BMSC-derived exosomes facilitate osteogenesis and ameliorate ageing-related bone loss through restoring Th17/Treg homeostasis via the miR-21/Skp2/FoxO1 axis

Abstract

Osteoporosis (OP) is among the most prevalent systemic skeletal disorders worldwide and is characterized by decreased bone mass and microarchitectural deterioration, leading to increased fracture risk and significant impairment of quality of life, particularly among elderly individuals. Recently, exosomes derived from bone marrow mesenchymal stem cells (BMSCs), termed BMSC-exosomes, have emerged as promising therapeutic agents for OP because of their regenerative and immunomodulatory potential. In this study, we used senescence-accelerated mouse prone 6 (SAMP6) mice, MC3T3-E1 osteoblastic cells, and CD4(+) T cells to investigate the effects of BMSC-exosomes on osteogenesis and to elucidate the underlying molecular mechanisms. Our results demonstrate that BMSC-derived exosomes enhance osteogenic differentiation in vitro and ameliorate age-related bone loss in vivo. We identified miR-21-5p as a highly enriched microRNA within BMSC-exosomes, which plays a central role in mediating their pro-osteogenic effects and protecting against OP progression. Flow cytometry analysis revealed that BMSC-exosome treatment effectively restored the imbalance between T helper 17 cells (Th17) and regulatory T cells (Treg cells)—a key immune dysregulation observed in OP—in both SAMP6 mice and cultured CD4(+) T cells. Through integrated bioinformatics analysis and experimental validation, we showed that BMSC-derived miR-21-5p directly targeted S-phase kinase-associated protein 2 (SKP2), leading to its downregulation. SKP2 then promotes the ubiquitination and subsequent degradation of Forkhead Box O1 (FoxO1), a transcription factor essential for maintaining Th17/Treg homeostasis. By suppressing SKP2, miR-21-5p stabilizes FoxO1, thereby promoting immune balance and enhancing osteogenic activity. Collectively, these findings indicate that miR-21-5p-enriched BMSC-exosomes alleviate OP by modulating the SKP2/ubiquitination/FoxO1 signalling axis and restoring the Th17/Treg balance. This dual action—promoting bone formation and correcting immune dysfunction—highlights the therapeutic potential of BMSC-exosomes. Thus, the use of miR-21-5p-loaded BMSC-exosomes represents a novel and promising strategy for the prevention and treatment of OP.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-026-04927-4.

Introduction

Osteoporosis (OP) is a widespread bone disorder characterized by decreased bone mass and compromised skeletal strength, resulting in an increased risk of fractures [1]. As people age, the prevalence of OP increases, making it a significant health concern among older adults [2]. Drug interventions typically target only a single pathophysiological aspect of OP, which limits their therapeutic efficacy and often leads to adverse effects [3]. In contrast, stem cell therapy offers a multimodal mechanism of action and has significant advantages over conventional pharmacological treatments for OP [4]. Mesenchymal stem cells (MSCs) possess multilineage differentiation potential and exhibit key characteristics, such as self-renewal capacity, low immunogenicity, and the ability to promote tissue repair, making them promising candidates for novel OP therapies [5]. Emerging evidence indicates that MSC-based therapies for OP exert their effects largely through the secretion of exosomes [6, 7].

MicroRNAs (miRNAs) are small noncoding RNAs that regulate gene expression by silencing target mRNAs and play critical roles in various pathological processes [8]. More importantly, miRNAs have been identified in exosomes. Exosomal miRNAs are closely related to various biological and functional aspects of human health [9]. Accumulating evidence indicates that exosomal miRNAs derived from MSCs play pivotal roles in the prevention and treatment of OP. For instance, exosomal miR-935 derived from bone marrow mesenchymal stem cells (BMSCs) enhances osteoblast proliferation and differentiation in osteoporotic rats [10]. Among the various miRNAs implicated in bone metabolism, miR-21-5p has been among the most extensively studied. Mounting evidence suggests that miR-21-5p is closely associated with the progression and therapeutic management of OP [11]. Notably, recent studies have demonstrated that BMSC-derived exosomal miR-21-5p can ameliorate OP [12]. However, the specific regulatory factors responsible for the therapeutic benefits of BMSC-derived exosomal miR-21-5p in the treatment of OP remain unclear.

The immune system plays a pivotal role in the onset and progression of OP [13]. Immunological components, particularly T cells, are central to regulating immune responses. Among these, T helper 17 cells (Th17) represent a distinct subset of T cells that predominantly secrete proinflammatory cytokines, whereas regulatory T cells (Treg cells) serve as key anti-inflammatory modulators [14]. The balance between Th17 and Treg cells (Th17/Treg homeostasis) is essential for maintaining bone homeostasis, and disruption of this equilibrium is implicated in the pathogenesis of OP [15]. Evidence suggests that targeted regulation of Th17/Treg homeostasis can effectively inhibit OP progression [16]. In addition, exosomes play a significant role in modulating the Th17/Treg balance [17]. However, whether exosomes derived from BMSCs regulate Th17/Treg homeostasis in OP remains unclear. Further investigations into the functional role and cooperative interaction between BMSC-derived exosomes and the Th17/Treg balance may provide novel insights for the development of therapeutic strategies against OP.

Nevertheless, whether the effect of BMSC-exosomes on bone is mediated by the Th17/Treg balance remains unclear. In this study, we demonstrated that BMSC-derived exosomes exert osteogenic effects on both SAMP6 and MC3T3-E1 cells. We further identified miR-21-5p as highly enriched in these exosomes and showed that miR-21-5p-rich exosomes ameliorate the Th17/Treg imbalance in SAMP6 mice. Mechanistically, S-phase kinase-associated protein 2 (SKP2), a key component of the ubiquitin‒proteasome system (UPS), was validated as a direct target of miR-21-5p. Exosomal miR-21-5p suppresses SKP2, thereby inhibiting the degradation of Forkhead Box O1 (FoxO1), a key regulator in maintaining Th17/Treg homeostasis, ultimately promoting immune balance and supporting bone health. Collectively, our findings reveal a novel mechanism linking exosomal miR-21-5p to immune and bone regulation and suggest a promising therapeutic strategy for OP based on BMSC-derived exosomes enriched with miR-21-5p.

