presence of CD9, CD63, and tumor susceptibility gene 101 (TSG101), along with the absence of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Fig. S4c), indicating successful Exos purification.

To optimize mechanical parameters, CTS duration (2-12 h), frequency (0.5-2 Hz), and strain amplitude (2.5-12.5%) were systematically varied. Exos yield increased steadily up to 10 h of stimulation and plateaued thereafter, while excessive duration reduced Exos purity (Fig. S5). Under optimal conditions (1 Hz, 10% strain, 10 h), CTS stimulation enhanced Exos yield by 8.3-fold and purity by 7.8-fold compared to static culture (Fig. S6). Additionally, treatment with GW4869, an Exos release inhibitor, significantly reduced Exos secretion, confirming the vesicles’ MSC origin (Fig. S7).

To elucidate the molecular changes induced by CTS, we conducted high-throughput miRNA sequencing of CTS-Exos and STA-Exos (Fig. S8). This analysis identified nine differentially expressed miRNAs between CTS-Exos and STA-Exos, among which miR-330-3p was one of the most significantly enriched candidates in CTS-Exos (Fig. 1a–c, Fig. S9). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses revealed enrichment in pathways related to cell proliferation, ECM remodeling, autophagy, and FoxO signaling (Fig. S10–S13). RT-qPCR further validated the enrichment of miR-330-3p in CTS-Exos and confirmed its increased abundance in CTS-Exos-treated chondrocytes (Fig. 1d). Bioinformatics analysis predicted FKBP4 as a conserved target of miR-330-3p (Fig. 1e–f), which was confirmed by dual-luciferase reporter assays (Fig. 1g). Western blot further showed that FKBP4 protein levels were downregulated upon CTS-Exos treatment (Fig. 1h, Fig. S14).

Fig. 1.

Upregulated miR-330-3p in CTS-stimulated MSC-derived Exos targets FKBP4. (a) Venn diagram showing the distribution of differentially expressed miRNAs between CTS-Exos and STA-Exos groups: 168 shared miRNAs, 94 specific to CTS-Exos, and 18 unique to the control group. (b) Heatmap of selected differentially expressed miRNAs, with hsa-miR-330-3p markedly upregulated in the CTS-Exos group (highlighted in red). (c) Volcano plot illustrating significantly altered miRNAs between groups, with hsa-miR-330-3p prominently upregulated in CTS-Exos (highlighted in red). (d) Bar graph comparing the relative expression of selected miRNAs in Ctrl, STA-Exos, and CTS-Exos groups. hsa-miR-330-3p exhibits the highest upregulation in the CTS-Exos group. Data are presented as mean ± SD. ns, not significant, ∗∗∗∗P < 0.0001. (n = 3 biological replicates per group). (e) Venn diagram of predicted target genes of hsa-miR-330-3p based on overlap among miRDB, miRTarBase, miRWalk, and TargetScan databases. (f) Schematic diagram showing the predicted interaction between miR-330-3p and the 3′UTR of FKBP4 in both wild-type (WT) and mutant (MUT) constructs. (g) Dual-luciferase reporter assay showing that miR-330-3p significantly reduces luciferase activity in cells transfected with WT FKBP4 3′UTR, but not with the MUT construct. (n = 3 biological replicates per group). (h) Western blot analysis demonstrating reduced FKBP4 protein levels in cells treated with CTS-Exos compared to other groups. Data are presented as mean ± SD. ns, not significant, ∗∗∗∗P < 0.0001. (n = 3 biological replicates per group).

2.2. CTS-Exos restore mitochondrial quality control in chondrocytes

Mitophagy is a critical mechanism for maintaining mitochondrial homeostasis under cellular stress conditions [21,22], and has been shown to exert protective effects in chondrocytes exposed to oxidative injury [23,24]. To model oxidative injury, chondrocytes were exposed to tert-butyl hydroperoxide (TBHP) [25,26], and subsequently treated with STA-Exos or CTS-Exos. Pathway enrichment analysis revealed that CTS-Exos were associated with mitophagy-related signaling (Fig. S10). Transmission electron microscopy showed that TBHP induced marked mitochondrial disruption, whereas CTS-Exos treatment preserved mitochondrial integrity and increased the presence of mitophagosome-like structures (Fig. 2a, top).

