Exosomal miR-212-5p promotes tendon repair via targeting FOXO1 to activate PP1A/YAP1 signaling

Abstract

Tendon injury, resulting from repetitive strain or acute trauma, often leads to pain, reduced mobility, and impaired healing due to the limited regenerative capacity of tendon tissue. Adipose-derived stem cells (ADSCs) exosomes show therapeutic promise, though their mechanisms are unclear. We demonstrated that ADSC-Exos delivers miR-212-5p to tendon-derived stem cells (TDSCs), thereby enhancing their proliferation, migration, and tenogenic differentiation. miR-212-5p directly suppresses forkhead box protein O1 (FOXO1) by binding to its 3′UTR. This downregulation relieves transcriptional repression of protein phosphatase 1A (PP1A), thereby increasing its expression and leading to dephosphorylation and activation of Yes-associated protein 1 (YAP1) signaling. In vivo, ADSC-derived exosomal miR-212-5p promotes tendon repair in male C57BL/6 mice by downregulating FOXO1 and activating YAP1 signaling. Taken together, these findings demonstrate that ADSC-derived exosomal miR-212-5p promotes tendon repair by downregulating FOXO1 to modulate the PP1A/YAP1 axis, highlighting a exosome-based regulatory mechanism and suggesting potential therapeutic targets for tendon injury management.

Exosomal miR-212-5p from adipose stem cells promotes tendon repair by regulating FOXO1 and activating YAP1 signaling, highlighting a functional axis involved in tendon stem cell proliferation and differentiation.

Introduction

Tendon injury is a common musculoskeletal condition, often caused by repetitive strain or intense physical activity. It results in pain, impaired mobility, and a reduced quality of life¹. With more than 30 million cases reported annually, the incidence of tendon injury is rising². Due to the poor regenerative capacity of tendons, healing is typically slow and incomplete, often resulting in compromised tissue function³. Current treatments, including physical therapy, medication, and surgery, are often insufficient for full recovery and preventing reinjury⁴. Recently, tendon-derived stem cells (TDSCs) have emerged as a promising therapeutic option due to their low immunogenicity, rapid proliferation, and strong tenogenic differentiation potential^5,6. This has fueled growing interest in understanding the molecular mechanisms of tendon repair to develop more effective regenerative strategies.

Adipose-derived stem cells (ADSCs) have demonstrated great potential in tissue healing due to their abundant, easily accessible sources and lower risk of immune rejection compared to other stem cells⁷. ADSC-derived exosomes (ADSC-Exos) contain bioactive molecules like proteins, lipids, and microRNAs that enhance tendon repair by promoting the proliferation and differentiation of TDSCs⁸. Their regenerative properties make them a promising approach for tendon injury treatment. For example, one study indicated that ADSC-Exos promoted tendon regeneration by enhancing the proliferation, migration, and tenogenic differentiation of tendon stem cells while inhibiting early inflammatory responses via SMAD signaling pathways, thereby improving tendon healing in vivo⁹. Another study found that ADSC-Exos increased rotator cuff repair by promoting TDSC growth, migration, and differentiation both in vitro and in a rat injury model¹⁰. However, the mechanisms by which ADSC-Exos regulate TDSCs are not yet fully understood. Further research is required to clarify these molecular processes and maximize their therapeutic potential in treating tendon injury.

Among the key components of ADSC-Exos are miRNAs, which have garnered considerable attention due to their stability and potential applications in early disease diagnosis and targeted therapies¹¹. Increasing evidence suggests that miRNAs carried by ADSC-Exos play a pivotal role in modulating cellular processes and mitigating the progression of various diseases⁸. It was reported that miR-212-5p was enriched in ADSCs and their derived exosomes¹². Zheng et al. demonstrated that exosomes derived from miR-212-5p-overexpressing synovial mesenchymal stem cells suppressed chondrocyte degeneration and inflammation, and simultaneously promoted extracellular matrix synthesis¹³. Furthermore, miR-212 promoted the proliferation and inhibited the apoptosis of precartilaginous stem cells¹⁴. However, the specific role of miR-212-5p in regulating TDSC function and its impact on tendon injury repair remains largely unexplored.

One known mechanism of miRNA action is its ability to bind to the 3′-UTR region of target genes, promoting their degradation¹⁵. Using the Starbase database, we identified a potential interaction between miR-212-5p and forkhead box O1 (FOXO1), a member of the forkhead box O (FOXO) transcription factor family. FOXO1 is known to bind DNA elements in the promoter regions of target genes, inhibiting their transcription¹⁶. Notably, studies have shown that FOXO1 expression was significantly elevated in tendon rupture models, and high cholesterol has been shown to induce apoptosis in TDSCs through the protein kinase B (AKT)/FOXO1 signaling pathway¹⁷. Despite these findings, the specific relationship between miR-212-5p and FOXO1 in tendon injury and repair mechanisms remains unclear and requires further investigation.

We also discovered potential FOXO1 binding sites on the protein phosphatase 1 alpha (PP1A) gene in mice using the JASPAR database. Previous studies have shown that PP1A can bind directly to yes-associated protein (YAP1) in the Hippo signaling pathway, inhibiting its phosphorylation and promoting its translocation into the nucleus¹⁸. Additionally, YAP1 is known to support the regenerative potential of TDSCs, facilitating tendon repair^19,20. For instance, irisin has been reported to enhance the proliferation and differentiation of rat TDSCs in vitro by activating the YAP1/transcriptional coactivator with the PDZ-binding motif (TAZ) pathway²¹.

In this study, we uncovered that ADSC-derived exosomal miR-212-5p promoted PP1A transcription by downregulating FOXO1, thereby enhancing YAP1 dephosphorylation and nuclear translocation. This activation of YAP1 signaling could stimulate TDSC proliferation and differentiation, contributing to tendon injury repair. These findings provide mechanistic insight into the regulatory role of exosomal miR-212-5p in tendon regeneration and suggest a clinically relevant strategy for enhancing tendon healing through ADSC-derived exosomal therapy.

Results

Exosomes derived from ADSCs were uptaken by TDSCs

To understand the roles of ADSC-Exos in tendon injury, they were isolated from ADSCs and characterized using TEM, NTA, and Western blot. TEM confirmed the characteristic cup-shaped morphology of ADSC-Exos (Fig. 1A), while NTA demonstrated a size distribution ranging from 50 to 150 nm (Fig. 1B). Western blot revealed high expression of exosome markers CD9, TSG101, and HSP70, with minimal Calnexin (Fig. 1C). TDSCs were isolated from mice and subsequently characterized. Flow cytometry analysis showed that the cells were positive for mesenchymal stem cell surface markers CD90, CD105 and CD44, and negative for the hematopoietic markers CD11b and CD45 (Supplementary Fig. 1A). Furthermore, the multipotent differentiation potential of TDSCs was confirmed through induction of adipogenic, osteogenic, and chondrogenic lineages (Supplementary Fig. 1B). PKH-67 labeling demonstrated that ADSC-Exos were effectively internalized by TDSCs (Fig. 1D). These findings demonstrated that ADSC-Exos were taken up by TDSCs, potentially influencing their regenerative functions.

Fig. 1. Exosomes derived from ADSCs were uptaken by TDSCs.

A, B Exosomes were visualized by transmission electron microscopy (TEM), and their particle size was analyzed by nanoparticle tracking analysis (NTA). Scale bar: 200 nm. C Western blot was conducted to detect exosomal markers CD9, TSG101, and HSP70, along with the negative marker Calnexin. n = 3. D PKH-67-labeled ADSC-Exos (green) were internalized by TDSCs. Scale bar: 25 μm. n = 3.