Materials and methods

Cell culture and characterization

The murine preosteoblast cell line MC3T3-E1 were obtained from the Cell Bank of the Chinese Academy of Sciences (https://cellbank.org.cn/search-detail.php?id=939) (Shanghai, China). The cells were cultured in α minimal essential medium (α-MEM; 12571063; Gibco, USA) supplemented with 10% (v/v) fetal bovine serum (FBS; 10099158; Gibco, USA), 100 mg/L streptomycin, and 100 U/mL penicillin (P1400; Solarbio, China) and cultivated in a 5% CO₂ incubator at 37 °C.

After mice were sacrificed, bone marrow was harvested from the femurs, and BMSCs were isolated and purified using the whole bone marrow adherence method. The femoral bone marrow cavity was repeatedly flushed with Dulbecco's modified Eagle’s medium/nutrient mixture F-12 (DMEM/F-12; 11320033; Gibco, USA), and the collected cells were filtered through a 200-mesh sieve and supplemented with 10% FBS. The cells were subsequently centrifuged at 1500 rpm for 3 min, resuspended in fresh culture medium, and seeded into culture flasks. The cells were then cultured in a 5% CO₂ incubator at 37 °C.

Characterization of BMSCs

Flow cytometric analysis was performed to characterize BMSC surface markers. Briefly, BMSCs were resuspended in PBS and incubated with anti-CD29-FITC (11-0291-82, eBioscience), anti-CD106-FITC (11-1061-82, eBioscience), anti-CD34-PE (MA5-17831, eBioscience), and anti-CD45-PE (12-0451-82, eBioscience) antibodies. After 30 min of incubation in a 5% CO₂ environment, the supernatant was removed, and the cells were resuspended in phosphate-buffered saline (PBS) for flow cytometric detection.

Mouse T-cell isolation

Spleen and lymph node cells were isolated from mice. CD4(+) T cells were isolated using a CD4(+) T-cell isolation kit (130-104-454; Miltenyi Biotec, Germany) according to the manufacturer’s instructions [18].

BMSC-exosome isolation and characterization

To avoid contamination by bovine exosomes, BMSCs were cultured in exosome-depleted medium. After 48 h, the culture medium was collected. Exosomes were isolated from the supernatant using an ExoQuick-TC kit (EXOTC50A-1; SBI, USA) according to the manufacturer's instructions. Briefly, the medium was first centrifuged at 3000 × g for 15 min to remove dead cells and debris. Large vesicles were then eliminated by centrifugation at 10,000 × g for 70 min. The exosome precipitation solution was then added, and the mixture was refrigerated overnight, followed by centrifugation at 10,000 × g for 30 min. The resulting pellet was resuspended in PBS and stored at − 80 °C until further use.

The size distribution of the exosomes was analysed using nanoparticle tracking analysis (NTA). Morphology was confirmed by transmission electron microscopy (TEM), and the exosomal marker proteins TSG101, CD9 and CD36 were identified via western blotting [19].

Cell transfection

BMSCs, MC3T3-E1 cells and CD4 (+) T cells were seeded into 6-well plates and cultured for 48 h. Transfection was carried out using Lipofectamine 2000 (Invitrogen, USA) according to the manufacturer’s protocol. The plasmids pcDNA3 and pcDNA3-FoxO1, along with a miR-21-5p inhibitor and its negative control (NC inhibitor; GenePharma, Shanghai, China), were transfected into the cells. Twenty-four hours post-transfection, the cells were collected and used for subsequent experiments.

Osteoblastic differentiation assays

Alkaline phosphatase (ALP) staining and quantitative reverse transcriptase polymerase chain reaction (qRT‒PCR) were used to assess osteoblastic differentiation. The cells were seeded into 6-well plates and pretreated with BMSC-exosomes at a concentration of 50 µg/mL. MC3T3-E1 cells were randomly divided into the following groups: (1) MC3T3-E1 cells treated with PBS; (2) MC3T3-E1 cells with BMSC-exosome treatment; (3) MC3T3-E1 cells treated with exosomes derived from NC inhibitor-transfected BMSCs; (4) MC3T3-E1 cells treated with exosomes derived from miR-21-5p inhibitor-transfected BMSCs; (5) MC3T3-E1 cells treated with exosomes derived from miR-21-5p inhibitor-transfected BMSCs and DMSO; (6) MC3T3-E1 cells treated with exosomes derived from miR-21-5p inhibitor-transfected BMSCs and SKPin C1 (50 μM, S8652, Selleck, China); (7) pcDNA3 vector-transfected MC3T3-E1 cells treated with exosomes derived from miR-21-5p inhibitor-transfected BMSCs; and (8) pcDNA3-FoxO1-transfected MC3T3-E1 cells treated with exosomes derived from miR-21-5p inhibitor-transfected BMSCs.

For the ALP assay, the cells were fixed with 4% paraformaldehyde for 20 min, incubated with BCIP/NBT solution (C3206; Beyotime, China) in the dark, and then washed with PBS three times to stop the reaction. Images were captured using a camera and microscopy. Each sample was measured in triplicate.

qRT‒PCR was performed to detect osteogenic markers. Total RNA was extracted using TRIzol reagent (Invitrogen, USA) according to the manufacturer’s instructions, followed by cDNA synthesis using Reverse Transcription 10X Buffer (Promega, USA). qRT‒PCR was subsequently performed using a real-time PCR system (Bio-Rad, USA) with a SYBR Green PCR Kit (Takara, Japan). Relative transcript expression was calculated using the 2^−ΔΔCt method after normalization against that of GAPDH. The expression data are presented as the fold increase relative to the endogenous control. Each sample was measured in triplicate. The primers used were as follows:

ALP-F: CCAACTCTTTTGTGCCAGAGA; ALP-R: GGCTACATTGGTGTTGAGCTTTT; BGLAP-F: TTCTGCTCACTCTGCTGACC; BGLAP-R: ATAGCTCGTCACAAGCAGGG; RUNX2-F: TTCAACGATCTGAGATTTGTGGG; RUNX2-R: GGATGAGGAATGCGCCCTA GAPDH-F: AGGTCGGTGTGAACGGATTTG; GAPDH-R: GGGGTCGTTGATGGCAACA.

qRT‒PCR

A miRNA real-time PCR kit (RiboBio, China) was used to reverse transcribe the RNA into cDNA for miR-21-5p detection. The transcript levels of the miRNAs were normalized to those of U6. Each sample was measured in triplicate.

