Abstract Background Osteoarthritis (OA) is a prevalent degenerative joint disorder affecting over 240 million people worldwide, yet no disease-modifying therapies currently exist, with clinical management limited to symptomatic relief or joint replacement. Exosomes (Exos) from bone marrow mesenchymal stem cells (Exo^BMSC) play positive role in the treatment of cartilage damage. Parathyroid hormone (PTH) (1–34) can enhance cartilage repair. Here, We found Exos from Exo^BMSC reduces cartilage damage during treatment. Meanwhile, the Exos of PTH(1–34)-preconditioned BMSCs (Exo^PTH) can alleviate OA better than Exo^BMSC. Through MicroRNA (miRNA) sequencing analysis, this study aims to reveal the effects and potential mechanism of miRNA (let-7a-5p) in Exo^PTH to repair OA cartilage. Methods Differential centrifugation was used for isolating Exo^BMSC and Exo^PTH. Extract bone marrow mesenchymal stem cells from rats and utilize the C28/I2 chondrocytes line, the OA model was established using lipopolysaccharide (LPS; 1 µg/mL) in vitro. OA was induced in rats with intra-articular injection with collagenase-2. By performing a miRNA array, RNA-seq, in addition to bioinformatic analysis, the miRNA and the potential regulatory mechanism were detected. We compared in vitro let-7a-5p effects on the ability of OA chondrocytes to proliferate, migrate, apoptosis, and form the extracellular matrix (ECM). Histological and immunohistochemical assessments were used for evaluating cartilage pathology in vivo. Results We extracted Exo^BMSC and Exo^PTH and established the OA model in vitro. Compared with Exo^BMSC group, Exo^PTH group has a stronger effect on promoting the proliferation and migration of chondrocytes. Exo^BMSC and Exo^PTH can inhibit the apoptosis of chondrocytes, but there was no significant difference between the two groups. The two most significant differences in groups Exo^BMSC and Exo^PTH are let-7a-5p. Let-7a-5p promotes OA chondrocytes proliferation and migration by inhibiting the expression of IL-6 in vitro experiments. For in vivo experiments, let-7a-5p delays the progression of OA. Conclusion Our study shows that Exo^PTH may improve the regulatory inflammatory responses to delays the progression of OA by shuttling let-7a-5p. Let-7a-5p promoted chondrocytes migration and proliferation to suppress OA pathology by inhibiting IL-6/STAT3 pathway. Supplementary Information The online version contains supplementary material available at 10.1186/s13287-025-04416-0. Keywords: Osteoarthritis, PTH (1–34) preconditioned-BMSCs exosomes, Let-7a-5p, Chondrocyte, IL-6 Background Osteoarthritis (OA) has emerged as a widespread degenerative joint disease with serious effect on the life quality of the patients and significant socio-economic impacts. An epidemiological study revealed that OA impacts 30% of people over the age of 60 and increases to 40% among those over 70 [[38]1]. Cartilage is made up of chondrocytes, collagens, and proteoglycans. Because cartilage lacks innervation and vascularization, it has a limited capacity for repair [[39]2]. Nowadays, despite using anti-inflammatory drugs, analgesics, and viscosupplementation to treat OA, there is a lack of OA disease-modifying therapy. Bone marrow mesenchymal stem cells (BMSCs) have showed potential in preclinical studies for cartilage repair [[40]3, [41]4]. However, its clinical translation remains limited, with only a few early-phase trials revealing symptomatic improvement through intra-articular injection of bone marrow concentrate (BMC) or aspirate (BMA) [[42]5]. Notably, disease-modifying effects of BMSCs in OA are still debated. Their clinical application remains limited due to challenges in standardization and inconsistent therapeutic outcome [[43]6]. To delay OA progression, noncellular therapies may represent a new research direction. Recent decades have seen an explosion of interest in exosomes (Exos) for the treatment of OA. There are several active substances in Exos, including DNA, RNA, and proteins [[44]7, [45]8]. Exos exhibit potential advantages such as targeted delivery and enhanced bioavailability. Preclinical studies suggest they may mitigate OA progression by modulating inflammation and chondrocyte metabolism [[46]9, [47]10], though evidence for true tissue regeneration remains