Environ. Sci. Technol. 2007, 41, 315-320
Occurrence of Perfluorooctane Sulfonate and Other Perfluorinated Alkylated Substances in Harbor Porpoises from the Black Sea K R I S T I N I N N E K E V A N D E V I J V E R , * ,† LUDO HOLSBEEK,‡ KRISHNA DAS,§ RONNY BLUST,† CLAUDE JOIRIS,‡ AND WIM DE COEN† Department of Biology, University of Antwerp, Groenenborgerlaan 171, 2020 Antwerpen, Belgium, Laboratory for Ecotoxicology en Polar ecology, University of Brussels (VUB), Pleinlaan 2, 1050 Brussels, Belgium, and MARE Center, Laboratory for Oceanology, Lie`ge University B6, B-4000, Lie`ge Belgium
Perfluorooctane sulfonate (PFOS) and other perfluorinated alkylated substances (PFAS) were determined in liver, kidney, muscle, brain, and blubber samples of 31 harbor porpoises (Phocoena phocoena relicta) of different age and sex stranded along the Ukrainian coast of the Black Sea. In all individuals and in all tissues, PFOS was the predominant PFAS, accounting for on average 90% of the measured PFAS load. PFOS concentrations were the highest in liver (327 ( 351 ng/g wet wt) and kidney (147 ( 262 ng/g wet wt) tissue, and lower in blubber (18 ( 8 ng/g wet wt), muscle (41 ( 50 ng/g wet wt), and brain (24 ( 23 ng/g wet wt). No significant differences could be determined between males and females, nor between juvenile and adult animals (p > 0.05). Perfluorononanoic acid, perfluorodecanoic acid, perfluoroundecanoic acid, and perfluorododecanoic acid could be detected in liver tissue of approximately 25% of the individuals. Perfluorobutane sulfonate, perfluorobutanoic acid, and perfluorooctanoic acid were not detected in any of the porpoise livers. Although we investigated a potential intraspecies segregation according to the source of prey, using stable isotopes, no statistically significant correlation between PFOS concentrations and stable isotopes could be determined. It is, however, noteworthy that the contamination by PFOS in the Black Sea harbor porpoises is comparable to levels found in porpoises from the German Baltic Sea and from coastal areas near Denmark and, therefore, might pose a threat to this population.
capacities (4). These chemicals are commonly used as surfactants; they were applied in fire foam extinguishers until 2001, but they were also used for stain and/or wetness resistance of paper and textiles. Perfluorooctane sulfonate (PFOS) is the predominant perfluoroalkyl compound in biotic samples and is found worldwide in a great diversity of wildlife species (5, 6). PFOS has been detected in abiotic compartments such as water, air, sediment, and biota (5, 7-9). Different research studies have focused on PFAS in marine mammals like seals, dolphins, whales, and polar bears from a wide range of geographical regions: from remote regions such as the Canadian and Norwegian Arctic to more industrialized, coastal areas like Florida and the Baltic Sea (5, 10-14). Among marine mammals with the highest concentrations of PFOS measured in liver tissue are polar bears from Greenland (up to 6340 ng/g wet wt) (15, 16), ringed seals (up to 1100 ng/g wet wt) (5), and harbor porpoises from the Baltic Sea (up to 1149 ng/g wet wt) (14) and bottlenose dolphins from the United States (up to 1520 ng/g wet wt) (5). Much of the variability of PFAS concentrations between species is due to interspecies differences (feeding habits, metabolism, migration routes, and trophic level) and temporal and spatial differences. Eastern Europe is one of the regions with a lack of data on the degree of PFOS pollution and distribution in the environment (6). The Black Sea being a closed sea in between the European and Asian content receives most of its pollutant load through river input from the surrounding countries, including heavily industrialized regions such as Bulgaria, Ukraine, and Russia. The Black Sea ecosystem is considered to be under constant threat by inputs of insufficiently treated sewage and oil as a result of accidental and operational marine and land based discharges and by dumping of toxic and industrial wastes (17-19). Though some studies have investigated the environmental quality of the Black Sea, marine mammals like the harbor porpoise (Phocoaena phocoena relicta) are poorly studied. Previous studies reported on contaminant burdens of organochlorine, butyltin, and heavy metals in harbor porpoises from the Black Sea (20-26). Being a top predator, the harbor porpoise ought to be a target species for accumulation of perfluorinated compounds (23, 27). This study presents the results from analyses of PFAS and their distribution in liver, kidney, muscle, brain, and blubber tissues of 31 harbor porpoises from the Ukrainian coast of the Black Sea and its relation to gender and age. PFAS concentrations are also put in relation to stable isotope measurements (δ13C and δ15N) so that a potential intraspecies segregation according to the source of prey can be investigated. Finally, a geographical comparison with existing data from the same species originating from other regions will be made.