Fig. 2.

CTS-Exos promote mitophagy via PINK1/Parkin signaling and interact with the FKBP4/FoxO3a axis. Unless otherwise indicated, oxidative stress was induced by treating chondrocytes with 50 μM TBHP for 24 h. The TBHP-only group served as the control, whereas the STA-Exos and CTS-Exos groups were subsequently treated with STA-Exos or CTS-Exos, respectively. All quantitative data were derived from three independent biological replicates per group. (a) TEM and fluorescence images showing restored mitochondrial morphology, integrity (MitoTracker Red), and membrane potential (JC-1) in CTS-Exos-treated chondrocytes under TBHP-induced oxidative stress. (b) Western blot showing increased LC3-II, PINK1, and Parkin, and reduced p62 after CTS-Exos treatment. (c) RT-qPCR quantification of mitophagy-related genes (LC3, p62, PINK1, Parkin). Data are presented as mean ± SD. ns, not significant, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001. (d–g) Immunofluorescence and line-scan analysis showing enhanced co-localization of Parkin (red)/TOM20 (green) and LC3 (red)/TOM20 (green), confirming CTS-Exos-induced mitophagy. (h) Predicted structures and docking model illustrating FKBP4-FoxO3a interaction. (i) Co-immunoprecipitation confirming physical binding between FKBP4 and FoxO3a. (j) Immunofluorescence showing co-localization of FKBP4 (green) and FoxO3a (red), and TEM revealing autophagosome formation after CTS-Exos treatment; inhibition by AS attenuated these effects. (k) Western blot analysis showing FKBP4-dependent LC3 activation and p62 degradation in the TBHP model.

Consistent with these observations, MitoTracker and JC-1 staining demonstrated that TBHP markedly reduced mitochondrial membrane potential, whereas CTS-Exos partially restored mitochondrial polarization, as reflected by normalization of the JC-1 aggregate-to-monomer ratio (Fig. 2a, bottom). CTS-Exos treatment was further associated with increased expression of mitophagy-related markers microtubule-associated protein 1 light chain 3-II (LC3-II), PTEN-induced kinase 1 (PINK1), and Parkin RBR E3 ubiquitin protein ligase (Parkin), accompanied by reduced p62 levels, as determined by western blotting and RT-qPCR (Fig. 2b and c). Immunofluorescence co-localization of translocase of outer mitochondrial membrane 20 (TOM20) with Parkin or LC3 further supported enhanced engagement of PINK1/Parkin-mediated mitophagy following CTS-Exos treatment (Fig. 2d–g).

Previous studies have suggested that FKBP4 negatively regulates autophagy, although its underlying mechanism remains incompletely defined [27,28]. To explore the upstream regulatory events linking CTS-Exos to mitophagy activation, KEGG pathway analysis highlighted potential involvement of the FoxO3a signaling axis. FoxO3a is a stress-responsive transcription factor known to promote mitophagy under oxidative conditions [29,30]. Molecular docking analysis predicted an interaction between FKBP4 and FoxO3a (Fig. 2h), which was supported by co-immunoprecipitation and immunofluorescence co-localization under TBHP stimulation (Fig. 2i and j). CTS-Exos treatment was associated with enhanced nuclear localization of FoxO3a, consistent with activation of its transcriptional activity.

To further assess the functional contribution of FoxO3a, chondrocytes were treated with AS1842856, a selective FoxO3a inhibitor [31]. FoxO3a inhibition attenuated CTS-Exos-associated formation of double-membrane autophagosomes surrounding damaged mitochondria, as observed by transmission electron microscopy (TEM) (Fig. 2j). CTS-Exos increased FoxO3a abundance and nuclear localization, whereas AS1842856 markedly reduced these effects. CTS-Exos also suppressed FKBP4 expression, while FoxO3a inhibition did not further alter FKBP4 levels (Fig. 2k, Fig. S15). Western blot analysis showed that CTS-Exos-induced increases in LC3 and reductions in p62 were largely abolished upon FoxO3a inhibition.

Collectively, these results indicate that CTS-Exos promote FoxO3a-dependent mitophagy in stressed chondrocytes, at least in part by alleviating FKBP4-associated repression of mitochondrial quality control pathways.