limited. Exos from different sources play different roles in different contexts. There are, however, some side effects associated with the Exos application [[48]11]. BMSC-derived exosomes (Exo^BMSC) may promote chondrocyte hypertrophy and calcification, a process resembling endochondral ossification rather than functional cartilage repair [[49]6]. Based on our previous research, in contrast to Exo^BMSC, Exos from a parathyroid hormone (PTH)-preconditioned BMSCs (Exo^PTH) increased OA chondrocytes migration and proliferation [[50]12]. Despite the similar effects of two Exos types, Exo^PTH shows better effects than Exo^BMSC, so it can alleviate OA better. The reason for the effects of two different Exos was unexplored. The primary objective of this research is to identify effective therapeutic drugs and targets that address core pathological mechanisms. Multiple miRNAs have demonstrated crucial roles in delaying the progression of OA [[51]13, [52]14, [53]15]. An in-depth investigation of microRNA (miRNA) variations within two types of Exos is conducted, presenting new possibilities for the treatment of OA. In particular, the comparative analysis of exosomal miRNA profiles from distinct origins revealed let-7a-5p as a promising therapeutic target, capable of impeding OA advancement by inhibiting the IL-6/STAT3 signaling pathway. Methods BMSCs and chondrocyte characterization Rat BMSCs [[54]16] and C28/I2 chondrocytes line [[55]17] were procured from the Shanghai GuanDao Biological Engineering Co., Ltd and the Wuhan Procell Life Sci&Tech Co., Ltd, respectively. BMSC-specific surface markers (CD44, CD90, and CD45) were analyzed by flow cytometry. A collagen-type II (COL-2) immunofluorescence (IF) stain was used to confirm chondrocytes. The chondrocytes were treated with LPS 1 µg/mL for 24 h for the construction of a cell OA model, followed by drug treatment (let-7a-5p mimics 20 nM) for an additional 12, 24 and 48 h. Exos extraction and identification PTH (1–34) was used for pretreating BMSCs for 6 h, followed by replacing the medium again, which was collected after 48 h, and by collecting 200 mL cell culture media, Exos were harvested, followed by removing both cell debris as well as non-Exo vesicles and any possible apoptotic bodies from the cell culture supernatant by being centrifuged at 300 and 2000 g, for 10 min each as well as at 10,000 g for 40 min, respectively at 4 °C. Eventually, Exos were obtained at 70 min of 120,000 g centrifugation at 4 °C, to be either utilized immediately or kept at 80 °C for subsequent analysis. Exos were labeled with PKH67 (Sigma, USA) for 15 min at 37 °C, followed by ultracentrifugation to remove excess dye. Labeled Exos were incubated with chondrocytes for 24 h before imaging. Transmission electron microscopy (TEM) analysis was performed on Exos with a Hitachi HT7700 transmission electron microscope (Japan). By performing antibodies against CD63, TSG101 and CD9 (abCAM, USA), the Exo protein markers were determined. Exos were validated for BMSCs origin using antibodies against CD90 and CD14 (CST, USA). After 100 times dilution and using the nanoparticle tracking analysis (NTA), size distribution was determined. Treating the primary chondrocytes with Exos (10 µg/mL) for 24 h followed previous studies [[56]18, [57]19]. Western blot (WB) assays We modified the WB assays slightly from previous publications [[58]20], starting by lysing the samples in cell lysis buffer without prior fixation, incubating them on ice for 5 min, and scraping them off of the plates. A microplate reader was used to measure protein concentrations after centrifugation to remove cell debris. Protein concentration was measured using a BCA assay kit (Thermo Fisher), with absorbance at 562 nm normalized to a bovine serum albumin (BSA) standard curve. After nearly 30 min of 180 V electrophoresis on a 10% gradient gel, the proteins were electrotransferred to PVDF membranes and incubated for a whole night with primary antibodies at 4 °C after blocking the membranes overnight with 5% skim milk at room temperature, then succeeded by washing the membranes in PBS for 5–7 min, followed by 1 h of blotting with secondary antibodies. LAS-3000 Luminescent Image Analyzer (Fuji Film, Tokyo, Japan) was used to analyze the labeled proteins. Cell proliferation assay