Environ. Sci. Technol. 2008, 42, 4344–4349
Polybrominated Diphenyl Ethers, Perfluorinated Compounds and Chlorinated Pesticides in Swordfish (Xiphias gladius) from the Mediterranean Sea SIMONETTA CORSOLINI,* CRISTIANA GUERRANTI, GUIDO PERRA, AND SILVANO FOCARDI Dipartimento di Scienze Ambientali G. Sarfatti, University of Siena, via P.A. Mattioli, 4, I-53100 Siena, Italy
Received December 7, 2007. Revised manuscript received March 5, 2008. Accepted March 27, 2008.
The relative isolation of the Mediterranean population, their feeding habits, and the widespread use of their fillets for human consumption make the Mediterranean swordfish, Xiphias gladius, an interesting species from an ecotoxicological and commercial point of views. High resolution gas chromatography and tandem mass spectrometry detected 19 PBDE congeners, perfluorooctane sulfonate (PFOS), and perfluorooctanoic acid (PFOA), hexachlorobenzene (HCB), p,p’ and o,p’ isomers of DDT, DDE, and DDD in all samples. The presence of PBDEs was reported for the first time in Mediterranean swordfish from the South Tyrrhenian Sea; total PBDE concentrations were 2218 ( 3291 and 612 ( 598 pg/g wet wt in the liver and in the muscle, respectively. Significant correlations were identified between BDE47 and ΣPBDE liver concentrations versus sex and sexual maturity of specimens. The lipid-normalized concentrations ratio BDEnliver/BDEnmuscle+liver suggested that this species mostly accumulates POPs in the liver. PFOS and PFOA were below the LOD (1.5 and 3 ng/g wet wt, respectively) in all the samples. The ΣDDTs was 155 ( 125 and 309 ( 273 ng/g wet wt in the muscle and the liver, respectively. The estimated daily ingestion of PBDEs and DDTs through diet was lower than the acceptable weekly intakes proposed by the World Health Organization.
Introduction The swordfish Xiphias gladius Linnaeus 1758 is a highly migratory species and shows a wide distribution from tropical to temperate oceans within 61°N to 50°S and 180°W to 180°E, including the Mediterranean Sea (www.fishbase.org). Swordfish is an important species from both ecological and commercial points of view, and it is fished worldwide. The relative isolation of the Mediterranean population, their feeding habits, and the widespread use of their fillets for human consumption make the Mediterranean swordfish an interesting species from an ecotoxicological point of view as well. Genetic analyses have revealed that differences exist between populations inhabiting the Atlantic Ocean and those in the Mediterranean basin and that little genetic exchange may occur between them (1); besides, Mediterranean sword* Corresponding author phone: ++39 0577 232939; fax: ++39 0577 232806; e-mail:
[email protected]. 4344
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 12, 2008
fish display limited movement to areas in the adjacent North Atlantic (2). Therefore, studies conducted on specimens collected in the Mediterranean basin are important to assess the status of this population. The Mediterranean Sea receives a heavy pollutant input from the urban and industrial areas along its coasts, and from cultivated land through rivers. Its turnover time is approximately 100 years (3); this limited water exchange increases the chemical residence time and allows bioaccumulation. The Stockholm Convention recognized that persistent organic pollutants (POPs) are chemical substances that persist in the environment, bioaccumulate through the food web, and pose a risk of causing adverse effects to human health and the environment (www.pops.int). They include chemicals that have been used worldwide for many decades, such as chlorinated pesticides (e.g., hexachlorobenzene, HCB, and DDTs). Emerging pollutants like perfluorinated compounds (PFCs) and brominated flame retardants (BFRs) also show bioaccumulation, biomagnification, and toxic properties that have already been reported by many authors (4–10). The most important contribution to the dietary BFR intake in humans is from fish and seafood, representing approximately one-third of the total (6). Several chlorinated pesticides, polychlorinated biphenyls, polychlorinated dibenzo-dioxins and -furans, and perfluorinated compounds have already been detected in this species (e.g., 7–9, 11, 12), and high concentrations of POPs were reported to induct vitellogenin (Vtg) and Zona radiata proteins (Zrp) in male Mediterranean swordfish (13). Following these results, it has been assessed that that 14% of the central Mediterranean male swordfish are likely undergoing sex inversion (14). This study reports the accumulation of nineteen congeners of PBDEs in the Mediterranean swordfish from the South Tyrrhenian Sea; to the authors’ knowledge, this is the first time that their presence is reported in this species. Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA), HCB, p,p’ and o,p’ isomers of DDT, DDE, and DDD were also determined. The levels and patterns of these POPs (paying special attention to the PBDEs), the POP distribution in the liver and muscle tissues, and the possible continued input of DDTs in the area were also examined. It is important to keep monitoring DDTs because the Stockholm Convention allows some countries to use DDT for disease vector control (15); thus, from those areas, residues can reach the Mediterranean through global transport (16–18). Despite the fact that swordfish fillets are consumed worldwide as a delicious seafood, not many studies have been published on the their quality; the daily ingestion of PBDEs and DDTs through fillet consumption was estimated, and compared with estimated acceptable daily ingestion (EDI) proposed by the World Health Organization (WHO).
