Munoz, G; Labadie, P; Geneste, E; Pardon, P; Tartu, S; Chastel, O; Budzinski, H
We report on a fast, accurate and rugged analytical procedure to determine a wide span of perfluoroalkyl and polyfluoroalkyl substances (PFASs) in seabird plasma. The 26 investigated compounds included perfluoroalkyl carboxylates (C5-C14 PFCAs), perfluoroalkyl sulfonates (C4, C6, C7, C8, C10 PFSAs), perfluorooctane sulfonamide (FOSA) and N-alkyl derivatives (MeFOSA, EtFOSA), N-alkyl perfluorooctane sulfonamido acetic acids (MeFOSAA, EtFOSAA), fluorotelomer sulfonates (4:2 FTSA, 6:2 FTSA, 8:2 FTSA), polyfluoroalkyl phosphate diesters (diPAPs) and perfluorooctane sulfonamide phosphate diester (diSAmPAP). The method described herein requires a reduced sample amount (25μL) and involves rapid and simple sample preparation (protein precipitation with acetonitrile but without acidification) prior to analysis by on-line solid phase extraction (Oasis HLB sorbent) coupled to high performance liquid chromatography negative electrospray ionization tandem mass spectrometry. The optimization was conducted using experimental designs to account for potential interactions between variables. Out of the 26 target analytes, 23 compounds showed excellent accuracy (±25% of the expected values). Intermediate precision and matrix effects remained acceptable for most analytes thanks to efficient internal standardization. A human serum standard reference material (NIST SRM 1957) was included in the validation scheme to evaluate method trueness, which proved satisfactory (│Z-scores│<2 for most compounds). Notwithstanding the small initial sample intake, limits of detection as low as 0.003-0.1ngg(-1) plasma were obtained. This allowed the determination of 11 target PFASs in Antarctic seabird plasma samples. ΣPFASs in Antarctic seabird plasma ranged from 0.37 to 19ngg(-1), with a predominance of PFOS (>54% of ΣPFASs on average). The reduced plasma amount required implies that the present method could also be applied to the analysis of PFASs in the plasma of smaller biological models.