A CCK-8 assay was utilized for measuring cell proliferation. It began with seeding chondrocytes in 96-well plates, where cells were cultivated at 37 °C for 4 h and then instilled with CCK-8 reagents. In the next step, the addition of 10 µL CCK-8 solution took place into each 96-well plate, incubated for 1 h at 37 °C, and eventually measured the absorbance at 450 nm by a microplate reader. Cell migration assay The scratch assay was used for performing the cell migration. At 90–100% confluence, serum-free DMEM medium was used for treating the cells for 24 h, succeeded by linear scratching using a 100-µL sterile pipette tip. A microscope was utilized to obtain images. Migration ratio (%) = [(Initial scratch width - Final scratch width) / Initial scratch width] * 100%. Images were analyzed using Image J. Apoptosis assay To conduct the apoptosis assays, after collecting 3 × 10^5 cells from each sample, cells underwent PBS wash and resuspension in 200 µL Binding Buffer. Annexin V - FITC and Propidium Iodide (PI) solution addition took place for 10 min at room temperature. Eventually, a flow cytometer was used to analyze the cells. Analysis of mRNAs by real-time quantitative reverse transcriptase polymerase chain reaction (RT-qPCR) TransScript II All-in-One First-Strand cDNA (GeneCopoeia, Guangzhou, China) and ReverTra Ace qPCR RT Master Mix (Toyobo, Japan) were used for mRNA. Afterward, qRT-PCR was performed by SYBR^® Green Real-time PCR Master Mix (TOYOBO, Japan) for mRNA. mRNA were controlled using β-actin. Primer syntheses were performed by Guangzhou IGE Biotechnology Corporation (Guangzhou, China). Using the 2^−△△Ct method, quantitative analysis was conducted. Primer sequences can be found in Table [59]1. Table 1. List of the five PCR primer pairs chosen for in vitro PCR validation Primer for RT-PCR Forward(5'-3') Reverse(5'-3') β-actin CACCATTGGCAATGAGCGGTTC AGGTCTTTGCGGATGTCCACGT IL-6 AGACAGCCACTCACCTCTTCAG TTCTGCCAGTGCCTCTTTGCTG Collagen II CCTGGCAAAGATGGTGAGACAG CCTGGTTTTCCACCTTCACCTG MMP-13 CCTTGATGCCATTACCAGTCTCC AAACAGCTCCGCATCAACCTGC ADAMTS 5 CCTGGTCCAAATGCACTTCAGC TCGTAGGTCTGTCCTGGGAGTT [60]Open in a new tab RNA-seq and data analysis Small RNA library preparation, sequencing [[61]21] and bioinformatic analyses took place at BGI (Shenzhen, China). To prepare the library, polyacrylamide gel electrophoresis was used for purifying an 18–40 nt small RNA and then ligated by proprietary adaptors, which was then reversely transcribed to cDNA and amplified by PCR. Eventually, BGISEQ-500 technology was used for sequencing the libraries. BGI Company performed the miRNA microarray assay on the BGISEQ-500 platform using TPM to normalize the raw data. Then, DESeq2 was applied by adjusting |fold change| ≥ 2.0 and p < 0.05 to define differentially expressed miRNAs. Using Sangerbox, we created the heat map. TargetScan, miRanda, and RNAhybrid databases were used to predict miRNA-predicted targets. Animals Ethics approval and consent to participate the animal experiment was approved by the Institutional Animal Care and Use Committee of the Shenzhen TopBiotech Co. Ltd. We housed and fed 12-week-old SD rats (n = 24; Vital River Experimental Animal Technical Co., Ltd., Beijing, China) a standard rodent diet. We randomly divided the rats into 4 groups (n = 6) as follows: saline joint injection (Control group), the collagenase + saline injection group (OA group); the collagenase + mimics negative control (NC) treatment group (OA + NC-mimics group); the collagenase + let-7a-5p mimics treatment group (OA + let-7a-5p mimics group). The rats were anesthetized using 3% isoflurane. Through joint cavity injection of collagenase-2 [[62]22], we have established a rat OA model. Each injection of let-7a-5p mimics or NC-mimics (20 nM) in the articular cavity was administered once a week for four consecutive weeks [[63]14]. We housed all rats under standard laboratory conditions in temperature-controlled rooms and provided them with 12 h of light and 12 h of darkness every day. The animals were normally raised without any abnormal deaths. After 6 weeks, rats were euthanized using CO[2] asphyxiation and subsequently necropsied. Histology After fixing the knees in 4% formalin solution for 48 h, they were decalcified for 5 weeks in 10% Na[2]EDTA solution (PH = 7.3) at room