2.3. CTS-Exos improve chondrocyte survival and reparative capacity

To determine whether CTS-Exos are internalized by chondrocytes, PKH26-labeled Exos were co-cultured with chondrocytes. Both STA-Exos and CTS-Exos were efficiently taken up by recipient cells, as evidenced by intracellular red fluorescence (Fig. S16).

Enhancement of chondrocyte proliferation and migration is critical for effective cartilage repair [32]. Chondrocytes were treated with TBHP to establish an oxidative stress-induced cartilage injury model. Under TBHP-induced oxidative stress, chondrocyte proliferation was markedly suppressed, whereas treatment with STA-Exos or CTS-Exos partially restored proliferative capacity, with CTS-Exos exerting a more pronounced effect. This was supported by increased expression of the proliferation markers proliferating cell nuclear antigen (PCNA) and Cyclin D1, as well as enhanced DNA synthesis assessed by 5-ethynyl-2′-deoxyuridine (EdU) incorporation (Fig. 3a,b,e and Fig. S17a,d).

Fig. 3.

CTS-Exos enhance cell proliferation, migration, and ECM homeostasis while reducing apoptosis and senescence under oxidative stress. Unless otherwise indicated, oxidative stress was induced by treating chondrocytes with 50 μM TBHP for 24 h. The TBHP-only group served as the control, whereas the STA-Exos and CTS-Exos groups were subsequently treated with STA-Exos or CTS-Exos, respectively. All quantitative data were derived from three independent biological replicates per group. (a) Western blot analysis of PCNA and Cyclin D1 expression in control (Ctrl), TBHP, STA-Exos, and CTS-Exos groups. (b) Quantification of relative mRNA expression of PCNA and Cyclin D1 by RT-qPCR under different treatment conditions. Data are presented as mean ± SD. ns, not significant, ∗∗P < 0.01, ∗∗∗P < 0.001. (c) Representative images of wound healing assays at 0 h, 24 h, and 48 h showing chondrocyte migration capacity. (d) Transwell migration assays at 12 h and 24 h confirming improved migratory capacity in the CTS-Exos group. (e) EdU staining (red) for proliferation and TUNEL staining (red) for apoptosis, with nuclei counterstained with DAPI (blue). CTS-Exos significantly promoted proliferation and reduced apoptosis compared to TBHP and STA-Exos. (f) Western blot analysis showing the apoptosis-related proteins (Bcl-2, Bax, cleaved caspase-3) and ECM-related proteins (Agg, Col-II, ADAMTS-5, MMP-13) in chondrocytes under different treatments. (g) Immunofluorescence staining of Col-II (green) and MMP-13 (red), and SA-β-Gal staining (blue) in chondrocytes under different treatments, with nuclei counterstained with DAPI (blue), and merged images shown below.

Chondrocyte migration, which is often impaired after cartilage injury [33], was significantly improved by CTS-Exos treatment. Wound-healing assays performed at 24 h and 48 h, together with transwell migration assays conducted at 12 h and 24 h, demonstrated that CTS-Exos markedly enhanced migratory capacity compared with TBHP treatment or STA-Exos (Fig. 3c,d and Fig. S17b and c).

Chondrocyte apoptosis is a major pathological feature of cartilage degeneration and contributes to matrix loss and tissue deterioration [[34], [35], [36], [37]]. Consistent with this, TBHP exposure increased Bax and cleaved caspase-3 levels while reducing Bcl-2 expression. Treatment with STA-Exos or CTS-Exos attenuated these apoptotic changes, with CTS-Exos exhibiting a stronger protective effect, as confirmed by western blotting, RT-qPCR, and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining (Fig. 3e,f and Fig. S17e–g, Fig. S19a). In parallel, TBHP-induced chondrocyte senescence, another hallmark of cartilage degeneration [38], was significantly reduced following CTS-Exos treatment, as demonstrated by decreased senescence-associated β-galactosidase (SA-β-gal) staining (Fig. 3g and Fig. S19b).