ADSC-Exos induced the proliferation, migration, and tenogenic differentiation of TDSCs

To evaluate the effects of ADSC-Exos on TDSC behavior, a series of functional assays was conducted. The MTT and EdU staining assays demonstrated that ADSC-Exos significantly enhanced TDSC viability and proliferation in a dose-dependent manner, ranging from 0 μg/mL to 80 μg/mL (Fig. 2A, B). For migration, both wound healing and Transwell assays demonstrated enhanced migration rates with increasing ADSC-Exo concentrations (Fig. 2C, D). Furthermore, the ADSC-Exos upregulated the expression of tenogenic differentiation markers, including TNC, TNMD, Scx, and COL1A1, in TDSCs in a dose-dependent fashion (Fig. 2E–G). To further evaluate the effects of ADSC-Exos on the multipotent differentiation potential of TDSCs, Alizarin Red, Oil Red O, and Alcian Blue staining, along with the analysis of endothelial marker expression, were performed as shown in Fig. 2H, Alizarin Red staining and Oil Red O staining revealed no significant differences in osteogenic and adipogenic differentiation among TDSCs treated with varying concentrations of ADSC-Exos. In contrast, Alcian Blue staining, which reflects chondrogenic matrix production, demonstrated a dose-dependent reduction in matrix deposition with increasing concentrations of ADSC-Exos (Fig. 2H). Furthermore, ADSC-Exos did not significantly alter the expression levels of VEGFR2, CD31, or VE-cadherin in TDSCs, indicating limited effects on vasculogenic gene expression (Fig. 2I). Collectively, while ADSC-Exos promoted TDSC proliferation, migration, and tenogenic differentiation.

Fig. 2. ADSC-Exos promoted the proliferation, migration, and tenogenic differentiation of TDSCs.

TDSCs were treated with 0, 20, 40, or 80 μg/mL of ADSC-Exos, and functional assays were performed to assess proliferation and migration. A Cell viability was evaluated using the MTT assay. B Cell proliferation was assessed by EdU incorporation. Scale bar: 100 μm. C Migration was measured by scratch wound assay. Scale bar: 500 μm. D The Transwell migration assay was used to quantify migratory capacity. Scale bar: 100 μm. E–G The mRNA and protein levels of tenogenic markers, including TNC, TNMD, Scx, and COL1A1, were determined by qPCR and Western blot. H Staining assays for osteogenesis (Alizarin Red), adipogenesis (Oil Red O), and chondrogenesis (Alcian Blue) of TDSCs. Scale bar: 100 μm. I qPCR analysis of relative expression levels of angiogenesis-related genes (VEGFR2, CD31, VE-cadherin) in TDSCs. Data are presented as mean ± standard deviation (SD), and statistical analysis was conducted using one-way ANOVA followed by Tukey’s post hoc test. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

ADSC-Exos facilitated functional changes in TDSCs through the delivery of miR-212-5p

It is well-accepted that exosomes facilitate intercellular communication by transferring bioactive molecules, such as proteins, miRNAs, and mRNAs, to recipient cells, where they modulate cellular behavior and activate key signaling pathways²². Therefore, we aimed to explore the underlying mechanism by which ADSC-Exos regulate TDSCs. ADSC-Exos increased miR-212-5p expression in TDSCs in a dose-dependent manner (Fig. 3A). Transfection of ADSCs with a miR-212-5p inhibitor significantly suppressed miR-212-5p expression in both ADSC cells and ADSC-Exos (Fig. 3B). Next, exosomes were isolated from ADSCs transfected with miR-212-5p inhibitor and subsequently used to treat TDSCs to investigate the effects of ADSC-derived exosomes on TDSC function. When miR-212-5p was knocked down in ADSC-Exos, the enhancement of TDSC proliferation observed in the MTT and EdU assays was significantly reduced (Fig. 3C, D). For migration, miR-212-5p knockdown in exosomes also significantly impaired the promotion of TDSC migration, as shown in both the wound healing and Transwell assays (Fig. 3E, F). miR-212-5p knockdown in ADSC-Exos resulted in markedly reduced expression of tenogenic differentiation markers, including TNC, TNMD, Scx, and COL1A1 (Fig. 3G). Furthermore, treatment with ADSC-Exos reduced chondrogenic differentiation in TDSCs, as indicated by Alcian Blue staining. Notably, inhibition of miR-212-5p in ADSC-Exos restored chondrogenic differentiation to levels comparable to the control group (Fig. 3H). To directly assess the role of miR-212-5p in regulating TDSC function, TDSCs were treated with ADSC-Exos and simultaneously transfected with a miR-212-5p inhibitor. ADSC-Exos significantly enhanced TDSC viability, proliferation, and migration, and these effects were reversed by miR-212-5p inhibition (Supplementary Fig. 2A–D). Furthermore, the upregulation of tenogenic differentiation markers, TNC, TNMD, Scx, and COL1A1, induced by ADSC-Exos was attenuated by miR-212-5p inhibition (Supplementary Fig. 2E). Our findings supported that miR-212-5p played a central role in the regulatory effects of ADSC-Exos on TDSC behavior, including proliferation, migration, and tenogenic differentiation.

Fig. 3. ADSC-Exos facilitated functional changes in TDSCs through the delivery of miR-212-5p.

A qPCR analysis of miR-212-5p expression in TDSCs treated with increasing concentrations (0–80 μg/mL) of ADSC-Exos. B Verification of miR-212-5p knockdown in ADSCs and corresponding exosomes by qPCR after transfection with miR-212-5p inhibitor or negative control (NC). TDSCs were treated with exosome from miR-212-5p inhibitor-transfected ADSCs. Cell viability was evaluated by MTT assay (C), proliferation by EdU incorporation (D), migration by scratch wound assay (E), and Transwell assay (F). Scale bar: 100 or 500 μm. G Protein expression levels of tenogenic markers (TNC, TNMD, Scx, COL1A1) were analyzed by Western blot, with densitometric quantification normalized to GAPDH. H Alcian Blue staining was performed to evaluate cartilage-like matrix deposition in TDSCs. Scale bar: 100 μm. Data are presented as mean ± SD, and statistical analysis was conducted using a student t-test (for B) of one-way ANOVA followed by Tukey’s post hoc test (for A, C–G). n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

ADSC-derived exosomal miR-212-5p activated YAP1 signaling in TDSCs by targeting and inhibiting FOXO1

Next, we investigated the regulatory mechanism of ADSC-derived exosomal miR-212-5p in TDSCs. ADSC-Exos transfected with a miR-212-5p inhibitor and subsequently used to treat TDSCs. A series of experiments were then performed. Treatment with ADSC-Exos significantly inhibited FOXO1 mRNA and protein expression, as well as phosphorylated YAP1 levels in TDSCs, while knocking down miR-212-5p in ADSCs attenuated the regulatory effects of ADSC-Exos on FOXO1 expression and phosphorylated YAP1 levels (Fig. 4A, B). Bioinformatics analysis predicted an interaction between miR-212-5p and FOXO1 (Fig. 4C). The expression of miR-212-5p was overexpressed in TDSCs by transfecting them with miR-212-5p mimics, and the efficiency was validated by qPCR (Fig. 4D). Exosomes derived from ADSCs or overexpression of miR-212-5p remarkably suppressed luciferase activity in TDSCs transfected with the FOXO1-WT construct (Fig. 4E). This inhibitory effect was not observed in cells transfected with the FOXO1-MUT construct, indicating that miR-212-5p specifically targeted the FOXO1 binding site (Fig. 4E). In contrast, exosomes derived from miR-212-5p knockdown ADSCs (Exo-miR-212-5p inhibitor) failed to inhibit luciferase activity in FOXO1-WT-transfected TDSCs, further confirming the regulatory role of miR-212-5p (Fig. 4E). RNA pull-down assays further demonstrated direct binding between miR-212-5p and FOXO1 (Fig. 4F). Finally, FOXO1 was knocked down in TDSCs, and transfection with shFOXO1 inhibited FOXO1 expression in TDSCs (Fig. 4G). Treatment with miR-212-5p-depleted ADSC-Exos increased FOXO1 and YAP1 phosphorylation levels in TDSCs, while simultaneous knockdown of FOXO1 in TDSCs reversed these effects (Fig. 4H). Additionally, TDSCs were treated with ADSC-Exos and transfected with a miR-212-5p inhibitor. miR-212-5p inhibition reversed ADSC-Exo-induced suppression of FOXO1 mRNA and protein, and increased phosphorylation of YAP1 at serine 127 (Supplementary Fig. 3A, B). Dual-luciferase reporter assays showed that luciferase activity in FOXO1-WT-transfected TDSCs was reduced by ADSC-Exos and then rescued by miR-212-5p inhibition, but not FOXO1-MUT constructs (Supplementary Fig. 3C). Moreover, PP1A upregulation by ADSC-Exos was attenuated upon miR-212-5p inhibition (Supplementary Fig. 3D, E). Overall, ADSC-derived exosomal miR-212-5p activated YAP1 signaling in TDSCs through the suppression of FOXO1.

