Investigating Lipid and Energy Dyshomeostasis Induced by Per- and Polyfluoroalkyl Substances (PFAS) Congeners in Mouse Model Using Systems Biology Approaches

Background: Exposure to per- and polyfluoroalkyl substances (PFAS, including 7H-Perfluoro-4-methyl-3,6-dioxaoctanesulfonic acid (PFESA-BP2), perfluorooctanoic acid (PFOA), and hexafluoropropylene oxide (GenX), has been associated with liver dysfunction. While previous research has characterized PFAS-induced hepatic lipid alterations, their downstream effects on energy metabolism remain unclear. This study investigates metabolic alterations in the liver following PFAS exposure to identify mechanisms leading to hepatoxicity. Methods: We analyzed RNA sequencing datasets of mouse liver tissues exposed to PFAS to identify metabolic pathways influenced by the chemical toxicant. We integrated the transcriptome data with a mouse genome-scale metabolic model to perform in silico flux analysis and investigated reactions and genes associated with lipid and energy metabolism. Results: PFESA-BP2 exposure caused dose- and sex-dependent changes, including upregulation of fatty acid metabolism, β-oxidation, and cholesterol biosynthesis. On the contrary, triglycerides, sphingolipids, and glycerophospholipids metabolism were suppressed. Simulations from the integrated genome-scale metabolic models confirmed increased flux for mevalonate and lanosterol metabolism, supporting potential cholesterol accumulation. GenX and PFOA triggered strong PPARα-dependent responses, especially in β-oxidation and lipolysis, which were attenuated in PPARα−/− mice. Mitochondrial fatty acid transport and acylcarnitine turnover were also disrupted, suggesting impaired mitochondrial dysfunction. Additional PFAS effects included perturbations in the tricarboxylic acid (TCA) cycle, oxidative phosphorylation, and blood–brain barrier (BBB) function, pointing to broader systemic toxicity. Conclusions: Our findings highlight key metabolic signatures and suggest PFAS-mediated disruption of hepatic and possibly neurological functions. This study underscores the utility of genome-scale metabolic modeling as a powerful tool to interpret transcriptomic data and predict systemic metabolic outcomes of toxicant exposure.