Trophodynamics of Some PFCs and BFRs in a Western Canadian

May 4, 2009 - Canada, and School of Earth and Ocean Sciences, University ... components of a marine food web from the western Canadian. Arctic...
0 downloads 0 Views 372KB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 4076–4081

Trophodynamics of Some PFCs and BFRs in a Western Canadian Arctic Marine Food Web ,†,‡

†

GREGG T. TOMY,* KERRI PLESKACH, STEVE H. FERGUSON,† JONATHON HARE,† GARY STERN,† GORDIA MACINNIS,§ CHRIS H. MARVIN,§ AND LISA LOSETO| Fisheries and Oceans, Canada, Arctic Aquatic Research Division, Winnipeg, Manitoba R3T 2N6 Canada, Department of Environment and Geography, University of Manitoba, Winnipeg, Manitoba R3T 2N2 Canada, Environment Canada, Water Research Directorate, Burlington, Ontario L7R 4A6 Canada, and School of Earth and Ocean Sciences, University of Victoria, Victoria, British Columbia V8W 3 V6 Canada

Received January 17, 2009. Revised manuscript received April 15, 2009. Accepted April 22, 2009.

The trophodynamics of per- and polyfluorinated compounds and bromine-based flame retardants were examined in components of a marine food web from the western Canadian Arctic. The animals studied and their relative trophic status in the food web, established using stable isotopes of nitrogen (δ15N), were beluga (Delphinapterus leucas) > ringed seal (Phoca hispida) > Arctic cod (Boreogadus saida) > Pacific herring (Clupea pallasi) ≈ Arctic cisco (Coregonus autumnalis) > pelagic amphipod (Themisto libellula) > Arctic copepod (Calanus hyperboreus). For the brominated diphenyl ethers, the lipid adjusted concentrations of the seven congeners analyzed (Σ7BDEs: -47, -85, -99, -100, -153, -154, and -209) ranged from 205.4 ( 52.7 ng/g in Arctic cod to 2.6 ( 0.4 ng/g in ringed seals. Mean ∑7BDEs concentrations in Arctic copepods, 16.4 ng/g (n ) 2, composite sample), were greater than those in the top trophic level (TL) marine mammals and suggests that (i) Arctic copepods are an important dietary component that delivers BDEs to the food web and (ii) because these compounds are bioaccumulative, metabolism and depletion of BDE congeners in top TL mammals is an important biological process. There were differences in the concentration profiles of the isomers of hexabromocyclododecane (HBCD) in the food web. The most notable difference was observed for beluga, where the R-isomer was enriched (accounting for ∼90% of the ΣHBCD body burden), relative to its primary prey species, Arctic cod, where the R-isomer accounted for only 20% of the ΣHBCD body burden (β: 4% and γ: 78%). For the C8-C11 perfluorinated carboxylic acids, the trophic magnification factors (TMFs) were all greater than unity and increased with increasing carbon chain length. PFOS and its neutral precursor, PFOSA, also had TMF values greater than one. There were also pronounced differences in the PFOSA to PFOS ratio in ringed seal (0.04) and

* Corresponding author phone: 204-983-5167; fax: 204-984-2403; e-mail: [email protected]. † Arctic Aquatic Research Division. ‡ University of Manitoba. § Environment Canada. | University of Victoria. 4076

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 11, 2009

in beluga (1.4) and suggests that, in part, there are differences in the efficacy of biotransforming PFOSA by whale and seal top predators that both preferentially feed on Arctic cod.

Introduction The circumpolar Arctic is now widely recognized as a repository for anthropogenic organic compounds. Persistent chemicals produced in regions far removed from the Arctic can be delivered to the region via the atmosphere either in the gaseous phase and/or bound to particulates or, if nonvolatile, by ocean currents (1). Prevailing conditions such as reduced levels of solar radiation and relatively low annual temperatures make the Arctic an ideal environment for persistent chemicals to remain chemically unchanged for long-periods which, in turn, can lead to accumulation in wildlife. Chemicals of emerging concern (CECs) is a term used to describe chemicals for which there is little information available on their environmental distribution, fate, and persistence. Considering that there are over 20 000 chemicals in commerce today, it is perhaps not too surprising that the list of CECs has the potential to be quite extensive. Based on empirical physical and chemical properties, Muir and Howard (2007) and Brown and Wania (2008) have compiled a comprehensive list of potential CECs (2, 3). In the current study we investigate the bioaccumulation processes and pathways of two classes of CECs, the fluorine-based surfactants and the bromine-based flame retardants, in a marine food web from the Beaufort Sea in the western Canadian Arctic. From an ecological perspective, the western Canadian Arctic is a unique geographic region because it is undergoing rapid environmental and industrial changes such as sea ice reduction and petroleum exploration (4). The effect of these stressors on food web communities, structure and contaminant processes are unknown. This region provides habitat to high trophic level (TL) consumers such as the Beaufort Sea beluga whale (Delphinapterus leucas) and the western Arctic ringed seal (Phoca hispida). Both beluga and ringed seal are important country food species to the local Inuvialuit. The beluga population is thought to predominantly prey on Arctic cod (Boreogadus saida) in addition to near shore forage fish and ringed seal are thought to incorporate both Arctic cod and crustaceans such as Themisto libellula in their diet (5, 6). The impetus for the work described here was borne out of results from our earlier studies on the behavior of brominated and fluorinated compounds in a food web from the eastern Canadian Arctic (7, 8). Two findings from those studies were particularly noteworthy: (i) lower TL organisms such as zooplankton and Arctic cod contained greater concentrations of brominated diphenyl ethers (BDEs) than higher TL animals which is in contrast to what has been observed in lower TL animals from Scandinavian Arctic waters and (9, 10), (ii) perfluorooctane sulfonate (PFOS) was shown to biomagnifiy throughout the entire food web based on a trophic magnification factor (TMF) of 3.1. Taken together, those observations raise another research question: is the trophodynamics of BFRs and per- and polyfluorinated compounds (PFCs) consistent across different geographical regions of the Canadian Arctic? In our view, we felt that the food web from the western Arctic examined in this study would provide insights into this. 10.1021/es900162n CCC: $40.75

