Environ. Sci. Technol. 2008, 42, 3465–3471
Predatory Bird Species Show Different Patterns of Hydroxylated Polychlorinated Biphenyls (HO-PCBs) and Polychlorinated Biphenyls (PCBs) V E E R L E L . B . J A S P E R S , * ,† A L I N C . D I R T U , ‡,† M A R C E L E E N S , † HUGO NEELS,‡ AND ADRIAN COVACI‡ Department of Biology, Campus Drie Eiken, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk, Belgium, Toxicological Center, Campus Drie Eiken, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk, Belgium, and Department of Inorganic and Analytical Chemistry, ”Al.I.Cuza” University of Iassy, Carol I Bvd. No 11, 700506 Iassy, Romania
Received December 6, 2007. Revised manuscript received January 17, 2008. Accepted January 17, 2008.
Hydroxylated metabolites of polychlorinated biphenyls (HOPCBs) have previously been associated with endocrine disrupting effects. Since metabolic capacity may differ among species, we investigated the levels and profiles of HO-PCBs and PCBs in livers of four predatory bird species from Belgium. Maximum concentrations for sum HO-PCBs were found in the common buzzard (Buteo buteo) up to 13 700 pg/g wet weight (ww). The most prominent HO-PCB congener in all bird species was 4-HOCB 187 (up to 6420 pg/g ww in buzzard liver), followed by 4-HOCB146inthebuzzard(upto1820pg/gww),sparrowhawk(Accipiter nisus), and grey heron (Ardea cinerea), and by 3′-HOCB138 in long eared owl (Asio otus) and in one grey heron (up to 985 pg/g ww and 3450 pg/g ww, respectively). The mean profile of the grey heron differed from the other species with 3′HO-CB138 and 4-HO-CB163 contributing more to the sum HOPCBs. This indicates that aquatic and terrestrial predatory bird species may show differences in their HO-PCBs profiles. Variation in the diet and species-specific accumulation and metabolism of PCBs are probably the most important causes for these differences. Correlations between HO-PCBs and their parent PCBs were only found significant for buzzards.
Introduction Polychlorinated biphenyls (PCBs) are widespread contaminants that were formerly used as cooling and insulating fluids in electronic circuits. Although they have been banned in the 1970s, concentrations in the environment are only slowly decreasing due to their high persistency (1, 2). PCBs have been associated with various toxic effects, such as endocrine disruption, immunotoxicity, neurotoxicity, and carcinogenesis (3). It has been shown that PCB metabolites can be formed in different organisms, including humans and birds * Corresponding author phone: +32-3 820 22 85; fax: +32-3 820 22 71; e-mail:
[email protected]. † Department of Biology, University of Antwerp. ‡ Toxicological Center, University of Antwerp. † University of Iassy. 10.1021/es7030455 CCC: $40.75
Published on Web 03/28/2008
2008 American Chemical Society
of prey (4). Hydroxylated polychlorinated biphenyls (HOPCBs) are generated through CYP enzyme-mediated phase I metabolism of PCBs via (1) direct insertion of an OH-group in the meta-position, or (2) via the formation of an arene-epoxide intermediate followed by a 1,2-shift of the meta-hydrogen bond and the para-chlorine (NIH-shift), resulting in a meta- or a para-HO-PCB metabolite (4). HOPCBs are formed in the liver and subsequently retained in the blood, and possibly other body compartments, through binding with proteins (5). Since HO-PCBs show structural similarities with thyroxin (T4), several HO-PCBs compete for binding sites on the transport protein transthyretin (TTR). As a result, decreased levels of circulating plasma T4 have been shown in laboratory animals following PCB exposure (6). Furthermore, some HO-PCBs display estrogenic activity as well (7). Predatory birds are situated at the top of the food chain and have been shown to accumulate high concentrations of PCBs and other persistent organic pollutants (8, 9). Recently, some studies have also reported HO-PCBs levels in avian biota, such as albatrosses (Diomedea immutabilis and Diomedea nigripes) (10), white tailed sea eagle (Haliaeetus albicilla) (11), and glaucous gull (Larus hyperboreus) (12). Hasegawa et al. (13) found similar sum PCBs levels among three investigated species, whereas levels of sum HO-PCBs varied. Metabolic capacity may differ among species and may therefore play an important role in controlling bioaccumulation and toxicity (4, 13). Here, we examine for the first time the levels and profiles of HO-PCBs in liver from different predatory bird species representative for terrestrial and aquatic food chains in Belgium. Furthermore, relations among the parent PCBs and their corresponding hydroxylated metabolites were investigated in these species. In addition, other phenolic compounds, such as tribromophenols (TBPs) and pentachlorophenol (PCP), were also analyzed in the liver samples.
