Exploring Toxicity of Per- and Polyfluoroalkyl Substances (PFAS) Mixture Through ADMET and Toxicogenomic In Silico Analysis: Molecular Insights
Katarina Baralić, Teodora Petkovski, Nađa Piletić, Đurđica Marić, Aleksandra Buha Djordjević, Biljana Antonijević, Danijela Đukić-Ćosić
Abstract
This study aimed to explore the health impacts, mechanisms of toxicity, and key gene biomarkers of a mixture of the most prominent perfluoroalkyl/polyfluoroalkyl substances (PFAS) through in silico ADMET and toxicogenomic analysis. The following databases and tools were used: AdmetSAR (2.0), ADMETlab (2.0), Comparative Toxicogenomic Database, ToppGene Suite portal, Metascape (3.5), GeneMANIA server, and CytoHubba and CytoNCA Cytoscape (3.10.3) plug-ins. ADMET analysis showed that PFAS compounds pose risks of organ-specific toxicity, prolonged retention, and metabolic disruptions. Forty mutual genes were identified for all the tested PFAS. The mutual gene set was linked to disruption of lipid metabolism, particularly through nuclear receptors. The most important gene clusters identified were nuclear receptor signaling and PPAR signaling pathways, with kidney and liver diseases, diabetes, and obesity as the most significant related diseases. Phenotype data showed that PFAS compounds impact cell death, growth, inflammation, steroid biosynthesis, and thyroid hormone metabolism. Gene network analysis revealed that 52% of the 40 mutual genes showed co-expression, with co-localization as the next major interaction (18.23%). Eight key genes were extracted from the network: EHHADH, APOA2, MBL2, SULT2A1, FABP1, PPARA, PCK2, and PLIN2. These results highlight the need for further research to fully understand the health risks of PFAS mixtures.