Tetrabromobisphenol-A and Hexabromocyclododecane in Birds from

Jul 12, 2010 - Levels and distribution of hexabromocyclododecane (HBCD) in environmental samples near manufacturing facilities in Laizhou Bay area, Ea...
2 downloads 7 Views 281KB Size
Environ. Sci. Technol. 2010, 44, 5748–5754

Tetrabromobisphenol-A and Hexabromocyclododecane in Birds from an E-Waste Region in South China: Influence of Diet on Diastereoisomer- and Enantiomer-Specific Distribution and Trophodynamics M I N G - J I N G H E , †,‡ X I A O - J U N L U O , * ,† L E - H U A N Y U , †,‡ J U A N L I U , †,‡ X I U - L A N Z H A N G , †,‡ S H E - J U N C H E N , † DA CHEN,§ AND BI-XIAN MAI† State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China, Graduate University of Chinese Academy of Sciences, Beijing, 100049, China, and Department of Environmental and Aquatic Animal Health, Virginia Institute of Marine Science, The College of William and Mary, Gloucester Point, Virginia 23062

Received May 4, 2010. Revised manuscript received June 25, 2010. Accepted June 29, 2010.

Tetrabromobisphenol-A (TBBPA) and three diastereomers of hexabromocyclododecane (HBCD) were examined in the muscles of six bird species and their diet collected from an e-waste recycling region (Qingyuan) in South China. Stable isotope ratios (δ15N and δ13C) were analyzed to measure the diet source and trophic levels (TLs) of the birds. The median TBBPA and HBCD concentrations in the birds ranged from 28 to 173 and not detectable to 1995 ng/g lipid weight, respectively. The diastereoisomeric pattern shows the predominance of R-HBCD in birds feeding in an aquatic environment and that of γ-HBCD in birds feeding in a terrestrial environment, whereas no clear preference for R isomer or γ isomer was found in birds that inhabited freshwater wetland. A significant positive correlation between δ13C and percentage of R-HBCD was observed, indicating the importance of diet exposure pathways in the determination of HBCD diastereoisomer pattern. The enantiomer fractions (EFs) for R-HBCD differed substantially between aquatic and terrestrial bird species with a significant enrichment of (+) R-HBCD enantiomer for aquatic birds and a preferential enrichment of (-) R-HBCD enantiomer for terrestrial birds. The similarity in diastereoisomer profiles of HBCD and the EFs of R-HBCD between prey (fish) and predator (Chinese pond heron) also suggested that dietary exposure is an important contributor for the observed diastereoisomer- and enantiomer-specific distribution of HBCD in birds. Trophic magnification was observed for R-HBCD and TBBPA as concentrations increased with the TLs of the birds defined by * Corresponding author phone: +86-20-85290146; fax: +86-2085290706; e-mail: [email protected]. † Guangzhou Institute of Geochemistry, CAS. ‡ Graduate University of the Chinese Academy of Sciences. § Virginia Institute of Marine Science. 5748

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 15, 2010

δ15N, but only R-HBCD showed a strong positive relationship (p ) 0.001). The biomagnification factors for R- and γ-HBCD and TBBPA were calculated based on individual predator/prey feeding relationships for two species.