Total RNA was isolated from tissue using TRIzol reagent. Then, the mRNA levels of IL17, RORγt, FOXP3 and IL-10 were measured by qRT‒PCR.

The primers used were as follows:

miR-21-5p-F: ACACTCCAGCTGGGTAGCTTATCAGACTGA; miR-21-5p-R: CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGTCAACATC; U6-F: AGAGAAGATTAGCATGGCCCCTG; U6-R: ATCCAGTGCAGGGTCCGAGG; IL-17-F: TCGATGAATTGGACAAAATGACAGG; IL-17-R: TTGCGAGCAGCACGATTTAG; RORγt-F: CAGAGACACCACCGGACATC; RORγt-R: CCCAGATGACTTGTCCCCAC; FOXP3-F: GGTTTACTCGCATGTTCGCC; FOXP3-R: CCACTCGCACAAAGCACTTG; IL-10-F: ACCTGGTAGAAGTGATGCCC; IL-10-R: TGTAGACACCTTGGTCTTGGA.

Luciferase activity assay

The wild-type (WT) or mutant (MUT) sequences of SKP2 containing the miR-21-5p binding sites were subsequently cloned and inserted into the psiCHECK2 vector. After transfection with the established reporter vector together with or without miR-21-5p mimics/NC mimics using Lipofectamine 2000 (Thermo Fisher, USA) for 48 h, the luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega, USA). Each sample was measured in triplicate.

Animals and administration

The senescence-accelerated mouse prone 6 (SAMP6) mouse strain serves as a model for OP. Female SAMP6 mice and control senescence-accelerated mouse/resistant 1 (SAMR1, 6 months) mice were obtained from Vital River Laboratory Animal Technology (Beijing, China). The mice were maintained under standard specific pathogen-free (SPF) conditions. All studies involving animals were in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines for reporting experiments involving animals [20] and Institutional Animal Care and Use Committee (IACUC) guidelines. Animal experiment design in this study was reviewed and authorized by the Animal Ethics Committee of Kunming Medical University (Approval no. kmmu20241120). The mice were housed at room temperature (25 ± 2 °C), a relative humidity of 65 ± 2%, and a 12-h light/dark cycle. The mice were reared in an animal facility for 1 week before the experiment.

The mice were randomly separated into the following groups: (1) the SAMR1 group (n = 6); (2) the SAMP6 group (n = 6); (3) the SAMP6 + PBS group (n = 6, tail vein injection of 100 µL of PBS once per week); (4) the SAMP6 + BMSC-exosome group (n = 6, tail vein injection of 100 ng of BMSC-exosomes dissolved in 100 µL of PBS once per week); (5) the SAMP6 + NC-inhibitor-exosome group (n = 6, tail vein injection of 100 ng of exosomes derived from BMSCs transfected with the NC inhibitor dissolved in 100 µL of PBS once per week); and (6) the SAMP6 + miR-21-5p inhibitor-exosome group (n = 6, tail vein injection of 100 ng of exosomes derived from BMSCs transfected with the miR-21-5p inhibitor dissolved in 100 µL of PBS once per week). After 10 weeks, anesthetized mice (pentobarbital, 70 mg/kg, intraperitoneal injection. P3761; Sigma, USA) were euthanized by cervical dislocation. The femoral tissue of the mice was collected for the next experiment. Blind measurements were carried out to avoid unconscious biases.

Western blot

The indicated cells and tissues were harvested and lysed with radioimmunoprecipitation assay (RIPA) lysis buffer (P0013B; Beyotime, China) supplemented with protease inhibitors and phosphatase inhibitors. Equal amounts of protein extracts were loaded per lane, separated by sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE) and then transferred onto polyvinylidene fluoride (PVDF) membranes (ISEQ00010, Millipore, USA). The membranes were blocked with 5% non-fat milk. After being washed with Tris-buffered saline Tween (TBST) three times, the membranes were incubated with anti-CD9 (ab223052; Abcam), anti-CD63 (ab68418; Abcam), anti-TSG101 (ab30871; Abcam), anti-SKP2 (ab68455; Abcam), and anti-FOXO1 (18592-1-AP; Proteintech) antibodies overnight at 4 °C. After being washed with TBST, the membranes were incubated with the corresponding HRP-conjugated secondary antibodies (Proteintech) at room temperature for 2 h. Finally, the membranes were visualized with a chemiluminescence system. The expression of proteins was normalized to that of GAPDH as a reference and quantified using ImageJ [21, 22].

Flow cytometry

Spleen and bone marrow were collected to isolate mononuclear cells. Treg and Th17 cells were identified using flow cytometry. FITC-conjugated anti-CD4 (11-0041-82, eBioscience) and PE-conjugated anti-CD25 (12-0251-82, eBioscience) antibodies were used to identify Treg cell surface markers, and APC-FOXP3 (17-5773-82, eBioscience) was used to identify intracellular markers in Treg cells. PE-CD4 (12-0041-82; eBioscience) was used to indicate surface markers in Th17 cells, and APC-IL-17A (17-7177-81; eBioscience) was used to indicate intracellular markers in Th17 cells [23]. After the collected cells were centrifuged, the pellets were resuspended in PBS, and anti-CD4 and anti-CD25 antibodies were added. The cell suspension was incubated at room temperature in the dark for 20 min. The cells were then washed with 1 mL of PBS and centrifuged at 500 × g for 5 min, after which the supernatant was carefully discarded. Following the manufacturer’s instructions for the Transcription Factor Fixation/Permeabilization Kit (00-5521-00, eBioscience), the cells were fixed and permeabilized prior to incubation with anti-IL-17 and anti-FOXP3 antibodies at 4 °C in the dark for 20 min. After the incubation, the cells were washed with 1 mL of PBS and centrifuged at 500 × g for 5 min, after which the supernatant was discarded, and the washing step was repeated twice with PBS. Finally, the cells were resuspended in 400 μL of PBS, mixed thoroughly, and analysed immediately by flow cytometry [24, 25].

