Environ. Sci. Technol. 2008, 42, 2650–2655
Hexabromocyclododecane in White-Sided Dolphins: Temporal Trend and Stereoisomer Distribution in Tissues A A R O N M . P E C K , * ,†,§ R E B E C C A S . P U G H , † AMANDA MOORS,† MICHAEL B. ELLISOR,† BARBARA J. PORTER,‡ PAUL R. BECKER,† AND JOHN R. KUCKLICK† National Institute of Standards and Technology, Hollings Marine Laboratory, 331 Fort Johnson Road, Charleston, South Carolina 29412, and National Institute of Standards and Technology, 100 Bureau Drive, Gaithersburg, Maryland 20899
Received August 16, 2007. Revised manuscript received December 14, 2007. Accepted January 07, 2008.
Hexabromocyclododecane (HBCD) is a brominated flame retardant used primarily in expanded polystyrene foams and other styrene resins. Samples of blubber (n ) 57) and liver (n ) 16) from Atlantic white-sided dolphins, Lagenorhynchus acutus, that stranded on the eastern coast of United States between 1993 and 2004 were obtained from the National Marine Mammal Tissue Bank (NMMTB). Blubber samples from most of these animals (n ) 47) were previously analyzed for polybrominated diphenyl ethers (PBDE), polychlorinated biphenyls (PCB), and several toxaphene congeners. The three most abundant diastereomers in the technical HBCD mixture (RHBCD, β-HBCD, and γ-HBCD) and their enantiomers were determined using liquid chromatography-triple quadrupole mass spectrometry (LC-MS/MS). R-HBCD was found in all blubber and liver samples while β-HBCD and γ-HBCD were not detected in any samples. The R-HBCD concentration in blubber and liver ranged from 14 ng/g wet mass (19 ng/g lipid) to 280 ng/g wet mass (380 ng/g lipid) and 0.051 ng/g wet mass (2.9 ng/g lipid) to 3.6 ng/g wet mass (140 ng/g lipid), respectively. Concentrations of R-HBCD were 2 to 3 orders of magnitude lower than previously reported PBDE, PCB, and toxaphene concentrations in these same animals. There was not a significant temporal trend for these compounds in whitesided dolphin blubber. The enantiomeric fractions (EF) measured in blubber and liver were not statistically different and ranged from 0.34 to 0.53. Blubber EFs were significantly correlated with both R-HBCD concentrations and whitesided dolphin body length. In general, concentrations of HBCDs were lower in these white-sided dolphins than in cetaceans from Western Europe. * Corresponding author phone: 912.598.2315; fax: 912.598.2310; e-mail:
[email protected]. † National Institute of Standards and Technology, Hollings Marine Laboratory. ‡ National Institute of Standards and Technology, Gaithersburg, Maryland. § Current address: Skidaway Institute of Oceanography, 10 Ocean Science Circle, Savannah, GA 31411. 2650 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008
Introduction Hexabromocyclododecane (1,2,5,6,9,10-hexabromocyclododecane; HBCD) is a brominated flame retardant used primarily in expanded and extruded polystyrene foams and textiles (1, 2). The technical mixture consists predominantly of three diastereomers: γ-HBCD (≈82%) with smaller amounts of R-HBCD (≈12%), β-HBCD (≈6%), and other compounds (3). Of the three major diastereomers found in technical HBCD, each has two enantiomers that can be chromatographically separated (4). In 2000, approximately 600000 t of polystyrene foam containing from 0.8% to 4% technical HBCD was produced (5). Europe (9500 t), the United States (2800 t), and Japan (2200) used most of the HBCD in 2001 (2). The difference in use between the European Union and the United States is partially due to restrictions placed on the use of polybrominated diphenyl ethers (PBDE) in the European Union (6, 7). HBCD is now used in place of PBDE for some applications in the European Union (6, 7). A similar shift may occur in the United States, as the use of PBDEs is restricted there. Covaci and co-workers recently reviewed the distribution and trends of HBCD in the environment and in humans (7). Relative to PBDEs, there are few reports of environmental concentrations or temporal trends of HBCD, especially in North America (7). HBCD has been shown to bioaccumulate (4, 8–13) and biomagnify (14) in aquatic organisms. These compounds have also been detected in eggs of peregrine falcons, Falco peregrinus (15, 16), guillemots, Uria aalge (17, 18), little owls, Athene noctua (19), and in tissues of glaucous gulls, Larus glaucescens (20) and polar bears, Ursus maritimus (20). Concentrations in biota are largely dependent on their proximity to sources and range from below detection limits to ∼20 µg/g lipid weight. While γ-HBCD is the predominant diastereomer in the technical mixture, R-HBCD is found in higher concentrations in biota. In addition to shifts in diastereomer distributions in the technical HBCD mixture compared to those in biological samples, nonracemic HBCD distributions have been reported in liver and muscle tissue from several fish species collected in The Netherlands (4). Changes in enantiomer ratios indicate biological transformation, stereoselective accumulation, or a combination of these processes. Consequently enantiomeric distributions can be useful tools for studying processes controlling the food web transfer (21) and metabolism (22–27) of environmental organic contaminants. In the present study, 57 blubber samples and 16 liver samples from Atlantic white-sided dolphins, Lagenorhynchus acutus, archived by the National Institute of Standards and Technology (NIST) in the National Marine Mammal Tissue Bank (NMMTB) (28) were analyzed for HBCD. Concentrations of R-HBCD, β-HBCD, and γ-HBCD and their enantiomeric distribution in the tissues of these dolphins were evaluated. The primary objectives were to assess the temporal trend of HBCD concentrations in white-sided dolphin blubber, evaluate factors affecting these concentrations, compare HBCD concentrations to previously reported concentrations of PCBs, PBDEs, and toxaphene in these same animals (29, 30), and evaluate the HBCD enantiomeric distributions in blubber and liver to gain insight into metabolic and accumulation pathways for HBCD in these animals.