Introduction Tetrabromobisphenol-A (TBBPA) and hexabromocyclododecane (HBCD) are two high-production-volume, current-use, and nonregulated brominated flame retardants (BFRs). TBBPA is the most widely used BFR and is mainly used in printed circuit boards as a reactive agent and in acrylonitrile butadiene styrene as an additive. HBCD is primarily used as an additive in thermal insulation material (polystyrene roof isolation) in the building industry, in upholstery textiles, and, to a minor extent, in electrical equipment housings (1). The size of the global TBBPA and HBCD market was 119 600 and 16 700 tons in 2001, respectively (2). Due to their wide use, TBBPA and HBCD have become ubiquitous environmental contaminants that bioaccumulate in the tissues of wildlife and humans (3, 4). Although there are no restrictions on the production and usage of TBBPA and HBCD, HBCD has been considered as a candidate of persistence organic pollutants, and both TBBPA and HBCD are included in the OSPAR list of chemicals for priority action under the Hazardous Substances Strategy (5). Despite its extensive use, TBBPA has been rarely studied, and little data on it are available for use in environmental assessments. Most TBBPA data in biota published so far come from Europe, where the levels of TBBPA in biological matrices and humans have been shown to be low or undetectable (4). Asia has the highest consumption of TBBPA, and most electronic items using circuit boards are produced in this area. The worst-case scenario for the possible environmental impacts of TBBPA is likely to be found in this region (6). However, there has been little work undertaken to date on TBBPA in Asia (7). Therefore, studies on TBBPA in this region are urgently needed. The distribution and trends of HBCD in the environment and humans have been reviewed recently by Covaci et al. (3) and Law et al. (8). The bioaccumulation and biomagnification of HBCD in the freshwater and marine food web have been demonstrated in several studies (9–13). In isomer-specific studies in aquatic organisms and birds, R-HBCD predominates, although γ-HBCD is the dominant diastereomer in technical mixtures (3). In addition, nonracemic R-HBCD distributions have also been reported in fish (14), marine mammals (15), and bird samples (16). Biological transformation, stereoselective uptake/metabolism, or a combination of these processes is considered as the main reason for the observed diastereoisomer- and enantiomer-specific accumulation in biota. However, there have been few studies that investigated the influence of the habitat and food choice of biota on diastereoisomer- and enantiomer-specific accumulation. Therefore, the roles played by these factors are not fully understood. In the present study, we collected birds with different habitat and feeding habits, including aquatic birds feeding on aquatic organisms, terrestrial birds feeding on grain and plant leaf, and wetland birds feeding on both aquatic and terrestrial food, and their diet from an e-waste region in South China. TBBPA and HBCD were examined in the muscles of the birds, in the birds’ diet such as fish, grain, and plant leaves, and in environmental samples like water and soil using LC-MS/MS. Nitrogen- and carbon-stable isotopes (δ15N and δ13C) were also analyzed to measure the diet source and 10.1021/es101503r

 2010 American Chemical Society

Published on Web 07/12/2010

trophic levels (TLs) of the birds. The primary objectives were to determine the levels of TBBPA and HBCD in six bird species in the e-waste site, evaluate the influence of the habitat and feeding habits on the observed diastereoisomer- and enantiomer-specific distribution of HBCD, and evaluate the potential biomagnification of TBBPA and HBCD.