 2009 American Chemical Society

Published on Web 05/04/2009

Materials and Methods Chemicals. The full list of compounds, native and mass labeled, used in this study can be found in Supporting Information Table SI1. Standard reference materials of fish (SRM, WMF-01) and beluga (SRM, 1941) were obtained from Wellington Laboratories (Guelph, ON) and the National Institute for Science and Technology (NIST, Gaithersburg, MD), respectively. Distilled in glass grade isooctane, isopropanol, dichloromethane (DCM), hexane, and acetone were obtained from Caledon (Edmonton, AB). Optima grade water and methanol were obtained from Fisher Scientific (Nepean, ON). Samples. The animals selected were from the sample archived repository at Fisheries and Oceans, Canada. Blubber and liver of beluga (n ) 10, all males,) from Hendrickson Island (69° 32′, 133° 36′) and ringed seal (n ) 10, all males) from Holman Island (70° 38′, 117° 43′) were collected in 2007 and 2004, respectively. Tissue samples were collected in the field, placed into separate WhirlPak bags and shipped to the Institute on ice where they were transferred to -30 °C freezers. Morphometric information such as age, length, sex, blubber thickness, and weight were available for most of the animals. Three fish species and two zooplankton species were collected from Amundsen Gulf and the Mackenzie Delta within the eastern Beaufort Sea ecosystem. Fish species collected in 2004 and 2005 included the marine pelagic Arctic cod (n ) 10, mean size: 15 cm) from the Amundsen Gulf, the marine coastal Pacific herring (Clupea pallasi, n ) 10, mean size: 21 cm) from the Mackenzie Shelf and the anadromous Arctic Cisco (Coregonus autumnalis, n ) 9, mean size: 25 cm) from the Mackenzie estuary. The marine pelagic amphipod Themisto libellula (pooled samples, n ) 2) and the marine Arctic copepod Calanus hyperboreus (pooled samples, n ) 5) were collected in 2004 from the eastern Beaufort Sea and Amundsen Gulf region. Fluorinated chemicals were measured in subsamples of the livers of the marine mammals and pelagic fish and in pooled whole-body composites of T. libellula and C. hyperboreus. The brominated compounds were measured in the blubber of ringed seal and beluga, in the whole organism minus liver for the pelagic fish, and pooled composites for the invertebrates. Extraction and Clean-Up. Details of the sample workup for the analysis of BFRs and PFCs were as previously published (7, 8, 11) and as described in the Supporting Information (SI) section. Instrumental Analysis. BDEs were analyzed by high resolution gas chromatography electron capture negative ion mass spectrometry and HBCD and the PFCs by high performance liquid chromatography tandem mass spectrometry (HPLC/MS/MS), as described previously (7, 8, 12, 13) and in the SI. Quality Assurance/Quality Control (QA/QC). Detailed information on our QA/QC approach can be found in the SI. Stable Isotope Analysis. Stable isotopes of nitrogen, expressed as δ15N were analyzed at the stable isotope laboratory at the University of Winnipeg (Manitoba, Canada; University of Winnipeg Isotope Laboratory, UWIL) (see ref 5). Carbon and nitrogen isotopic analyses on the muscle were achieved by continuous flow, isotopic ratio mass spectrometry using a GV-Instruments IsoPrime attached to a peripheral, temperature-controlled, EuroVector elemental analyzer. Data Analysis. Owing to the highly significant positive relationship between lipid content and wet weight contaminant concentrations for BFRs in the samples (∑3HBCD and ∑7BDEs; r2 ) 0.563 and 0.190, respectively, p < 0.002 in both cases), lipid normalized data are presented and included in calculations. Per and poly-fluorinated compound concentrations are presented on a wet weight basis. Trophic levels (TL), trophic magnification factors (TMFs), and TL-adjusted

biomagnification factors (BMFTLs) were calculated using previously published eq (8). Subscripts for whole organism (w), blubber (b), and liver (l) are added to define what tissues are being compared in the BMFTL calculation. As an example, BMFTL(w:l) would represent the TL-adjusted ratio of the concentration in whole-organism of the predator to that of the concentration in the liver of its prey. Statistical analysis of the data was done using SigmaStat (Version 9.01, Systat Software Inc.) and outliers in the data set determined using the Q-test (14). Analysis of variance (ANOVA) and two-tailed t test were used to test for differences in concentrations among species. Pearson correlation were used to test for relationships among BFR and PFC congeners as well as with biological parameters (e.g., age, size).

