RAB5A Governs Synovial Fibroblast Extracellular Vesicles and Mediates Chondrocyte Degeneration in Osteoarthritis

Abstract

Osteoarthritis is a degenerative joint disease characterized by synovial inflammation and cartilage destruction. This study aimed to investigate the function of RAB5A in regulating human fibroblast-like synoviocytes (HFLS) and its subsequent impact on chondrocyte degeneration. Differential expression and pathway analyses were performed according to the Gene Expression Omnibus (GEO) dataset. HFLS were stimulated with IL-1β, and RAB5A was knocked down via transfection. Inflammatory signaling, cytokine expression, and extracellular vesicles (EVs) secretion were assessed. Isolated EVs were characterized by nanoparticle tracking analysis, transmission electron microscopy, and Western blotting. The functional effects of HFLS-derived EVs on chondrocyte inflammation, extracellular matrix metabolism, apoptosis, and viability were evaluated. Bioinformatics analysis identified RAB5A as a key gene linked to the endocytosis pathway. In IL-1β-stimulated HFLS, RAB5A expression was upregulated. RAB5A knockdown reduced the activation of p38 and NF-κB pathways and suppressed pro-inflammatory factors and matrix degradation mediators. Furthermore, RAB5A deficiency impaired EV secretion and altered their cargo. Crucially, while EVs from control HFLS promoted chondrocyte catabolism, inflammation, and apoptosis, EVs from RAB5A-knockdown HFLS mitigated these degenerative phenotypes. Our findings demonstrate that RAB5A, by modulating the biogenesis and composition of HFLS-derived EVs, plays a critical role in driving chondrocyte degeneration, highlighting its importance as a key molecular regulator in osteoarthritis.

Keywords

chondrocyte extracellular vesicle exosome osteoarthritis synovial

Introduction

Knee osteoarthritis (KOA) is characterized by progressive degeneration of articular cartilage, subchondral bone sclerosis, osteophyte formation, and synovial inflammation (Yunus et al., 2020). It is one of the leading causes of chronic pain and disability in middle-aged and elderly people worldwide, and its epidemiological burden is increasing with the aging of the global population (Zhu et al., 2024). Currently, clinical treatment strategies for KOA primarily aim to relieve symptoms, including nonsteroidal anti-inflammatory drugs, intra-articular injections, and physical therapy (Gelber, 2024), while end-stage patients can only undergo joint replacement surgery (He et al., 2025). However, these methods cannot effectively slow the progression of the disease. Recent studies have increasingly revealed that its pathogenesis involves complex biological processes, including persistent low-grade inflammation, extracellular matrix (ECM) metabolic imbalance, abnormal apoptosis of multiple cell types, and dysregulation of intercellular communication (Nedunchezhiyan et al., 2022; Sanchez-Lopez et al., 2022; Veronesi et al., 2023). Despite significant progress in cartilage biology (Hodgkinson et al., 2022), disease-modifying therapies for KOA remain scarce, prompting us to expand the research to other intra-articular tissues, particularly the synovium, to explore novel pathogenic mechanisms and therapeutic targets (Kloppenburg, 2024).

The synovium, as the lining tissue of the joint, maintains joint homeostasis by secreting synovial fluid. Its inflammatory response is one of the core pathological features of KOA, closely related to pain severity and disease progression (Sanchez-Lopez et al., 2022). Activated synovial fibroblasts are key effector cells mediating synovitis, directly participating in and exacerbating cartilage destruction by secreting large amounts of pro-inflammatory factors and proteases (Knights et al., 2023; Zhao et al., 2023). To elucidate the molecular regulatory network of the synovium in KOA, a key differentially expressed gene (DEG), RAB5A, was successfully screened through bioinformatics analysis of synovial transcriptome data from KOA patients and healthy controls. Data showed that RAB5A was significantly upregulated in KOA synovial tissue. RAB5A is an important member of the Ras superfamily of small GTPases and, as a major regulator of early endosomes, plays a dominant role in cellular memory transport, signal transduction, and cell membrane receptor turnover (Jin et al., 2021). It is noteworthy that RAB5A has been shown to play a crucial pathogenic role in various inflammatory and proliferative diseases, including rheumatoid arthritis (Ye et al., 2024) and cancer (Qiao et al., 2024), by regulating the signaling pathways of growth factor receptors (Zhang et al., 2022). Its upregulation in the synovium of patients with KOA strongly suggests that RAB5A may disrupt the normal endocytic balance of synovial cells, thereby driving pathological events in the KOA environment.