Materials and Methods Sample Collection. Between October 2003 and June 2004, seven grey herons (Ardea cinerea), five barn owls (Tyto alba), three long-eared owls (Asio otus), one tawny owl (Strix aluco), 15 buzzards (Buteo buteo), and five European sparrowhawks (Accipiter nisus) were collected by Wildlife Rescue Centers in Flanders (Belgium). Birds collected for this study were found dead or had died shortly after collection. Frequent causes of death included traffic accidents, natural causes, and starvation. No birds were killed for the purpose of this study. Livers were excised and stored at -20 °C until sample preparation. Chemicals and Materials. The analysis of PCB congeners was already reported elsewhere (9, 14). The investigated HOPCBs, measured as methoxylated derivatives (MeO-PCBs), together with the IUPAC abbreviations and possible PCB precursors, are given in Table 1. All individual standards of PCBs, TBPs, and PCP were obtained from Dr. Ehrenstorfer Laboratories (Augsburg, Germany), whereas standards of HOPCBs and MeO-PCBs were received in 2002 as a gift from Courtney Sandau (Centers for Disease Control, Atlanta, GA). All standards were of g98% purity. All solvents used for the analysis (n-hexane, acetone, ethanol, methyl-tert butyl ether (MTBE), dichloromethane (DCM) iso-octane) were of SupraSolv grade (Merck, Darmstadt, Germany). Anhydrous Na2SO4, KOH and HCl were Analytical grade (Merck). KOH 2 M was prepared in water: ethanol (1:1, v/v). Na2SO4 was used after prewashing with hexane and heated overnight at 120 °C. Extraction thimbles VOL. 42, NO. 9, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3465
TABLE 1. Reference Hydroxy-Compounds Measured as Methoxylated Derivatives in Liver Samples of Predatory Birds. Potential PCB Precursors Are Also Givena
hydroxy-compounds
abbreviation
PCB retention precursors PCB time quantification qualification by direct precursors (min) Ion (m/z) Ions (m/z) insertion by NIH shift
2,4,6-tribromophenol 2,4,5-tribromophenol 2,3,4-tribromophenol 2,3,6-tribromophenol pentachlorophenol unknown hydroxyl-tetrachlorobiphenyl 3-hydroxy-2,3′,4,4′,5-pentachlorobiphenyl 4-hydroxy-2,3,3′,4′,5-pentachlorobiphenyl 3-hydroxy-2,2′,4,4′,5,5′-hexachlorobiphenyl 4-hydroxy-2,2′,3,4′,5,5′-hexachlorobiphenyl 4′-hydroxy-3,3′,4,5,5′-pentachlorobiphenyl 3′-hydroxy-2,2′,3,4,4′,5′-hexachlorobiphenyl 4′-hydroxy-2,2′,3,3′,4,5′-hexachlorobiphenyl 4-hydroxy-2,3,3′,4′,5,6-hexachlorobiphenyl 4-hydroxy-2,2′,3,4′,5,5′,6-heptachlorobiphenyl 3′-hydroxy-2,2′,3,4,4′,5,5′-heptachlorobiphenyl 4′-hydroxy-2,2′,3,3′,4,5,5′-heptachlorobiphenyl 4-hydroxy-2,3,3′,4′,5,5′,6-heptachlorobiphenyl unknown hydroxy-heptachlorobiphenyl tetrabromobisphenol A 4′-hydroxy-2,3,3′,4,5,5′-hexachlorobiphenyl
246-TBP 245-TBP 234-TBP 236-TBP PCP HO-T4CB 3-HO-CB118 4-HO-CB107 3-HO-CB153 4-HO-CB146 4′-HO-CB127 3′-HO-CB138 4′-HO-CB130 4-HO-CB163 4-HO-CB187 3′-HO-CB180 4′-HO-CB172 4-HO-CB193 HO-H7CB TBBP-A 4′-HO-CB159 (IS)
8.32 8.51 9.04 9.32 9.43 19.82 22.69 22.77 23.50 23.60 24.66 24.85 24.97 25.29 25.95 28.32 28.45 28.98 25.50 33.43 26.38
a
81 81 81 81 282 318 318 356, 354 356, 390 390, 388 354 356, 390 390, 354 390, 388 411, 424 388, 424 411, 424 411, 424 411, 424 81 388
na na na na na ? CB 118 CB 107 CB 153 CB 146 CB 127 CB 138 CB 130 CB 163 CB 187 CB 180 CB 172 CB 193 ? na CB 159
na na na na na ? ? CB 105, CB 118 CB 146, CB 128 CB 138, CB 153 ? CB 130, CB 157 CB 128, CB 138 CB 158 CB 183 CB 172 CB 170, CB 180 CB 191 ? na CB 156
na, not applicable; ?, not known.
