Environ. Sci. Technol. 2005, 39, 7439-7445
Perfluorinated Alkyl Substances in Plasma, Liver, Brain, and Eggs of Glaucous Gulls (Larus hyperboreus) from the Norwegian Arctic J O N A T H A N V E R R E A U L T , * ,†,‡ M A G A L I H O U D E , §,| GEIR W. GABRIELSEN,† URS BERGER,⊥ M A R I A N N E H A U K Å S , †,‡ ROBERT J. LETCHER,# AND DEREK C. G. MUIR| Norwegian Polar Institute, Tromsø, NO-9296, Norway, Department of Aquatic BioSciences, University of Tromsø, Tromsø, NO-9037, Norway, Department of Environmental Biology, University of Guelph, Guelph, Ontario N1G 2W1, Canada, National Water Research Institute, Environment Canada, Burlington, Ontario L7R 4A6, Canada, Norwegian Institute for Air Research, Tromsø, NO-9296, Norway, and Canadian Wildlife Service, National Wildlife Research Centre, Environment Canada, Ottawa, Ontario K1A 0H3, Canada
Recent environmental surveys have ascertained the widespread occurrence of perfluorinated alkyl substances (PFAS) in tissues of wildlife from the Arctic. In the present study, we investigated the distribution of a suite of PFAS in plasma, liver, brain, and egg samples from adult glaucous gulls (Larus hyperboreus), an apex scavengerpredator seabird breeding in the Norwegian Arctic. Perfluorooctane sulfonate (PFOS) was the predominant PFAS in all samples and was present at concentrations that are the highest reported thus far in any arctic seabird species and populations. Among the body compartment/ tissue samples analyzed, PFOS was highest in plasma (48.1349 ng/g wet weight (ww)), followed by liver ≈ egg > brain. Perfluorocarboxylic acids (PFCAs) with 8-15 carbon (C) atoms were found, with the highest concentrations determined in plasma (sum PFCA: 41.8-262 ng/g ww), whereas 5C- and 6C-PFCAs were below the limits of detection. Perfluorobutane sulfonate, perfluorooctane sulfonamide, and four saturated (8:2 FTCA and 10:2 FTCA) and unsaturated (8:2 FTUCA and 10:2 FTUCA) fluorotelomer carboxylic acids were not detected in any samples. Perfluorohexane sulfonate was measured at concentrations up to 2.71 ng/g ww. The accumulation profiles of PFCAs were characterized by high proportions of the long and oddnumbered carbon-chain-length compounds, namely perfluoroundecanoic (11C) and perfluorotridecanoic acid (13C), although their individual contribution differed between the matrixes analyzed. Current PFAS concentrations * Corresponding author phone: (+47) 77 75 05 42; fax: (+47) 77 75 05 01; e-mail:
[email protected]. † Norwegian Polar Institute. ‡ University of Tromsø. § University of Guelph. || National Water Research Institute, Environment Canada. ⊥ Norwegian Institute for Air Research. # Canadian Wildlife Service, Environment Canada. 10.1021/es051097y CCC: $30.25 Published on Web 09/02/2005
2005 American Chemical Society
suggest a bioaccumulation potential in Norwegian arctic glaucous gulls that needs to be assessed as part of a broad organohalogen contaminant cocktail with potential for mediating biological processes in this vulnerable top-predator marine species.
