Per- and polyfluoroalkyl substances (PFAS) are persistent environmental pollutants commonly involved in multiple health risks. Epidemiological studies have confirmed a direct connection between PFAS exposure(including PFOA and PFOS) with the prevalence of osteoarthritis(OA), but the underlying mechanisms and key targets remain unclear. This study integrates network toxicology, transcriptomic analysis, machine learning, and experimental validation to systematically investigate the role of PFOA/PFOS in OA pathogenesis. We identified 38 common targets shared between PFOA/PFOS exposure and OA, enriched in oxidative stress, endocrine metabolism, and senescence pathways. We further screened seven diagnostic biomarkers using machine learning to construct a nomogram for predicting the early onset of OA. Among these, ESR1 emerged as the core target associated with PFOA/PFOS exposure-related OA through multi-algorithm integration. Moreover, molecular docking and molecular dynamics simulations established that PFOA and PFOS could directly bind to ESR1 in a stable manner. In vitro experiments and single-cell transcriptomic analysis indicated that PFOA and PFOS exacerbate oxidative stress and senescence by downregulating ESR1 expression on chondrocytes, resulting in cartilage matrix degradation. Importantly, the ESR1 agonist raloxifene effectively reversed these detrimental effects to maintain cartilage homeostasis. Overall, these findings revealed ESR1 dysfunction as a critical molecular event in PFAS exposure-related OA, providing novel insights into environmental etiologies of OA and potential therapeutic strategies.