Materials and Methods Collection of Samples. Seventeen samples (10 males, 7 females) of muscle and 12 samples of liver were excised from 17 swordfish (Xiphias gladius) caught in April 2005 by longline fishing off the coasts of Calabria and the Aeolian Islands in the Southern Tyrrhenian Sea (Supporting Information Figure S1). Unfortunately, five liver samples arrived in bad condition to our laboratory (problems occurred during shipping), and they where discarded. The length (144 ( 30 cm) and weight (39 ( 23 kg) of each specimen were measured as soon as the fish were on board the fishing vessel. Sex identification of the swordfish was carried out in the field immediately after capture by direct observation of gonads. After dissection, samples were 10.1021/es703057f CCC: $40.75
2008 American Chemical Society
Published on Web 05/17/2008
wrapped in aluminum foil previously cleaned with solvents (acetone and n-hexane), stored in dry ice, and kept in the dark until transportation to the laboratory, where they were stored at -20 °C until analysis. The data available for the specimens analyzed are available in the Supporting Information (Table S1). Analytical Methods for Chlorinated and Brominated Chemicals. All the samples were extracted for the determination of HCB, DDTs, and PBDEs. All chemicals were analyzed following a method described elsewhere, with some modifications (8, 11). Briefly, samples were homogenized with anhydrous sodium sulfate salt and Soxhlet extracted with hexane:dichloromethane (40:60). Interferences were removed by fractionation by a multilayer silica gel column. HCB and the p,p’ and o,p’ isomers of DDT, DDE, and DDD were identified and quantified using a gas chromatograph (Perkin-Elmer mod. Autosystem) equipped with a 63Ni electron capture detector (HRGC-ECD); a capillary column coated with DB-5 ((5%-phenyl)-methylpolysiloxane, 30 m × 0.25 mm i.d., film thickness 0.25 µm; Supelco Inc.) was used. For GC-ECD conditions see Corsolini et al. (11). The limit of detections (LODs) of individual compounds were evaluated as the mean blank+3SD and they were 1-75 pg/g wet weight. Blanks were run with each set of five samples, and their values ranged from 1 to 10 pg. Nineteen BDE congeners (IUPAC numbers BDE3, BDE7, BDE5, BDE17, BDE28, BDE49, BDE71, BDE47, BDE66, BDE77, BDE100, BDE119, BDE99, BDE85, BDE126, BDE154, BDE153, BDE138, BDE156) were identified and quantified using a GC/ MS (ion trap mass spectrometer) from ThermoFinnigan (Trace GC 2000/GC Polaris), equipped with an AS2000 autosampler (Rtx-5MS capillary column, 30 m × 0.25 mm i.d., film thickness 0.25 µm; Restek). A 2 µL aliquot of sample in isooctane was injected in splitless injection mode with helium as the carrier gas. The injector temperature was 275 °C. The ramp program was as follows: the initial temperature of the oven was 80 °C, held for two minutes; it increased to 200 °C at a rate of 25 °C/min, then to 300 at 4 °C/min and held for 10 min. The excitation voltages were 4.75V for triand tetra-BDEs, 4.60V for penta-BDEs, and 4.70V for hexaBDEs. The internal standard was PCB1413C in isooctane, from Cambridge Isotope Laboratories; the PBDE calibration standard solution was from Wellington Laboratories, Inc. The compounds detected in the blanks were PBDEs 99 and 154. Detection limits, calculated as the mean blank +3SD, were 4 pg/g tissue. Throughout this manuscript PBDEs are represented by their IUPAC numbers. The accuracy and precision of HCB and DDT analyses were tested by intercomparison exercises for their determination in biota (IAEA-432, 2002). The quality assurance is based on the analyses of Certified Reference Material (CRM CARP-2) from the National Research Council, Canada (NRC). Our results were in good agreement with the certified values. The quality control of PBDE analyses was based on regular analyses of