Introduction Worldwide surveys of perfluorinated alkyl substances (PFAS) have ascertained their ubiquitous presence in wildlife and human populations from urban and remote locations. Due to their unique chemical and biological stability with respect to abiotic and biotic degradation, PFAS exhibit a high propensity for persistence and bioaccumulation in wildlife, including species occupying high trophic positions in the marine food web. In recent years, particular concern has arisen, as PFAS have been reported in tissues of marine mammal, seabird, and fish species inhabiting various regions of the Arctic (1-7). The predominant PFAS reported in arctic biota has been perfluorooctane sulfonate (PFOS), while perfluorohexane sulfonate (PFHxS), perfluorooctane sulfonamide (PFOSA), and 8-15 carbon-chain length perfluorocarboxylic acids (PFCAs) have also been reported. Measurements of PFAS in the polar bear (Ursus maritimus), a widely distributed arctic marine apex predator, have shown that this class of organohalogens is present at similar or higher concentrations than persistent organochlorines (OCs) in this species (2, 7, 8). In a circumpolar assessment of polar bears, individuals from the Norwegian Arctic were found to accumulate some of the highest PFAS concentrations among the populations investigated (8). Because PFAS are anionic, nonvolatile chemicals and thus not ideal candidates for long-range atmospheric transport, mechanisms of global transport to remote arctic regions are unclear. Recently, Ellis et al. (9) suggested that atmospheric oxidative degradation of airborne fluorotelomer alcohols detected in urban and rural air samples (10, 11) is likely to contribute to the widespread dissemination of homologous series of PFCAs found in arctic animals. The fluorotelomer carboxylic acids, which are primary degradation products in microbial degradation and oxidative atmospheric decomposition of fluorotelomer alcohols (9, 12), have for instance been reported in samples of Canadian arctic ringed seals (Phoca hispida), beluga whales (Delphinapterus leucas), and northern fulmars (Fulmarus glacialis) (3). It has also been argued that atmospheric degradation is expected to occur for all industrial volatile polyfluorinated chemicals, adding additional complexity to the possible sources in the Arctic (9). Furthermore, the oceanic circulation was also put forward as a possible transport pathway for the dispersal of partially water soluble PFAS in remote geographical regions, as documented for instance in the western North Pacific Ocean (13). In the present study, we investigated the distribution and concentrations of a suite of PFAS in plasma, liver, brain, and egg samples collected from glaucous gulls (Larus hyperboreus) breeding at Svalbard and Bear Island in the Norwegian Arctic. The glaucous gull from the Norwegian Arctic, which has an annual distribution entirely within the North Atlantic region (14), exhibits a typical scavenging-predatory behavior in the marine food web and utilizes a wide range of food items, such as eggs, chicks, fish, carrion, crustaceans, and adult birds (15). This species has previously been reported to achieve some of the highest tissue/blood levels of OC contamination of any northern circumpolar seabird species VOL. 39, NO. 19, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7439
Sigma-Aldrich (Oakville, Ontario, Canada or Steinheim, Germany), Fluka (Buchs, Switzerland), Interchim (Montluc¸ on Cedex, France), and ABCR (Karlsruhe, Germany). PFOSA was supplied by the 3M Company (St. Paul, MN) or ABCR. Standard purity for individual PFAS is described by Smithwick et al. (7, 8) and Berger and Haukås (22). The saturated (8:2 and 10:2 FTCA) and unsaturated (8:2 and 10:2 FTUCA) fluorotelomer carboxylic acid standards (all > 95% purity) were synthesized according to methods of Achilefu et al. (23) by the working group of Prof. Scott Mabury (Department of Chemistry, University of Toronto, Toronto, Ontario, Canada).
FIGURE 1. Map of the Norwegian Arctic showing sampling locations (9) of glaucous gulls. and population (16). Several biological effects on reproduction, behavior, development, thyroid hormone balance, and immunity have been reported in glaucous gulls (17, and references therein). These effects are potentially mediated via the toxicological actions of organohalogen contaminants. However, the complexity of organohalogens in glaucous gulls as a consequence of ecosystem and food web exposure is at present not fully defined. Other persistent and bioaccumulative chemicals may also have health and population status impacts in this species. In laboratory studies, chronic exposure to PFOS and analogous fluorochemicals has been shown to cause adverse treatment-related effects such as alteration in steroid hormone levels in fish (18) and subtle toxicological and reproductive endpoints in birds (19, 20).