Fig. 4. ADSC-derived exosomal miR-212-5p activated YAP1 signaling in TDSCs by targeting and inhibiting FOXO1.

A qPCR was performed to assess FOXO1 expression in TDSCs after treatment with exosomes derived from miR-212-5p-depleted ADSCs. B Western blot was conducted to evaluate FOXO1 and p-YAP1 levels after treatment with exosomes derived from miR-212-5p-depleted ADSCs. C Bioinformatics analysis using the Starbase database was performed to predict the target interaction between miR-212-5p and FOXO1. D qPCR was used to measure miR-212-5p levels in TDSCs after transfection with miR-212-5p mimics. E Dual-luciferase reporter assay was used to assess luciferase activity after treatment with exosomes derived from miR-212-5p-depleted or miR-212-5p-overexpressing ADSCs. F RNA pull-down assay was conducted using biotin-labeled miR-212-5p to evaluate its binding interaction with FOXO1. G qPCR was performed to assess FOXO1 expression after transfection with shFOXO1 in ADSCs. H Western blot analysis was used to evaluate FOXO1 and p-YAP1 levels after knockdown of miR-212-5p in ADSC-Exos and simultaneous knockdown of FOXO1. Data are presented as mean ± SD, and statistical analysis was conducted using a student t-test (for D, F–G) of one-way ANOVA followed by Tukey’s post hoc test (for A, B, E, H). n = 3.*p < 0.05, **p < 0.01, ***p < 0.001.

FOXO1 directly bound the PP1A promoter to suppress its expression, thereby promoting YAP1 phosphorylation and inhibiting YAP1 signaling

PP1A is known to regulate YAP1 signaling by dephosphorylating YAP1, leading to its nuclear activation and transcriptional activity^23,24. Therefore, we next explored the relationship between exosomal miR-212-5p, FOXO1, and PP1A in TDSCs. ADSC exosomes significantly increased PP1A mRNA and protein expression in TDSCs, but this effect was diminished upon knockdown of exosomal miR-212-5p, indicating its essential role in PP1A upregulation (Fig. 5A, B). Notably, FOXO1 depletion reversed the suppression of PP1A expression caused by miR-212-5p knockdown (Fig. 5A, B). Bioinformatic analysis predicted binding sites for FOXO1 on the PP1A promoter (Fig. 5C). Furthermore, ChIP results showed that the FOXO1 antibody enriched binding site 1 on the PP1A promoter, indicating that FOXO1 bound to site 1 of the PP1A promoter (Fig. 5D). FOXO1 was overexpressed in TDSCs and the efficiency of FOXO1 overexpression was confirmed by qPCR and Western blot (Fig. 5E, F). Knockdown of FOXO1 in TDSCs increased PP1A transcriptional activity, while FOXO1 overexpression had the opposite effect (Fig. 5G). qPCR data further supported these findings, showing that FOXO1 knockdown led to increased PP1A expression, whereas FOXO1 overexpression reduced it (Fig. 5H). To confirm the regulatory role of PP1A, PP1A was knocked down in TDSCs by transfection with shPP1A. As confirmed by qPCR, transfection of shPP1A effectively inhibited PP1A mRNA expression (Fig. 5I). FOXO1 knockdown increased PP1A expression and decreased YAP1 phosphorylation levels in TDSCs, whereas co-knockdown of PP1A reversed these changes (Fig. 5J, K). These findings suggested that FOXO1 directly interacted with the PP1A promoter, repressing its transcription and subsequently enhancing YAP1 phosphorylation, which led to the inhibition of YAP1 signaling.

Fig. 5. FOXO1 directly bound the PP1A promoter to suppress its expression, thereby promoting YAP1 phosphorylation and inhibiting YAP1 signaling.

A, B qPCR and Western blot were performed to analyze PP1A levels in TDSCs transfected with shNC or shFOXO1 after treatment with exosomes derived from miR-212-5p-depleted ADSCs. C The JASPAR database was used to predict FOXO1 binding sites on the PP1A promoter. D ChIP assay was conducted using a FOXO1 antibody to assess its binding to the PP1A promoter. E, F qPCR and Western blot were performed to evaluate FOXO1 expression in TDSCs after transfection with FOXO1 overexpression vectors. G Dual-luciferase reporter assay was used to investigate the impact of FOXO1 knockdown and overexpression on PP1A transcriptional activity. H qPCR was performed to assess PP1A expression following FOXO1 knockdown and overexpression in TDSCs. I qPCR was conducted to evaluate PP1A expression after transfection with shPP1A in TDSCs. J Western blot was performed to study the effects of FOXO1 and PP1A knockdown on PP1A and p-YAP1 levels. K Immunofluorescence was used to examine nuclear YAP1 levels (green) in TDSCs after FOXO1 and PP1A knockdown. Scale bar: 100 μm. Data are presented as mean ± SD, and statistical analysis was conducted using a student t-test (for D–F, I) of one-way ANOVA followed by Tukey’s post hoc test (for A, B, G, H, J). n = 3 **p < 0.01, ***p < 0.001.

Exosomal miR-212-5p derived from ADSCs promoted TDSC proliferation, migration, and tenogenic differentiation by suppressing FOXO1

To validate the regulatory role of FOXO1 in exosomal miR-212-5p-mediated regulation of TDSC behavior, miR-212-5p was knocked down in ADSC-Exos, and/or FOXO1 was depleted in TDSCs. Treatment with miR-212-5p-depleted ADSC-Exos significantly reduced TDSC proliferation and migration compared to those treated with normal ADSC-Exos (Fig. 6A–D). However, FOXO1 knockdown reversed these inhibitory effects of miR-212-5p-depleted ADSC-Exos, restoring TDSC proliferation and migration (Fig. 6A–D). Similarly, treatment with miR-212-5p-depleted ADSC-Exos reduced the protein expression of tenogenic differentiation markers, including TNC, TNMD, Scx, and COL1A1, in TDSCs, which was also reversed by FOXO1 knockdown (Fig. 6E). TDSCs treated with ADSC-Exos in which miR-212-5p was inhibited exhibited enhanced chondrogenic differentiation, as indicated by increased Alcian Blue staining compared to the Exo-inhibitor NC group. However, knockdown of FOXO1 in TDSCs reversed this pro-chondrogenic effect, resulting in reduced Alcian Blue staining to levels comparable with the control group (Fig. 6F). Altogether, exosomal miR-212-5p isolated from ADSCs mediated TDSC proliferation, migration, and tenogenic differentiation by targeting FOXO1.

Fig. 6. Exosomal miR-212-5p derived from ADSCs promoted TDSC proliferation, migration, and tenogenic differentiation by suppressing FOXO1.

A–D MTT, EdU staining, scratch, and Transwell assays were used to evaluate the proliferation, migration, and tenogenic differentiation of TSDCs transfected with shNC or shFOXO1 upon treatment with exosomes derived from miR-212-5p-depleted ADSCs Scale bar: 100 or 500 μm. E The expression of TNC, TNMD, Scx, COL1A1, and GAPDH in TSDCs was assessed by Western blot. F Alcian Blue staining was performed to evaluate cartilage-like matrix deposition in TDSCs. Scale bar: 100 μm. Data are presented as mean ± SD, and statistical analysis was conducted using one-way ANOVA followed by Tukey’s post hoc test. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001.