procedural blanks, solvent blanks, spiked samples, and injection of standards. Results are given on a wet weight basis (wet wt), and on a lipid basis when specified. Where needed for calculations, concentrations 168.8, 218.8, and 498.8 > 368.9, respectively. The instrumental limit of detection (LOD) for PFOS was determined as three times the signal-to-noise (S/N) ratio. The LODs for PFOS and PFOA were 1.5 and 3 ng/g wet weight (wet wt), respectively. Data quality assurance and quality control protocols included matrix spikes, laboratory blanks, and continuing calibration verification. Matrix spikes were analyzed for each type of tissue sample. Blanks were analyzed with each set of five tissue samples as a check for possible laboratory contamination and interferences. Estimated Daily Ingestion (EDI). The EDI of PBDEs and DDTs via fish consumption was calculated by multiplying the contaminant concentrations in fish fillets by the estimated daily intake for the general population (31.8 g/day, (20)); the resulting values were then used to evaluate the daily intake per kg of body weight, considering an average body weight of 50 kg for women and 70 kg for men. The results were compared with the minimal risk level (MRL) accepted for PBDE intake via food (10, 21) and the acceptable daily intake (ADI) of DDTs through ingestion proposed by the World Health Organization (10). When the PBDE or DDT isomer concentrations were under the respective limit of detection (LOD), daily intakes were calculated assuming that respective values would be equal to one-half of the LOD. Statistical Analysis. As data sets were not normally distributed, differences between samples were examined using the Mann-Whitney, Kolmogorov-Smirnov, and WaldWolfowitz nonparametric tests; correlations were evaluated using the Pearson coefficient (STATISTICA v7.1, StatSoft Italia). Values below the LOD were considered equal to the LOD or empty values or zero, depending on the analyses.
Results and Discussion The presence of nineteen PBDE congeners, PFOS and PFOA, HCB, and the p,p’ and o,p’ isomers of DDT, DDE, and DDD was detected in the swordfish tissue samples analyzed, and the results are shown in Tables 1 and 2. The values of the standard deviation is high in most of the cases and this may be due to the undetectable levels of chemicals in many samples and to the natural variability usually found when analyzing organisms belonging to different sex and class of age; the individual diet and foraging areas may also contribute to very different accumulation levels in tissues. All the chemicals were detected in most of the specimens, except PFOS and PFOA, which were below the LOD (1.5 and 3 ng/g wet wt, respectively). It cannot be excluded that PFOA and PFOS could be detected in our samples using more sensitive instruments, or in other specimens. In fact, 3 and 36 ng/g wet wt of PFOS and PFOA, respectively, were detected in the livers of seven swordfish collected in the Strait of Messina, Italy (7). PBDE concentrations (sum of mono- to hexa-BDEs) were 2218 ( 3291 pg/g wet wt (13 862 ( 20 568 pg/g lipids) in the liver, and 612 ( 598 pg/g wet wt (6689 ( 6536 pg/g lipids) in the muscle (Table 1). To the authors’ knowledge, this is the first time that PBDEs have been detected in Mediterranean swordfish. Luksemburg et al. (23) analyzed PBDE concentrations in fillets of three wild swordfish purchased VOL. 42, NO. 12, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4345
TABLE 1. Concentrations of PBDEs (pg/g), PFOS, PFOA, HCB, and DDTs (ng/g) on a Wet Wt and Lipid Basis(*) in the Muscle and Liver of Swordfish from the Southern Thyrrenian Seaa and Percentage of Fish Positive for Chemicals in Both the Tissues
n Lipid % ΣPBDEs ΣPBDEs* PFOS PFOA HCB o,p’-DDE p,p’-DDE o,p’-DDD p,p’-DDD o,p’-DDT p,p’-DDT ΣDDTs a
muscle
liver
17 8.8-9.5 612 ( 598 (