Experimental Section Sample Collection. Samples of plasma (n ) 20), liver (n ) 5), and brain (n ) 8) were collected during the breeding season (i.e., incubation and chick rearing period) of 2004 from an equal number of male and female glaucous gulls at Svalbard (ice edge) and Bear Island in the Norwegian Arctic (Figure 1). Glaucous gulls included in the present study consisted solely of sexually mature and breeding individuals of similar and good body condition. Individuals were either live-captured on their nest using a trap or shot at sea while foraging. Individuals captured alive were sampled for blood only and released on-site. In addition, freshly laid eggs (n ) 10) of glaucous gulls were collected randomly from major breeding colonies at Bear Island. Samples were collected and processed according to methods described elsewhere (21). Capture and handling methods were approved by the Norwegian Animal Research Authority (P.O. Box 8147 Dep., Oslo, NO-0033, Norway) and the Governor of Svalbard (Box 633, Longyearbyen, NO-9171, Norway). Chemical Analyses. Chemical analyses of glaucous gull samples were carried out by two laboratories, i.e., the National Water Research Institute (Burlington, Ontario, Canada) (method A) and the Norwegian Institute for Air Research (Tromsø and Kjeller, Norway) (method B). To ascertain and compare PFAS concentrations, an interlaboratory test was performed in these laboratories based on extraction, cleanup, and quantification of a suite of PFAS in glaucous gull liver samples. This test was not designed as an interlaboratory method validation, and the quantitative comparisons should be interpreted accordingly (see also QA/QC and Method Comparison). Standards. Authentic standards of perfluorobutane sulfonate (PFBS), PFHxS, PFOS, and PFCAs were purchased from 7440
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 19, 2005
Method A. Samples of plasma, livers, and eggs were extracted using an ion-pairing procedure adapted from Hansen et al. (24). Briefly, 1.0 g of homogenized sample was added to 3 mL of aqueous Na2CO3 (0.25 M) and 2 mL of aqueous ion-pairing agent tetrabutylammonium hydrogen sulfate (0.5 M adjusted to pH 10), and spiked with 20 ng of internal standard (ISTD) 1,2-13C perfluorooctanoic acid (13CPFOA) (PerkinElmer Life and Analytical Sciences, Woodbridge, Ontario, Canada). Samples were extracted twice with 5 mL of methyl-tert-butyl-ether (MTBE), and the combined organic phase was evaporated to dryness under a gentle flow of high-purity nitrogen. The sample extracts were reconstituted in 1 mL of high-performance liquid chromatograph (HPLC) grade methanol, vortexed, and filtered through 0.2 µm nylon filters. Aliquots (10 µL) of the final extracts were injected automatically on a HPLC (Agilent 1100; Agilent Technologies, Palo Alto, CA) coupled to a mass spectrometer (MS) (API 2000; Applied Biosystems/MDS SCIEX, Streetsville, Ontario, Canada) with negative electrospray ionization (ESI) (HPLC-ESI/MS/MS). Compounds were separated on a Luna C8 column (50 × 2 mm, 3 µm particle size) (Phenomenex, Torrance, CA) using a gradient of 0.01 M ammonium acetate (NH4OAc) in both methanol and water delivered at a flow rate of 250 µL/min. The elution gradient started at 20:80 (volume:volume) methanol/water, followed by a 5 min ramp increase to 100% methanol, a 5 min hold at 100% methanol, and reversion to the initial condition after min 12. PFAS were determined by HPLC-MS/MS ESI using multiple reaction monitoring (MRM) of optimum parent and daughter ions (Table 1). Quantification of compounds was performed using a standard curve composed of 8 dilutions of a PFAS standard mix extracted in a manner similar to the matrix of interest. Because authentic standards were not available for perfluorotridecanoic acid (PFTriA) and perfluoropentadecanoic acid (PFPA), standard curves from perfluorododecanoic acid (PFDoA) and perfluorotetradecanoic acid (PFTA), respectively, were used for quantification of these compounds. Method B. Samples of liver and brain tissues were analyzed on the basis of a method described by Berger and Haukås (22). Briefly, homogenized samples (1.0 g) were added to 2.7 mL of methanol/water (50:50, 2 mM NH4OAc) and spiked with the ISTD 7H-perfluoroheptanoic acid (7HPFHpA; ABCR). Samples were sonicated for 30 min to allow complete extraction. The sample extracts were then passed through facial tissue covering the tip of a Pasteur pipet, and the solution was further cleaned up using Microcon YM-3 centrifugal filters at 14 000 rpm. The final extracts were added to a recovery standard (3,5-bis(trifluoromethyl)phenyl acetic acid (3,5-BTPA)), and aliquots (25 µL) were injected automatically on a HPLC (Agilent 1100; Agilent Technologies, Palo Alto, CA) coupled to ESI time-of-flight-high-resolution MS in the negative ion mode (HPLC-ToF-HRMS ESI) (LCT, Micromass, Manchester, England). Compounds were separated on a ACE C18 column (150 × 2.1 mm, 3 µm particle size) (ACT, Aberdeen, U.K.) using a gradient of 200 µL/min methanol and water (both with 2 mM NH4OAc). The initial mobile phase condition was 50:50 methanol/water, followed by a 5 min ramp increase to 85:15, a 5 min hold at 85:15, a
TABLE 1. Meana Concentration (ng/g wet weight), Standard Error (SE), and Range of a Suite of Perfluorinated Alkyl Substances in Egg and Plasma Samples of Glaucous Gulls from the Norwegian Arcticb egg (n ) 10) compound perfluorosulfonates perfluorohexane sulfonate perfluorooctane sulfonate perfluorocarboxylic acids perfluorooctanoic acid perfluorononanoic acid perfluorodecanoic acid perfluoroundecanoic acid perfluorododecanoic acid perfluorotridecanoic acid perfluorotetradecanoic acid perfluoropentadecanoic acid
mass % of acronym transition samples >MDL PFHxS PFOS PFCA PFOA PFNA PFDA PFUnA PFDoA PFTriA PFTA PFPA ∑PFCA
mean ( SE
plasma (n ) 20) range
% of samples >MDL
399f99 499f99
30 100