FOXO1 regulated TDSC proliferation, migration, and tenogenic differentiation by promoting YAP1 phosphorylation

Similarly, to demonstrate the role of YAP1 signaling in FOXO1-mediated TDSC behavior, we introduced an S127A mutation in YAP1. Transfection with the wild-type YAP1 expression vector increased both total YAP1 and phosphorylated YAP1 levels in TDSCs (Fig. 7A). In contrast, transfection with the S127A mutant YAP1 expression vector increased YAP1 levels but did not affect phosphorylated YAP1 (Fig. 7A), indicating that the S127A mutation prevents YAP1 phosphorylation. Overexpression of FOXO1 significantly inhibited TDSC proliferation, migration, and the expression of tenogenic differentiation markers, including TNC, TNMD, Scx, and COL1A1 (Fig. 7B–F). However, simultaneous overexpression of the S127A mutant YAP1 reversed the inhibitory effects of FOXO1 overexpression, restoring TDSC proliferation, migration, and differentiation. Overexpression of FOXO1 slightly enhanced chondrogenic matrix production compared to the vector control, andco-overexpression of YAP1-S127A attenuated this effect (Fig. 7G). In summary, FOXO1 suppressed TDSC function by inducing YAP1 phosphorylation.

Fig. 7. FOXO1 regulates TDSC proliferation, migration, and tenogenic differentiation by promoting YAP1 phosphorylation.

A Western blot was performed to analyze the effects of transfection with wild-type YAP1 (YAP1-WT) expression vectors or S127A mutant YAP1 expression vectors (YAP1-S127A) on YAP1 and p-YAP1 levels. B–F MTT, EdU staining, scratch assay, Transwell assay, and Western blot were used to evaluate TDSC proliferation, migration, and the expression of tenogenic differentiation markers, including TNC, TNMD, Scx, COL1A1, and GAPDH, after FOXO1 overexpression with simultaneous overexpression of the S127A mutant YAP1. Scale bar: 100 or 500 μm. G Alcian Blue staining was performed to evaluate cartilage-like matrix deposition in TDSCs. Scale bar: 100 μm. Data are presented as mean ± SD, and statistical analysis was conducted using one-way ANOVA followed by Tukey’s post hoc test. n = 3. **p < 0.01, ***p < 0.001.

ADSC-derived exosomal miR-212-5p alleviated tendon injury in mice by regulating the FOXO1/PP1A/YAP1 signaling axis

To further validate the in vitro results in vivo, a tendon injury model was established in mice. Treatment with ADSC-Exos significantly improved tendon healing, as evidenced by the more orderly arrangement of fibrous tissue in the exosome-treated group (Fig. 8A). In contrast, knockdown of exosomal miR-212-5p showed no such effect. Safranin O/Fast Green Staining was subsequently performed to assess cartilage-like matrix formation during tendon healing (Fig. 8B). ADSC-Exo treatment markedly reduced metachromatic matrix deposition compared to the control group, while inhibition of miR-212-5p in ADSC-Exos reversed this suppressive effect. Expression of tenogenic markers TNMD, Scx, and COL1A1 was significantly upregulated in ADSC-Exos-treated group but not elevated in the miR-212-5p knockdown group (Fig. 8C). Similarly, the expression of cartilage-associated marker SOX9 was downregulated in ADSC-Exos treatment group, whereas knockdown of miR-212-5p attenuated this regulatory effect (Fig. 8C). Furthermore, ADSC-Exos treatment increased the expression of TNC, TNMD, Scx, COL1A1, and PP1A, while reducing the levels of FOXO1 and phosphorylated YAP1 (Fig. 8D, E). These molecular changes were not observed following miR-212-5p silencing (Fig. 8D, E). Moreover, tendon failure load and stiffness were significantly enhanced in the ADSC exosome-treated group, indicating improved tendon strength (Fig. 8F, G). However, no improvements were observed in the miR-212-5p knockdown group (Fig. 8F, G). These data indicated that ADSC-derived exosomal miR-212-5p promoted tendon repair by regulating the FOXO1/PP1A/YAP1 signaling axis.

Fig. 8. ADSC-derived exosomal miR-212-5p alleviated tendon injury in mice by regulating the FOXO1/PP1A/YAP1 signaling axis.

A Hematoxylin and eosin (HE) staining was performed to evaluate the arrangement of fibrous tissue in tendons from mice treated with exosomes derived from normal or miR-212-5p-depleted ADSCs. Scale bar: 50 or 20 μm. B Safranin O/Fast Green staining of tendon sections to evaluate ectopic cartilage-like matrix formation after injury in different groups of mice. Scale bar: 50 or 20 μm. C Immunohistochemistry (IHC) was used to assess the levels of TNMD, Scx, COL1A1, and SOX9 in tendon tissues from different groups of mice. Scale bar: 20 μm D, E Western blot analysis was conducted to examine the expression of tenogenic differentiation markers, including TNC, TNMD, Scx, and COL1A1, as well as PP1A, FOXO1, and p-YAP1 levels in tendon tissues from different groups of mice. F, G Biomechanical testing was performed to measure the failure load and stiffness of tendon tissues from different groups of mice. Data are presented as mean ± SD, and statistical analysis was conducted using one-way ANOVA followed by Tukey’s post hoc test. n = 6. *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

Tendon injury, caused by repetitive strain or intense activity, leads to pain, limited mobility, and reduced quality of life, while current treatments often fail to ensure full recovery despite its rising incidence²⁵. ADSCs have shown promise in promoting tendon repair due to their regenerative potential, but the precise mechanisms by which ADSCs aid in tendon healing remain unclear^26,27. In this study, we demonstrated that ADSC-Exos delivered miR-212-5p to TDSCs, promoting proliferation, migration, and tendon differentiation to improve tendon injury by regulating the FOXO1/PP1A/YAP1 axis. Mechanistically, exosomal miR-212-5p bound to FOXO1 and suppressed its expression levels, subsequently reducing PP1A expression. This led to increased YAP1 phosphorylation and deactivation of the YAP1 signaling pathway, thereby affecting TDSC function and tendon repair. This study offers valuable insights into the potential of miR-212-5p as a therapeutic target for promoting tendon repair and regeneration.

Exosomes are crucial mediators of intercellular communication, transferring a variety of bioactive molecules, including proteins, lipids, and RNAs such as miRNAs, to regulate numerous physiological processes²⁸. Among these, ADSC-Exos have garnered increasing attention for their regenerative potential, particularly in tissue repair^29,30. In the context of tendon injury, ADSC-Exos exert diverse effects, which can be categorized into four primary functions: promoting angiogenesis, reducing inflammation, stimulating tendon cell proliferation and migration, and accelerating collagen synthesis, which are essential for effective tendon healing^9,29–31. For instance, ADSC-Exos modulate inflammation by reducing inflammatory cell infiltration and promoting anti-inflammatory factor release, enhancing tendon healing³². Liu et al. further showed that ADSC-Exos were absorbed by TDSCs, promoting proliferation, migration, and tenogenic differentiation through SMAD2/3 and SMAD1/5/9 signaling pathways essential for tendon regeneration⁹. However, despite these promising findings, current research on ADSC-Exos in tendon injury and the underlying mechanism remains limited. In our studies, consistent with previous studies, we demonstrated that ADSC-Exos significantly enhanced TDSC function by promoting proliferation, migration, and tenogenic differentiation, leading to improved tendon repair. These results highlight the potential of ADSC-Exos as a therapeutic strategy for tendon injuries.