temperature as previously described [[64]23–[65]24]. An embelimination procedure was used to embed paraffin in the tibia and femur, and the lengths were cut to a thickness of 5 μm. Hematoxylin and eosin (HE) and Safranin O/fast green (SOF) staining was applied to the coronal sections of each sample. Then, section images were recorded under an optical microscope (Olympus BX53, Olympus, Japan). According to the OARSI score system, five characteristics of articular cartilage were determined: articular cartilage structure, tidemark integrity chondrocyte density, cell cloning, tidemark integrity, and interterritorial SOF [[66]25]. Immunohistochemical assessments The primary antibodies to metalloproteinase-13 (MMP-13) (1: 300; Gene Tex Inc., USA) and collagen II (COL-2) (1: 200; Gene Tex Inc., USA) were used to perform immunohistochemistry. A brief summary of the procedure is as follows: all sections were deparaffinized, rehydrated, and repaired with 0.05% trypsin. A ten-minute inactivation with H[2]O[2]at room temperature was performed on endogenous peroxidases. With the previous target protein, overnight incubation at 4 °C was performed. A PV-6000 DAB detection kit and a ZLI-9018 DAB kit (both from ZSGB-BIO Corp., China) were used for the remaining experiments. Tibia plateau cartilage, intensity of optical density (IOD), expressed as IOD/mm^2, was determined by dividing the integrated optical density by the area of cartilage tissue in the region of interest (ROI) with a magnification of 5 times. Immunostaining intensity was quantified as IOD per mm² cartilage area using Image-Pro Plus (IPP) 6.0 (Media Cybernetics, Rockville, Maryland, USA). Then, the IOD values of each group are divided by the average IOD of the Control group, using this as the relative value between groups to compare the differences among them, as previously described [[67]26]. Immunofluorescence Immunofluorescence was performed following the steps below. The paraffin sections of the knees were baked at 65 °C for one hour, then transferred quickly to xylene I and xylene II for 10 min each, and subsequently soaked in gradient alcohol 100%, 95%, 85% and 75% for hydration. The slices were subjected to antigen retrieval using sodium citrate, blocked in 5% BSA at room temperature (RT) for one hour and incubated with the anti-MMP-9 (1:200; Proteintech, Wuhan, China) antibody and anti-aggrecan (AGG) (1:200; Servicebio Technology, Wuhan, China) antibody at 4 °C overnight. After washed in PBST for 3 times, the sections were incubated with Alexa Flour 594-conjugated secondary antibodies (1:500; Thermo Scientific, USA) at RT for one hour, next counterstained with DAPI and mounted for observation under the fluorescence microscope. Immunofluorescence was quantified as IOD per mm² cartilage area using IPP. Then, the IOD values of each group are divided by the average IOD of the Control group, using this as the relative value between groups to compare the differences among them, as previously described [[68]26]. Data analysis and statistics Data were reported as mean and SD. One-way analysis of variance (ANOVA) was utilized for testing the differences in Gaussian distribution data between groups while using the Fisher’s least significant difference (LSD) t-test or Kruskal-Wallis H-test for pairwise comparisons that depend on variance homogeneity. By performing the Kruskal-Wallis H-test, OARSI scores, and non-Gaussian distributed data were analyzed. Two-tailed p < 0.05 indicated statistically significant. R, version 3.5.3 (R Project for Statistical Computing), was utilized to perform the analyses. Results BMSCs and chondrocyte characterization The chondrocytes used for this study were tested for COL-2 (Supplementary Fig. [69]1A). Supplementary Fig. [70]1B shows the observed BMSCs under an inverted microscope. Supplementary Fig. [71]1C–E show that the majority of BMSCs are negative for CD45 and positive for CD44 and CD90. Identification of Exo^BMSC and Exo^PTH A flow diagram showed that the two kind of Exos were extracted and analyzed by RNA-seq (Supplementary Fig. [72]2A). TEM observations revealed that Exo^BMSC and Exo^PTH had the typical Exos morphology by having a saucer or cup shape (Supplementary Fig. [73]2B). The Exo^BMSC and Exo^PTH particle concentration was approximately 4.26 × 10^11 and 