H&E staining

The collected femurs were fixed in 4% paraformaldehyde for 48 h and then incubated in 15% EDTA for decalcification. Afterwards, the specimens were embedded in paraffin, and the femurs were cut into 5 μm-thick sections and stained with a H&E staining Kit (Solarbio, China). Photographs were taken under a light microscope.

Microcomputed tomography analysis

To evaluate the differences in bone mass and microarchitecture among the groups, microcomputed tomography (micro-CT) was performed using a SkyScan1176 micro-CT system (Bruker, Germany). The femurs were dissected and fixed with 4% paraformaldehyde. CT-AN was used for scanning, and Dataviewer was used to correct femur placement. The scanning parameters included 50 kV peaks, 500 μA, an effective pixel size of 35.52 μm, and a total scanning time of 65 ms. The resolution of each femur was 18 μm. The bone mineral density (BMD), trabecular number (Tb.N), bone volume to tissue volume (BV/TV), and trabecular thickness (Tb.Th) were quantitatively analysed.

Ubiquitination assay

For the ubiquitination assay, HEK293T cells were transfected with Flag-tagged FoxO1, His-tagged ubiquitin and Myc-tagged SKP2 plasmids. Ubiquitinated FoxO1 and SKP2 were purified from HEK293T cells. Then, ubiquitinated FoxO1 protein was incubated with SKP2 in deubiquitination buffer (50 mM Tris–HCl [pH 8.0], 50 mM NaCl, 1 mM EDTA, 10 mM DTT, and 5% glycerol) at 37 °C. The level of ubiquitinated FoxO1 was analysed by immunoblotting.

Statistical analysis

The data are expressed as the means ± SDs. Comparisons between two groups were performed by unpaired, double-tailed t tests. For multiple comparisons, ANOVA or repeated-measures ANOVA was used, and then the Bonferroni post mortem test was performed using GraphPad Prism^® 9.0 software. p < 0.05 was considered to indicate statistical significance.

Results

Characterization of BMSCs and BMSC-exosomes

BMSCs were isolated, and their phenotypes were identified using flow cytometric analysis (Fig. 1a). TEM revealed that BMSC-exosomes exhibited typical round bilayer membrane vesicular structures and a cup-shaped morphology (Fig. 1b). NTA revealed that the size distribution of the exosomes in both groups was predominantly within the 50–150 nm range, which aligns well with the known biophysical characteristics of exosomes (Fig. 1c). Protein markers of exosomes, including CD63, CD9 and TSG101, were detected; in contrast, calnexin was barely detected (Fig. 1d). Taken together, these results confirm that the morphology, size distribution, and surface marker expression of the isolated vesicles are characteristic of exosomes.

Fig. 1.

Characterization of BMSC and BMSC-exosomes. A Flow cytometric analysis was used to identify BMSC surface markers; B TEM was used to examine the morphology of the exosomes; C the particle size distribution of the exosomes was analysed by NTA; D western blot analysis confirmed the presence of exosomal marker proteins

BMSC-exosomes facilitate osteogenic differentiation of MC3T3-E1 cells and ameliorate age-related bone loss in SAMP6 cells

To evaluate the functional effects of BMSC-exosomes in vitro, MC3T3-E1 cells were treated with exosomes. ALP activity was significantly greater in the exosome-treated group than in the control group (Fig. 2a). Consistently, qRT‒PCR analysis revealed upregulated expression levels of key osteogenic markers, including ALP, BGLAP (osteocalcin), and Runx2, following coculture with BMSC-exosomes (Fig. 2b).

Fig. 2.

BMSC-exosomes facilitate osteogenic differentiation of MC3T3-E1 cells and ameliorate age-related bone loss in SAMP6 cells. A ALP staining of MC3T3-E1 cells treated with PBS or BMSC-exosomes; B mRNA levels of osteogenesis-related markers in PBS- or BMSC-exosome-treated MC3T3-E1 cells, as determined by qRT‒PCR analysis. C representative micro-CT images of femurs from SAMR1 mice, SAMP6 mice, and PBS- or BMSC-exosome-treated SAMP6 mice; D–G quantitative analysis of bone parameters; H representative H&E-stained femoral sections from SAMR1 mice, SAMP6 mice, and PBS- or BMSC-exosome-treated SAMP6 mice across different groups. n = 6 per group for mice, n = 3 per group for cells; the data are expressed as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001

We further investigated the therapeutic potential of BMSC-derived exosomes in SAMP6 mice, a well-established animal model of OP. Micro-CT analysis was performed to assess bone mass and microarchitectural parameters in the femur. Representative images are shown in Fig. 2c. Compared with age-matched control SAMR1 mice, SAMP6 mice exhibited significant reductions in bone mineral density (BMD; Fig. 2d), bone volume/total volume (BV/TV; Fig. 2e), trabecular thickness (Tb.Th; Fig. 2f), and trabecular number (Tb.N; Fig. 2g). Notably, the administration of BMSC-derived exosomes effectively restored these parameters in SAMP6 mice to normal levels. Furthermore, H&E staining revealed severe trabecular bone loss in untreated SAMP6 mice, which was markedly attenuated following exosome treatment (Fig. 2h). These results demonstrate that BMSC-derived exosomes promote osteogenic differentiation in vitro and enhance bone formation in vivo, effectively counteracting age-related bone loss. Together, these findings highlight the potential of BMSC-exosomes as a promising therapeutic strategy for OP.