(25 × 100 mm, Whatman, England) were pre-extracted for 1 h with hexane:acetone (3:1, v/v) and dried at 100 °C for 1 h. For the cleanup of neutral and phenolic fractions, two types of polypropylene columns were used: 25 mL (Alltech, Lokeren, Belgium) and 3 mL (Supelco, Bornem, Belgium). Sample Preparation. The method used for the analysis of neutral and phenolic compounds was adapted from previously described methods for the determination of PCBs in predatory birds tissues (9, 14) and for the determination of phenolic compounds in plasma (15, 16) and briefly presented below. Homogenized liver samples (∼2.5 g) were mixed with anhydrous Na2SO4 (∼5 g) and spiked with 2 ng of 4′-HO-CB159, used here as internal standard for quantification. Further, the extraction of the analytes was performed using 100 mL hexane:acetone (3:1, v/v) in an automated Soxhlet extractor (Büchi, Flawil, Switzerland) in hot extraction mode for 2 h. The extract was concentrated under a gentle nitrogen stream to approximately 1.5 mL. A partitioning step with 2 × 1.5 mL of alcoholic solution KOH 2 M was applied in order to separate the neutral and phenolic fractions. The organic phase (upper layer) contained the neutral organohalogenated contaminants, while the lower aqueous layer was kept for the analysis of phenolic compounds. Phenolic Fraction. The aqueous solution was neutralized with 2 mL of HCl 8 M and then extracted three times with 2 mL MTBE:hexane (1:1, v/v). The organic phase was dried over a 3 mL cartridge filled with anhydrous Na2SO4. The cartridges were eluted with 2 mL of MTBE:hexane (1:1, v/v) and 2 mL of hexane:DCM (1:1, v/v). The combined organic extract was evaporated until dryness and then redissolved in 250 µL of DCM. For methylation, extracts were treated with an excess of a diazomethane solution in diethyl ether at room temperature for 30 min. Afterward, the excess of reagent was evaporated under nitrogen and the extract was redissolved in 0.5 mL of DCM. A further cleanup was applied by passing the extracts over a 25 mL cartridge filled with 6 g of acidified silica (25%, w/w) and elution with 50 mL of hexane:DCM (1:1, v/v). The eluate was concentrated using a rotavapor and further under nitrogen until dryness. The extract was redissolved in 50 µL iso-octane and 50 µL of PCB 3466
79 79 79 79 280 320 320 341 354 375 356 354 375 375 409 390 409 409 409 79 390
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 9, 2008
143 (100 pg/µL), used here as syringe standard to determine the recovery of the internal standard for quality control purposes. Analysis. For the analysis of MeO-PCBs, an Agilent 6890 GC (equipped with a programmable-temperature vaporizer) connected with an Agilent 5973 MS operated in electron capture negative ionization (ECNI) mode was equipped with a 20 m × 0.18 mm × 0.20 µm AT-5 capillary column (Alltech, Lokeren, Belgium). Methane was used as reagent gas, and the ion source, quadrupole, and interface temperatures were set at 170, 150, and 300 °C, respectively. A 10 µL amount of the cleaned extract was injected in solvent vent mode (injector temperature 90 °C (0.22 min) then to 300 °C with 700 °C/ min), vent time 0.20 min, vent flow 100 mL/min, purge time 1.50 min. Helium was used as carrier gas at constant flow (0.8 mL/min). The temperature of the AT-5 column was held at 90 °C for 1.50 min, then increased to 180 °C at a rate of 15 °C/min, further increased to 255 °C at a rate of 3 °C/min, further increased to 300 °C at a rate of 30 °C/min, and then held for 6 min. Analytical Characteristics. Method Specificity. The MS system was used in selected ion monitoring (SIM) mode with minimum two specific ions monitored for each HO-PCB congener and PCP in specific windows, while ions m/z ) 79 and 81 were monitored for TBPs and TBBP-A during the entire run (Table 1). The most abundant ions were chosen for each methoxylated derivative of the phenolic compounds, which corresponded to the ion cluster [M] in the case of 4′-HO-CB127 and 4′-HO-CB159, [M-CH3] for the following derivatives: 4-HO-CB107, 4-HO-CB146, 4′-HO-CB130, 4-HOCB163, 4-HO-CB187, 4′-HO-CB172, 4-HO-CB193, and [M-Cl] for 3-HO-CB118, 3-HO-CB153, 3′-HO-CB138, 3′-HO-CB180. For partial coelutions between HO-pentaCB and HO-hexaCB or between HO-hexaCB and HO-heptaCBs (all MeO-PCBs), the ions which were absent in the spectra of the higher chlorinated HO-PCBs were chosen for the analysis of the lower chlorinated HO-PCBs. PBDEs could not be completely separated during the partitioning with an alcoholic solution of KOH. Therefore, BDE 154, which partly coeluted on the AT-5 capillary column with dimethylated TBBP-A, was still present in the phenolic
TABLE 2. Median Concentrations and Range of Phenolic Organic Compounds (in pg/g ww) and Sum PCBs (in ng/g ww (*)) in Liver Samples of Predatory Birds from Belgium grey heron (n ) 7)
barn owl (n ) 5)
long eared owl (n ) 3)
tawny owl (n ) 1)
common buzzard (n ) 15)
sparrowhawk (n ) 5)
Lipid %
2.5–3.7
3.5–6.6
3.3–4.2
4.1
0.5–6.1
1.1–4.1
HO-T4CB 3-HO-CB118 4-HO-CB107 3-HO-CB153 4-HO-CB146 4′-HO- CB127 3′-HO-CB138 4′-HO-CB130 4-HO-CB163 HO-H7CB 4-HO-CB187 3′-HO-CB180 4′-HO-CB172 4′-HO-CB193 Σ HO-PCBs
10 35 30 35 700