5.15 × 10^11 particles/mL (Supplementary Fig. [74]2C–D). Supplementary Figs. [75]2E–F revealed that most Exo^BMSC and Exo^PTH particles were 60–150 nm in size. The distribution of Exos within chondrocytes was seen in the cytosol (Supplementary Fig. [76]2G), indicating that Exo^BMSC and Exo^PTH were successfully taken up by these cells. According to WB analysis of the Exos (Supplementary Fig. [77]2H and [78]3A), CD9, TSG101, and CD63 were expressed specifically. To further validate that the Exos originate from BMSCs, we examined the expression of CD90 and CD14 in both types of Exos. The results demonstrated consistent expression of CD90 while no detectable expression of CD14 was observed (Supplementary Fig. [79]3A). Exo^BMSC and Exo^PTH promote OA chondrocyte migration and proliferation, and Exo^PTH showes stongerly Figure [80]1 illustrates the scratch wound assay, demonstrating enhanced chondrocyte migration in the Exo^PTH group compared to Exo^BMSC and OA groups. In Fig. [81]1A-C, compared to the Control group, the rate of migration were dramatically reduced in the OA group. The rate of migration in the Exo^BMSC group and Exo^PTH group was increased compared to the OA group. The Exo^PTH group had a higher increase compared to the Exo^BMSC group. In Fig. [82]1D, compared with the other three groups, there was a significant reduction in cell migration in the OA Group. The number of cell migration was significantly increased in the Exo^PTH group than in the Exo^BMSC group. In Supplementary Fig. [83]4A-B, compared with the other three groups, there was a significant reduction in cell proliferation in the OA Group at 24 and 48 h. The number of cell proliferation was significantly increased in the Exo^PTH group than in the Exo^BMSC group at 24 and 48 h. Fig. 1. [84]Fig. 1 [85]Open in a new tab Exo^BMSC and Exo^PTH promote chondrocyte migration, and Exo^PTH showed stongerly. Exo^BMSC and Exo^PTH promote chondrocyte migration, Exo^PTH showed stongerly. (A). The scratch wound assays of chondrocytes. (B). Statistical results chart of the scratch wound assays of chondrocytes at 24 h. (C). Statistical results chart of the scratch wound assays of chondrocyte at 48 h. (D). Chondrocyte migration assays. Exo^BMSC: Exosomes derived from BMSCs; Exo^PTH: Exosomes derived from PTH-preconditioned BMSCs; BMSCs: bone marrow mesenchymal stem cells. bars = 100 μm. All data are expressed as the mean ± SD. *p < 0.05, **p < 0.01 Exo^BMSC and Exo^PTH inhibiting OA chondrocyte apoptosis In Supplementary Fig. [86]4C-D, cell apoptosis was significantly increased in the OA group than any other group. However, there was no significant difference between Exo^BMSC group and Exo^PTH group. Let-7a-5p was associated with OA Based on RNA-seq assay in Exo^BMSC and Exo^PTH, three overexpressed miRNAs (let-7a-5p/miR-133c/miR-1b) were identified, with a negative or positive fold change > 2 (all p < 0.05) (Fig. [87]2A). Analyses of enriched KEGG pathways have been performed (Fig. [88]2B). KEGG analysis revealed enrichment for pathways associated with immune disease, cell growth and death and cell motility. Let-7a-5p showed the highest fold change in the two kind of Exos. After PCR validation (Supplementary Fig. [89]3B-D), compared with the Exo^BMSC group, the expression level of let-7a-5p was significantly increased in the Exo^PTH group, but no significant differences were observed between Exo^BMSC and Exo^PTH group for the other two miRNAs. For the subsequent analysis, we focused only on the let-7a-5p (Fig. [90]2C), which showed the most significant differences between the two kinds of Exos. Meanwhile, for target genes of the miRNA, KEGG pathway enrichment analysis was performed utilizing Sangerbox. KEGG pathway analysis of let-7a-5p was utilized for identifying significant pathways correlated to OA, including MAPK, TGF-β and JAK-STAT signaling pathways(Fig. [91]2D). Fig. 2. [92]Fig. 2 [93]Open in a new tab RNA-seq assay was performed in Exo^BMSC and Exo^PTH, and let-7a-5p showed the most significant differences. RNA-seq assay was performed in Exo^BMSC and Exo^PTH, and let-7a-5p showed the most significant differences. (A). Heat map of miRNA-seq. (B). KEGG enrichment analysis of three miRNA targets. (C). TargetScan, miRanda, and RNAhybrid databases were used to predict let-7a-5p