The delivery of miR-21-5p by BMSC-exosomes promotes osteogenic differentiation and ameliorates age-related bone loss

Accumulating evidence indicates that miR-21-5p plays a regulatory role in OP pathogenesis and can be transported via exosomes. In this study, quantitative analysis confirmed the significant enrichment of miR-21-5p in BMSC-derived exosomes (Fig. 3a). Upon coculture with these exosomes, MC3T3-E1 cells exhibited a marked increase in miR-21-5p expression (Fig. 3b). To elucidate the functional contribution of exosomal miR-21-5p, BMSCs were transfected with either a miR-21-5p inhibitor or a negative control (NC) inhibitor. Efficient knockdown of miR-21-5p in BMSCs was validated by qRT‒PCR (Fig. 3c). Notably, significantly reduced expression of miR-21-5p was observed in exosomes isolated from miR-21-5p-inhibited BMSCs (Fig. 3d). Functional assays demonstrated that these modified exosomes exhibited diminished pro-osteogenic activity, as evidenced by significantly reduced ALP activity and downregulated expression of key osteogenic markers (Fig. 3e, f). In vivo, the administration of BMSC-derived exosomes increased miR-21-5p expression in the bone tissues of SAMP6 mice (Fig. 3g). Consistent with these findings, the ability of exosomes derived from miR-21-5p-deficient BMSCs to stimulate bone formation was impaired in SAMP6 mice, as reflected by compromised improvements in bone structural parameters (Fig. 3h–m).

Fig. 3.

miR-21-5p delivered by BMSC-exosomes promotes osteogenic differentiation and ameliorates ageing-related bone loss. The expression levels of miR-21-5p were determined by qRT‒PCR in A BMSCs and BMSC-derived exosomes; B MC3T3-E1 cells treated with either PBS or BMSC-derived exosomes; C NC inhibitor- or miR-21-5p inhibitor-transfected BMSCs; D exosomes isolated from NC inhibitor- or miR-21-5p inhibitor-transfected BMSCs. E ALP staining of MC3T3-E1 cells following treatment with exosomes derived from NC inhibitor- or miR-21-5p inhibitor-transfected BMSCs. F protein expression levels of osteogenesis-related markers in MC3T3-E1 cells treated with exosomes from inhibitor-transfected BMSCs, as assessed by western blotting. G the expression levels of miR-21-5p were determined by qRT‒PCR in femoral tissues from SAMP6 and SAMR1 mice, as well as from SAMP6 mice treated with PBS or BMSC-derived exosomes. H representative micro-CT images of femurs from SAMP6 mice treated with exosomes derived from BMSCs transfected with either the NC inhibitor or the miR-21-5p inhibitor. I–L quantitative analysis of bone morphometric parameters; M representative H&E-stained femoral sections from the corresponding treatment groups. n = 6 per group for mice, n = 3 per group for cells; the data are presented as the means ± SDs; statistical significance was set at *p < 0.05, **p < 0.01, and ***p < 0.001

Effects of BMSC-derived exosomal miR-21-5p on the proportions of Treg cells and Th17 cells in SAMP6 mice

Considering that the imbalance between Th17 cells and Tregs plays critical roles in the pathogenesis of OP, we evaluated the influence of BMSC-derived exosomes on the Th17/Treg ratio in the bone marrow and spleen. Flow cytometry analysis revealed a significantly greater Th17/Treg ratio in the bone marrow of SAMP6 mice than in that of SAMR1 control mice. Notably, treatment with BMSC-exosomes substantially ameliorated this imbalance, as evidenced by a reduced Th17/Treg cell ratio (Fig. 4a). Upon treatment with BMSC-exosomes, the mRNA levels of FOXP3 (Fig. 4b) and IL-10 (Fig. 4c) increased, whereas those of IL-17 (Fig. 4d) and RORγt (Fig. 4e) decreased.

Fig. 4.

Effects of BMSC-derived exosomal miR-21-5p on the proportions of Tregs and Th17 cells in bone marrow. A Proportions of Treg cells and Th17 cells were analysed in SAMR1 mice, SAMP6 mice, and SAMP6 mice treated with PBS or BMSC-derived exosomes; the Th17/Treg cell ratio was also determined. The mRNA levels of FOXP3 (B), IL-10 (C), IL-17 (D), and RORγt (E) across the experimental groups are shown. F proportions of Treg and Th17 cells in SAMP6 mice treated with exosomes derived from BMSCs transfected with either an NC inhibitor or a miR-21-5p inhibitor; the Th17/Treg ratio was assessed. The expression levels of FOXP3 (G), IL-10 (H), IL-17 (I), and RORγt (J) in the respective groups of mice are presented. n = 6 per group; Data are expressed as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001

In the spleen, a similar trend was observed: BMSC-exosome administration led to a reduction in Th17 cell frequency and an increase in Treg cell frequency in SAMP6 mice. Additionally, the mRNA levels of IL-17 and RORγt were significantly downregulated in spleen tissues, whereas the expression of FOXP3 and IL-10 was upregulated (Additional file 3: Fig. S1). Collectively, these findings indicate that BMSC-derived exosomal miR-21-5p promotes Treg cell differentiation and suppresses Th17 cell polarization in both the bone marrow and the spleen.

BMSC-derived exosomal miR-21-5p promotes Treg differentiation in CD4(+) T cells

We further investigated whether BMSC-derived exosomal miR-21-5p could polarize naïve CD4(+) T cells into different subsets in vitro. Naïve CD4(+) T cells were isolated, and their purity was determined by flow cytometry (Fig. 5a). Following treatment with BMSC-exosomes, flow cytometry analysis was performed to evaluate the differentiation status of naïve CD4(+) T cells. Compared with the PBS-treated control group, the BMSC exosome-treated group exhibited a significantly greater proportion of Treg cells and a markedly reduced proportion of Th17 cells (Fig. 5b). The inhibition of miR-21-5p in BMSC-exosomes promoted the differentiation of naïve CD4(+) T cells into pathogenic Th17 cells while suppressing their polarization towards the Treg phenotype (Fig. 5c). Overall, these findings suggest that BMSC-derived exosomal miR-21-5p suppresses Th17 cell differentiation and enhances Treg cell generation in vitro.

Fig. 5.