To further explore the potential function of RAB5A in KOA, gene set enrichment analysis was performed on differentially upregulated genes containing RAB5A. The results revealed that these genes were significantly enriched in the endocytosis pathway. This finding is highly consistent with the classical function of RAB5A and directs the focus to a key area derived from the endocytic system—extracellular vesicles (EVs) (Gorji-Bahri et al., 2021a, 2021b). Exosomes are nanoscale vesicles released extracellularly after the fusion of intracellular multivesicular bodies with the cell membrane, serving as important carriers of intercellular communication (Krylova and Feng, 2023). EVs or exosomes can carry “cargo” such as proteins, lipids, and nucleic acids and deliver them to recipient cells, thereby reshaping the phenotype of recipient cells (Nguyen et al., 2021). In the joint environment, synovial cell-derived exosomes have been shown to enter cartilage and regulate chondrocyte metabolism and survival (Liu et al., 2024). Therefore, our team hypothesizes that the upregulation of RAB5A may disrupt homeostasis by altering the biogenesis, secretion, or content loading of synovial fibroblast exosomes.

Based on the above analysis, a hypothesis is proposed. The expression of RAB5A in synovial fibroblasts is abnormally elevated, altering the number, composition, or biological characteristics of exosomes produced by the cells. These reprogrammed exosomes are subsequently taken up by chondrocytes, promoting the cartilage degeneration process in KOA by transmitting specific pathogenic signals. This study aims to verify this hypothesis through in vitro experiments, providing a new perspective for understanding cross-tissue dialogue in KOA and offering mechanistic insights into the regulatory role of RAB5A in KOA treatment.

Materials and Methods

Bioinformatics analysis and key gene screening

The datasets GSE39340, GSE82107, and GSE55235 used in this study were derived from the Gene Expression Omnibus (GEO) database, containing 17 normal and 27 human KOA samples. The downloaded data were in MINiML format. The Limma package (version 3.40.2) in R software (version 4.0.3) was used to study differential mRNA expression. The threshold for screening DEGs was defined as “adjusted P value <0.05 and |log2(Fold Change (FC))| > 1.” To further explore the functions of potential targets, functional enrichment analysis including Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) term was performed on the data. GO was used to annotate gene functions, including molecular function, biological process, and cellular composition. KEGG enrichment was used to analyze gene function and related higher-order genomic functional information. These functional enrichment results were obtained from the R package ClusterProfiler (version 3.18.0). Genes from the KEGG TOP1 up and down pathways were input into String, and the string-interations-short.tsv file was downloaded. The file was then input into Cytoscape, and the map was analyzed to obtain the degree. The data are sorted by degree, with larger circles indicating higher degrees.

Cell culture and transfection

Human fibroblast-like synoviocytes (HFLS; YS2325C, YaJi Biological, Shanghai, China) were maintained in OriCell^® Complete Medium (HXXFB-90011, Cyagen Biosciences, Suzhou) supplemented with 1% penicillin/streptomycin at 37°C under 5% CO₂. HFLS were transfected with short hairpin RNA (sh)RAB5A or negative control (sh-NC) (Genepharma) using Lipofectamine 3000 (Invitrogen) for 48 h in accordance with manufacturer’s instruction. To simulate an inflammatory environment, HFLS was treated with IL-1β. Human chondrocytes (CP-H107, Procell, Wuhan) were maintained in Complete Medium (CM-H107, Procell, Wuhan). They were both passaged at 80% confluence at ratio of 1:2.

EV isolation and identification

EVs were isolated from the supernatant of HFLS maintained in serum-free medium for 24 h. Briefly, the supernatant was centrifuged (3,000g for 15 min and 10,000g for 20 min, 4°C) to remove cellular debris. The resulting supernatant was then mixed thoroughly with an EV extraction reagent (Bestbio, Nanjing, China) and incubated at 4°C overnight. Following incubation, the mixture was centrifuged (10,000g, 60 min, 4°C). The pellet was collected and suspended in EV storage solution for subsequent analysis. For morphological assessment, they were fixed in 2.5% glutaraldehyde, dehydrated through a graded ethanol series, embedded in epoxy resin, and stained with uranyl acetate and lead citrate before being visualized under a transmission electron microscope (TEM, JEOL, Japan). For biomarker identification, EVs were lysed with Radio-Immunoprecipitation Assay (RIPA) buffer, and the presence of specific EV markers (CD9, CD63, TSG101) and the absence of the negative control calnexin were confirmed by Western blotting. The particle size distribution and concentration were determined using nanoparticle tracking analysis (NTA) on a ZetaView system (Particle Metrix, Germany). Finally, the total protein concentration of the EV preparation was quantified using a Bicinchoninic Acid (BCA) assay kit.