Results and Discussion (a) δ15N and Food Web Structure. δ15N results for the fish and zooplankton subsets were ordered as reported for the larger ecosystem study whereby the copepod C. hyperboreus was at the bottom of the food web representing the herbivorous trophic position feeding on phytoplankton (SI Table SI2) (5). T. libellula followed with higher δ15N, and is a known predator of herbivorous copepods such as C. hyperboreus (15). Among the fish species, Arctic cod had significantly higher δ15N (p < 0.05), supporting a diet at a higher TL relative to Pacific herring and Arctic cisco. The greater δ15N (geometric mean (GM) ( 1 × standard error (SE): 16.5 ( 0.12 ‰) in beluga relative to ringed seal (15.6 ( 0.21) implies that beluga whales feed at a slightly higher TL. Because the difference in δ15N values among the top predators of this food web were e3‰, this indicates that they both fed at a similar TL, or some combination of TLs (16). (b) Concentrations and Profiles of PFCs in Biota. SI Table SI2 presents the wet weight concentrations of PFCs in animals from the food web examined in this study. Total (Σ5) concentrations of PFCAs in beluga liver [GM ( SE: 43.6 ( 5.3 ng/g] were greater than in any of the other animals studied. Based on the median values, the rank order of the concentrations of individual PFCAs in beluga were PFDA > PFUA > PFNA > PFOA > PFDoDa. Concentrations of Σ5PFCA in ringed seal were significantly different (ca. 1.4 smaller) to those in beluga (p < 0.05) with concentrations again driven largely by PFNA, PFDA, and PFUA. Concentrations of Σ5PFCA in the three fish species were not significantly different (p < 0.05); ranging from 6.9 ( 2.5 ng/g in Pacific herring to 4.7 ( 1.2 ng/g in Arctic cisco. Despite similar concentrations of Σ5PFCA, the order of concentrations of individual PFCAs differed among species. Σ5PFCA in T. libellula were similar to concentrations in fish and were just over double the concentration measured in C. hyperboreus (5.7 ( 2.6 and 2.0 ( 0.2 ng/g, respectively). PFOS was the dominant PFCs measured in ringed seals and T. libellula while in Arctic cod and pacific herring PFOS was undetectable. In beluga, median wet weight concentrations of PFOSA were ca. 1.4 times greater than that of PFOS. Conversely, in ringed seal, median concentrations of PFOSA were significantly smaller (ca. 25 times, p < 0.05) than that of PFOS (p < 0.05). In addition, PFOSA concentrations in ringed seal (range: 0.2-1.1 ng/g) were significantly different (ca. 25 times smaller, p < 0.001)) than in beluga (range: 7.8-24.5 ng/g). The small amounts of PFOSA in ringed seal from this study is consistent with the observations made by Butt et al. (2008) for animals collected in 2005 from the same location (17). An earlier study reported on concentrations of ΣPFCs in a truncated food web from the same region collected in 2004 (18). In invertebrates, Powley et al. PFOS was detected in only one T. Libellula sample that was ca. two times smaller than what we measured in our study (18). PFOS was also VOL. 43, NO. 11, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4077

detectable in their Arctic cod samples (whole-body determination) while in our study PFOS was undetectable in the liver. PFOA was detected in 90% of our Arctic cod samples, whereas Powley et al. were unable to measure PFOA in any of their Arctic cod samples (18). Reasons for this discrepancy remain unclear. There was, however, generally good agreement between liver-based PFOS concentrations in ringed seals in both studies. Butt et al. reported on concentration of PFOS in ringed seals from Sachs Harbor collected in 2005 that were ca. 1.7 times smaller than what was measured in this study (17). The inputs or loadings of a chemical to a remote region like the Canadian Arctic may vary with time. As such, differences in sampling years can influence the interpretation of food web transfer processes. However, every attempt was made in this study to minimize this potential temporal variation by selecting animals from our archive that were collected within a 3 year time-span. Other factors such as age, sex, lipid content, length, mass, and proximity to exposure of contaminants are also known to affect contaminant burdens in biota. Some of these variables were regressed against individual PFCs concentrations measured in beluga and ringed seal. The rationale for selecting these animals is that we have the most field morphometric information, i.e., variables, to make the test more meaningful. Surprisingly, there were no correlations between adjacent PFCAs concentrations (e.g., PFOA/PFNA, PFNA/PFDA, etc.) in either of the species. The fact that fluorotelomer alcohols (FTOHs), the volatile precursors of PFCAs, photodegrade to adjacent chain-length PFCAs in known ratios provides us with a useful tool for discerning the importance of atmospheric transport in delivering these compounds to the Arctic. For example, the 8:2 FTOH photodegrades to produce near equimolar amounts of PFOA and PFNA (19). The fact that we do not see a correlation between adjacent PFCA isomers suggests that other delivery mechanisms of these compounds to the region are perhaps more important than that of atmosphere. A similar conclusion was made by Powley et al. (18). In ringed seal, there were no correlations between liverbased concentrations of PFCs and age, blubber thickness, weight or δ15N. There was a positive correlation between concentrations of PFDA and length (r ) 0.64, p < 0.001) which may support differences in dietary sources as ringed seal may feed differently in relation to size. This is substantiated by the positive correlation between δ15N and length (r ) 0.661, p < 0.05). PFNA concentrations were positively correlated to PFOS (r ) 0.637, p < 0.05) which was also noted by Powley et al. (18). There was also a strong negative correlation between liver concentrations of PFOS and PFOSA (r ) -0.758, p < 0.01) suggesting that PFOSA is a major contributor to PFOS concentrations in ringed seals. There were also no relation between length of beluga and δ15N to concentrations of PFC. Unlike ringed seal, PFOSA and PFOS concentrations were not correlated in beluga. This last observation coupled with two others: (i) both mammals occupy similar TLs and, (ii) the PFOSA/PFOS ratios in these top TL animals, 1.4 (beluga) and 0.04 (for ringed seal), lead us to speculate that the efficacy of biotransforming PFOSA into PFOS is greater in ringed seal than in beluga. However, the role of other PFOS-precursor compounds and their contribution to the body burdens of PFOS and differences in diet of these top TL animals also needs to be considered. The profile of PFCs in the animals generated using median concentration values are shown in Figure 1. At first glance, there appears to be a disproportion in PFOA burden in Arctic cod and beluga. That Arctic cod is the primary prey species of beluga and that PFOA accounts for ∼70% of the ΣPFCA liver burden in cod, whereas in beluga the PFOA contribution is 6% may seem puzzling. However, longer-chained PFCAs 4078