It is well-known that exosomes exert their regulatory roles via the transfer of bioactive molecules, such as miRNAs, proteins, and lipids, which influence cellular functions and intercellular communication³³. Previous studies documented that miRNAs derived from ADSC-Exos have shown significant therapeutic effects in various diseases. For example, ADSC-Exos increased tenocyte proliferation, migration, and collagen production via miR-144-3p³⁴. Another study indicated that ADSC-Exos induced cartilage formation and reduced inflammation by increasing the levels of miR-145 and miR-221³⁵. These findings reflect the functional diversity of exosomal miRNAs, yet the mechanisms by which individual miRNAs exert their effects in tendon repair remain largely uncharacterized. Our focus on miR-212-5p was driven by its established enrichment in ADSCs and their exosomes¹²; its documented roles in suppressing inflammation and stem cell apoptosis, as well as enhancing collagen formation and stem cell proliferation^13,14; and the absence of any prior investigation into its function in tendon biology, positioning it as a high-potential candidate for novel discovery. We found that silencing miR-212-5p in ADSC-Exos significantly reduced their ability to promote TDSC function, identifying miR-212-5p as a key mediator. In a related context, interference with miR-212 and miR-384 has been shown to promote osteogenic differentiation by targeting runt-related transcription factor 2 (RUNX2) in osteoporosis³⁶. This study revealed the alleviating effect of ADSC-Exos miR-212-5p in tendon injury, highlighting its therapeutic potential for tendon repair and broader tissue regeneration applications.

FOXO1 is a transcription factor that plays a vital role in regulating various cellular processes, including oxidative stress response, apoptosis, and metabolism³⁷. Previous studies showed that FOXO1 was closely involved in tendon regulation, particularly influencing tendinopathy progression by modulating the transcriptional activity of peroxiredoxin 2 (PRDX2)³⁸. In tendon injury models, FOXO1 expression was significantly elevated, and high cholesterol levels were found to induce apoptosis of TDSCs via activation of the AKT/FOXO1 signaling pathway¹⁷. Despite these findings, the role of FOXO1 in tendon healing and the underlying mechanism remain poorly understood. Mechanistically, FOXO1 is post-transcriptionally regulated by various non-coding RNAs, including microRNAs, which suppress its activity and influence numerous biological and pathological processes³⁹. For example, miR-223 has been shown to promote the proliferation of several cancer cell lines by interacting with FOXO1 and inhibiting its expression⁴⁰. In our study, we observed that exosomal miR-212-5p from ADSCs targeted FOXO1 and suppressed its expression in TDSCs. Knockdown of exosomal miR-212-5p upregulated FOXO1, impairing TDSC proliferation, migration, and tenogenic differentiation, highlighting its role in regulating tendon repair.

PP1A is a serine/threonine phosphatase involved in dephosphorylating proteins, thereby regulating key cellular processes such as the cell cycle and gene expression⁴¹. In the YAP1 signaling pathway, PP1A plays a crucial role by dephosphorylating YAP1, enabling its translocation to the nucleus, where it activates genes related to cell proliferation and apoptosis²⁴. Previous studies showed that YAP1 was critical in regulating tendon injury. For example, YAP1 inhibition through phosphorylation reduced the expression of tendon-related genes⁴², while YAP1 activation through dephosphorylation promoted tendon healing by increasing the expression of genes essential for tissue repair⁴³. Small extracellular vesicles containing long non-coding RNA H19 promoted tendon healing by activating YAP1, which translocated to the nucleus and activated key tendon repair genes¹⁹. The upstream regulators of PP1A in tendon repair are unclear. Building on previous findings, we demonstrated that FOXO1 bound to the PP1A promoter, suppressing its expression. This regulation increased YAP1 phosphorylation, reduced nuclear YAP1 levels, and deactivated YAP1 signaling in TDSCs, highlighting the critical role of the PP1A/YAP1 axis in tendon repair.

In summary, ADSC-derived exosomal miR-212-5p promotes tendon injury repair by downregulating FOXO1, which enhances PP1A transcription, leading to YAP1 dephosphorylation and the stimulation of TDSC proliferation and differentiation. These findings provide insights into the molecular mechanisms of tendon repair and suggest that targeting exosomal miR-212-5p holds promise as a potential therapeutic agent for tendon injury in clinical settings. However, it is worth noting that such an acute tendon injury mode does not fully recapitulate the complex degenerative pathology of chronic tendinopathy resulting from overuse. Future studies incorporating extended time-course analyses and, importantly, validation in more chronic or repair-integrated models will be essential to fully elucidate the long-term effects of this treatment and better reflect the clinical spectrum of tendon disorders.

Methods

Animal ethics and compliance statement

All experimental procedures involving animals were conducted in accordance with the ARRIVE 2.0 guidelines and were approved by the Institutional Animal Care and Use Committee of Shengli Clinical Medical College of Fujian Medical University (approval number: IACUC-FPH-SL-20230724[0048]). We have complied with all relevant ethical regulations for animal use, including institutional and national guidelines for the care and use of laboratory animals.

Isolation and identification of ADSCs and TDSCs

Inguinal fat was freshly harvested from 4 to 6-week-old male C57BL/6 mice and immediately washed with phosphate-buffered saline (PBS, Gibco, Carlsbad, CA, USA). The adipose tissue was finely chopped into small pieces and digested with 2 mg/mL type I collagenase (Sigma-Aldrich, St. Louis, MO) for 30 min at 37 °C under constant agitation. After digestion, the cell suspension was centrifuged at 1500 rpm for 5 min, and the supernatant was filtered through a 40-μm nylon mesh filter (BD Falcon, Franklin Lakes, NJ, USA) to remove any undigested tissue. The filtered cells were rinsed three times with PBS and resuspended in alpha-modified Eagle’s medium (alpha-MEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco). The cells were then seeded into T75 tissue culture flasks (BD Falcon) and incubated in a humidified environment at 37 °C with 5% CO₂. Non-adherent cells were removed after 24 h, and the medium was changed after 3 days. Once the cells reached ~90% confluence, they were passaged for further use.

TDSCs were isolated following a previously established protocol⁴⁴. Intact flexor tendons were excised from both limbs of euthanized 4–6-week-old male C57BL/6 mice, ensuring collection of only midsubstance tendon tissue while avoiding the bone–tendon junction. The peritendinous connective tissues were carefully removed, and the tendons were stored in sterile PBS. The tendon tissues were finely minced and digested for 2.5 h at 37 °C using 3 mg/mL type I collagenase (Sigma-Aldrich). The digested material was then passed through a 70-μm cell strainer (Becton Dickinson, Franklin Lakes, NJ, USA) to obtain a single-cell suspension. These cells were washed with PBS and centrifuged at 300 × g for 5 min before being resuspended in Dulbecco’s Modified Eagle’s Medium (DMEM), supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 2 mM L-glutamine, and 100 µg/mL streptomycin (all from Invitrogen, Carlsbad, CA). The isolated cells were plated at varying densities and cultured under standard conditions (37 °C, 5% CO₂) to promote colony formation. Non-adherent cells were removed after 2 days by washing with PBS. On day seven, adherent cells were trypsinized and collected as passage 0 (P0). For subsequent experiments, cells from passages 1 to 3 (P1–P3) were used. The culture medium was refreshed every 3 days throughout the experiments.

Flow cytometry analysis of mouse TDSCs

TDSCs at passage 3 were harvested and washed with PBS, then incubated with fluorochrome-conjugated antibodies for 30 min at 4 °C in the dark. The antibodies used included: FITC-conjugated anti-CD90 (rat IgG2c monoclonal antibody, Abcam, ab25672; 1:2000), FITC-conjugated anti-CD44 (rat IgG2b monoclonal antibody, Abcam, ab25064; 1:1000), FITC-conjugated anti-CD11b (rat IgG2b monoclonal antibody, Abcam, ab24874; 1:1000), FITC-conjugated anti-CD105 (rabbit IgG monoclonal antibody, Abcam, ab314950; 1:2500), and FITC-conjugated anti-CD45 (rat IgG monoclonal antibody, Abcam, ab210225; 1:100). Afterward, the cells were washed with PBS containing 1% BSA and analyzed using a BD FACSCanto II flow cytometer. Isotype controls were included for each antibody, and data were processed using FlowJo software (version 10.6). The gating strategy was provided in Supplementary Fig. 4.