EV treatment of chondrocytes

For all chondrocyte functional assays, EVs were standardized by particle number (1 × 10⁹ particles/mL) (Forteza-Genestra et al., 2023) to ensure consistent EV dosing. The corresponding protein concentration was approximately 10 μg/mL. The supernatant remaining after EV isolation was retained as EV-depleted medium and applied to chondrocytes to assess the specificity of EV-mediated effects. For mechanistic rescue experiments, chondrocytes were treated with EVs derived from RAB5A-knockdown HFLS (shRAB5A-EVs) in the presence or absence of recombinant MMP3 (rMMP3, 25 ng/mL, TargetMol) (Ganguly et al., 2020). After 48 h of treatment, cells were harvested for subsequent analysis.

Quantitative real-time PCR

The relative mRNA levels of cytokines or ECM-related genes in HFLS or chondrocytes were quantified by quantitative real-time PCR (RT-qPCR). Briefly, total RNA was isolated using TRIpure reagent (ELK Biotechnology) and reverse-transcribed into cDNA with EntiLink™ 1st Strand cDNA Synthesis SuperMix (EQ031, ELK Biotechnology). PCR amplification was performed on a QuantStudio 6 system (Applied Biosystems) using EnTurbo™ SYBR Green PCR SuperMix (EQ001, ELK Biotechnology) under the following conditions: 95°C for 30 min; 40 cycles of 95°C for 10 s, 58°C for 30 s, and 72°C for 30 s. The 2^−ΔΔCT method was applied to calculate relative expression, using β-actin for normalization.

Western blotting

Protein extracts from HFLS were prepared using RIPA lysis buffer, and their concentrations were quantified with a BCA assay. After separation by sodium dodecyl sulfate - polyacrylamide gel electrophoresis, proteins were transferred to Polyvinylidene Difluoride membranes (Millipore). The membranes were blocked with 5% BSA for 1 h at room temperature and subsequently incubated overnight at 4°C with specific primary antibodies against RAB5A, p-p65, p65, p-p38, p38, and β-actin. Following Tris-buffered saline with Tween 20 washes, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies for 2 h at room temperature. Finally, protein bands were visualized using an ECL reagent (Aspen, Wuhan) and quantified by densitometry in Fiji software.

Enzyme-Linked Immunosorbent Assay (ELISA)

HFLS supernatant was collected after centrifugation (1,000g, 20 min, 4°C). Indicator detection was performed using interleukin (IL)-18 (YE18973), IL-6 (YE18970), MCP1 (YE23066), and MMP3 (YE28018, YUS Biotechnology) ELISA kits according to the instructions. The procedure included serial dilution of standards, incubation with biotinylated antibodies, horseradish peroxidase labeling, and chromogenic reaction termination. Absorbance was read at 450 nm using a microplate reader (Tecan).

Chondrocyte phagocytosis

To track the uptake of EVs by chondrocytes, the EVs were labeled with the fluorescent dye PKH26 (D0030, Solarbio, Beijing, China). Briefly, EVs were incubated with PKH26 working solution for 5 min in the dark. To remove unbound dye, the labeling reaction was stopped by adding phosphate-buffered saline (PBS), and the labeled EVs were reisolated using the exosome isolation reagent by incubation for 2 h, followed by centrifugation at 12,000g for 20 min at 4°C. The labeled EV pellet was resuspended and cocultured with chondrocytes. After 48 h, cells were washed with PBS to remove noninternalized EVs, and uptake was observed under a fluorescence microscope.

Cell counting kit-8 (CCK-8)

Chondrocytes were treated with EVs, and the viability was assessed using a CCK8 kit (40203ES76, Yeasen, Shanghai). Each well was added with 10 μL of CCK-8 solution, and they were incubated for 2 h. Absorbance was read at 450 nm using a microplate reader (Tecan).

Flow cytometry

Chondrocyte apoptosis was assessed with an Annexin V-FITC/PI detection kit (40302ES20, Yeasen, Shanghai) per the manufacturer’s instructions. Chondrocytes were washed with ice-cold PBS (300g, 5 min, 4°C) and suspended in 1× binding buffer. Following a 15-min incubation with 5 μL each of Annexin V-FITC and PI in the dark, the cells were washed, resuspended in fresh buffer, and immediately analyzed on a flow cytometer (BD Biosciences).

Statistics analysis

Data from three independent replicates are presented as mean ± SD. Statistical significance (P < 0.05) was determined using GraphPad Prism 8.0, with an unpaired two-tailed Student’s t-test for two-group comparisons and one- or two-way Analysis of Variance followed by Tukey’s test for multigroup analyses.