Maintenance of ECM integrity is essential for preserving cartilage structure and mechanical function [39,40]. Collagen II (Col-II) and aggrecan (Agg) are the principal components of the cartilage ECM, whereas matrix metalloproteinase-13 (MMP-13) and A disintegrin and metalloproteinase with thrombospondin motifs-5 (ADAMTS-5) drive matrix degradation during cartilage degeneration [41]. TBHP-induced oxidative stress reduced Col-II and Agg expression and increased catabolic enzyme levels, whereas CTS-Exos restored anabolic matrix markers and more effectively suppressed MMP-13 and ADAMTS-5 compared with STA-Exos (Fig. 3f and Fig. S19c). Safranin O staining and immunofluorescence analysis further confirmed enhanced matrix deposition and reduced matrix degradation in CTS-Exos-treated chondrocytes (Fig. S18 and Fig. S19d, e). Overall, these results demonstrate that CTS-Exos enhance chondrocyte proliferation and migration while mitigating apoptosis and senescence and preserving ECM homeostasis, thereby improving the reparative capacity of chondrocytes under oxidative stress conditions.

2.4. Silencing miR-330-3p abolishes the protective and regenerative effects of CTS-Exos

To determine whether miR-330-3p is required for CTS-Exos-mediated chondrocyte protection and regeneration, chondrocytes were transfected with a miR-330-3p inhibitor to selectively suppress endogenous miR-330-3p expression. Western blotting revealed that inhibition of miR-330-3p markedly attenuated the CTS-Exos-induced upregulation of FoxO3a, LC3, PINK1, and Parkin proteins in TBHP-treated chondrocytes (Fig. 4a and b and Fig. S20). Immunofluorescence co-localization further demonstrated that CTS-Exos enhanced the overlap of the mitochondrial marker TOM20 with Parkin and LC3 under oxidative stress, whereas this effect was abolished following miR-330-3p inhibition (Fig. 4c and Fig. S21). These findings indicate that CTS-Exos promote PINK1/Parkin-mediated mitophagy through exosomal miR-330-3p delivery.

Fig. 4.

miR-330-3p is required for the protective and regenerative effects of CTS-Exos. Unless otherwise indicated, oxidative stress was induced by treating chondrocytes with 50 μM TBHP for 24 h. The TBHP-only group served as the control, whereas the STA-Exos and CTS-Exos groups were subsequently treated with STA-Exos or CTS-Exos, respectively. All quantitative data were derived from three independent biological replicates per group. (a) Western blot analysis showing that CTS-Exos treatment upregulated Parkin, PINK1, and LC3-II expression in TBHP-treated chondrocytes, whereas transfection with anti-miR-330-3p abolished these effects. GAPDH was used as the loading control. (b) Western blot of cytoplasmic and nuclear fractions demonstrating that CTS-Exos increased nuclear translocation of FoxO3a in chondrocytes, while anti-miR-330-3p reversed this translocation. GAPDH and Lamin B served as cytoplasmic and nuclear markers, respectively. (c) Immunofluorescence staining showing enhanced colocalization of Parkin or LC3 (red) with the mitochondrial marker TOM20 (green) in the CTS-Exos group, indicating mitophagy activation; this effect was blocked by anti-miR-330-3p. Nuclei were counterstained with DAPI (blue). (d) Representative images of transwell migration assays showing that CTS-Exos promoted chondrocyte migration at 12 h and 24 h compared with controls, whereas anti-miR-330-3p treatment inhibited this effect. (e) Immunofluorescence staining and senescence analysis showing that CTS-Exos enhanced Col-II expression and increased EdU-positive cells and decreased SA-β-Gal-positive senescent cells, whereas inhibition of miR-330-3p reversed these protective effects.

We next assessed the functional consequences of miR-330-3p silencing on CTS-Exos-mediated chondrocyte repair. Western blotting showed that inhibition of miR-330-3p significantly reduced the expression of PCNA and Cyclin D1 compared with the CTS-Exos group (Fig. S22). Consistent with this, wound-healing and transwell assays revealed that CTS-Exos markedly enhanced chondrocyte migration at 24 h and 48 h, while this pro-migratory effect was abolished after co-treatment with the miR-330-3p inhibitor (Fig. 4d and Fig. S23, S24). Similarly, EdU staining demonstrated robust proliferation in the CTS-Exos group, which was substantially suppressed by miR-330-3p inhibition (Fig. S25).

