Temporal Trends of Hexabromocyclododecane, Polybrominated

---

ARTICLE pubs.acs.org/est

Temporal Trends of Hexabromocyclododecane, Polybrominated Diphenyl Ethers and Polychlorinated Biphenyls in Ringed Seals from East Greenland Katrin Vorkamp,*,† Frank F. Rig
et,‡ Rossana Bossi,§ and Rune Dietz‡ †

Department of Environmental Chemistry and Microbiology, National Environmental Research Institute (NERI), Aarhus University (AU), Frederiksborgvej 399, DK - 4000 Roskilde, Denmark ‡ Department of Arctic Environment, NERI, AU, Frederiksborgvej 399, DK - 4000 Roskilde, Denmark § Department of Atmospheric Environment, NERI, AU, Frederiksborgvej 399, DK - 4000 Roskilde, Denmark

bS Supporting Information ABSTRACT: Concentrations of hexabromocyclododecane (HBCD) were determined in a combination of archived and fresh blubber samples of juvenile ringed seals from East Greenland collected between 1986 and 2008. R-HBCD was the only diastereoisomer consistently above levels of quantification and showed a significant log-linear (exponential) increase from 2.0 to 8.7 ng/g lipid weight (median concentrations) with an annual rate of þ6.1%. The concentrations were up to several orders of magnitude lower than those reported for marine mammals from industrialized areas. Previously presented time trends on polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) have been extended with new data for 2006 and 2008. ΣPBDE in juvenile seals was the only parameter with a slight upward trend, however, dependent on the low 1986 concentration. Removing this data point resulted in a downward trend, which also was found for adult seals with a time trend starting in 1994. ΣPCB decreased significantly in juvenile seals, again due to the 1986 value, while no trend was found for the adult animals. This indicates stagnating PCB concentrations at a relatively high level, in some cases possibly exceeding tolerable daily intake rates for seal blubber as traditional Arctic food items.

’ INTRODUCTION The systematic monitoring of persistent organic pollutants (POPs) in Greenland was established in the early 1990s as part of the Arctic Monitoring and Assessment Program (AMAP) and has produced time series data for various combinations of compounds, species, and locations. Ringed seal (Pusa hispida) is a key species in this context as their POP body burden is generally high, as a result of bioaccumulation and biomagnification.1,2 As ringed seals occur throughout the Arctic, with limited migration, they have also been used in circumpolar assessments of the spatial distribution of POPs. High POP concentrations in blubber are also of concern from a human exposure point of view, blubber being part of the Arctic traditional diet. Organochlorine compounds (OCs) have been monitored in ringed seals from East Greenland since 1994 and have generally been found to decrease or stagnate, following bans and restrictions of OCs in most countries since the 1970s.1 As results from the Canadian Arctic indicated major concentration changes prior to the main monitoring activities in the 1990s,3 the Greenland time series were extended backward using archived samples.1,2 Today, the Stockholm Convention on POPs regulates twelve organochlorine compounds and compound groups, with the objective for parties to take measures to eliminate and reduce the release of POPs. Polybrominated diphenyl ethers (PBDEs) have been a research focus for 20 years due to their high volume production, ubiquitous occurrence in the environment, and their endocrine disruptive potential. The three technical mixtures are PentaBDE, OctaBDE, r 2011 American Chemical Society

and DecaBDE, named after the predominant homologue group, of which the Penta- and OctaBDE congeners were included in the Stockholm Convention in 2009. PBDEs are present in the Greenland environment, and mean concentrations up to 40 ng/g lipid weight (lw) were determined for ΣPBDE in juvenile ringed seals from East Greenland, but with no significant time trend.1,4 In West Greenland, the concentration in ringed seals increased significantly between 1982 and 2006.2 Different intercontinental application patterns and PBDE regulations in Europe might have had an effect on PBDE emissions at lower latitude, with consequences for the long-range transport to Greenland and the exposure of Greenland wildlife.2 Hexabromocyclododecane (HBCD) has mainly been used as an additive in textiles and polystyrene products. Its estimated worldwide usage increased by 32% between 2001 and 2003, and in contrast to the PBDE application pattern, the highest amounts were used in Europe.5 At present, there is no control on the production and use of HBCD worldwide,6 but voluntary emission reductions have been reported in Europe.7 Technical HBCD consists of approximately 12% R-, 6% β-HBCD, and 80% γ-HBCD.8 With a logKOW value of 5.6, technical HBCD has a high bioaccumulative potential, and there are indications of biomagnification comparable Received: August 12, 2010 Accepted: December 22, 2010 Revised: December 14, 2010 Published: January 19, 2011 1243

dx.doi.org/10.1021/es102755x | Environ. Sci. Technol. 2011, 45, 1243–1249

Environmental Science & Technology

ARTICLE

with or even exceeding that of BDE-47.9,10 HBCD data for the Arctic were reviewed recently, with time trend studies described as one of the present knowledge gaps.5 HBCD has previously been included in a time trend study on peregrine falcon (Falco peregrinus) eggs from Greenland but likely reflected an integrated exposure situation of the breeding and winter grounds of these migratory birds.11 For an assessment of HBCD in the Greenland environment, the temporal development of HBCD in ringed seals from Central East Greenland since the 1980s has been studied. New data have also become available on PBDEs and OCs to update previously published time series.1 Of the different OC compounds, PCBs have been selected for this update and comparison of time trends.