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 11, 2009

FIGURE 1. Profiles of PFCs in biota from the western Canadian Arctic. were infrequently detected in Arctic cod, and as such, PFOA accounted for the bulk of the PFCA body burden in this species. The same is not true in beluga where other PFCAs were routinely detected. In Pacific herring, PFOA was also the dominant PFCA followed by PFNA, whereas in Arctic cisco PFOA concentrations were much smaller. Pacific herring and Arctic cod spend more time in the marine system relative to the anadromous Arctic cisco that prefers freshwater and brackish environments. Differences in PFOA body-burdens among these species may support differences in sources/ transport in marine and freshwater based aquatic food webs. In our invertebrate samples, PFOA and PFNA contributed near equal amounts to the ΣPFCA whole-body burden. (c) Concentrations and Profiles of BFRs in Biota. SI Table SI3 presents the lipid adjusted concentrations of BFRs in animals from the food web examined in this study. The rank order of Σ7BDEs was: Arctic cod > cisco > C. hyperboreus > herring > beluga > T. libellula > ringed seal with a range of 205 ( 53 ng/g (Arctic cod) to 2.6 ( 0.4 ng/g (ringed seal). Among the fish, Arctic cod and Arctic cisco (99.2 ( 39.1 ng/g) had significantly greater Σ7BDE concentrations than Pacific herring (15.1 ( 7.9 ng/g, p < 0.05). Concentrations of Σ7BDE in C. hyperboreus (16.4 ng/g) were ca. 3 times greater than those measured in T. libellula (5.6 ng/g). Like the BDEs, median Σ3HBCD concentrations were greatest in Arctic cod (11.8 ng/g) relative to all the other animals. Median concentrations of Σ3HBCD in pacific herring (1.7 ng/g) were ca. 2 times greater than those observed in Arctic cisco (0.9 ng/g). In the top TL mammals, Σ3HBCD were ca. 1.7 times smaller in ringed seal (1.1 ng/g) than beluga (1.9 ng/g). Of particular interest was the detection of the β-isomer in our samples. The frequency of detection was slightly greater in the pelagic fish species than in the marine mammals supporting biotransformation of this isomer at higher TLs (see below for discussion). Concentrations of Σ7BDEs in blubber of beluga from this study (9.3 ( 1.9 ng/g) were similar to those found in animals from the eastern Canadian Arctic (12 ( 2 ng/g) (7) but much smaller than concentrations measured in male belugas from Norway (161 ( 23 ng/g) (9). Because the Σ7BDEs and TL of the animals from both geographic regions were similar (TL: eastern Arctic ) 3.9; western Arctic ) 3.8) it might be tempting to suggest that there are no broad spatial differences in BDE concentrations in the Canadian Arctic. However, Muir et al. measured BDEs in male belugas from Hudson Strait and eastern Hudson Bay that were ca. 2-3 times greater than the measured values in both our studies (20). Measured concentrations of Σ7BDEs in zooplankton in this study were ca. 5 times smaller than what was measured in the eastern Canadian Arctic (72.9 ( 10.1 ng/g) but similar to that measured by Morris et al. (Σ14BDEs: 18.6 ng/g) in a food web from Barrow Strait (Nunavut, Canada) (21). Conversely, Σ7BDEs concentrations in Arctic cod in this study were significantly greater (p < 0.05) than that measured in

FIGURE 2. Profiles of BDEs in biota from the western Canadian Arctic.

