Systematic Exploration of Potential Toxicity Targets and Molecular Mechanisms of Emerging Short-Chain PFAS Substitutes: PFBA- and PFBS-Induced Hepatocellular Carcinoma Based on Toxicity Network Analysis, Machine Learning, and Biomimetic Calculations.

Abstract

Perfluorobutanoic acid (PFBA) and perfluorobutanesulfonic acid (PFBS) are short-chain alternatives to traditional perfluoroalkyl and polyfluoroalkyl substances (PFASs). Long-term exposure to these pollutants is closely associated with hepatocellular carcinoma (HCC). However, the toxic targets and mechanisms underlying PFBA- and PFBS-induced HCC remain unclear. To address this knowledge gap, this study employed a multifaceted approach encompassing network toxicology, molecular docking, and molecular dynamic simulation. Thirty-six core targets associated with PFBA- and PFBS-induced HCC were identified, and 12 key genes were initially screened through network toxicity analysis. Subsequently, based on the TCGA and ICGC datasets, three classical algorithms were applied to screen key genes: PPARG, ESR1, and ALB. Further exploration of the HCC-related dataset from the GEO database identified six critical genes: PPARG, ESR1, CD36, ABCA1, ACACA, and ALB. Survival analysis and ROC analysis based on the TCGA dataset revealed and validated the strong association between the expression levels of key genes (PPARG, ESR1, and ACACA). Single-gene GSEA showed that these three key genes may induce HCC through multiple biological pathways via interfering with the normal growth and development of hepatocytes and promoting inflammation and cell proliferation. Ultimately, molecular dynamics demonstrated the strong binding affinities between PFBA, PFBS, and the three protein receptors, with the best stability and flexibility of the interaction between PFBS and PPARG. These findings provide insights into the theoretical foundation for applying network toxicology, molecular docking, and molecular dynamic simulations in environmental pollutant research.