Materials and Methods Field Sampling. Specimens (n ) 40) from six bird species, including Ardeidae (Chinese pond heron, Ardeola bacchus; n ) 5), Rallidae (white-breasted waterhen, Amaurornis phoenicurus; n ) 11; slaty-breasted rail, Gallirallus striatus n ) 4), Scolopacidae (common snipe, Gallinago gallinago; n ) 8), Columbidae (spotted dove, Streptopelia chinensis; n ) 9), and Phasianidae (Chinese francolin, Francolinus pintadeanus; n ) 3) families, were collected between 2005 and 2008 from Qingyuan County, the second largest e-waste recycling region in Pearl River Delta. Detailed information about the sampling site and birds is provided elsewhere (17), and the habitat and diet type of the sampled species are given in Table S1 in the Supporting Information. All birds collected were found dead or dying from various causes (hunting, trauma, poisoning, distress, etc.). Wild fish were collected from abolished fish pools located in the e-waste recycling region using electric fishing devices (18). The fish samples with body weight less than 100 g (crucian carp, n ) 7) were analyzed to represent the diet of the fish-eating birds in this study. Grain samples were collected from different (n ) 5) rice fields in the study region, and they were mixed together to obtain three composite samples for analyses. Three composite leaf samples of eucalyptus (Eucalyptus spp), the prevailing species of vegetation in the study region, were collected in two locations in the same region. Three water samples (from fish pools) and four composite soil samples were also collected to represent the aquatic and terrestrial environmental matrix of the study area. The bird samples were immediately transferred to the laboratory. Various tissues were excised and stored at -20 °C until chemical analysis. Pectoral muscle was used in the present study. Sample Extraction and Cleanup. A homogenized muscle tissue (5 g) was mixed with 10 g of Na2SO4 and ground into powders. The labeled recovery internal standards 13C12TBBPA (20 ng) and 13C12-R-, β-, γ-HBCD (40 ng), bought from Cambridge Isotope Laboratories, were added to the samples before extraction. The samples were extracted using a mixture of acetone/n-hexane (1:1, v/v) in a Soxhlet apparatus for 48 h. The lipid content was determined gravimetrically from an aliquot of the extract. Another aliquot of the extract, used for chemical analysis, was subjected to gel permeation chromatography using a glass column packed with 40 g of SX-3 Bio-Beads (Bio-Rad Laboratories, Hercules, CA), and eluted with dichloromethane/n-hexane (1:1, v/v) for lipid removal. Eluate from 90 to 280 mL containing organohalogen compounds was collected and concentrated to 2 mL and further purified on a 1 mm i.d. multilayer silica column packed with neutral silica (8 cm, 3% deactivated) and 44% sulfuric acid silica (8 cm). The fraction containing the targets was eluted using 30 mL of 50% dichloromethane in n-hexane (volume fraction). The eluate was concentrated to near dryness under gentle nitrogen, redissolved in 200 µL of methanol, and spiked with known amounts of labeled instrument performance and matrix internal standards d18R-HBCD (40 ng), d18-β-HBCD (40 ng), and d18-γ-HBCD (40 ng) (Wellington Laboratories Inc.) prior to instrument analysis. The detailed procedures of the sample pretreatment and cleanup for the water, grain, plant leaf, and soil samples are shown in the Supporting Information. Nitrogen- and Carbon-Isotope Measurements. The subsamples for nitrogen- and carbon-stable isotopes analysis were lyophilized and ground into ultrafine powder. Ap-

FIGURE 1. Stale isotope ratio of nitrogen and carbon in collected birds. proximately 1 mg of the ground samples was weighed in tin capsules and analyzed by a Flash EA 112 series elemental analyzer interfaced with a Finigan MAF ConFlo 111 isotope ratio mass spectrometer. The stable isotope ratios of the samples were assessed against the reference standards ammonium sulfate for δ15N and carbon black for δ13C. They are expressed as δX(values [‰]), with δX ) [(Rsample/Rstandard - 1) × 1000], where X is δ15N or δ15C, and R is the corresponding ratio of 15N/14N or 13C/12C. The precision for this technique is 0.5‰ (2 SD) for δ15N and 0.2‰ (2 SD) for δ13C. Instrumental Analysis. Quantitative determination of TBBPA and the HBCD diastereoisomers and enantiomers was performed on an Agilent 1200 series liquid chromatography and an Agilent 6410 triple quadrupole mass spectrometer with an electrospray interface working in negative ionization mode. More details about the instrumental analysis can be found in the Supporting Information. QA/QC. The procedural blank and spiking blank were analyzed for each batch of the samples. Trace HBCDs were found in the procedural blanks, and they were not subtracted from the samples. The mean recoveries of TBBPA, R-HBCD, β-HBCD, and γ-HBCD in the spiking blanks were 80%, 95%, 94%, and 93%, respectively. The average recoveries of internal standards were as follows: 74.8% ( 8.9% for 13C12-TBBPA, 95.0% ( 7.3% for 13C12-R-HBCD, 91.4% ( 4.7% for 13C12-βHBCD, and 84.8% ( 3.7% for 13C12-γ-HBCD. Method detection limits (MDLs) were determined by three times the standard deviation of the target value in blanks. The MDLs of quantification for TBBPA, R-HBCD, β-HBCD, and γ-HBCD were 0.27, 0.26, 1.20, and 1.53 ng/g, respectively. Statistics. The correlations between δ13C and the percentage of total HBCD that was the R isomer and between δ15N and levels of HBCD and TBBPA were investigated by a simple linear regression analysis using the software SSPS 11.5. The samples below the detection limits were omitted during data analysis.

