Network toxicology and immune-metabolic dysregulation: linking per- and polyfluoroalkyl substances exposure to osteoarthritis pathogenesis

Abstract Per- and polyfluoroalkyl substances (PFAS) are emerging environmental contaminants linked to various health conditions. However, the molecular mechanisms by which PFAS contribute to OA remain unclear. This study integrates network toxicology and bioinformatics to explore PFAS-related toxicity targets and their roles in OA pathogenesis. Transcriptomic data from the GSE48556 dataset were analyzed to identify differentially expressed genes (DEGs). PFAS-related genes (PSRGs) were retrieved from the CTD. Cross-analysis revealed overlapping genes, which were further evaluated via protein–protein interaction (PPI) networks, pathway enrichment, immune infiltration analysis, and nomogram construction. A total of 1,703 DEGs (910 upregulated, 793 downregulated) were identified in OA. Cross-analysis with 346 PSRGs yielded 26 overlapping genes, highlighting PFAS-OA molecular links. Enrichment analysis implicated IL-17 signaling, Th1/Th2 differentiation, and fatty acid metabolism as key pathways disrupted by PFAS. Immune-inflammatory pathways were robustly enriched, with CD3E, CARD11, and IFNG driving synovial inflammation. A nomogram incorporating five core targets (CARD11, IFNG, PAX8, PLD1, ZNF609) predicted OA risk and demonstrated clinical utility via decision curve analysis. Immune profiling revealed elevated infiltration of T cells, Th1 cells, and NK CD56dim cells in OA, alongside upregulated antigen presentation and TCR/BCR signaling. Core PFAS-related targets correlated significantly with immune dysregulation. PFAS exposure exacerbates OA by dysregulating immune-inflammatory axes and metabolic pathways, promoting synovitis and cartilage degradation. The identified genetic targets and nomogram provide mechanistic insights and translational tools for OA risk prediction in PFAS-exposed populations. This study establishes a systems-level framework linking PFAS toxicity to OA progression, offering actionable targets for therapeutic intervention.