Multilineage differentiation and staining of TDSCs

TDSCs were seeded in 24‑well plates and induced to differentiate using lineage-specific media. Cells were differentiated in osteogenic induction medium consisting of high-glucose DMEM supplemented with 10% FBS, 0.1 μM dexamethasone, 50 μg/mL ascorbic acid, and 10 mM β‑glycerophosphate. For adipogenic differentiation, cells were cultured in adipogenic induction medium containing high-glucose DMEM with 10% FBS, 1 μM dexamethasone, 0.5 mM IBMX, 10 μg/mL insulin, and 100 μM indomethacin. Chondrogenic induction was carried out for 21 days using a pellet culture system in high-glucose DMEM supplemented with 1% ITS (insulin–transferrin–selenium), 100 nM dexamethasone, 50 μg/mL ascorbate-2-phosphate, and 10 ng/mL TGF‑β3. After induction, cells were fixed in 4% paraformaldehyde (P6148, reagent grade; CAS30525‑89‑4; Sigma-Aldrich) for 30 min. For osteogenic mineralization, fixed cells were rinsed with deionized water and stained with 2% Alizarin RedS (A5533, Sigma-Aldrich) for 30 min. For adipogenic differentiation, cells were rinsed with 60% isopropanol and stained with freshly prepared Oil RedO solution (0.3% in isopropanol; O0625, Sigma-Aldrich) for 15 min. For chondrogenic matrix detection, cells were stained with 1% Alcian Blue8GX (A5268, Sigma-Aldrich) for 30 min. Stained cells from all assays were visualized and imaged using the Olympus IX73 microscope.

Exosome isolation

Exosomes were isolated using the ExoQuick-TC system (System Bioscience, Palo Alto, CA, USA) according to the manufacturer’s protocol. In brief, ADSCs, either treated/transfected or untreated/transfected, were cultured until they reached ~70% confluence. Afterward, the cells were washed with PBS and incubated for 24 h in alpha-MEM supplemented with 10% exosome-depleted FBS (Gibco). After the incubation period, the culture medium was centrifuged at 2000 × g for 30 min to remove cell debris. The supernatant was filtered through a 0.22 μm filter (Sigma-Aldrich) to further eliminate impurities. Exosome isolation reagent (Invitrogen) was then added to the filtered medium, and the mixture was incubated overnight at 4 °C. Following incubation, the samples were centrifuged at 10,000 × g at 4 °C for 60 min. The supernatant was discarded, and the exosome pellet was resuspended in 100 μL of PBS for further analysis.

Exosome characterization

To confirm the successful isolation of exosomes, we performed a series of characterization assays as per MISEV2018 guidelines. ADSC-Exos were analyzed utilizing nanoparticle tracking analysis (NTA), western blot, and transmission electron microscopy (TEM). For TEM, the exosomes were first fixed in 2.5% glutaraldehyde in calcium carbonate buffer (Sigma-Aldrich) for 1 h, then negatively stained with 2% phosphotungstic acid (Sigma-Aldrich) for ~2 min. Imaging was performed using a transmission electron microscope (FEI Tecnai G2 Spirit, Thermo Fisher Scientific, Waltham, MA, USA) operating at 80 kV. For NTA, ADSC-Exo suspended in PBS were diluted to a final concentration of ~0.05 μL/mL and introduced into the sample chamber. The particles were visualized utilizing the Nanosight NS 300 system (NanoSight Technology, Malvern, UK), and their size and concentration were analyzed with the NTA software (version 2.3). Western blot was also conducted to detect exosome surface markers, including tumor susceptibility gene 101 (TSG101), cluster of differentiation 9 (CD9), and heat shock protein 70 (HSP70), as well as the negative marker calnexin.

Exosome uptake experiment

ADSC-Exos were labeled with PKH-67 fluorescent dye (Sigma-Aldrich) following an established protocol⁴⁵. The PKH-67-labeled exosomes were then added to TDSCs and incubated for 24 h. After incubation, the TDSCs were fixed with 4% paraformaldehyde (PFA) and subsequently stained with 4′,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich) for 15 min to label the nuclei. Fluorescent images were captured using a microscope.

Treatment of TDSCs with ADSC-Exos

To evaluate the impact of ADSC-Exos treatment on TDSCs, 1 × 10⁶ TDSCs were plated in six-well culture plates and allowed to adhere for 24 h. Exosomes extracted from different ADSC groups were added to an exosome-free medium at concentrations of 0, 20, 40, or 80 μg/mL, which was then used to replace the existing TDSC culture medium. After 24 h of incubation, TDSCs were collected for western blot analysis or following experiments.

Cell transfection

Short hairpin RNAs (shRNAs) targeting FOXO1 (shFOXO1: 5′-CGGAGGATTGAACCAGTATAA-3′) and PP1A (shPP1A: 5′-CCGGAGAATTTCTTTCTACTT-3′), along with a non-targeting control shRNA (shNC: 5′-CAACAAGATGAAGAGCACCAA-3′), were designed by Genesee Biotech and then cloned into GV102 vectors by Genepharma (Shanghai, China). For overexpression studies, the full coding sequence of FOXO1 was amplified by PCR and inserted into the pLVX-Puro lentiviral vector (Takara Bio, Kusatsu, Japan) to create the FOXO1 overexpression plasmid. An empty vector was used as the negative control. The cDNA ORF coding full sequences for YAP1 (Origene, Shanghai, China) mutant serine-to-alanine mutation at position 127 (S127A) was generated through PCR-based site-directed mutagenesis using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, Sacramento, CA, USA) following the manufacturer´s protocol, and was subcloned into an overexpressing vector. For lentiviral production, 293 T cells (5 × 10⁶ cells per well) were transfected with lentiviral vectors containing shFOXO1, shPP1A, shNC, FOXO1, or the empty vector using Lipofectamine 3000 (Invitrogen). After 48 h, the supernatant containing lentiviral particles was collected, filtered, and used to infect TDSCs (5 × 10⁶ cells per well). Two days post-infection, the TDSCs were subjected to selection with 2.5 µg/mL puromycin. After selection, puromycin was removed, and the cells were cultured under standard conditions until fully recovered.

For miR-212-5p modulation, miR-212-5p mimics (sense: 5′-ACCUUGGCUCUAGACUGCUUACU-3′, antisense: 5′-AGUAAGCAGUCUAGAGCCAAGGU-3′), inhibitor (5′-AGUAAGCAGUCUAGAGCCAAGGU-3′), and their respective negative control (mimics NC sense: 5′-UUCUCCGAACGUGUCACGUTT-3′, antisense: 5′-ACGUGACACGUUCGGAGAATT-3′; inhibitor NC: 5′-CAGUACUUUUGUGUAGUACAA-3′) oligonucleotides were obtained from RiboBio (#R10034.8, Guangzhou, Guangdong, China). ADSCs or TDSCs were seeded into flasks at 70–80% confluency. After washing with serum-free and antibiotic-free medium, the cells were transfected with either miR-212-5p mimics, miR-212-5p inhibitor, or control oligonucleotides using Lipofectamine 3000 (20 µL). Six hours post-transfection, the cells were further processed for downstream analyses.

3-(4,5)di-methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay

Cell viability was assessed using the MTT assay (Millipore, Billerica, MA, USA). TDSCs were seeded into 96-well plates and grown to confluence. MTT reagent was added to each well, which was incubated for an additional 4 h at 37 °C. The resulting formazan crystals were dissolved, and the absorbance at 570 nm was measured using a MicroQuant Plate Reader (Bio-Tek Instruments Inc., Vermont, USA).

5-ethynyl-2′-deoxyuridine (EdU) assay

To assess cell proliferation, TDSCs were treated with ADSC-Exos for 24 h and then exposed to 50 μM EdU using an EdU Assay Kit (Abcam, Cambridge, UK, ab219801) for 4 h. After incubation, the cells were fixed in 4% PFD and stained following the EdU assay kit’s protocol. DAPI was used to stain the nuclei for visualization. Fluorescent images of the labeled cells were captured using an inverted fluorescence microscope to evaluate proliferation rates.