Materials and Methods Introduction Among a large number of man-made chemicals, perfluorinated alkylated substances (PFAS) are of great concern due to their bioaccumulative nature, their persistence, and their toxic biological effects such as membrane-related effects (1), developmental problems (2, 3), and peroxisome proliferating * Corresponding author phone: +32 3 265 33 50; fax: +32 3 265 34 97; e-mail:
[email protected]. † University of Antwerp. ‡ University of Brussels. § Lie ` ge University. 10.1021/es060827e CCC: $37.00 Published on Web 11/17/2006
2007 American Chemical Society
Sample Collection and Storage. During 1997 and 1998 harbor porpoises (Phocoena phocoena relicta) were by-caught incidentally along the Ukrainian coast of the Black Sea and collected by the local marine mammal network (coordination A. Birkun). Liver, kidney, muscle, and brain of 31 harbor porpoises were analyzed for PFAS along with 10 fat samples. Within the 31 samples, there was a mother with a fetus. The same animals were analyzed for heavy-metal and PCB content (22). Tissue samples were kept at -20 °C. PFAS Analysis. Tissue extracts were analyzed using highperformance liquid chromatography (HPLC) combined with electrospray tandem mass spectrometry (LC-MS/MS) as VOL. 41, NO. 1, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
315
described by Hansen et al. (28) with minor modifications as described by Van de Vijver et al. (13). Perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA), PFOS, perfluorodecanoic acid (PFDA), perfluoroundecanoic acid (PFUA), and perfluorododecanoic acid (PFDoA) were measured by the same procedure: LC-MS/MS was done on a CapLC system (Waters, Millford, MA) connected to a Quattro II triple quadrupole mass spectrometer (Micromass, Manchester, UK). Aliquots of 5 µL of extract were loaded on an Optiguard C18 precolumn (10 mm × 1 mm i.d., Alltech, Sercolab, Belgium), followed by a Betasil C18 column (50 mm × 1 mm i.d., Keystone Scientific, Bellefonte, PA) at a flow rate of 40 µL/min. The mobile phase was 2 mM NH4OAc/methanol, starting at 10% methanol, increasing to 90% in 8 min. After 10 min initial conditions were resumed. However, for the measurement of perfluorobutane sulfonate (PFBS) and perfluorobutanoic acid (PFBA), 0.1% of formic acid was used instead of 2 mM NH4OAc. All components were measured under negative electrospray ionization using the following transitions: 213 f 169 (PFBA), 299 f 99 (PFBS), 413 f 369 (PFOA), 463 f 419 (PFNA), 499 f 99 (PFOS), 513 f 469 (PFDA), 563 f 519 (PFUA), and 613 f 569 (PFDoA). The internal standard, 1H,1H,2H,2H-perfluorooctane sulfonate (Sigma-Aldrich Chemical Co., Milwaukee, WI) was measured under the same conditions (427 f 81). No other standards were included. The dwell time was 0.1 s. The EScapillary voltage was set at -3.5 kV and the cone voltage was 24 V (35 V for measurements of PFBA and PFBS). The source temperature was 80 °C. The pressure in the collision cell was 3.3 10-5 mmHg (Ar). These settings were the same for all measurements. Data quality assurance and quality control protocols included matrix spikes of all tissues, laboratory blanks, and continuing calibration verification for each block of eight samples. This way, changes in instrument sensitivity could be monitored and matrix effects on ESI suppression/ enhancement could be minimized. Recoveries of spiked samples based on duplicate analysis varied from 71% to 113%. The used standards of PFBS, PFOS, and perfluorocarboxylic acids (PFCA) were purchased from Sigma-Aldrich (SigmaAldrich Chemical Co., Milwaukee, WI) and Interchim (Montluc¸ on Cedex, France). The purity of the standards ranged from 95% to 99%. Recoveries of the internal standard in samples ranged from 64% to 83%. The instrumental limit of detection (LOD) for individual PFAS was determined as 3 times the signal-to-noise ratio (S/N). The LOD of PFOS was 1.5 ng/g wet weight (wet wt), whereas for the other measured compounds it varied from 1.4 to 3.2 ng/g wet wt. Concentrations were evaluated versus an unextracted standard curve composed of seven dilutions of a PFAS standard mix and were not corrected for the recoveries