predicted targets. (D). The top 20 most enriched pathways fortargeting proteins of let-7a-5p by KEGG pathway analysis Let-7a-5p promoting chondrocyte proliferation Figure [94]3A demonstrated the uptake of let-7a-5p by chondrocytes. Figure [95]3B indicates that concentrations of 50 nM and 100 nM significantly inhibited chondrocyte proliferation at 24 and 48 h after treatment, compared to the Control group. Figure [96]3C revealed that the cell survival rate in the OA group was significantly lower than that in the Control group. Conversely, the 20 nM concentration group exhibited a significantly higher cell survival rate than the OA group. However, there were no significant differences in cell survival rates between the OA group and the NC-mimics or 10 nM groups. Thus, we chose 20 nM as a non-toxic and effective concentration for ensuing experiments. Figure [97]3D-E showed that there was a significant difference in the Edu^+ cells in the OA group compared with controls, but there was no difference in the OA group compared with OA + NC-mimics group. Compared to the OA + NC-mimics group, the OA + let-7a-5p mimics group was significantly increased. Fig. 3. [98]Fig. 3 [99]Open in a new tab Let-7a-5p promotes proliferation in OA chondrocytes. (A). Chondrocytes absorb fluorescently tagged let-7a-5p. (B). The CCK-8 assay assessed the impact of different concentrations of let-7a-5p mimics on cell viability. (C). Cell proliferation rates of various groups as measured by CCK-8 assay. (D-E). Cell proliferation rates of various groups as measured by flow cytometry Let-7a-5p promoting OA chondrocyte migration Figure [100]4A showed the scratch assay. As shown in Fig. [101]4C, the migration ratio of chondrocytes was signifcantly increased in the OA + let-7a-5p mimics group than the OA + NC-mimics group at 24. Compared with the Control group, the migration ratio of chondrocytes in the OA group signifcantly decreased at two time points. There was no noticeable difference among the other groups at 12 h (Fig. [102]4B). Fig. 4. [103]Fig. 4 [104]Open in a new tab Let-7a-5p enhanced OA chondrocyte migration. (A). scratch wound-healing migration assay was performed. (B). A scratch wound-healing migration assay statistical results at 12 h. (C). A scratch wound-healing migration assay statistical results at 24 h. bars = 100 μm. All data are expressed as the mean ± SD. *p < 0.05, **p < 0.01 Let-7a-5p delaying the progression of OA in vivo Figure [105]5A showed the experimental design. Figures [106]5B-C showed cartilage histological observation and OARSI scoring. White arrows indicated areas of lesion. The OA group had more significant histological changes, including cartilage surface irregularities, reduced chondrocyte density, and Safranin-O/Fast green staining of the hyaline chondrocyte layer. The total OARSI score in the OA group showed a significantly high score than in the Control group, while the total OARSI score in the OA + NC-mimics group had a significantly high score than in the OA + let-7a-5p mimics group. Fig. 5. [107]Fig. 5 [108]Open in a new tab Histological analysis and OARSI scoring between groups, and let-7a-5p delaysed OA progression. (A). The Flowchart in vivo experiment. (B). OARSI microscopic scoring between groups (HE and SOF staining). (C). The result of statistical analysis of OARSI score in each group. White triangles indicate the site of damaged cartilage tissues. bars = 100 μm. All data are expressed as the mean ± SD. **p < 0.01. HE: Hematoxylin and eosin; SOF: safranin-O/fast green Let-7a-5p not inhabiting chondrocyte apoptosis In Supplementary Fig. [109]5, chondrocyte apoptosis was assessed by flow cytometry. Compared with the Control group, the rate of OA chondrocyte apoptosis showed a substantial increase in the OA group. However, there was no noticeable difference between the OA + NC-mimics group and the OA + let-7a-5p mimics group. Let-7a-5p reducing the degradation of the ECM To further investigate the role of let-7a-5p in cartilage repair after OA, we assessed the expression of the main cartilage components COL-2 and AGG, as well as the ECM degrading enzymes MMP-9 and MMP-13. In the Fig. [110]6A, immunostained images demonstrated that the cartilage with brownish yellow was determined to be MMP-13 and COL-2 positive. In the Fig. [111]6B, the