To further explore the role of miR-330-3p in apoptosis regulation, we examined apoptosis-related markers. CTS-Exos significantly increased the anti-apoptotic protein Bcl-2 and decreased Bax and cleaved caspase-3 expression compared with controls, whereas these effects were reversed by miR-330-3p inhibition (Fig. S26). TUNEL staining confirmed that CTS-Exos markedly reduced TBHP-induced apoptosis, while the miR-330-3p inhibitor restored extensive apoptotic cell death (Fig. S27). SA-β-gal staining showed that CTS-Exos effectively suppressed chondrocyte senescence, whereas miR-330-3p inhibition restored senescence-associated staining (Fig. S28).

In addition to their anti-apoptotic and anti-senescent effects, CTS-Exos promoted ECM homeostasis. Western blotting revealed that CTS-Exos upregulated Col-II and Agg while reducing MMP-13 and ADAMTS-5 expression under oxidative stress; these changes were largely reversed by miR-330-3p inhibition (Fig. S29). Immunofluorescence staining corroborated these findings, showing that the CTS-Exos-mediated recovery of Col-II and suppression of MMP-13 were abolished following miR-330-3p silencing (Fig. S30). These results demonstrate that miR-330-3p is indispensable for the protective and regenerative effects of CTS-Exos. Loss of miR-330-3p abolishes CTS-Exos-mediated mitophagy activation and compromises chondrocyte proliferation, migration, survival, and matrix preservation under oxidative stress conditions.

2.5. CTS-Exos promote miR-330-3p-dependent cartilage regeneration in vivo

To evaluate the therapeutic effects of CTS-Exos in vivo, a full-thickness cartilage defect model was established in the knee joints of Sprague-Dawley rats (Fig. 5a). Animals were randomly assigned to six groups: sham (SM), defect control (DF), static exosomes (SE), CTS-Exos (CE), miR-330-3p Agomir (AG), and combined treatment (AG-CE). Cartilage repair was evaluated over the indicated experimental period.

Fig. 5.

CTS-Exos synergize with miR-330-3p to promote cartilage regeneration in vivo. For all in vivo analyses, n = 8 animals per group unless otherwise indicated. (a) Schematic illustration of the cartilage defect model with weekly intra-articular injections. (b) Representative gross morphology (macroscopy) and micro-CT reconstructions of the defect sites at the indicated time point. Yellow dashed circles indicate the defect region. (c–e) Quantitative analysis of BV/TV, Tb.Th, and Tb.N in the defect regions. Data are presented as mean ± SD; ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001. (f) ICRS histological scores for cartilage repair evaluation. ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001. (g) Representative histological images of H&E, Safranin O/Fast Green, and toluidine blue staining showing cartilage regeneration across different groups. (h) Immunohistochemical staining of Agg, Col-II and FoxO3a demonstrating enhanced ECM deposition in CTS-Exos and Agomir-treated groups.

Macroscopic examination at the early stage of repair (4 weeks) revealed minimal defect filling in the DF and SE groups, whereas partial repair was observed in the CE and AG groups, with the most pronounced defect refilling in the AG-CE group (Fig. S31). At the later evaluation (8 weeks), the DF group exhibited irregular tissue formation with visible gaps, while CE and AG treatments resulted in improved tissue coverage and integration with adjacent cartilage. Notably, combined treatment with AG and CTS-Exos produced a smooth cartilage surface with superior structural continuity compared with either treatment alone (Fig. 5b). Consistent trends were also observed at the 4-week time point, as supported by histological scoring and micro-computed tomography analyses performed at 4 weeks.

Micro-computed tomography analysis further demonstrated enhanced tissue regeneration within the defect region in the CE, AG, and AG-CE groups relative to DF and SE controls (Fig. 5b and Figs. S32 and S33). Quantitative analysis revealed significantly increased bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular number (Tb.N) in all three treatment groups, with the highest values observed in the AG-CE group (Fig. 5c–e and Fig. S34).