Experimental Section Sample Extraction and Preparation. All samples were collected from animals stranded in good condition on the eastern United States coast from 1993 to 2004 using 10.1021/es072052v
Not subject to U.S. Copyright. Publ. 2008 Am. Chem. Soc.
Published on Web 03/04/2008
established collection protocols (31) and stored at -150 °C or below in a clean-room environment. The samples, sampling methods, and age determinations have been described in detail previously (31). Blubber (n ) 57) and liver (n ) 16) samples from Atlantic white-sided dolphins were obtained from the NMMTB. All samples were cryogenically homogenized in a clean-room environment (ISO Class 5 clean room) using established protocols (31). Blubber subsamples collected from animals stranded between 1993 and 2000 (n ) 47) were previously analyzed for PCBs, toxaphene, and PBDEs (29, 30). The blubber samples from animals stranded between 2001 and 2004 (n ) 10) were not analyzed for these other halogenated compounds. More information about these samples is listed in Tables S-1 and S-2 in the Supporting Information. Five aliquots of NIST Standard Reference Material (SRM) 1945 Organics in Whale Blubber and replicate aliquots of three dolphin blubber samples were analyzed for quality control. Eight method blanks were also used to monitor potential sample contamination during processing. Each tissue sample (≈0.5 g of blubber or ≈3 g of liver) was mixed with ∼30 g of Na2SO4 and loaded into pressurized fluid extraction (PFE) cells. An internal standard solution containing 13C12-R-HBCD (≈5 ng), 13C12-β-HBCD (≈ 5ng), and 13C12-γ-HBCD (≈2 ng) (Wellington Laboratories, Guelph, ON) was added to each sample cell. Samples were extracted with dichloromethane at 100 °C and 13.79 MPa (2000 psi) in a pressurized fluid extractor (ASE-200, Dionex, Sunnyvale, CA). The programmed extraction heating and static times were both 5 min and three extraction cycles were used for each sample. Lipids and other matrix interferences were removed from each extract by gel permeation chromatography (GPC) using a 600 mm × 25 mm i.d. PLGel column and 110 mm × 25 mm i.d. PLGel guard column (10 µm particle size with 100 Å diameter pores; Polymer Laboratories, Amherst, MA) with 10 mL/min dichloromethane mobile phase. Interferences were removed in the first 173 mL of GPC eluent and the next 107 mL was collected. Prior to GPC, a subsample of each liver extract and blubber extract from animals stranded between 2001 and 2004 was used for gravimetric lipid content determination. The lipid content for animals stranded between 1993 and 2000 was determined previously using the same gravimetric method (29, 30). Following GPC, solid phase extraction using 5% deactivated alumina (Acros Organics, Geel, Belgium) was used to remove proteins and other potential matrix interferences. Extracts were loaded onto the alumina in ∼0.5 mL of isooctane. The HBCDs were eluted using 9 mL of 35% dichloromethane in hexane (volume fraction). Extracts were solvent exchanged into acetonitrile and reduced to ∼200 µL for analysis. Determination of Diastereomers. An Agilent 1100 series liquid chromatography system coupled to an API 4000 triple quadrupole mass spectrometer (Applied Biosystems, Foster City, CA) was used for all analyses. The column temperature was 27 °C, the autosampler temperature was 20 °C, and 10 µL injections were used. The liquid chromatography method was based on one reported previously (4). Separation of the three HBCD diastereomers was performed using a C18 column (Agilent ZORBAX Eclipse XDB; 250 mm × 5 mm; 5 µm) and a trisolvent system: methanol, acetonitrile, and water. A constant 300 µL/min flow rate was used. The initial solvent composition was 60% water/30% methanol/10% acetonitrile (volume fraction). The solvent mixture was changed linearly over 4 min to a final solvent composition of 50% methanol/50% acetonitrile. This was achieved using a binary pump with the solvent mixtures 50% methanol/ 50% acetonitrile (A) and 75% water/25% methanol (B). An initial flow composition of 20% A/80% B was ramped linearly over 4 min to 100% A. The final conditions were held for 10 min before a linear ramp over 4 min returned to the initial