OA group showed significantly higher MMP-13 expression in the cartilage than the Control group. The OA + NC-mimics group showed significantly lesser MMP-13 expression in the cartilage than the OA + let-7a-5p mimics group. In the Fig. [112]6C, the OA group showed significantly less COL-2 expression in the cartilage than the Control group. The OA + NC-mimics group showed significantly less COL-2 expression in the cartilage than the OA + let-7a-5p mimics group. Moreover, the expression of AGG and MMP-9, a key indicator of cartilage matrix integrity, were further detected. Figure [113]7A displays immunofluorescence staining images, where red signals in cartilage were identified as positive expression of MMP-9 and AGG. In Fig. [114]7B, the expression of AGG in OA group cartilage was significantly lower than that in the Control group. Furthermore, AGG expression in the OA + NC-mimics group was markedly reduced compared to the OA + let-7a-5p mimics group. In Fig. [115]7C, MMP-9 expression in OA group cartilage was markedly higher than that in the Control group. Notably, MMP-9 expression in the OA + NC-mimics group was significantly lower than in the OA + let-7a-5p mimics group. Let-7a-5p can inhibit the upregulation of matrix degrading enzymes and protect the integrity of ECM, thereby delaying cartilage damage. Fig. 6. [116]Fig. 6 [117]Open in a new tab Let-7a-5p promoted the expression of COL-2 and inhibited the expression of MMP-13. (A). This figure demonstrates the expression of MMP-13 and COL-2 in load bearing areas of cartilage among groups. (B). The IHC statistical results of MMP-13. (C). The IHC statistical results of COL-2. The positive expression of MMP-13 and COL-2 was defined as brown-yellow stain. Data are presented as the mean ± SD. Long bars = 200 μm. Short bars = 100 μm. All data are expressed as the mean ± SD. **p < 0.01 Fig. 7. [118]Fig. 7 [119]Open in a new tab Let-7a-5p promoted the expression of AGG and inhibited the expression of MMP-9. (A). This figure demonstrates the expression of AGG and MMP-9 in load bearing areas of cartilage among groups. (B). The IF statistical results of AGG. (C). The IF statistical results of MMP-9. The positive expression of AGG and MMP-9 was defined as red stain. Data are presented as the mean ± SD. Long bars = 200 μm. Short bars = 40 μm. All data are expressed as the mean ± SD. **p < 0.01 The inhibition of IL-6 expression by let-7a-5p Figure [120]8A illustrated the putative binding sites between IL-6 and let-7a-5p. Subsequently, Fig. [121]8B showed the protein expression levels of IL-6, p-STAT3, and STAT3 as determined by WB analysis. Additionally, Fig. [122]8C presented the statistical findings pertaining to IL-6, revealing a notable elevation in IL-6 protein expression in the OA group in comparison to the Control group. Conversely, a significant reduction in IL-6 protein expression was observed in the let-7a-5p mimics group relative to the OA group. Furthermore, Fig. [123]8D illustrated the statistical findings regarding the p-STAT3/STAT3 ratio, revealing a significant increase in the OA group compared to the Control group, and a significant decrease in this ratio in the let-7a-5p mimics group compared to the OA + NC-mimics group. Additionally, Fig. [124]8E demonstrated a significant increase in IL-6 gene expression in the OA group compared to the Control group, and a significant decrease in IL-6 gene expression in the let-7a-5p mimics group compared to the OA + NC-mimics group. Figure [125]8F demonstrated a significant decrease in COL-2 gene expression in the OA group compared to the Control group, with a significant increase observed in the let-7a-5p group relative to the OA + NC-mimics group. Figures [126]8G-H illustrated a significant increase in MMP-13 and ADAMTS-5 gene expressions in the OA group compared to the Control group, while a significant decrease is observed in the let-7a-5p group relative to the OA + NC-mimics group. Fig. 8. [127]Fig. 8 [128]Open in a new tab Let-7a-5p inhibiting the IL-6/STAT3 signaling pathway by targeting IL-6 protected chondrocytes. (A). Potential binding sites of IL-6 to let-7a-5p. (B-D). Western blot showed the protein expression of IL-6, phospho-STAT3, and STAT3 in chondrocytes. (E-H). PCR showed the mRNA expression of IL-6, COL-2, MMP-13 and ADAMTS5 in