Experimental characterization of the mechanism of perfluorocarboxylic acids' liver protein bioaccumulation: The key role of the neutral species

Perfluorocarboxylic acids (PFCAs) of chain length greater than seven carbon atoms bioconcentrate in the livers of fish. However, a mechanistic cause for the empirically observed increase in the bioconcentration potential of PFCAs as a function of chain length has yet to be determined. To this end, recombinant rat liver fatty acid‐binding protein (L‐FABP) was purified, and its interaction with PFCAs was characterized in an aqueous system at pH 7.4. Relative binding affinities of L‐FABP with PFCAs of carbon chain lengths of five to nine were established fluorimetrically. The energetics, mechanism, and stoichiometry of the interaction of perfluorooctanoic acid (PFOA) with L‐FABP were examined further by isothermal titration calorimetry (ITC) and electrospray ionization combined with tandem mass spectrometry (ESI‐MS/MS). Perfluorooctanoic acid was shown to bind to L‐FABP with an affinity approximately an order of magnitude less than the natural ligand, oleic acid, and to have at least 3:1 PFOA:L‐FABP stoichiometry. Two distinct modes of PFOA binding to L‐FABP were observed by ESI‐MS/MS analysis; in both cases, PFOA binds solely as the neutral species under typical physiological pH and aqueous concentrations of the anion. A comparison of their chemical and physical properties with other well‐studied biologically relevant chemicals showed that accumulation of PFCAs in proteins as the neutral species is predictable. For example, the interaction of PFOA with L‐FABP is almost identical to that of the acidic ionizing drugs ketolac, ibuprofen, and warfarin that show specificity to protein partitioning with a magnitude that is proportional to the KOW (octanol–water partitioning) of the neutral species. The experimental results suggest that routine pharmacochemical models may be applicable to predicting the protein‐based bioaccumulation of long‐chain PFCAs. Environ. Toxicol. Chem. 2010; 29:1669–1677. © 2010 SETAC