solvent composition. The column was allowed to equilibrate for 7 min between analyses. Calibration and internal standard solutions were prepared in hexane from R-HBCD, β-HBCD, γ-HBCD, 13C12-R-HBCD, 13C -β-HBCD, 13C -γ-HBCD from commercial standard 12 12 solutions (Wellington Laboratories, Guelph, ON). Gravimetrically measured aliquots of each calibrant solution (≈1 mL) and the internal standard solution (≈1 mL) were extracted by PFE and processed in a similarly to the samples and the resulting solutions were used as calibrants. The r2 value for all calibration curves (five points) was >0.998. The calibration mass range was 0.320-101 ng for R-HBCD, 0.353-111 ng for β-HBCD, and 0.353-111 ng for γ-HBCD. The approximate calibration concentration range for blubber was 0.7-200 ng/g for each diastereomer. The approximate calibration concentration range for liver was 0.1-30 ng/g for each diastereomer. Determination of Enantiomers. The method for separation of the six HBCD stereoisomers was based on one reported previously (4). A cyclodextrin column (Phenomonex Nucleosil β-PM, 200 mm × 4 mm) and a solvent system consisting of water, methanol, and acetonitrile were used. A constant 500 µL/min flow rate was used. The initial solvent composition was 49% water/30% methanol/21% acetonitrile. The solvent mixture was changed linearly over 4 min to a final solvent composition of 30% methanol/70% acetonitrile. This was achieved using a binary pump with the solvent mixtures 30% methanol/70% acetonitrile (A) and 70% water/30% methanol (B). An initial flow composition of 30% A/70% B was ramped linearly over 14 min to 100% A. The final conditions were held for 6 min before a linear ramp over 7 min returned to the initial solvent composition. The column was allowed to equilibrate for 8 min between analyses. The enantiomeric distribution resulting from this analysis was described by the enantiomeric fraction (EF). The EF for each enantiomer pair was determined from the area of each enantiomer and corresponding labeled enantiomer in the chiral analysis (eq ); (+)area, (-)area, (+)areaIS, and (-)areaIS are the areas of the (+) and (-) enantiomer of each unlabeled and labeled HBCD to elute (32). EF )
(+)area ⁄ (+)areaIS (+)area ⁄ (+)areaIS + (-)area ⁄ (-)areaIS
Chromatograms of the enantiomeric HBCD separation using the Nucleosil β-PM column are shown for a calibration standard and a blubber sample in Figure 1. Both enantiomers from each enantiomeric pair were baseline separated, but slight interferences occurred between diastereomers. Nearbaseline resolution of all enantiomers was achieved for a mixture of all three diastereomers. Because of the absence of the β-diastereomer in blubber and liver samples, baseline resolution of the R-diastereomers was achieved in all samples.
Results and Discussion Quality Assurance/Quality Control. NIST SRM 1945 was analyzed with each batch of blubber samples. The average of five replicate measurements of R-HBCD in SRM 1945 was 6.0 ng/g wet mass ( 0.28 ng/g wet mass (mean ( standard deviation). This is consistent with the reference concentration for R-HBCD in SRM 1945 of 6.3 ng/g wet mass ( 1.8 ng/g wet mass (33). The relative standard deviation (RSD) for three replicate measurements of R-HBCD in a blubber sample (ID No. NM3B042) was 0.93% (360 ng/g lipid ( 3.4 ng/g lipid). The concentrations determined for duplicate measurements of R-HBCD in two blubber samples were 160 ng/g lipid and 180 ng/g lipid (ID No. NM10B107C) and 55 ng/g lipid and 56 ng/g lipid (ID No. NM8B235). HBCD Concentrations in Blubber and Liver. R-HBCD was quantified in all blubber and liver samples analyzed. VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2651
FIGURE 1. Chromatograms of the enantiomeric HBCD separation using the Nucleosil β-PM column. Identification of r (-) and r (+) is from Heeb et al. (3) and was confirmed for the solvent system used in this study as described in the text. The elution order for the enantiomers of β-HBCD and γ-HBCD were not verified using our solvent system. See Heeb et al. (3) for elution order of those enantiomers. β-HBCD and γ-HBCD were not detected (