Results

DEGs in KOA and pathway enrichment analysis

Based on the dataset, DEGs were identified, and pathway enrichment analysis was performed. A volcano plot illustrates the fold change in gene expression and its statistical significance (Fig. 1A). Red dots indicate genes that simultaneously meet the FC and P value screening thresholds, meaning their expression changes are significant and substantial. Green dots indicate genes whose FC meets the threshold but whose P values are not statistically significant. Compared with the control group, 1,048 downregulated genes and 1,617 upregulated genes were identified in KOA. A heatmap further illustrates the expression patterns of DEGs (Fig. 1B), with samples arranged from the outside in by group, the outermost circle representing the control group, and the innermost circle representing the KOA group. Given the large number of DEGs, 50 upregulated genes and 50 downregulated genes with the most significant expression changes were selected for visualization. Functional enrichment analysis covered both KEGG pathways and GO terms. In the GO enrichment analysis (Fig. 1C), the figure shows the top 15 entries with the highest significance. The results showed that downregulated differentially regulated genes were mainly enriched in the ribonucleoprotein complex export from nucleus pathway, while upregulated genes were significantly enriched in the regulation of mRNA metabolic process pathway. In the KEGG pathway enrichment map (Fig. 1D), the intensity of the color indicates the significance of the enrichment, and the size of the circle represents the number of enriched genes. Downregulated genes were significantly enriched in the amyotrophic lateral sclerosis pathway, while upregulated genes were mainly enriched in the endocytosis pathway.

FIG. 1.

DEGs in KOA and pathway enrichment analysis. (A) A volcano plot illustrates the fold change in gene expression and its statistical significance. Each dot in the plot represents a gene. Red dots indicate genes that simultaneously meet the FC and P value screening thresholds. Green dots indicate genes whose FC meets the threshold but whose P values are not statistically significant. (B) A heatmap illustrates the expression patterns of DEGs, including 50 upregulated and 50 downregulated genes. (C) GO enrichment analysis. Different colors represent different types of GO terms, and the bar length indicates the number of genes. (D) KEGG pathway enrichment map. The intensity of the color indicates the significance of the enrichment (P value), and the size of the circle represents the number of enriched genes. DEGs, differentially expressed genes; GO, Gene Ontology; KOA, knee osteoarthritis.

Screening of DEGs and validation of cellular expression

To further screen key genes, two interaction networks of genes within the KEGG Top1 pathway were constructed based on the String database, and visualization analysis was performed using Cytoscape (Fig. 2A, B). Among the upregulated genes, RAB5A and CDC42 were identified as core genes. Considering that CDC42 has been extensively studied in macrophages and cytoskeleton, while the NDUF family, as a key downregulated gene, has been reported to be closely related to mitochondrial function, RAB5A was selected for subsequent functional validation. To simulate an inflammatory environment, HFLS was stimulated with IL-1β. RT-qPCR and Western blotting results showed that IL-1β stimulation significantly upregulated RAB5A expression (Fig. 2C, D). Subsequently, RAB5A expression was knocked down in HFLS using transfection technology, and both RT-qPCR and WB results confirmed the significant knockdown effect (Fig. 2E, F).

FIG. 2.

Screening of DEGs and validation of cellular expression. (A, B) Two interaction networks of genes within the KEGG Top1 pathway were constructed based on the String and Cytoscape. (C) RT-qPCR and (D) Western blotting show RAB5A expression in HFLS upon IL-1β stimulation. (E) RT-qPCR and (F) Western blotting confirm RAB5A knockdown in HFLS using transfection technology. *P < 0.01, ***P < 0.001. HFLS, human fibroblast-like synoviocytes.

RAB5A knockdown inhibits the activation of inflammatory signaling pathways

To investigate the role of RAB5A in the inflammatory response, the activation status of key inflammatory signaling pathways was examined. Western blotting results showed that IL-1β stimulation induced increased phosphorylation levels of p38 and NF-κB p65 in HFLS, while RAB5A knockdown significantly reduced the phosphorylation levels of both (Fig. 3A). Further RT-qPCR detection of intracellular inflammation-related factor expression revealed that IL-1β induction significantly increased the mRNA levels of IL-6, CXCL8, MMP3, and PTGS2, while RAB5A knockdown effectively reversed this trend (Fig. 3B). Furthermore, ELISA results showed that IL-1β stimulation significantly promoted the secretion of IL-6, CXCL8, MCP-1, and MMP3 in HFLS conditioned medium, while RAB5A knockdown significantly inhibited the release of these factors (Fig. 3C).

FIG. 3.

RAB5A knockdown inhibits the activation of inflammatory signaling pathways. (A) Western blotting shows the phosphorylation levels of p38 and NF-κB p65 in HFLS upon IL-1β stimulation. (B) RT-qPCR detection of intracellular inflammation-related factor IL-6, CXCL8, MMP3, and PTGS2. (C) ELISA shows the secretion of IL-6, CXCL8, MCP-1, and MMP3 in HFLS conditioned medium. *P < 0.05, **P < 0.01, ***P < 0.001.