Histological evaluation using the International Cartilage Repair Society (ICRS) scoring system confirmed superior cartilage repair in the AG-CE group compared with CE or AG treatment alone (Fig. 5f). Consistently, Safranin O/Fast Green, hematoxylin and eosin, and toluidine blue staining revealed limited cartilage regeneration in the DF group, whereas CE and AG treatments promoted matrix deposition, with the most robust cartilage-like tissue formation observed in the AG-CE group (Fig. 5g and Fig. S35). Immunohistochemical staining further revealed coordinated upregulation of the cartilage matrix markers Col-II and Agg, together with FoxO3a, in the CE, AG, and AG-CE groups relative to DF, with the most pronounced staining observed in the combined treatment group (Fig. 5h and Fig. S36). Together, these results demonstrate that CTS-Exos enhance cartilage regeneration in vivo in a miR-330-3p-dependent manner and that combined delivery of CTS-Exos and miR-330-3p agomir produces synergistic reparative effects.

3. Discussion

Osteochondral degeneration remains a major clinical challenge due to the limited intrinsic repair capacity of cartilage and the increasing burden of age- and activity-related joint disease. In this study, we demonstrate that CTS enhances the regenerative efficacy of MSC-derived Exos by selectively enriching mechano-responsive microRNAs. Among these, miR-330-3p emerges as a key mediator linking mechanical stimulation to chondrocyte repair.

Mechanistically, miR-330-3p targets FKBP4, thereby facilitating FoxO3a nuclear activation and the subsequent activation of PINK1/Parkin-mediated mitophagy. FKBP4 downregulation may relieve FKBP4-associated inhibitory regulation of FoxO3a, allowing its accumulation in the nucleus and promoting mitochondrial quality-control programs. Consistent with this model, CTS-Exos treatment enhanced FoxO3a nuclear localization, increased PINK1 and Parkin expression, promoted the recruitment of Parkin and LC3 to mitochondria, and restored mitochondrial homeostasis (Fig. 6). Preservation of mitochondrial homeostasis was accompanied by improved chondrocyte survival, enhanced ECM maintenance, and superior regenerative outcomes both in vitro and in vivo, providing insight into how mechanical cues can be translated into sustained cellular repair responses.

Fig. 6.

Schematic illustration of the mechanism by which CTS-induced Exos promote cartilage regeneration. CTS stimulation enhances the secretion of MSC-derived Exos enriched in miR-330-3p. Following intra-articular administration, CTS-induced Exos are internalized by chondrocytes, where miR-330-3p suppresses FKBP4 expression. Reduced FKBP4 levels facilitate FoxO3a nuclear localization and activation of a PINK1/Parkin-mediated mitophagy program, thereby restoring mitochondrial quality control. Enhanced mitochondrial homeostasis is associated with reduced oxidative stress, improved chondrocyte survival and matrix synthesis, and ultimately accelerated cartilage repair in osteochondral defects.

Beyond this specific mechanism, our findings highlight Exos as mechano-sensitive carriers whose molecular cargo and biological activity can be modulated by physical stimulation. In contrast to biochemical or genetic manipulation, CTS represents a physiologically relevant and potentially scalable approach to modulate exosomal function through selective microRNA enrichment, offering a versatile strategy for Exos engineering.

From a translational perspective, CTS-induced Exos represent a promising cell-free therapeutic strategy for osteochondral repair, potentially overcoming key limitations of direct MSC transplantation, such as poor engraftment and phenotypic instability. Nevertheless, several challenges remain before clinical translation can be realized. First, scalable and standardized manufacturing processes are required to ensure consistent exosome yield, purity, and bioactivity. Because mechanical stimulation may influence both exosome production and cargo composition, rigorous quality-control criteria will be needed to minimize batch-to-batch variability, including particle size distribution, particle-to-protein ratio, expression of characteristic surface markers, and miR-330-3p content. Second, the storage stability of CTS-Exos requires systematic investigation, as long-term preservation, freeze–thaw cycles, and storage conditions may affect exosome integrity and biological function. Third, although intra-articular injection is clinically feasible, freely injected exosomes may undergo rapid clearance from the joint cavity. Future delivery approaches, such as hydrogel-based retention systems, scaffold-assisted implantation, and cartilage-targeting modifications, may enhance local retention and improve therapeutic efficacy.