BMSC-derived exosomal miR-21-5p promotes Treg differentiation in CD4(+) T cells. A The purity of naïve CD4(+) T cells was determined using MACS. B Th17 and Treg subsets from BMSC-exosome-treated naïve CD4(+) T cells were detected by flow cytometry. C Th17 and Treg subsets from naïve CD4(+) T cells treated with exosomes derived from BMSCs transfected with either the NC inhibitor or the miR-21-5p inhibitor were detected by flow cytometry. n = 3 per group; Data are expressed as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001

BMSC exosomal miR-21-5p directly targets SKP2

To identify potential targets of exosomal miR-21-5p, bioinformatics analysis was performed to predict candidate genes. Among these genes, SKP2 was identified as a direct target of miR-21-5p and is implicated in the pathogenesis of OP. A luciferase reporter assay confirmed that miR-21-5p directly binds to the 3′ untranslated region (3′UTR) of SKP2 (Fig. 6a). SKP2 expression was downregulated in MC3T3-E1 cells treated with BMSC-derived exosomes and in the femoral tissue of SAMP6 mice (Fig. 6b, c). Furthermore, the inhibition of miR-21-5p led to the restoration of SKP2 expression in both MC3T3-E1 cells and SAMP6 mice (Fig. 6d, e). The inhibition of SKP2 mediated by BMSC-derived exosomal miR-21-5p was further confirmed in CD4(+) T cells. In CD4(+) T cells treated with BMSC-derived exosomes, SKP2 expression was significantly downregulated; however, upon inhibition of miR-21-5p expression within the exosomes, SKP2 expression was restored (Additional file 3: Fig. S2).

Fig. 6.

BMSC exosomal miR-21-5p directly targets SKP2. A Predicted binding site of miR-21-5p within the 3′UTR of SKP2 mRNA using TargetScan; luciferase activity assays validating the direct interaction between miR-21-5p and SKP2. Western blot analysis of SKP2 expression in B MC3T3-E1 cells treated with PBS or BMSC-derived exosomes; C femoral tissue from SAMR1, SAMP6, and PBS- or BMSC-exosome-treated SAMP6 mice; D MC3T3-E1 cells treated with exosomes derived from cells transfected with the NC inhibitor or miR-21-5p inhibitor; E femoral tissue from SAMP6 mice treated with exosomes derived from NC inhibitor- or miR-21-5p inhibitor-transfected BMSCs. F SKP2 protein levels in MC3T3-E1 cells treated with exosomes from miR-21-5p inhibitor-transfected BMSCs in combination with DMSO or SKPin C1. G ALP staining of MC3T3-E1 cells under the same treatment conditions. H protein expression levels of osteogenic markers in MC3T3-E1 cells following the indicated treatments. n = 6 per group for mice, n = 3 per group for cells; the data are expressed as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001

To investigate the functional role of SKP2 in osteoblast differentiation, MC3T3-E1 cells were treated with the SKP2-specific inhibitor SKPin C1, and effective suppression of SKP2 expression was confirmed (Fig. 6f). As shown in Fig. 6g, h, the administration of SKPin C1 partially reversed the inhibitory effects of miR-21-5p knockdown on osteogenic differentiation, as evidenced by ALP staining and osteogenic marker expression.

BMSC-derived exosomal miR-21-5p stabilizes FoxO1 by inhibiting Skp2-mediated ubiquitination and degradation

As a critical component of the ubiquitin‒proteasome system (UPS), SKP2 regulates the ubiquitination and subsequent proteasomal degradation of its target proteins. Accumulating evidence indicates that SKP2 promotes the degradation of FoxO1, a transcription factor essential for maintaining immune cell development and differentiation, thereby modulating inflammatory responses [26]. Following treatment with BMSC-derived exosomes, FoxO1 expression was significantly upregulated in MC3T3-E1 cells (Fig. 7a). In contrast, exosomes isolated from BMSCs transfected with a miR-21-5p inhibitor resulted in reduced FoxO1 expression in recipient cells (Fig. 7b). Consistently, the administration of BMSC-derived exosomes increased FoxO1 levels in vivo (Fig. 7c), whereas the delivery of exosomes from miR-21-5p-inhibited BMSCs led to decreased FoxO1 expression in SAMP6 mice (Fig. 7d). In CD4(+) T cells, we also observed that BMSC-derived exosomal miR-21-5p regulated the expression of FoxO1 (Additional file 3: Fig. S3).

Fig. 7.

BMSC-derived exosomal miR-21-5p stabilizes FoxO1 by inhibiting Skp2-mediated ubiquitination and degradation. Western blot analysis was used to assess FoxO1 expression in A MC3T3-E1 cells treated with PBS or BMSC-derived exosomes; B MC3T3-E1 cells treated with exosomes derived from control inhibitor- or miR-21-5p inhibitor-transfected BMSCs; C femurs of SAMR1, SAMP6, and SAMP6 mice treated with PBS or BMSC-derived exosomes; D femurs of SAMP6 mice treated with exosomes from control inhibitor- or miR-21-5p inhibitor-transfected BMSCs. The ubiquitination levels of FoxO1 were analysed in E MC3T3-E1 cells treated with PBS or BMSC-derived exosomes and F MC3T3-E1 cells treated with exosomes from control inhibitor- or miR-21-5p inhibitor-transfected BMSCs. G FoxO1 expression in MC3T3-E1 cells treated with exosomes from miR-21-5p inhibitor-transfected BMSCs in the presence or absence of SKPin C1 was determined by western blotting. n = 6 per group/for mice; n = 3 per group for cells; the data are presented as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001

To further investigate the regulatory effect of BMSC-derived exosomes on modulating FoxO1 stability, ubiquitination assays were performed. Coculture with BMSC-derived exosomes markedly reduced FoxO1 ubiquitination in MC3T3-E1 cells (Fig. 7e), whereas silencing of miR-21-5p in donor BMSCs attenuated this inhibitory effect (Fig. 7f). Notably, pharmacological inhibition of SKP2 using SKPin C1 effectively reversed the suppressive effect of the miR-21-5p inhibitor on FoxO1 expression (Fig. 7g). These findings suggest that BMSC-derived exosomal miR-21-5p suppresses FoxO1 ubiquitination by targeting SKP2.