Statistical Analysis. Concentrations below Limit of quantification (LOQ) were treated as zero in calculations of mean and median. The analyses of temporal trends followed the procedure used by ICES for temporal trend assessments. The method is a robust regression-based analysis to detect temporal trends13 and uses the median as yearly contaminant index value. The total variation over time is divided into a linear and a nonlinear component. Log-linear regression analysis was applied to describe the linear component and a loess smoother (locally weighted regression smoother) with a window width of 7 years was applied to describe the nonlinear component.14,15 The linear and nonlinear components were tested by ANOVA. The statistical trend analyses were performed using the free software R, version 2.8.1.16

’ EXPERIMENTAL SECTION

’ RESULTS AND DISCUSSION

Samples. The ringed seal samples originated from Ittoqqortoormiit in Central East Greenland (70°280 N, 21°950 E). They were obtained from local hunters and shipped to Denmark for age determination and chemical analyses as described previously.1 HBCD was determined in archived and recently collected blubber samples. In all cases, new subsamples were taken and homogenized prior to analysis. A total of 50 samples were analyzed with 5 samples from each of the analysis years (1986, 1994, 1999-2004, 2006, 2008) (Table S1, Supporting Information). In general, juvenile animals were chosen for analysis as the gender specific compound patterns seen among adults were not yet expected to be present. For 2008, one older individual was included (Table S1, Supporting Information). Chemical Analysis. The PBDE and OC analysis followed previously described methods2 which remained largely unchanged throughout the entire time series. Approximately 0.5 g of homogenized blubber was dried with diatomaceous earth, spiked with the recovery standards CB-3, CB-40, CB-198, BDE-77, and 13Ctranschlordan, and Soxhlet extracted with hexane:acetone (4:1). The preconcentrated extracts were cleaned on a multilayer column consisting of aluminum oxide, silica, H2SO4-impregnated silica, and sodium sulfate. The compounds were eluted with hexane and concentrated to 1 mL. The same extract was used for PBDE and PCB analysis by gas chromatography (GC)-mass spectrometry (MS) with electron capture negative ionization and dual column GC-electron capture detection (ECD), respectively. A list of individual compounds and details on quality assurance are given in the Supporting Information. Samples from 1999 were analyzed by the National Laboratory for Environmental Testing (NLET) of Environment Canada.2 HBCD analysis was based on a previously described method,12 which has subsequently been modified. A separate blubber subsample was homogenized, and approximately 0.5 g was spiked with 20 ng of each 13C12-labeled HBCD diastereoisomer. The sample was Soxhlet extracted with 350 mL of hexane:acetone (4:1) and preconcentrated by rotary evaporation. The same cleanup columns were used as for PBDE and PCB analysis but eluted with hexane: dichloromethane (1:1). The extracts were evaporated to dryness by rotary evaporation and under nitrogen and reconstituted to 500 μL using methanol. Diastereoisomer-specific analysis of HBCD was performed by high performance liquid chromatography (HPLC)MS-MS.12 Electrospray ionization (ESI) in negative mode was used for ionization of the analytes. The transition ions monitored were 641/79 and 641/81 for native HBCDs; transitions 653/79 and 653/ 81 were monitored for 13C12-labeled HBCDs. Quantification was based on a double seven point calibration curve.

HBCD Concentrations in a Global Context. R-HBCD was the only diastereoisomer generally above LOQ, at concentrations between 90% of ΣHBCD (e.g. refs 17 and 18). This has been related to bioisomerization, differences in water solubility and thus bioavailability, and to the rapid metabolization of the β- and γ-diastereoisomers.8,20,21 However, comparable percentages of R-HBCD and γ-HBCD in air and dust8 indicate additional abiotic factors. Some thermal rearrangement toward R-HBCD might also occur during production of flame-retarded materials, possibly leading to emission patterns that are different from the original technical mixture.8 Recent reviews have confirmed the ubiquitous presence of HBCD in the Arctic, probably as a consequence of long-range atmospheric transport.5,8 HBCD has been detected in air, the freshwater environment, and in virtually all levels of the marine food web.5 Results on HBCD concentrations in ringed seals are available from previous studies in Greenland,12,21 Svalbard,22 and the Canadian Arctic.23 Ringed seals from East Greenland collected in 2002 had a mean concentration of R-HBCD of 19 ( 2 (standard error) ng/g lw, at a mean age of 5.4.21 Samples from the same year had a mean concentration of only 6.4 ( 1.1 ng/g lw in our study. Another study on ringed seals from East Greenland found comparable HBCD-concentrations of 5.8 ( 1.1 ng/g lw, although determined differently, i.e. by GC-MS.12 The same study yielded concentrations of 1.7 ( 0.22 ng/g lw for ΣHBCD in ringed seals from West Greenland, i.e. approximately 4 times below the East Greenland concentrations.12 The mean HBCD concentrations in ringed seals from Svalbard collected in 2003 was 19.6 ng/g lw and exceeded our levels by approximately a factor of 4.22 Concentrations were below the LOQ of 0.1 ng/g lw in the Canadian samples.23 Although the circumpolar HBCD data set is very limited, there might be indications of the same spatial pattern as that established for most POPs, i.e. highest concentrations in the European Arctic, in this case Svalbard and East Greenland, a clear East Greenland > West Greenland gradient and further decreasing concentrations in the Canadian Arctic.5,24 Values from more industrialized areas exceed the levels in Arctic wildlife by several orders of magnitude. However, large geographical differences in HBCD concentrations of marine mammals have been found among industrialized regions, probably due to 1244

dx.doi.org/10.1021/es102755x |Environ. Sci. Technol. 2011, 45, 1243–1249

Environmental Science & Technology

ARTICLE

Table 1. Temporal Trend Statistics for HBCD, PBDEs, and PCBs in Ringed Seals from East Greenlanda parameter

annual change (%)

p (log-linear)

p (nonlinear)