In addition, although the rat cartilage defect model provides valuable in vivo evidence supporting the regenerative potential of CTS-Exos, it does not fully recapitulate the chronic inflammatory, catabolic, and mechanically altered microenvironment characteristic of human degenerative cartilage disease. Therefore, the therapeutic benefits observed in this acute injury model should be interpreted within this context. Future studies should further optimize mechanical loading parameters and dosing regimens, evaluate long-term safety and biodistribution, and assess efficacy in chronic osteoarthritis and large-animal models that more closely resemble human joint degeneration.

Collectively, these findings demonstrate that mechanical stimulation can be leveraged to enhance the therapeutic potential of MSC-derived Exos through miRNA-dependent regulation of mitochondrial homeostasis, supporting the potential of a mechano-responsive Exos strategy for regenerative therapies targeting mechanically active tissues.

4. Methods

4.1. Ethics statement

This study was approved by the Animal Ethics Committee of Shanghai Tenth People's Hospital (approval number: SHDSYY-2022-1501). All animal experiments and surgical procedures were conducted in accordance with institutional guidelines for the care and use of laboratory animals.

4.2. Cell isolation and culture

UMSCs (PCS-500-010, ATCC, USA) were obtained and cultured in culture medium (PCS-500-030, ATCC, USA) containing Low Serum (PCS-500-040, ATCC, USA). Primary articular cartilage chondrocytes were isolated using the following procedure: Newborn rats were euthanized under deep anesthesia, and their skin and soft tissues were removed using scissors and forceps. The cartilage tissue was then isolated and washed with PBS. After being cut into small pieces, the cartilage tissue was digested overnight in Dulbecco's modified Eagle's medium (DMEM)/F12 (10092001, Corning, USA) supplemented with 0.2% type II collagenase (17101015, Invitrogen, USA) and 1% penicillin/streptomycin (P/S) at 37 °C. The UMSCs and chondrocytes were cultured in a thermal incubator at 37 °C with 5% CO₂.

4.3. Isolation and identification of Exos

Exos were collected according to an optimized protocol [42]. Briefly, the supernatants of UMSCs (cultured with or without CTS treatment) were collected. Then, the supernatants underwent a series of centrifugations: first centrifuged at 300g for 10 min to remove the cells, then 2000g for 20 min to remove the dead cells, 10000g for 10 min to remove the cell debris, and finally centrifuged 2 times (each time 100000g for 70 min) to obtain STA-Exos and CTS-Exos, which were subsequently washed twice with filtered PBS. STA-Exos and CTS-Exos were quantified by measuring the protein concentrations via a BCA Protein Assay Kit (P0009, Beyotime, P.R. China).

4.4. Application of CTS

CTS stimulation of UMSCs was performed using the Flexcell FX-5000™ tension system. Briefly, UMSCs were first seeded into Flexcell six-well BioFlex plates (25 mm diameter, Flexcell International Corporation, Hillsboro, NC). When the cells reached nearly 100% confluence, the cells were subjected to CTS for 2, 4, 6, 8, 10, and 12 h using a computer-controlled Flexcell FX-5000™ tension system according to the manufacturer's instructions, with amplitudes of 2.5%, 5%, 7.5%, 10%, and 12.5%, and frequencies of 0.5, 1, and 2 Hz, respectively. After induction, the supernatant was collected from the culture plate and prepared for the next step of isolating Exos. Unless otherwise specified, CTS was applied at 10%, 1 Hz for 10 h.

4.5. Experimental design

In the in vitro experiments, chondrocytes were exposed to tert-butyl hydroperoxide (TBHP, 50 μM) for 24 h to establish an oxidative stress-induced injury model. Following TBHP treatment, the medium was removed, and the cells were gently washed with PBS. Fresh complete medium containing STA-Exos or CTS-Exos was subsequently added, and then cells were cultured for the indicated time periods prior to further analyses.