Scratch assay

TDSCs (1 × 10⁵ cells) were plated in six-well plates. A straight scratch was made across the cell monolayer using the tip of a P200 pipette. The wells were then rinsed three times with PBS to remove loose cells and debris. Following this, a serum-free medium containing different concentrations of ADSC-Exos was added to the wells. Cell migration was observed and imaged at 0 and 48 h using a microscope. The extent of wound closure was calculated by measuring the ratio of the healed area to the initial wound area, with the migration index calculated using ImageJ software.

Transwell assay

1 × 10⁴ TDSCs per group were seeded into the upper chamber of the transwell insert (Corning Inc., NY, USA) in 100 μL of serum-free medium. 500 μL of medium containing 10% serum, with or without different concentrations of ADSC-Exos, was added to the lower chamber. After 24 h of incubation, the TDSCs that had adhered to the membrane in the upper chamber were fixed with anhydrous ethanol and stained with crystal violet. The wells were then washed three times to remove excess dye, and non-migrated cells on the upper side of the membrane were carefully wiped away. The migrated cells on the underside of the membrane were quantified under a microscope.

RNA pull-down assay

To isolate biotin-labeled RNA complexes, custom biotinylated probes targeting miR-212-5p (GenePharma) were first incubated with cell lysates from TDSCs to allow the formation of complexes. Streptavidin-coated magnetic beads (Invitrogen) were then added to the mixture, selectively binding the biotin-labeled probes and associated complexes according to the manufacturer’s protocol. After incubation, the beads were thoroughly washed to remove nonspecific proteins and RNA. The enrichment of FOXO1 mRNA bound to miR-212-5p was subsequently evaluated by qPCR analysis.

Dual-luciferase reporter assay

The wild-type (WT) and mutant (MUT) sequences for FOXO1 3′-UTR containing the miR-212-5p binding site were amplified and cloned into the psiCHECK2 vector (Promega, Madison, WI, USA). TDSCs were transfected with either FOXO1-WT or FOXO1-MUT plasmids and treated with exosomes derived from normal ADSCs or ADSCs transfected with miR-212-5p mimics, miR-212-5p inhibitor, or their respective negative controls (mimic NC, inhibitor NC) using Lipofectamine 3000 (Invitrogen). Similarly, the WT and MUT sequences for the PP1A promoter site 1 containing the FOXO1 binding site were also cloned into the psiCHECK2 vector. TDSCs were transfected with either PP1A Site 1-WT or PP1A Site 1-MUT plasmids and treated with shNC, shFOXO1, vector control, or FOXO1 plasmid, using Lipofectamine 3000 (Invitrogen). After 48 h of incubation, luciferase activity was assessed using the Dual-Luciferase Reporter Assay Kit (Promega), following the manufacturer’s protocol.

Chromatin immunoprecipitation (ChIP) assay

TDSCs were collected and lysed using RIP lysis buffer (Millipore). The lysates were cross-linked and sonicated to shear the chromatin. The resulting lysates were then mixed with ChIP Dilution Buffer, Protease Inhibitor Cocktail, and Protein A Agarose/Salmon Sperm DNA beads, followed by incubation to facilitate binding. After centrifugation, the supernatant was transferred to a fresh tube. For immunoprecipitation, either an anti-FOXO1 antibody (mouse IgG1 monoclonal, Thermo Fisher Scientific, MA5-17078, 1:20) or a control IgG antibody (Abcam, ab205718, 1:1000) was added to the supernatant, and the mixture was incubated overnight at 4 °C. The next day, the samples were washed, and RNase A was added to digest RNA, followed by incubation at 37 °C for 1 h. Next, EDTA, Tris-HCl, and proteinase K were added to degraded proteins, and the samples were incubated at 45 °C for 2 h. The DNA was then purified, and its concentration was measured using quantitative PCR (qPCR).

Immunofluorescent (IF) staining

TDSCs were transfected with either shFOXO1, shFOXO1 combined with shPP1A, or shNC. After the transfection, the cells were fixed in 3.7% buffered formalin and incubated for 1 h in a blocking solution containing 1% bovine serum albumin (BSA). Following blocking, the cells were incubated overnight at 4 °C with an anti-YAP1 antibody (rabbit polyclonal, Abcam, ab52771; 1:250 dilution). The next day, the cells were thoroughly washed with PBS and then incubated for 2 h at room temperature with Alexa Fluor® 488 conjugated goat anti-rabbit IgG (H + L) secondary antibody (Abcam, ab150077; 1:500). Nuclear staining was performed using DAPI (Life Technologies, Waltham, CA, USA). Fluorescent images of the labeled cells were captured using the fluorescence microscope for analysis.

Construction of a mouse tendon injury model

Male C57BL/6 mice (Mus musculus), weighing 20–25 g, aged 10–12 weeks, were purchased from the human SJA Laboratory Animal Co., Ltd (Hunan, China). The mice were housed in groups of three to four per cage under standard laboratory conditions, including a temperature of 22 ± 2 °C, relative humidity of 50–60%, and a 12-h light/dark cycle. They had ad libitum access to standard rodent chow and water.

A total of 24 mice were randomly divided into four groups, including control, Exo, Exo-inhibitor NC, and Exo-miR-212-5p inhibitor, with six mice in each group. The experiments were performed as described in a previously published study⁴⁶. Before the surgical procedure, mice were anesthetized with a mixture of isoflurane (3%) and oxygen (1%), and both hind limbs were shaved. During surgery, anesthesia was maintained via a nose cone, with the isoflurane level reduced to 1% with oxygen. The left Achilles tendon was exposed through a longitudinal skin incision, and a complete transverse cut was made at the midpoint of the tendon using scissors, without any attempt at repair. In the Exo, Exo-inhibitor NC, and Exo-miR-212-5p inhibitor groups, 3500 ng of exosomes, derived from normal ADSCs, ADSCs transfected with inhibitor NC, or ADSCs transfected with a miR-212-5p inhibitor, were injected into the tendon defect site on days 1 and 7 post-transection, following a previously published paper⁴⁷. Mice were euthanized on day 14 post-transection, and the repaired Achilles tendons were harvested for analysis. The investigator was blinded to group allocation during treatment and outcome assessment.

Biomechanical testing

The tests were performed following a previously reported protocol⁴⁸. The Achilles tendon was carefully separated from the underlying bone and released by making a transverse cut across the midpoint of the muscle belly at the proximal end and a cut through the calcaneus at the distal end. Any remaining adherent soft tissue was meticulously removed. The gastrocnemius muscle fibers were bluntly dissected and removed, leaving only the intramuscular tendon fibers intact. The prepared tendon construct was mounted on an Instron Mechanical Testing Machine (Model 5542, Instron Corp., Canton, MA, USA) and preloaded to 0.02% of the mouse’s body weight. The gauge length of the tendon was measured using precision sliding calipers, and the tendons were subjected to a load-to-failure test at a strain rate of 100% strain per second (calculated as the change in length relative to gauge length). This strain rate was selected because it closely represents the physiological loading rates experienced in vivo. Force versus extension data were recorded at a sampling frequency of 20 Hz. Two key parameters were calculated for each tendon: maximum load to failure (structural strength) and structural stiffness (determined as the slope of the force versus extension curve). For each animal, data from the operated tendon were normalized against the sham-side tendon data before statistical analysis. After testing, tendons were inspected to assess the mode of failure.

Hematoxylin and Eosin (HE) staining

The tendon tissues were fixed in 3.7% formalin, embedded in paraffin, and sectioned into 5 µm thick slices. The sections were briefly stained with hematoxylin (Sigma-Aldrich) for 3 min, then rinsed under running tap water for 1 min. Next, eosin (Sigma-Aldrich) was applied for 45 s to stain the sections. After mounting, images were taken utilizing a Biozero BZ-9000 Series microscope (KEYENCE, Osaka, Japan). The stained patellar tendons were assessed using a previously established scoring method based on parallel collagen fiber alignment⁹. The scoring criteria were defined as follows: 0, 0–25% of fibers aligned in parallel; 1, 25–50% alignment; 2, 50–75% alignment; and 3, 75–100% alignment.

