Objective To investigate the regulatory role and molecular mechanisms of TSPAN9-mediated mitocytosis in an interleukin-1β (IL-1β)-induced rat chondrocyte senescence model, and to identify novel therapeutic targets for osteoarthritis (OA). MethodsPrimary knee articular chondrocytes were isolated from the cartilage of knee joints harvested from 1-week-old male Sprague-Dawley rats and maintained in culture. Cell viability was assessed using the cell counting kit 8 (CCK-8) assay, and cell-cycle distribution was analyzed by flow cytometry to determine the optimal concentration and exposure time of IL-1β for model induction, thereby establishing an in vitro chondrocyte senescence model [early-OA (E-OA), middle-OA (M-OA), late-OA (L-OA)] and grouped. Cellular senescence was evaluated by senescence-associated β-galactosidase (SA-β-gal) staining. Real-time quantitative PCR (qRT-PCR) was used to quantify the mRNA expression levels of senescence markers [cyclin-dependent kinase inhibitor 2A (Cdkn2a) and Cdkn1a], extracellular matrix (ECM) catabolic-anabolic genes [collagen type Ⅱ alpha 1 chain (Col2a1), Aggrecan (Acan), matrix metalloproteinase 13 (MMP-13), and a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5)], and key mitocytosis-related genes [kinesin family member 5B (KIF5B), TSPAN9, and TSPAN4]. Mitochondrial ultrastructure and mitocytosis-related morphological features were examined by transmission electron microscopy, and changes in mitochondrial membrane potential were assessed using the JC-1 fluorescent probe. To activate TSPAN9 expression, chondrocytes were transduced with a TSPAN9-overexpressing lentivirus (LV-TSPAN9). The experiments included four groups: control group, M-OA group, lentivirus negative control group, and LV-TSPAN9 group. After confirming overexpression efficiency, differences in cellular senescence, mitochondrial homeostasis, and key ECM metabolic indices were compared among the groups. Results Using CCK-8 assays and cell-cycle analysis, the staged rat chondrocyte senescence model induced by IL-1β (5 ng/mL), and defined 12, 24, and 48 hours as the E-OA, M-OA, and L-OA senescence phenotypes were successfully established, respectively. Model validation demonstrated that, with prolonged induction, the proportion of SA-β-gal-positive cells increased markedly, mitochondrial membrane potential progressively declined (P<0.05), and mitochondrial ultrastructural damage became increasingly severe. qRT-PCR showed progressive upregulation of the senescence markers Cdkn2a and Cdkn1a, as well as the ECM catabolic genes MMP-13 and ADAMTS5 (P<0.05), together with sustained downregulation of the ECM anabolic genes Col2a1 and Acan (P<0.05). Notably, the key mitocytosis-related genes KIF5B, TSPAN4, and TSPAN9 displayed a biphasic pattern, with early compensatory upregulation followed by decompensatory decline at the middle-to-late stages. In the M-OA model, TSPAN9 overexpression markedly reversed these pathological changes, restored mitochondrial membrane potential (P<0.05), ameliorated the senescent phenotype and ECM metabolic imbalance, and significantly upregulated the expression of mitocytosis-related genes and proteins (P<0.05), with transmission electron microscope revealing morphological structures suggestive of mitocytosis-related events. Conclusion Impaired mitocytosis is an important pathological mechanism underlying IL-1β-induced chondrocyte senescence, and TSPAN9 overexpression may delay chondrocyte senescence by promoting mitocytosis in concert with KIF5B, thereby facilitating the clearance of damaged mitochondria and restoring intracellular homeostasis.