or for the purity of the PFAS standards. The repeatability and reproducibility were done in triplicate and were 86% and 81%, respectively. Stable Isotope Measurements. Stable isotope ratios were measured in muscle tissue. Details on stable isotope measurements were described by Das et al. (23). Briefly, after drying at 50 °C (48 h), samples were ground into a homogeneous powder and treated with a 2:1 chloroform: methanol solution to remove lipids. Carbon dioxide and nitrogen gas were analyzed as described by Das et al. (29) on a V.G. Optima (Micromass) IR-MS coupled to a N-C-S elemental analyzer (Carlo Erba). Routine measurements are precise to 0.3% for both 13-carbon and 15-nitrogen. Stable isotope ratios were expressed in δ notation according to the following,
δX ) [(Rsample/Rstandard) - 1] × 1000 where X is 13C or 15N and R is the corresponding ratio 13C/12C or 15N/14N. Carbon and nitrogen ratios are expressed relative 316
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 1, 2007
FIGURE 1. Concentrations of perfluorooctane sulfonate in different tissues of harbor porpoises drowned in gillnets in the Black Sea. The straight line is the median and the dotted line represents the mean. The 25th and 75th percentiles define the boxes. The whiskers represent the 10th and 90th percentiles, while the dots represent the 5th and 95th percentiles. The “/” indicates the significant difference (p < 0.05) between the liver and the other tissues. to the v-PDB (Vienna Peedee Belemnite) standard and to atmospheric nitrogen, respectively. Reference materials were IAEA-N1 (δ15N ) +0.4 ( 0.2‰) and IAEA CH-6 (sucrose) (δ13C ) -10.4 ( 0.2%). Data Treatment. The normality of the data was analyzed by means of a Kolmogorov-Smirnov test. Because of a lack of normality in the distribution of the PFOS concentration of some species, a nonparametric Kruskal Wallis analysis of variance was used for the statistical comparison of PFOS concentrations between tissues. A Dunn’s Multiple Comparisons Test was performed as post-hoc criterion. The significance levels were taken as p < 0.05. To look at the influence of gender, a nonparametric Mann Whitney U-Test was used. ANOVA followed by post hoc multiple comparison tests (Tukey test) were used to compare the data between the different age groups (juveniles from neonate to the age of 3.5, adults from the age of 4-9 years old). To find relationships between the data of the PFOS concentrations and the stable isotopes ratios, a Spearman Rank correlation was used. Samples with a compound concentration below the LOD were assigned a value of 50% of the LOD. All analyses were performed using the software package Statistica (Statsoft Inc., Tulsa, OK).
Results PFAS Concentrations. In all individuals and in all tissues, PFOS was the predominant perfluorinated alkylated substance, accounting for on average 90% of the measured PFAS load. Concentrations of PFOS in liver tissue differed statistically significantly from the other tissues (p < 0.05), with mean values in the liver (range from 33 to 1790 ng/g wet wt, n ) 31) being 2.2 times as high as in the kidney (range from 2.6 to 1371 ng/g wet wt, n ) 29), and at least 8 times as high as those in muscle and brain tissue (Figure 1). Due to the small sample size of blubber tissue, no differences between blubber tissue and the other tissues could be investigated here. It is noteworthy that the highest PFOS concentrations in kidney (1371 ng/g wet wt) and the all but one highest in brain tissue (92 ng/g wet wt, maximum 100 ng/g wet wt) were measured in the fetus sample. No differences were found in PFOS concentrations in adults (n ) 15) and juveniles (n ) 16) in any of the tissues (p > 0.05) (Figure 2). Juveniles are considered animals up to the age of 3, given that at the age of 3 years, a female porpoise becomes ready to give birth (30).
FIGURE 2. Mean perfluorooctane sulfonate concentrations (ng/g wet wt) and standard deviations in different tissues of harbor porpoises (Phocoena phocoena relicta): juveniles (0-3.5 years old; n ) 16) versus adults (4-9 years old; n ) 15). None of the tissues tested statistically significant (T-test: p > 0.05).