Upregulation of FoxO1 expression rescues the impairment of osteogenic differentiation and Th17/Treg balance induced by miR-21-5p suppression

To investigate the role of FoxO1 in mediating the pro-osteogenic effects of exosomal miR-21-5p, we overexpressed FoxO1 in MC3T3-E1 cells and confirmed its expression by western blot analysis (Fig. 8a). As shown in Fig. 8b, c, FoxO1 overexpression effectively reversed the inhibitory effects of miR-21-5p knockdown on osteogenic differentiation, as evidenced by enhanced ALP staining and increased protein levels of key osteogenic markers, including ALP, Bglap, and Runx2.

Fig. 8.

Upregulation of FoxO1 expression rescues the impairment of osteogenic differentiation and Th17/Treg balance caused by miR-21-5p suppression. A Western blot analysis of FoxO1 expression in MC3T3-E1 cells transfected with an empty vector or a FoxO1 overexpression plasmid. B ALP staining of MC3T3-E1 cells under the indicated treatment conditions. C the protein expression levels of osteogenic markers (ALP, Bglap, and Runx2) were measured by western blotting. D western blot validation of FoxO1 expression in CD4⁺ T cells transfected with empty vector or FoxO1 overexpression plasmids. E flow cytometric analysis of Th17 and Treg cell populations among FoxO1-overexpressing CD4(+) T cells treated with exosomes derived from BMSCs transfected with a miR-21-5p inhibitor. n = 3; Data are presented as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001

We further explored the function of FoxO1 in regulating the Th17/Treg balance. CD4(+) T cells were transfected with FoxO1 overexpression plasmids, and successful overexpression was validated by western blotting (Fig. 8d). Flow cytometry analysis revealed that upregulation of FoxO1 expression reduced the proportion of Th17 cells and promoted Treg differentiation in CD4(+) T cells treated with exosomes isolated from miR-21-5p inhibitor-transfected BMSCs (Fig. 8e).

Discussion

In this study, we elucidated the role of BMSC-exosomes in promoting osteogenesis and preventing age-related bone loss. We found that miR-21-5p was highly enriched in BMSC-derived exosomes and played a key role in mediating osteogenic effects both in vitro and in vivo. The knockdown of miR-21-5p expression in BMSCs significantly impaired their ability to promote osteogenic differentiation. Further mechanistic analysis revealed that SKP2, a key component of the UPS, is a direct target of miR-21-5p. Notably, inhibition of SKP2 partially reversed the suppressive effects on osteogenesis caused by miR-21-5p silencing. Moreover, the downregulation of SKP2 prevented the ubiquitination-mediated degradation of FoxO1. We further demonstrated that miR-21-5p-enriched exosomes derived from BMSCs contribute to the maintenance of Th17/Treg homeostasis. Importantly, upregulation of FoxO1 expression was able to reverse the osteogenic impairment induced by miR-21-5p inhibition in BMSCs. Collectively, these results suggest that miR-21-5p-loaded BMSC-derived exosomes represent a promising therapeutic approach for ageing-related OP by simultaneously enhancing osteogenesis and restoring immune balance through the miR-21-5p/SKP2/FoxO1 axis and modulating Th17/Treg dynamics (Fig. 9).

Fig. 9.

Schematic diagram showing the role of BMSC-derived exosomal miR-21-5p in osteoporosis

OP is a common age-related disease characterized by reduced bone density and strength, leading to an increased risk of fractures. As the global population continues to age, OP has emerged as a significant health issue that negatively impacts quality of life. MSCs have attracted increasing attention as a promising therapeutic approach for various diseases because of their abundant sources, self-renewal capacity, multidirectional differentiation potential, and immunomodulatory properties [27]. Accumulating evidence suggests that the therapeutic effects of MSC-based treatments are largely mediated by MSC-derived exosomes [28]. MSC-exosomes, primarily obtained from sources such as adipose-derived stem cells (ADSCs), bone marrow-derived mesenchymal stem cells (BMSCs), and human umbilical cord mesenchymal stem cells (hUC-MSCs), play important roles in wound healing [29], the treatment of autoimmune diseases [30], and bone regeneration [31]. Compared with traditional stem cell transplantation therapy, exosomes offer notable advantages, including lower toxicity and reduced immunogenicity [32]. In particular, BMSC-derived exosomes have garnered significant interest because of their potential in treating bone disorders, including OP [33].

In this study, we investigated the therapeutic potential of BMSC-derived exosomes in OP using both in vivo and in vitro models. Among various OP models, the SAMP6 mouse has several distinct advantages: its bone biological characteristics closely resemble those of humans, and extensive experimental data on its physiological, biochemical, morphological, and pharmacological properties are available, facilitating mechanistic studies and comparative analyses of osteoporotic pathology. Additionally, SAMP6 mice reproduce rapidly, enabling large-scale, reproducible experiments [34]. Notably, previous studies have demonstrated no significant sex-based differences in age-related bone changes in SAMP6 mice [35, 36]. Therefore, we utilized both SAMP6 and control SAMR1 mice to evaluate the inhibitory effect of BMSC-derived exosomes on bone loss. Furthermore, in vitro validation was performed using MC3T3-E1 cells, a well-established preosteoblastic cell line with robust osteogenic differentiation potential that is widely employed in OP research. Our results indicate that treatment with BMSC-derived exosomes promotes osteogenic differentiation of MC3T3-E1 cells.

Accumulating evidence suggests that miRNAs are differentially expressed in individuals with and without OP, highlighting their potential as diagnostic biomarkers and therapeutic targets [37]. However, free miRNAs are highly susceptible to enzymatic degradation in biological environments, limiting their therapeutic efficacy. Exosomes can encapsulate miRNAs, protecting them from degradation and facilitating intercellular delivery. The regulatory role of exosome-transported miRNAs in disease pathogenesis and treatment has been widely proposed, and exosomal miRNAs have emerged as key players in OP [38, 39]. Among the numerous candidate miRNAs, we focused on miR-21-5p, which has been identified as a potential diagnostic biomarker for OP [40] and is implicated in modulating age-related bone loss through exosomal transfer [11]. In this study, we revealed that miR-21-5p is highly expressed in BMSC-derived exosomes and mediates cell-to-cell communication. The levels of miR-21-5p in BMSC-derived exosomes correlated with their capacity to promote osteogenesis. Using a loss-of-function approach (transfection with a miR-21-5p inhibitor), we confirmed that miR-21-5p contributes to BMSC-exosome-induced osteogenesis both in vitro and in vivo.