In the in vivo study, a rat knee cartilage defect model was established. Animals were randomly assigned to six groups: (1) Sham group (SM), intact knee joint capsule was exposed without cartilage defect, followed by intra-articular injection of equal volume sterile PBS.; (2) Defect group (DF), cartilage defects receiving intra-articular PBS injection; (3) STA-Exos group (SE), cartilage defects treated with STA-Exos; (4) CTS-Exos group (CE), cartilage defects treated with CTS-Exos; (5) Agomir-330-3p group (AG), cartilage defects treated with miR-330-3p agomir; and (6) Agomir-330-3p + CTS-Exos group (AG-CE), cartilage defects co-treated with miR-330-3p agomir and CTS-Exos. For exosome administration, purified STA-Exos and CTS-Exos were quantified using a bicinchoninic acid (BCA) protein assay and adjusted to the same exosomal protein concentration prior to injection. Each rat received 100 μL of exosome suspension containing 100 μg of exosomal protein. Comparisons between STA-Exos and CTS-Exos were therefore performed using equivalent protein-based dosing. For agomir administration, chemically stabilized miR-330-3p agomir was dissolved in sterile PBS at a concentration of 20 μM. Each rat in the AG group received 50 μL of agomir solution (1 nmol per injection). In the AG-CE group, miR-330-3p agomir was combined with CTS-Exos immediately before administration, and the final injection volume was adjusted with PBS to maintain a consistent intra-articular injection volume across all treatment groups. Intra-articular injections were administered once weekly following surgery until the designated endpoint. Animals were euthanized at 4 and 8 weeks postoperatively, and cartilage specimens were harvested for gross evaluation and histological analysis.

4.6. Animal experiment

SD rats were used in this study. A cartilage defect model with a diameter of 2 mm and a depth of 1.5 mm was created on the femoral trochlear groove. Six groups were established in this study: 1) Sham(Knee capsule exposed without cartilage defect, equal-volume sterile PBS injected intra-articularly, SM), 2) Defect (PBS was injected into the cartilage defect, DF), 3) STA-Exos (STA-Exos was injected into the cartilage defect, SE), 4) CTS-Exos (CTS-Exos was injected into the cartilage defect, CE), 5) Agomir-330-3p (Agomir-330-3p was injected into the cartilage defect, AG), 6) Agomir + CTS-Exos (Agomir-330-3p and CTS-Exos were injected into the cartilage defect, AG-CE).

4.7. Micro-CT image analysis

The samples were fixed with 4% PFA for 48 h and then scanned via micro-computed tomography (micro-CT) (μCT 35 Scanco Medical; Switzerland). Sample scanning-related parameters are as follows: 70 kV (voltage), 200 μA (current), 20 μm (resolution), 300 ms (exposure time).

4.8. Histology and immunohistochemical analysis

Histological sections were scored by 3 independent blinded observers using the ICRS visual histological assessment. The embedded specimens were sliced serially (6 μm thick), and H&E staining, Safranin O staining, and toluidine blue staining were used to evaluate cartilage destruction. Immunohistochemistry assay was performed to evaluate cartilage matrix components/cartilage repair, including aggrecan (Agg) and type II collagen (Col-II).

4.9. Statistical analysis

Statistical analyses were performed using GraphPad Prism 8.0.2 (GraphPad Software, USA). Student's t-test was used for comparisons between two groups, while one-way analysis of variance (ANOVA) was used for comparison among multiple groups. Data were expressed as the means ± standard deviations (SDs) for n ≥ 3. A p-value of <0.05 was considered statistically significant.

Data and materials availability

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Ethics approval and consent to participate

This study was approved by the Animal Ethics Committee of Shanghai Tenth People's Hospital (approval number: SHDSYY-2022-1501). All animal experiments and surgical procedures were conducted in accordance with institutional guidelines for the care and use of laboratory animals.

CRediT authorship contribution statement

Zeng Lin: Conceptualization, Data curation, Investigation, Project administration, Writing – original draft. Chao Jia: Formal analysis, Investigation, Methodology. Hongwei Lu: Formal analysis, Project administration, Software. Lei Zhou: Investigation, Validation. Jie Guo: Formal analysis, Visualization. Borui Jin: Project administration. Jinjin Wang: Investigation, Resources. Abudula Aji: Software, Validation. Peng Luo: Investigation, Methodology. Ling Cai: Formal analysis, Methodology, Validation. Libo Jiang: Methodology, Software, Supervision. Zhenwu Wang: Conceptualization, Methodology, Validation, Visualization, Writing – review & editing. Xing Wu: Project administration, Resources, Supervision, Visualization, Writing – review & editing.

Declarations of competing interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article.