RAB5A knockdown affects EV secretion and composition

Enrichment analysis suggests the importance of the endocytic pathway, and the role of RAB5A in the generation and function of EVs was further investigated. First, EVs were isolated from HFLS supernatant. Western blotting confirmed that the marker proteins CD9, CD63, and TSG101 were positive, while calnexin was negative, confirming the extract as EVs (Fig. 4A). TEM and NTA showed that RAB5A knockdown significantly reduced the concentration of EVs secreted by HFLS and slightly broadened the particle size distribution (Fig. 4B). Further Western blot analysis revealed that the relative abundance of several key matrix degradation-related proteins, including MMP1, MMP3, and ADAMTS5, was significantly reduced in EVs derived from HFLS with RAB5A knockdown (Fig. 4C), suggesting that RAB5A participates in modulating the composition of EVs and may affect their biological function.

FIG. 4.

RAB5A knockdown affects EV secretion and composition. (A) EVs were isolated from HFLS supernatant. Western blotting confirms the marker proteins CD9, CD63, TSG101, and negative calnexin. (B) TEM shows the particle images, and NTA shows their size and concentration secreted by HFLS. (C) The expression levels of matrix degradation-related proteins MMP1, MMP3, and ADAMTS5 in EVs. **P < 0.01, ***P < 0.001. EVs, extracellular vesicles; NTA, nanoparticle tracking analysis; TEM, transmission electron microscope.

HFLS-derived EVs affect chondrocyte ECM metabolism and apoptosis

To investigate the effects of HFLS-derived EVs on chondrocytes, PKH26 fluorescent labeling was used to verify that EVs could be effectively taken up by chondrocytes (Fig. 5A). Subsequently, RT-qPCR results showed that, compared with shCtrl-EV, shRAB5A-EV significantly reduced the expression of inflammatory genes IL-6 and PTGS2 in chondrocytes (Fig. 5B). Furthermore, shRAB5A-EV also caused a decrease in the expression of catabolic genes MMP13 and ADAMTS5, while promoting an increase in the expression of anabolic genes ACAN and COL2A1 (Fig. 5C). CCK-8 assay showed that shCtrl-EV significantly inhibited chondrocyte viability, while shRAB5A-EV alleviated this inhibitory effect (Fig. 5D). Meanwhile, flow cytometry showed that shCtrl-EV significantly promoted chondrocyte apoptosis, while shRAB5A-EV did not cause the same degree of increased apoptosis (Fig. 5E). To validate EV dependency and cargo specificity, chondrocytes were treated with EV-depleted conditioned medium and rMMP3. EV-depleted medium did not significantly alter the expression of MMP13 or COL2A1 compared with the ctrl, confirming that the observed effects are EV-dependent. Compared with shCtrl-EV, shRAB5A-EV significantly decreased MMP13 and increased COL2A1 expression, an effect that was partially reversed by supplementation with rMMP3. These findings indicate that MMP3 is a functionally relevant cargo mediating the chondrocyte-protective effects of shRAB5A-EV (Fig. 5F).

FIG. 5.

HFLS-derived EVs affect chondrocyte extracellular matrix metabolism and apoptosis. (A) PKH26-labeled EVs were taken up by chondrocytes. (B) RT-qPCR shows the expression of inflammatory genes IL-6 and PTGS2 in chondrocytes. (C) RT-qPCR detection of catabolic genes MMP13 and ADAMTS5, and anabolic genes ACAN and COL2A1. (D) CCK-8 assay shows chondrocyte viability. (E) Flow cytometry shows chondrocyte apoptosis. (F) qPCR analysis of MMP13 and COL2A1 expression in chondrocytes treated with EV-depleted medium, shNC-EVs, shRAB5A-EVs, or shRAB5A-EVs supplemented with recombinant MMP3. *P < 0.05, **P < 0.01, ***P < 0.001. CCK-8, cell counting kit-8.

Discussion

This study first identified a significant upregulation of RAB5A in the synovial tissue of KOA using bioinformatics analysis. To simulate an inflammatory environment in vitro and validate this finding, HFLS were treated with IL-1β. RAB5A expression was confirmed to be upregulated by inflammatory signals. This result links bioinformatics prediction with experimental validation, establishing the correlation between RAB5A and KOA pathology. More importantly, loss-of-function experiments showed that RAB5A knockdown significantly inhibited the activation of IL-1β-triggered NF-κB and p38 Mitogen-Activated Protein Kinase (MAPK) signaling pathways, specifically manifested as a decrease in the phosphorylation levels of p65 and p38. Downstream, the expression of a series of key pro-inflammatory factors IL-6, CXCL8, MCP-1, matrix degradation mediator MMP3, and inflammatory enzymes PTGS2 also decreased. Previous studies have shown that p-p38 and NF-κB are upregulated in chondrocytes (Jeon et al., 2024; Ji et al., 2019); this study is the first to reveal the regulation of these pathways by RAB5A in the context of KOA. This finding establishes that RAB5A drives the inflammatory response by activating multiple signaling axes.