FIGURE 3. Mean perfluorooctane sulfonate concentrations (ng/g wet wt) and standard deviations in tissues of male (n ) 18) and female (n ) 13) harbor porpoises (Phocoena phocoena relicta). In none of the tissues male and female differences tested statistically significant (T-test; p > 0.05). PFOS concentrations did not differ between male (n ) 18) and female (n ) 13) harbor porpoises for any of the tissues (p > 0.05) (Figure 3). Female porpoises showed higher PFOS levels than males for all tissues. However, possibly due to a too low sample size and the large variation between the individuals, significance could not be determined. The highest concentration (1789 ng/g wet wt) was, however, found in the liver of a 2 year old male. In kidney, muscle, brain, and blubber tissue, no other PFAS besides PFOS could be detected. In liver tissue of approximately 25% of the individuals, however, other perfluorinated alkylated substances (PFNA, PFDA, PFUA, and PFDoA could be detected (Figure 4). These perfluorocarboxylic acids (PFCA) were, although at low concentrations, generally present at detectable concentrations in liver tissue of harbor porpoises with the highest PFOS concentrations. Three other perfluorinated alkylated substancessPFBS, PFBA, and PFOAswere not detected at detection limits of respectively 1.6, 3.2, and 2.1 ng/g wet wt in any of the porpoise livers. PFOS and Stable Isotope Measurements. Detailed information on the results of the stable isotope measurements was given in Das et al. (23). No significant correlations were observed between δ13C values or δ15N values measured in the muscle tissue and PFOS concentrations in any of the tissues analyzed. This was probably due to the small sample set.
Discussion Geographical Comparison. In the present study, we report for the first time on the exposure levels of PFOS and related chemicals in Black Sea marine mammals. Contrary to levels of for example metals and PCBs, which were reported fairly
FIGURE 4. Hepatic concentrations (ng/g wet wt) of perfluorononanoic acid, perfluorodecanoic acid, perfluoroundecanoic acid, and perfluorododecanoic acid in the liver of harbor porpoises from the Black Sea. The straight line is the median and the dotted line represents the mean. The 25th and 75th percentiles define the boxes. Data points which were below the detection limit were included in the calculations for the mean as 50% of the detection limit. low (22, 23), levels of PFAS, in particular PFOS in harbor porpoises from the Ukrainian coast of the Black Sea, are in the same order of magnitude as those found in porpoises from the German Baltic Sea and from coastal areas near Denmark, and therefore high compared to those in other areas such as the North Sea and Norwegian and Icelandic coastal waters (Table 1). Similarly high PFOS pollution burden in porpoises from Black and Baltic Sea might be explained by the fact that both coastal zones are known as densely populated and highly industrialized locations, showing high overall pollutant burdens (10, 31, 32). In comparison, Iceland and Norway are remote, less urbanized areas. Hepatic PFDA, PFUA, and PFDoA concentrations were, however, lower compared to those in animals from Denmark and the German Baltic Sea. These relatively low levels might be explained by differences in pollution sources (6). As we are comparing the same species (Phocoena phocoena) originating from different locations, the low levels detected in porpoises from the Black Sea might not be due to metabolic degradation mechanisms or to xenobiotic-metabolizing enzyme systems. However, no specific information on production plants of PFAS, or on the presence of paper mills and/or textile industries, which might use PFAS in their production processes, was found for the region around the Black Sea, nor from other regions such as the Baltic Sea. The lower PFOS concentrations in harbor porpoises from the North Sea compared to the higher PFOS levels in porpoises from the Black Sea might be explained by the character of both seas. The North Sea is an open sea with a constant water flow, whereas the Black Sea is the world’s most enclosed sea, and thus with little to no renewal of the water body. The only connection of the Black Sea to other marine water bodies is through the winding Istanbul (Bosporus) Strait, a 35 km natural channel, as little as 40 meters deep in some places. Therefore, probably little dilution or distribution to other areas of present contaminants can occur easily. Tissue Distribution of PFOS. As no significant differences were found between males and females, or juveniles and adults, all samples were pooled for a statistical tissue comparative PFOS analysis. PFOS concentrations decreased from liver > kidney > muscle > brain to blubber. It is known from previous studies that persistent organic pollutants such as PCBs accumulates preferentially in fatty tissues. However, PFOS has a different accumulation pattern, as it preferentially binds to blood proteins and accumulates VOL. 41, NO. 1, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
317
TABLE 1. Hepatic Concentrations of Different Perfluorinated Alkylated Substances in Harbor Porpoises from European and Black Sea Coastsa Norway ref 20
Iceland ref 20
Denmark ref 20
Baltic Sea ref 20
North Sea ref 9
Black Sea this study
PFOS
71-749 130.7 n ) 19
26-67 33.1 n)8
129-620 203.8 n)7
232-1149 349.9 n)7
12-395 62.7 n ) 48
33-1790 209.9 n ) 31
PFNA