Biologically, miR-21-5p binds to the 3′-UTRs of target mRNAs, leading to translational repression or mRNA degradation [41, 42]. By integrating bioinformatics predictions with the literature, we identified SKP2 as a direct target of miR-21-5p [43]. Notably, miR-21-5p has been implicated in the progression of OP through the modulation of SKP2 expression [44]. In this study, we further elucidated the functional interaction between miR-21-5p and SKP2. The administration of BMSC-derived exosomes reduced SKP2 expression in MC3T3-E1 cells, CD4(+) T cells and SAMP6 mice. However, inhibition of miR-21-5p expression in BMSCs abolished the suppressive effect of BMSC-derived exosomes on SKP2 expression, indicating that exosomal miR-21-5p plays a critical role in regulating SKP2. SKP2 functions as an E3 ubiquitin ligase and plays a key role in various cellular processes by promoting the ubiquitination and subsequent proteasomal degradation of specific target proteins [45, 46]. Previous studies have shown that SKP2 mediates the ubiquitination and degradation of FoxO1 [47]. Our findings demonstrate that FoxO1 is a downstream effector in the pro-osteogenic signalling pathway mediated by the BMSC-exosome-miR-21-5p-SKP2 axis, highlighting its importance in osteogenic regulation.

In recent years, the involvement of the immune system and immune-related factors in the pathogenesis of bone diseases has been increasingly well established [48]. Various immune cells, including T and B lymphocytes, cooperate to maintain immune homeostasis [49]. Among T-cell subsets, Th17 cells and regulatory T (Treg) cells represent two distinct lineages that reciprocally regulate each other and play pivotal roles in the initiation and progression of OP [15]. Restoring the balance between Th17 and Treg cells has therefore emerged as a promising therapeutic approach for OP [50, 51]. Notably, accumulating evidence indicates that FoxO1 is a key regulator of the maintenance of Th17/Treg homeostasis [52, 53]. Building on these insights, we hypothesized that exosomes derived from BMSCs might alleviate OP by modulating the Th17/Treg balance. Our findings revealed that compared with control mice, SAMP6 mice exhibited elevated Th17 cell levels and reduced Treg cell populations. The administration of BMSC-exosomes partially reversed these imbalances in the SAMP6 mice. Furthermore, we demonstrated that miR-21-5p, which is enriched in BMSC-exosomes, plays a critical role in mediating the restoration of Th17/Treg homeostasis. The regulation of Th17/Treg homeostasis mediated by BMSC-derived exosomal miR-21-5p was further validated in CD4(+) T cells. In vitro experiments revealed that overexpression of FoxO1 could rescue the impairment of osteogenic differentiation and Th17/Treg balance induced by suppression of miR-21-5p. These results suggest that BMSC-exosomes may exert therapeutic effects in OP through the miR-21-5p/FoxO1 axis by rebalancing the Th17/Treg ratio, highlighting a potential immunomodulatory mechanism underlying exosome-based interventions.

Although the role of exosomal miR-21-5p in OP has been documented in multiple studies, its impact on Th17 and Treg dynamics in OP remains unclear. In this study, we used SAMP6 mice, MC3T3-E1 cells, and CD4(+) T cells to investigate the functional role and underlying mechanisms of miR-21-5p. The dual effects—promoting osteogenesis and alleviating immune imbalance—underscore the therapeutic potential of BMSC-derived exosomes in OP. However, as with the majority of studies, the design of the current study is subject to limitations. First, the SAMP6 model was used to investigate the functional role of exosomes in the context of ageing, whereas MC3T3-E1 cells were not subjected to ageing conditions in vitro. Currently, we are endeavouring to isolate BMSCs from both SAMP6 and SAMR1 mice to further increase the scope and validity of our research. However, owing to the instability of isolated cells, no conclusive experimental data have been obtained thus far. Furthermore, limitations exist regarding the sample size and range of experimental animal models used. To advance preclinical research on BMSC-derived exosomal miR-21-5p and evaluate its therapeutic potential in OP, it is essential to validate findings across a diverse array of OP animal models, supported by long-term observation and systematic data recording. Moreover, the biological functions of BMSC-derived exosomal miR-21-5p are highly complex, and a comprehensive understanding of their multilevel and multitarget regulatory mechanisms in OP pathogenesis remains elusive.

In conclusion, although BMSC-derived exosomal miR-21-5p holds significant promise for OP treatment, multiple challenges hinder its clinical translation, including issues related to exosome delivery, long-term storage, and maintenance of biological activity. Therefore, standardized protocols and analytical methodologies must be established and validated through large-scale, multicentre studies involving well-characterized cohorts to substantiate their clinical utility.

Author contributions

The concept of the research was provided by Zeyu Wang and Jie liu. Animal experiments were carried out by Zeyu Wang and Jianhai Luo, cell experiments were performed by Tong Yu and Pengcheng Hu. Statistical analysis was done by Zeyu Wang. The manuscript was drafted by Jie Liu. All authors read and approved the final manuscript.

Funding

This work was supported by the Yunnan Province Key Laboratory of Digital Orthopedics (No. 202005AG070004), Yunnan Province Clinical Medical Center for Spine and Spinal Cord Diseases (No. ZX2022000101) and Yunnan Provincial Science and Technology Department key R&D projects (No. 202403AC100003).

Data availability

N/A.

Declarations

Ethics approval and consent to participate

The animal study was reviewed and approved by Ethics Committee of Kunming Medical University (Approval No. kmmu20241120). Title of the approved project: Research on the Mechanism of Yi Medicine in Improving Bone Aging. Date of approval: 14/3/2024. The original source (the Cell Bank of Chinese Academy of Sciences) has confirmed that there was initial ethical approval for collection of human cells, and that the donors had signed informed consent.

Competing interests

The authors declare no competing interests.