Given the importance of the endocytic pathway suggested by gene enrichment analysis, we further investigated the function of RAB5A at the EV level. Notably, knockdown of RAB5A significantly reduced the concentration of EVs secreted by HFLS, and the particle size distribution slightly widened. This suggests that RAB5A not only affects EV production but may also interfere with the uniformity of its generation. Given that RAB5A is a key regulator of endosome sorting (Gorji-Bahri et al., 2021a; Yin et al., 2024), its absence likely disrupts the normal formation, maturation, or fusion of multivesicular bodies as EV precursors with the cell membrane (Das et al., 2018; Schneider et al., 2022). Furthermore, the levels of MMPs and ADAMTS5 were found to be reduced in EVs from RAB5A-knockdown cells. This aligns with previous findings from the KOA study (Yin et al., 2022), which indicates that EVs derived from the synovium are not inert carriers but actively participate in cartilage destruction. Multiple studies (Mustonen and Nieminen, 2021; Yang et al., 2023) have confirmed that inflammatory synovial cells secrete EVs rich in pathogenic miRNAs and proteins, thereby promoting cartilage degradation. Our research builds upon this foundation by identifying RAB5A for the first time as a molecular switch capable of “reprogramming” the pathogenic contents of synovial EVs.

To directly verify the effects of synovial EVs on chondrocytes, we conducted subsequent experiments. The result confirmed that chondrocytes could effectively take up fluorescently labeled EVs from HFLS. A key finding was that, compared with the control group, chondrocytes that took up HFLS-derived EVs with RAB5A knockdown exhibited significantly reduced inflammatory responses and catabolism levels. Specifically, the expression of pro-inflammatory genes and matrix degradation genes MMP13 and ADAMTS5 was downregulated, while the expression of anabolic genes ACAN and COL2A1 was upregulated. Simultaneously, chondrocyte viability was increased, and apoptosis levels decreased. ECM is fundamental to maintaining cartilage structure and function. MMP13 and ADAMTS5 are core enzymes for degrading type II collagen and proteoglycans (Yang et al., 2017; Zhang et al., 2024; Zou et al., 2025), respectively, and their overexpression is a direct marker of cartilage destruction. COL2A1 is a major component of type II collagen, and ACAN is a major proteoglycan in the cartilage matrix; its reduction directly leads to a decrease in cartilage elasticity and compressive strength (Kuppa et al., 2023; Xian Bo et al., 2022). Previous studies have shown that synovitis drives cartilage degeneration (Liu et al., 2024; Mathiessen and Conaghan, 2017), but this study is novel in that it reveals a specific molecular pathway mediated by RAB5A and dependent on EVs. We propose for the first time that HFLS in KOA produce and release EVs through high expression of RAB5A. These EVs precisely deliver destructive signals to chondrocytes, disrupting their metabolic homeostasis and leading to their degradation and death. However, this study also has limitations. First, while in vitro model reveals a direct link, it cannot fully simulate the complex joint microenvironment in vivo; future in vivo validation in animal models is needed. Second, although MMP3 was identified as a functionally relevant cargo via add-back experiments, the complete repertoire of RAB5A-regulated pathogenic molecules within EVs remains to be fully elucidated through omics analysis. Third, additional approaches such as dose–response analysis and biochemical treatments (e.g., detergent/proteinase K with RNase) could provide further quantitative and mechanistic insights. These remain important directions for future investigation.

In summary, this study elucidates a novel mechanism of RAB5A in KOA pathogenesis. Inflammatory signals induce upregulation of RAB5A expression in HFLS, which in turn exacerbates local inflammation by activating the NF-κB/p38 pathway. Furthermore, by regulating EVs biogenesis and the loading of pathogenic contents, destructive signals are delivered to chondrocytes, ultimately leading to ECM degradation and chondrocyte apoptosis. This study establishes RAB5A as a critical regulator of HFLS-EV function in KOA, providing a mechanistic basis for understanding chondrocyte degeneration. Future research using in vivo models is warranted to validate the translational relevance of this mechanism and to explore its therapeutic potential.