FIGURE 3. Profiles of HBCDs in biota from the western Canadian Arctic. animals from the eastern Canadian Arctic (23 ( 13 ng/g). Overall, BDE concentrations in invertebrates and Arctic cod from the Canadian Arctic are much greater than those reported in the Scandinavian Arctic (9, 10). The BDE congener profile in the animals generated using median concentration values is shown in Figure 2. With the exception of C. hyperboreus in which BDE-99 was the dominant congener, BDE-47 dominated the profiles of the animals. This was particularly true in the top TL animals like beluga and ringed seal, in which BDE-47 accounted for greater than 60% of the total BDE body burden. BDE-47 also contributed to greater than 50% of the total BDE burden in Arctic cod and T. libellula; in pacific herring, BDE-47 and -99 contributed similar amounts (∼35%) to the overall BDE burden. In our earlier study, the profile of BDEs in beluga from the eastern Canadian Arctic was also dominated by BDE-47 and -100 which is consistent with what we report here (7). Interestingly, the BDE profile in zooplankton from this study which was dominated by BDE-99 and -47 was strikingly different to that observed in animals from the eastern Canadian Arctic where the profile was dominated by BDE-153 and -209 (7). Different sources of BDEs to both regions may partly explain these findings. The diastereoisomer profile of HBCD in the animals generated using median concentration values is shown in Figure 3. Of particular note is the great disparity in the contribution of R-isomer relative to the other isomers in the top TL marine mammals. This is especially evident in beluga where the R-isomer accounts for ∼95% of the overall burden of HBCDs. Arctic cod, the primary prey species of beluga in our food web, has a HBCD profile dominated by the γ-isomer (>77%) and in beluga this isomer contributes to less than 5% of the total HBCD body burden is further evidence that beluga can bioprocess the γ- to the R-isomer. These findings are

consistent with laboratory-based studies by Zegers et al. and Law et al. where it has been shown that selective isomer metabolism and isomer conversion is possible in biota (22, 23). Our earlier work in the eastern Canadian Arctic also illustrated that the R-isomer dominated the HBCD profile in beluga (7). Interestingly, in that study we showed that Arctic cod contained appreciable amounts of the R-isomer which differs to what we observe in our current study. Like the BDEs, we speculate that differences in source loadings of HBCD to the Arctic likely contribute to the geographical dissimilarities observed in HBCD profiles and concentrations in animals. In ringed seal, there were strong positive relationships between animal length and weight and BDE-100 and -153 concentrations (p < 0.05 in all cases). Concentrations of the β- and γ-isomers of HBCD were also positively correlated in ringed seal and beluga (p < 0.01). Finally, significant positive relationships were observed with BDE-47 concentrations and those of BDE-100 and -154 suggesting similar bioaccumulation pathways for these compounds. Based on our earlier study, it was hypothesized that low TL organisms like zooplankton and Arctic cod were directly driving BDE concentrations in animals at higher TL. The results presented here lend further support to that hypothesis. The fact that concentrations of BDEs are smaller in the top TL animals relative to the invertebrates suggests that metabolic depletion of BDEs is an important clearance mechanism in top TL animals. It is still unclear if HBCD behaves similar to BDEs as we were unable to measure concentrations in zooplankton in this study. In our study from the eastern Canadian Arctic (7), HBCD concentrations in zooplankton and Arctic cod were smaller than those measured in beluga and suggest that perhaps bioaccumulation mechanisms at the base of the food web are different for these two groups of flame retardants. (d) Trophodynamics. The trophic transfer of the CECs studied was examined by (i) regressing the log-normalized concentrations of the individual compounds against TL and (ii) examining the BMFTL of individual predator/prey feeding relationships. For the PFCs, two approaches were adopted to examine the relationship between TL and PFC concentrations. In the first approach, we chose to examine the relationship using the measured tissue concentrations, i.e., liver concentrations in marine mammals and fish and in whole body of the invertebrates. Using this approach, strong statistically significant (p < 0.001 in all cases) positive relationships were observed for all the PFC studied (see SI Table SI4; PFDoDa were purposely omitted because of the large number of nondetects). In fact, there was a linear increase in PFCATMF values (C8-C10 chain length) with increasing carbon chain length (regression equation: TMF ) 2.115 × (no. carbon atoms) - 14.95, r2 ) 0.9841, p < 0.01). Inclusion of PFUA into the regression only weakened the statistical significance of this relationship (p ) 0.058). It should be cautioned that the large number of nondetects for PFDA and especially for PFUA in the two pelagic fish species, Pacific herring and Arctic cod, and substitution of nondetects with 1/2 MDL values, may influence the observed carbon chain length-TMF trends. The TMF values for PFOS and PFOSA in this study were 6.3 and 1.9, respectively. The PFOS TMF value is ca. 2 times greater than what we measured in our eastern Canadian Arctic food web (8) but similar to that measured in a Lake Ontario food web (TMF: 5.9) (24). Although only two geographic areas could be compared, we tentatively suggest that there might be spatial differences in the extent of PFOS magnification in the Canadian Arctic. Clearly, this is an area of research that requires further work. To our knowledge, this is the first report of a TMF value for PFOSA in the Arctic; Houde et al. recently reported on a PFOSA-TMF value of 5.0 in a bottlenose dolphin food web VOL. 43, NO. 11, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4079

