Abstract
With an increasing number of health-related impacts of per- and polyfluoroalkyl substances (PFAS) being reported, there is a pressing need to understand PFAS transport within both the human body and the environment. As proteins can serve as a primary transport mechanism for PFAS, understanding PFAS binding to proteins is essential for predictive physiological models where accurate values of protein binding constants are vital. In this work we present a critical analysis of three common models for analyzing PFAS binding to bovine serum albumin (BSA) based on fluorescence quenching: the Stern-Volmer model, the modified Stern-Volmer model, and the Hill equation. The PFAS examined include perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), perfluorobutanesulfonic acid (PFBS), perfluorohexanesulfonic acid (PFHxS), perfluorooctanesulfonic acid (PFOS), and the replacement compound 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)propanoate (HFPO-DA or GenX). While all three models capture the general effects of hydrophobicity and steric limitations to PFAS binding, the Hill equation highlighted a unique relationship between binding cooperativity and the number of fluorinated carbons, with PFOA exhibiting the greatest binding cooperativity. The significance of steric limitations was confirmed by comparing results obtained by fluorescence quenching, which is an indirect method based on specific binding, to those obtained by equilibrium dialysis where PFAS binding directly correlated with traditional measures of hydrophobicity. Finally, the binding constants were correlated with PFAS physicochemical properties where van der Waals volume best described the steric limitations observed by fluorescence quenching.