Authors’ Contributions

Z.Z. and Y.W. contributed to the concept, design, investigation, and writing—draft. X.T., Z.L., and X.Y. contributed to the investigation, analysis, and verification. Z.G. contributed to the concept, design, investigation, and writing—revision. All the authors approved the final version of article.

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Footnotes

Author Disclosure Statement

The authors have no relevant financial or nonfinancial interests to disclose.

Funding Information

This study was supported by Medical and Health Science Program of Zhejiang Province (No. 2023KY1232).

References

Das

, Prasad

, Roy

, et al. The protease activated receptor2 promotes rab5a mediated generation of pro-metastatic microvesicles. Sci Rep 2018;8(1):7357; doi: 10.1038/s41598-018-25725-w

Forteza-Genestra

, Antich-Rosselló

, Ramis-Munar

, et al. Comparative effect of platelet- and mesenchymal stromal cell-derived extracellular vesicles on human cartilage explants using an ex vivo inflammatory osteoarthritis model. Bone Joint Res 2023;12(10):667–676; doi: 10.1302/2046-3758.1210.Bjr-2023-0109.R1

Ganguly

, Hostetter

, Tang

, et al. Notch3 promotes prostate cancer-induced bone lesion development via MMP-3. Oncogene 2020;39(1):204–218; doi: 10.1038/s41388-019-0977-1

Gelber

. Knee osteoarthritis. Ann Intern Med 2024;177(9):Itc129–Iitc144; doi: 10.7326/annals-24-01249

Gorji-Bahri

, Moghimi

, Hashemi

. RAB5A effect on metastasis of hepatocellular carcinoma cell line via altering the pro-invasive content of exosomes. Exp Mol Pathol 2021a;120:104632; doi: 10.1016/j.yexmp.2021.104632

Gorji-Bahri

, Moghimi

, Hashemi

. RAB5A is associated with genes involved in exosome secretion: Integration of bioinformatics analysis and experimental validation. J Cell Biochem 2021b;122(3–4):425–441; doi: 10.1002/jcb.29871

, Li

, Huang

, et al. Pain phenotype in knee osteoarthritis: Implications for mechanism-based therapy. Orthop Surg 2025;17(11):3007–3021; doi: 10.1111/os.70161

Hodgkinson

, Kelly

, Curtin

, et al. Mechanosignalling in cartilage: An emerging target for the treatment of osteoarthritis. Nat Rev Rheumatol 2022;18(2):67–84; doi: 10.1038/s41584-021-00724-w

Jeon

, Kim

, Kwon

, et al. Cannabiorcol as a novel inhibitor of the p38/MSK-1/NF-κB signaling pathway, reducing matrix metalloproteinases in osteoarthritis. Phytomedicine 2024;135:156141; doi: 10.1016/j.phymed.2024.156141

10.

, Ma

, Wang

, et al. Activation of the P38/CREB/MMP13 axis is associated with osteoarthritis. Drug Des Devel Ther 2019;13:2195–2204; doi: 10.2147/dddt.S209626

11.

Jin

, Wen

, Wu

, et al. NOTCH-induced rerouting of endosomal trafficking disables regulatory T cells in vasculitis. J Clin Invest 2021;131(1):e136042; doi: 10.1172/jci136042

12.

Kloppenburg

. Synovial inflammation in osteoarthritis. A treatable target? Semin Arthritis Rheum 2024;64s:152326; doi: 10.1016/j.semarthrit.2023.152326

13.

Knights

, Farrell

, Ellis

, et al. Synovial fibroblasts assume distinct functional identities and secrete R-spondin 2 in osteoarthritis. Ann Rheum Dis 2023;82(2):272–282; doi: 10.1136/ard-2022-222773

14.

Krylova

, Feng

. The machinery of exosomes: Biogenesis, release, and uptake. Int J Mol Sci 2023;24(2):1337; doi: 10.3390/ijms24021337

15.

Kuppa

, Kim

, Kang

, et al. Polynucleotides suppress inflammation and stimulate matrix synthesis in an in vitro cell-based osteoarthritis model. Int J Mol Sci 2023;24(15):12282; doi: 10.3390/ijms241512282

16.

Liu

, Xian

, Chen

, et al. Inflammatory fibroblast-like synoviocyte-derived exosomes aggravate osteoarthritis via enhancing macrophage glycolysis. Adv Sci (Weinh) 2024;11(14):e2307338; doi: 10.1002/advs.202307338

17.

Mathiessen

, Conaghan

. Synovitis in osteoarthritis: Current understanding with therapeutic implications. Arthritis Res Ther 2017;19(1):18; doi: 10.1186/s13075-017-1229-9

18.