from southeastern US (25). A TMF value of greater than 1 for PFOSA comes as a bit of a surprise as we have shown previously that this isomer is a metabolic precursor in fish and in top TL-ringed seal PFOSA is thought to be metabolized readily to PFOS (26). Based on our observations, we suggest that the overall uptake and tissue storage of PFOSA in organisms from this study is greater than the rate of metabolic depletion. A second approach to examine the trophic magnification of the PFC was to extrapolate measured liver concentrations in the marine mammals and the fish to whole-body concentrations. This approach was first proposed by Houde et al. who showed that for PFCs, liver-based TMFs can overestimate the whole-body based TMF values (25). Wholebody based beluga PFC concentrations for PFNA, PFDA, and PFOS were determined as described by Houde et al. using their measured PFC tissue concentration. For ringed seal, the total mass of the liver was estimated from the overall mass of the animals measured in the field (27, 28). Wholebody concentrations for PFNA, PFDA, and PFOS were then calculated using the relative PFC tissue distribution. (D. Muir and S. Sturman, personal communication). For fish, extrapolating liver-based concentrations to whole-body was done as described in the SI. The whole-body based TMF (wbTMFs) values are shown in SI Table SI4. In all cases, our calculated wbTMFs value exceeds that of the TMF value determined using the measured concentrations in the tissues analyzed. This would imply that the slopes of the lines used in calculating PFC wbTMFs are greater than those of the respective lines determined from the measured concentration data. Plausible reasons for this are that our fish liver to whole-body calculation underestimates the true whole-body burden concentrations and/or our whole-body marine mammal conversions are being overestimated. If the hypothesis that beluga have a reduced capacity to metabolize PFOSA to PFOS relative to ringed seal is true, then it is not surprising that all feeding relationships in which beluga is the predator, BMFTL(l:l) values are much greater than 1. For ringed seal, the lone BMFTL(l:l) for the ringed seal to Arctic cod feeding relationship is much smaller than unity. However, Arctic cod may be less important than T. Libellula as a diet item for ringed seal generally. A large PFOSA-BMFTL(l:l) was also observed in the beluga to Arctic cod feeding relationship from our study in the eastern Arctic food web (8). Admittedly, because PFOSA was at undetectable levels in Arctic cod in that study, a value equal to 1/2 the MDLs was used which may result in an artificially large BMFTL(l:l) value. Based on the regression analysis, none of the BFRs showed any statistically significant positive relationship with TL; BDE99 did show a negative relationship with TL (r2 ) 0.2078, p < 0.05, TMF ) 0.1) suggesting trophic dilution of this congener with increasing TL. Because of the large disparity in concentrations of BDE congeners in organisms at the base of our food web relative to the higher TL animals it is perhaps not too surprising that none of the BDE congeners magnified throughout the entire food web. For example, median concentrations of BDE-47 were ca. 100 greater in Arctic cod than ringed seal and in C. Hyperboreus the difference was ca. 3 times. BMFTL’s are presented in SI Table SI5. For the BFRs, feeding relationships involving the top TL-marine mammals, lipidbased BMFTL(b:w) values are typically smaller than 1 with the exception being BDE-85, -154, and R-HBCD for the beluga to pacific herring feeding relationship. In our eastern Arctic food web study, we also observed a high BMF value for BDE85 (BMFTL(b:w) ) 40) within the narwhal (a top TL predator) to Arctic cod feeding relationship (7). Perhaps the most compelling example of biomagnification observed in the current food web is between the Arctic cod to T. Libellula 4080

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 11, 2009

(where a concentration of 1/2 MDL value was used in the calculation) where BMFTL(:w:w) range from 4.5 (BDE-153) to 840 (BDE-85). Taken together, these BMFTL values suggest that metabolic depletion of BDEs and HBCD is likely taking place in the higher TL-animals, whereas in lower TL-animals these compounds are being accumulated and stored more quickly than they are being eliminated/metabolized. The objective of our study was to examine the trophodynamics of BFRs and PFCs in a food web from the western Arctic and to see if there were any similarities in the bioaccumulation tendencies of these compounds relative to an earlier study done in the eastern Canadian Arctic (7, 8). BDE concentrations in the lower TL animals were generally greater than those measured in the higher TL animals; a similar observation was made in our earlier study (7). Overall, the congener-specific bioaccumulation of BDEs was different in the two studies: BDE-47 and -209 were shown to have TMF values greater than 1 in the food web from the eastern Arctic, whereas TMF values were all smaller than 1 in the food web from the western Arctic. The bioprocessing of the HBCD isomers was strikingly similar in both studies, in particular, the contribution of the R-isomer to the ΣHBCD body burden increased significantly between the beluga and its primary prey species, Arctic cod. We also found evidence of biomagnification of the C8-C11 PFCA homologues in the current food web. Similar to the eastern Arctic food web, PFOS showed a TMF value greater than 1 and suggests a spatial wide biomagnification of PFOS in marine food webs in the Canadian Arctic.

Acknowledgments We are tremendously grateful to the Department of Indian and Northern Affairs, Canada and the Northern Contaminants Program (NCP) for funding the analysis portion of the study. Equal contribution to the project study design and writing by G.T.T. and L.L. is noted. We thank Dr. Jonathan Martin (University of Alberta, Edmonton, Canada) for helping with the liver to whole-body PFC calculations in fish. We are grateful to Derek Muir (Environment Canada, Burlington, Canada) and Sabrina Sturman (University of Guelph, Guelph, Canada) for their willingness to share their ringed seal PFC data. Lois Harwood (DFO, Northwest Territories) is thanked for collecting the ringed seal samples from Holman Island (2004). Beluga collections from Hendrickson Island was funded by the Fisheries Joint Management Committee.