chondrocytes Discussion OA is a multifactorial joint disorder influenced by aging, genetic predisposition, biomechanical stress, and systemic inflammation. The pathogenesis involves a complex interplay of cartilage degradation, synovial inflammation, and aberrant mechanotransduction. Until now, there is no method for OA treatment, and the existing therapies have side effects or require invasive procedures. Our study identifies let-7a-5p as the dominant functional miRNA in Exo^PTH, mediating its therapeutic superiority over Exo^BMSC. Mechanistically, let-7a-5p suppresses IL-6 expression, thereby inhibiting STAT3 phosphorylation (Fig. [129]8) and reducing MMPs-driven ECM degradation (Figs. [130]6 and [131]7). This miRNA modulation of the IL-6/STAT3 axis directly enhances chondrocyte proliferation (Fig. [132]3D-E) and migration (Fig. [133]4A-C), while attenuating OA progression in vivo (Fig. [134]5). Notably, PTH preconditioning enriches let-7a-5p in Exos (Fig. [135]2A), providing a novel strategy to optimize MSC-derived exosomes for OA therapy. The results of our experimental data indicate that let-7a-5p suppresses MMPs expression, leading to an increase in ECM expression, ultimately contributing to the retardation of OA progression. Prior research has demonstrated that extracellular vesicles containing let-7a-5p can attenuate inflammation in OA chondrocytes [[136]27] and safeguard OA cartilage ECM from harm [[137]28]. These findings align with the outcomes of our study. In order to delve deeper into the protective mechanism of let-7a-5p on OA chondrocytes, our investigation focused on its interaction with the IL-6/STAT3 axis. Our data suggest a correlation between let-7a-5p-mediated IL-6 downregulation and reduced STAT3 phosphorylation (Fig. [138]7B-D). The IL-6/STAT3 signaling pathway is known to play a pivotal role in the pathogenesis of OA, as evidenced by studies demonstrating that IL-6 signaling activation is linked to cartilage degeneration in OA models [[139]29] and elevated IL-6 levels are typically observed in OA patients [[140]30]. Downstream effectors of IL-6 signaling, notably STAT3, are integral to the pathogenesis of OA. Specifically, the upregulation of phosphorylated STAT3 (p-STAT3) in cartilage and synovium following joint injury correlates with worsened symptoms of OA [[141]31]. The results indicate that let-7a-5p may have a protective effect on cartilage through the inhibition of IL-6 expression, leading to the suppression of matrix degradation and the enhancement of chondrocytes proliferation and migration, as depicted in Fig. [142]9. While this aligns with known IL-6/STAT3 signaling in OA pathogenesis, future studies employing IL-6 overexpression or siRNA rescue experiments are required to establish direct causality. Fig. 9. [143]Fig. 9 [144]Open in a new tab Mechanistic diagram of our study. The let-7a-5p in BMSC exosomes stimulated by PTH can inhibit inflammation and delay OA by suppressing the IL-6/STAT3 pathway. OA: osteoarthritis; ECM: extracellular matrix; PTH: parathyroid hormone A limitation of this study is the lack of mechanistic validation through IL-6 overexpression- or loss-of-function experiments. To confirm whether IL-6 suppression is essential for let-7a-5p’s therapeutic effects, future work will include: (1) IL-6 overexpression in let-7a-5p-treated chondrocytes to assess STAT3 reactivation, and (2) IL-6 knockdown in OA chondrocytes to determine if it phenocopies let-7a-5p’s effects. Conclusions Let-7a-5p can suppress IL-6 expression, leading to enhanced chondrocyte proliferation and migration, reduced ECM degradation, and ultimately, attenuation of OA progression. The delivery of Exo-mediated let-7a-5p transfer via knee joint injection holds promise as a viable therapeutic approach for individuals with OA. Electronic supplementary material Below is the link to the electronic supplementary material. [145]Supplementary Material 1^ (128.2KB, xls) [146]Supplementary Material 2^ (13MB, jpg) [147]Supplementary Material 3^ (9.2MB, jpg) [148]Supplementary Material 4^ (539.5KB, jpg) [149]Supplementary Material 5^ (6.6MB, jpg) [150]Supplementary Material 6^ (3.3MB, tif) [151]Supplementary Material 7^ (1.3MB, jpg) [152]Supplementary Material 8^ (4.1MB, jpg) [153]Supplementary Material 9^ (14.5MB, tif) Acknowledgements