Mustonen

, Nieminen

. Extracellular vesicles and their potential significance in the pathogenesis and treatment of osteoarthritis. Pharmaceuticals (Basel) 2021;14(4):315; doi: 10.3390/ph14040315

19.

Nedunchezhiyan

, Varughese

, Sun

, et al. Obesity, inflammation, and immune system in osteoarthritis. Front Immunol 2022;13:907750; doi: 10.3389/fimmu.2022.907750

20.

Nguyen

, Duong

, Nguyen

, et al. Mesenchymal stem cell-derived extracellular vesicles for osteoarthritis treatment: Extracellular matrix protection, chondrocyte and osteocyte physiology, pain and inflammation management. Cells 2021;10(11):2887; doi: 10.3390/cells10112887

21.

Qiao

, Dong

, Jia

, et al. RAB5A in triple-negative breast cancer: A critical role in macrophage reshaping in an exosomal miR-21-dependent manner. Endocr Relat Cancer 2024;31(5):e230257; doi: 10.1530/erc-23-0257

22.

Sanchez-Lopez

, Coras

, Torres

, et al. Synovial inflammation in osteoarthritis progression. Nat Rev Rheumatol 2022;18(5):258–275; doi: 10.1038/s41584-022-00749-9

23.

Schneider

, Farr

, Pinter

, et al. The Nma1 protein promotes long distance transport mediated by early endosomes in Ustilago maydis. Mol Microbiol 2022;117(2):334–352; doi: 10.1111/mmi.14851

24.

Veronesi

, Costa

, Bellavia

, et al. Epigenetic modifications of MiRNAs in osteoarthritis: A systematic review on their methylation levels and effects on chondrocytes, extracellular matrix and joint inflammation. Cells 2023;12(14):1821; doi: 10.3390/cells12141821

25.

Xian Bo

, Yan Jie

, De Chao

, et al. An inducible nitric oxide synthase dimerization inhibitor prevents the progression of osteoarthritis. Front Pharmacol 2022;13:861183; doi: 10.3389/fphar.2022.861183

26.

Yang

, Li

, Fu

, et al. Extracellular vesicles in osteoarthritis of peripheral joint and temporomandibular joint. Front Endocrinol (Lausanne) 2023;14:1158744; doi: 10.3389/fendo.2023.1158744

27.

Yang

, Chanalaris

, Troeberg

. ADAMTS and ADAM metalloproteinases in osteoarthritis—looking beyond the ‘usual suspects’. Osteoarthritis Cartilage 2017;25(7):1000–1009; doi: 10.1016/j.joca.2017.02.791

28.

, Yang

, Guo

, et al. Internalisation of neutrophils extracellular traps by macrophages aggravate rheumatoid arthritis via Rab5a. RMD Open 2024;10(1):e003847; doi: 10.1136/rmdopen-2023-003847

29.

Yin

, Ni

, Witherel

, et al. Harnessing tissue-derived extracellular vesicles for osteoarthritis theranostics. Theranostics 2022;12(1):207–231; doi: 10.7150/thno.62708

30.

Yin

, Ma

, Tian

, et al. Downregulation of the m(6)A reader YTHDC2 upregulates exosome content in lung adenocarcinoma via inhibiting IFIT and OAS family members. J Biol Chem 2024;300(10):107783; doi: 10.1016/j.jbc.2024.107783

31.

Yunus

MHM

, Nordin

, Kamal

. Pathophysiological perspective of osteoarthritis. Medicina (Kaunas) 2020;56(11):614; doi: 10.3390/medicina56110614

32.

Zhang

, Yan

, Wang

, et al. Surface functionalization of exosomes for chondrocyte-targeted siRNA delivery and cartilage regeneration. J Control Release 2024;369:493–505; doi: 10.1016/j.jconrel.2024.04.009

33.

Zhang

, Chen

, Ling

, et al. EGFR tyrosine kinase activity and Rab GTPases coordinate EGFR trafficking to regulate macrophage activation in sepsis. Cell Death Dis 2022;13(11):934; doi: 10.1038/s41419-022-05370-y

34.

Zhao

, Ruan

, Nie

, et al. Effects of synovial macrophages in osteoarthritis. Front Immunol 2023;14:1164137; doi: 10.3389/fimmu.2023.1164137

35.

Zhu

, Qu

, He

. Evaluation and management of knee osteoarthritis. J Evid Based Med 2024;17(3):675–687; doi: 10.1111/jebm.12627

36.

Zou

, Pan

, Sun

, et al. AM1241 inhibits chondrocyte inflammation and ECM degradation through the Nrf2/HO-1 and NF-κB pathways and alleviates osteoarthritis in mice. Mol Med 2025;31(1):9; doi: 10.1186/s10020-024-01012-5