Supporting Information Available The following tables are available: Table SI 1. List of native and mass-labeled compounds used in this study; Table SI2. Species, sampling year, δ15N [mean ( 1 × standard error (SE)], range and median concentrations (ng/g, wet wt) and frequency of detection (expressed as a percentage) of individual perfluorocarboxylates (PFCAs: C8-C12), total (∑) PFCAs (geometric mean ( 1 SE), PFOS and PFOSA in components of a western Arctic marine food web1; Table SI3. Species, sampling year, % lipid [mean ( 1 × standard error (SE)], range and median concentrations (ng/g, lipid weight) and frequency of detection (expressed as a percentage) of BDE congeners and HBCD diastereoisomers, total (∑) BDEs and HBCD (geometric mean ( 1 SE) in components of a western Arctic marine food web1,2; Table SI4. Trophic magnification factors (TMFs), lower-upper 95% confidence intervals and regression parameters for the CECs examined in components of a western Arctic food web1; Table SI5. Trophic level (TL) adjusted biomagnification factors (BMFTLs) for the CECs examined in this study. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Macdonald, R. W.; Barrie, L. A.; Bidleman, T. F.; Diamond, M. L.; Gregor, D. J.; Semkin, R. G.; Strachan, W. J.; Li, Y. F.; Wania, F.;

(2)

(3)

(4) (5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14) (15)

Alaee, M.; Alexeeva, L. B.; Backus, S.; Bailey, R.; Brewers, J. M.; Gobeil, C.; Halsall, C. J.; Harner, T.; Hoff, P. T.; Jantunen, L. M. M.; Lockhart, L. W.; Mackay, D.; Muir, D. C. G.; Pudykiewicz, J.; Reimer, K. J.; Smith, J. N.; Stern, G. A.; Schroeder, W. H.; Wagemann, R.; Yunker, M. B. Contaminants in the Canadian Arctic: 5 years of progress in understanding sources, occurrence and pathways. Sci. Total Environ. 2000, 254, 93–234. Muir, D. C. G.; Howard, P. H. Are there other persistent organic pollutants? A challenge for environmental chemists. Environ. Sci. Technol. 2007, 40, 7157–7166. Brown, T. N.; Wania, F. Screening chemicals for the potential to be persistent organic pollutants: a case study of Arctic contaminants. Environ. Sci. Technol. 2008, 42, 5202–5209. Serreze, M. C.; Holland, M. M.; Stroeve, J. Perspective of the Arctic’s shrinking sea-ice cover. Science 2007, 315, 1533–1536. Loseto, L. L.; Stern, G. A.; Deibel, D.; Connelly, T. L.; Prokopowicz, A.; Lean, D. R. S.; Fortier, L.; Ferguson, S. H. Linking mercury exposure to habitat and feeding behaviour in Beaufort Sea beluga whales. J. Mar. Syst. 2008, 74, 1012–1024. Holst, M.; Stirling, I.; Hobson, K. A. Diet of ringed seals (Phoca hispida) on the east and west sides of the Northwater Polyna, norther Baffin Bay. Mar. Mammal Sci. 2001, 17, 888–908. Tomy, G. T.; Pleskach, K.; Oswald, T.; Halldorson, T. H. J.; Helm, P. A.; Marvin, C. H.; MacInnis, G. Enantioselective bioaccumulation of hexabromocyclododecane and congener-specific accumulation of brominated diphenyl ethers in an Eastern Canadian Arctic marine food web. Environ. Sci. Technol. 2008, 42, 3634–3639. Tomy, G. T.; Budakowski, W. R.; Halldorson, T. H. J.; Helm, P. A.; Stern, G. A.; Friesen, K. J.; Pepper, K.; Tittlemier, S. A.; Fisk, A. T. Fluorinated organic compounds in an eastern Arctic marine food web. Environ. Sci. Technol. 2004, 38, 6475–6481. Wolkers, H.; Van Bavel, B.; Derocher, A. E.; Wiig, Ø.; Kovacs, K. M.; Lydersen, C.; Lindstro¨m, G. Congener-specific accumulation and food chain transfer of polybrominated diphenyl ethers in two Arctic food chains. Environ. Sci. Technol. 2004, 38, 1667–1674. Sørmo, E. G.; Salmer, M. P.; Jenssen, B.; Hop, H.; Baek, K.; Kovacs, K. M.; Lydersen, C.; Falk-Petersen, S.; Gabrielsen, G. W.; Lie, E.; Skaare, J. U. Biomagnification of polybrominated diphenyl ether and hexabromocyclododecane flame retardants in the polar bear food chain in Svalbard, Norway. Environ. Toxicol. Chem. 2006, 25, 2502–2511. Tomy, G. T.; Halldorson, T. H. J.; Tittlemie, S. A. Methanolic extraction of poly- and perfluorinated alkyl acids from biota. In Proceedings of the 25th International Symposium on Halogenated Persistent Pollutants, 2005; p 787-789. Tomy, G. T.; Halldorson, T. H. J.; Danell, R. W.; Law, K.; Arsenault, G.; Alaee, M.; MacInnis, G.; Marvin, C. H. Refinements to the diastereoisomer-specific method for the analysis of hexabromocyclododecane. Rapid Commun. Mass Spectrom. 2005, 19, 2819–2826. Budakowski, W. R.; Tomy, G. T. Congener-specific analysis of hexabromocyclododecane by high-performance liquid chromatography/electrospray tandem mass spectrometry. Rapid Commun. Mass Spectrom. 2003, 17, 1399–1404. Dean, R. B.; Dixon, W. J. Simplified statistics for small numbers of observations. Anal. Chem. 1950, 23, 636–638. Pakhomov, E. A.; Perissinotto, R. Trophodynamics of the hyperiid amphipod Themisto gaudichaudii in the south Georgia region