Immunohistochemistry (IHC) staining

Tendon tissues were initially fixed in 3.7% buffered formalin and then embedded in paraffin for histological analysis. Thin sections, 5 µm in thickness, were cut from the paraffin blocks for IHC staining. These sections were incubated with tenomodulin primary antibodies, including rabbit polyclonal anti-tenomodulin (TNMD) antibody (rabbit polyclonal antibody, Abcam, ab203676, 1:200), scleraxis (Scx, rabbit polyclonal, Thermo Fisher Scientific, PA5-23943, 1:1000), collagen type I alpha 1 (COL1A1, rabbit polyclonal, Thermo Fisher Scientific, PA5-29569, 1:1000), and SRY-Box Transcription Factor 9 (SOX9, Mouse IgG1 monoclonal, Thermo Fisher Scientific, 14-9765-82, 1:1000) for 1 h at room temperature. After the primary antibody incubation, the sections were exposed to specific secondary antibodies. To visualize cell nuclei, hematoxylin counterstaining was performed, and the stained sections were imaged using a Leica confocal microscope.

Safranin O/Fast Green (SO) staining

To assess cartilage-like matrix deposition, tendon sections were deparaffinized, rehydrated through a graded ethanol series, and stained with Weigert’s iron hematoxylin for 10 min to visualize nuclei. After rinsing in running water, sections were stained with 0.02% Fast Green (Sigma-Aldrich, F7252) for 5 min, followed by 1% acetic acid differentiation for 10 s. Subsequently, sections were stained with 0.1% Safranin O solution (Sigma-Aldrich, S8884) for 5 min to visualize glycosaminoglycan-rich matrix. Slides were then dehydrated, cleared, and mounted with neutral resin. Images were captured using a light microscope (Olympus IX73), and metachromatic (Safranin O-positive) areas were quantified using ImageJ software.

RNA extraction and quantitative real-time polymerase chain reaction (qPCR)

RNA extraction was conducted utilizing the Trizol reagent (Invitrogen). Subsequently, reverse transcription was performed on 1 µg of RNA using the PrimeScript cDNA Synthesis Kit (Takara, Osaka, Japan), following the protocol provided by the manufacturer. Quantitative PCR was executed with the TaqMan® Universal PCR Master Mix (Thermo Fisher Scientific). Primer sequences, sourced from Origen Biotech (Wuxi, Jiangsu, China), are detailed in Table 1. The relative expression levels of mRNA were determined through the 2^−ΔΔCt method, with GAPDH serving as the normalization control.

Table 1.

Primers used for qPCR analysis

Genes	Primer sequences (5′-3′)
mus-miR-212-5p-R	GTCGTATCCAGTGCAGGGTCCGAGGTAT TCGCACTGGATACGACAGTAAG
mus-miR-212-5p-F	GCCGAGACCTTGGCTCTAGACTG
m-FOXO1-F	CCCAGGCCGGAGTTTAACC
m-FOXO1-R	GTTGCTCATAAAGTCGGTGCT
m-PP1A-F	AGAGAACGAGATCCGTGGTCT
m-PP1A-R	ACAGCCGTAGAAGGTCATAGT
m-TNC-F	GAGCCCCTTTGCCTCAACAA
m-TNC-178-R	CTTCGCCCGTGAAACCTTCTT
m-TNMD-F	ACACTTCTGGCCCGAGGTAT
m-TNMD-R	GACTTCCAATGTTTCATCAGTGC
m-Scx-F	CTGGCCTCCAGCTACATTTCT
m-Scx-R	GTCACGGTCTTTGCTCAACTT
m-COL1A1-F	GCTCCTCTTAGGGGCCACT
m-COL1A1-R	CCACGTCTCACCATTGGGG
m-CD31-F	GGAAAGCCAAGGCCAAACAG
m-CD31-R	TTACTGCTTTCGGTGGGGAC
m-VEGFR2-F	TACGCTGGTCATCCAAGCTG
m-VEGFR2-R	GTGCCAGCCTACTACAACACT
m-VE-cadherin-F	TGCTCACGGACAAGATCAGC
m-VE-cadherin-R	GTGCGAAAACACAGGCCAAT
m-GAPDH-F	AGCCCAAGATGCCCTTCAGT
m-GAPDH-R	CCGTGTTCCTACCCCCAATG

The names of genes are in italics.

Western blot

Proteins were extracted by incubating the samples in radio-immunoprecipitation assay (RIPA) buffer containing protease inhibitors for 30 min at 4 °C (Beyotime Inc., Haimen, Jiangsu, China). Protein concentrations were examined utilizing a bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific). Subsequently, 30 μg of total protein per sample was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore). The membranes were blocked to prevent nonspecific binding, rinsed with PBS, and incubated overnight at 4 °C with primary antibodies targeting the following proteins: CD9 (recombinant rabbit monoclonal, Thermo Fisher Scientific, MA5-31980, 1:1000), TSG101 (mouse IgG1 monoclonal, MA1-23296, Thermo Fisher Scientific, 1:1000), HSP70 (mouse monoclonal IgG1, Thermo Fisher Scientific, MA3-006, 1:1000), calnexin (mouse monoclonal IgG2a, Thermo Fisher Scientific, MA5-31501, 1:1000), tenascin-C (TNC, mouse monoclonal IgG1, Thermo Fisher Scientific, MA5-16086, 1:1000), TNMD (rabbit polyclonal, PA5-112767, Thermo Fisher Scientific, 1:1000), Scx (rabbit polyclonal, Thermo Fisher Scientific, PA5-23943, 1:1000), COL1A1 (rabbit polyclonal, Thermo Fisher Scientific, PA5-29569, 1:1000), FOXO1 (rabbit monoclonal, Thermo Fisher Scientific, MA5-14846, 1:1000), phosphorylated yes-associated protein 1 (p-YAP1, rabbit polyclonal, Thermo Fisher Scientific, PA5-17481, 1:1000), yes-associated protein 1 (YAP1, rabbit polyclonal, Thermo Fisher Scientific, PA1-46189, 1:1000), PP1A (mouse monoclonal IgG1, Thermo Fisher Scientific, 43-8100, 1:1000), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, mouse monoclonal IgG1, MA5-15738, 1:1000). After washing with PBS, the membranes were incubated with secondary antibodies (Invitrogen). Protein bands were detected using an enhanced chemiluminescence (ECL) kit (WBULS0100, Merck Millipore). Uncropped western blot images were presented in Supplementary Fig. 5.

Statistical and reproducibility

All experiments were independently repeated at least three times (n = 3) unless otherwise specified. For in vivo experiments, six biological replicates (n = 6 mice per group) were used. Replicates refer to biological replicates, defined as independent experiments conducted on separate cell or animal samples. Data analysis was performed using GraphPad Prism 7. Results are expressed as the mean ± standard deviation (SD). The assumptions of normality and homogeneity of variance were assessed on the model residuals using the Shapiro-Wilk test and Levene’s test, respectively. For comparisons between two groups, unpaired Student’s t-tests were applied, while comparisons involving more than two groups were evaluated using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. A P-value less than 0.05 was considered statistically significant.

Ethical approval and consent to participate

All experimental procedures in this study were conducted in accordance with the ARRIVE 2.0 guidelines and received approval from the Shengli Clinical Medical College of Fujian Medical University Institutional Animal Care and Use Committee, approval number: IACUC-FPH-SL-20230724[0048]. The experiments complied with all relevant institutional and national regulations concerning the care and use of laboratory animals.

Author contributions

Kefeng Lin: Conceptualization; Writing—Original Draft; Supervision; Methodology; Xu Hu: Conceptualization; Resources; Writing—Original Draft; Supervision; Jin Yan: Conceptualization; Validation; Investigation; Methodology; Renzhi Gao: Formal analysis; Data Curation; Shishui Lin: Visualization; Writing—Review and Editing; Shiguo Zhou: Writing—Review and Editing; Project administration; Funding acquisition.

Peer review

Peer review information

Communications Biology thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editor: Ophelia Bu.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information files, or from the corresponding author upon reasonable request. All source data underlying the graphs and charts are provided in supplementary Data 1.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s42003-025-09210-5.