(16) (17)

(18)

(19)

(20)

(21)

(22)

(23)

(24) (25)

(26)

(27) (28)

during the late summer. Mar. Ecol.: Prog. Ser. 1996, 134, 91– 100. Cabana, G.; Rasmussen, J. B. Modelling food chain structure and contaminant bioaccumulation using stable nitrogen isotopes. Nature 1994, 372, 255–257. Butt, C. M.; Mabury, S. A.; Kwan, M.; Wang, X.; Muir, D. C. G. Spatial trends of perfluoroalkyl compounds in ringed seals (Phoca Hispida) from the Canadian Arctic. Environ. Toxicol. Chem. 2008, 27, 542–553. Powley, C. R.; George, S. W.; Russell, M. H.; Hoke, R. A.; Buck, R. C. Polyfluorinated chemicals in a spatially and temporally integrated food web in the Western Arctic. Chemosphere 2008, 70, 664–672. Ellis, D. A.; Martin, J. W.; De Silva, A. O.; Mabury, S. A.; Hurley, M. D.; Andersen, M. P. S.; Wallington, T. J. Degradation of fluorotelomer alcohols: a likely atmospheric source of perfluorinated carboxylic acids. Environ. Sci. Technol. 2004, 38, 3316– 3321. Muir, D. C. G.; Alaee, M.; Braune, B.; Butt, C. M.; Chan, L.; Helm, P. A.; Mabury, S. A.; Stock, N. L.; Tomy, G. T.; Wang, X. 2004, New Contaminants in Arctic biota. Department of Indian and Northern Affairs, Northern Contaminants Program, Ottawa, Canada. Morris, A. D.; Muir, D. C. G.; Teixeira, C.; Epp, J.; Sturman, S.; Solomon, K. R. Bioaccumulation and distribution of brominated flame retardants and current-use pesticides in an Arctic marine food-web. 28th Annual Meeting of the Society for Environmental Toxicology and Chemistry, Milwaukee, WI, November 2007. Zegers, B. N.; Mets, A.; van Brommel, R.; Minkenberg, C.; Hamers, T.; Kamstra, J. H.; Pierce, G.; Boon, J. P. Levels of hexabromocyclododecane in Harbor Porpoises and Common Dolphins from western European seas, with evidence for stereoisomerspecific biotransformation by cytochrome P450. Environ. Sci. Technol. 2005, 39, 2095–2100. Law, K.; Palace, V. P.; Halldorson, T. H. J.; Danell, R. W.; Wautier, K.; Evans, R. E.; Alaee, M.; Marvin, C. H.; Tomy, G. T. Dietary accumulation of hexabromocyclododecane diastereoisomers in juvenile rainbow trout (Oncorhynchus mykiss) I: bioaccumulation parameters and evidence of bioisomerization. Environ. Toxicol. Chem. 2006, 25, 1757–1761. Martin, J. W.; Whittle, D. M.; Muir, D. C. G.; Mabury, S. A. Perfluoroalkyl contaminants in a food web from Lake Ontario. Environ. Sci. Technol. 2004, 38, 5379–5385. Houde, M.; Bujas, T. A. D.; Small, J. M.; Wells, R. S.; Fair, P. A.; Bossart, G. D.; Solomon, K. R.; Muir, D. C. G. Biomagnification of perfluoroalkyl compounds in the Bottlenose dolphin (Tursiops truncatus) food web. Environ. Sci. Technol. 2006, 40, 4138– 4144. Tomy, G. T.; Tittlemier, S. A.; Palace, V. P.; Budakowski, W. R.; Braekevelt, E.; Brinkworth, L.; Friesen, K. J. Biotransformation of N-ethyl perfluoroooctane sulfonamide (N-EtPFOSA) by rainbow trout (Onchorhynchus mykiss) liver microsomes. Environ. Sci. Technol. 2004, 38, 758–762. Lydersen, C.; Ryg, M. S.; Hammill, M. O.; O’Briend, P. J. Oxygen stores and aerobic dive limit of ringed seals (Phoca hispida). Can. J. Zool. 1992, 70, 458–461. Gwynn, J. P.; Brown, J. E.; Kovacs, K. M.; Lydersen, C. The derivation of radionuclide transfer parameters for and doserates to an adult ringed seal (Phoca hispida) in an Arctic environment. J. Environ. Radioact. 2006, 90, 197–209.

ES900162N

VOL. 43, NO. 11, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4081