Results and Discussion Stable Isotope Ratios (δ15N and δ13C) of the Birds. The stable isotope ratios of nitrogen and carbon are shown in Figure 1. The stable isotope ratio of nitrogen (δ15N) increased with an increase in TL; thus, δ15N is used to ensure the position of the organism’s TL (19). The stable isotope ratio of carbon (δ13C) also increased with an increase in TL; however, the increment is so small (about 1‰) (20) that δ13C is generally employed to analyze the diet composition and carbon source of the organisms. The relative trophic status of the collected birds, defined by means of δ15N, increased in the following VOL. 44, NO. 15, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5749

5750

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 15, 2010

Pooled samples; only mean values were present due to the nonsignificant difference for the three analyses for water, grain, and plant leaf. b

Median (range). a

EFγ-HBCD

0.61 nd 0.6 nd 0.58 (0.55-0.58) 0.63 0.53 (0.51-0.56) nd nd nd 0.53 (0.47-0.58) 0.41 (0.39-0.43) 0.41 (0.34-0.48) 0.44 (0.42-0.49) nd 0.61 (0.56-0.63) 0.62 (0.59-0.67) 0.4 (0.38-0.44) 0.47 nd nd 0.53 (0.51-0.54)

EFr-HBCD ΣHBCD

1995 (460-5058) 73.4 (nd-394) 52 (nd-344) nd-216 73.6 (19-492) 47.5 (30.5-99.7) 112.4 (21.0-378) 52 19.3 60 45.8 (7.1-120) nd (nd-64) 27 (nd-305) 14.1 (nd-27.1) nd (nd-216) 54.3 (16.6-229) 22 (20.2-75.2) 13.4 (2.2-67.8) 7 7.1 32.1 27.7 (5.6-73)

γ-HBCD β-HBCD

nd (nd-8.7) nd (nd-22.9) nd (nd-27.1) nd nd (nd-19.1) 0.3 (nd-4.7) 2 (0.4-25.5) 3 5.5 13.7 5.6 (0.3-16) 1995 (420-5058)a 27.5 (nd-66.6) 6.9 (nd-169) nd (nd-42.5) 14.2 (nd-243) 19.8 (10-25.6) 95.7 (35.0-284) 42 6.2 14.2 12.5 (1.1-29)

r- HBCD TBBPA

173 (133-243) 170 (28.1-1482) 54.3 (10.6-149) 103.4 (9.0-139) 90.1 (39.2-503) 28.2 (24.2-54.8) 1.14 (0.23-1.74) 68 3618 8917 295 (2.9-780) Chinese pond heron (n ) 5) white-breasted waterhen (n ) 11) common snipe (n ) 8) slaty-breasted rail (n ) 4) spotted dove (n ) 9) Chinese francolin (n ) 3) fish (n ) 7) water (n ) 3b, pg/l) grain (n ) 3b, pg/g) plant leaf (n ) 3b, pg/g) soil (n ) 4b)

TABLE 1. Concentration of HBCDs and TBBPA in Birds, Fish (ng/g lipid weight), Water (pg/L), Grain, Plant Leaf (pg/g dry weight), and Soil (ng/g dry weight)

order: spotted dove (6.0‰) < slaty-breasted rail (6.9‰) and Chinese Francolin (7.0‰) < common snipe (7.8‰) < whitebreasted waterhen (9.5‰) < Chinese pond heron (11.1‰). The Chinese pond heron has a higher δ13C (mean of -22.4‰) than the spotted dove (-25‰) and Chinese francolin (-27‰), which is in line with its habitat-specific foraging. The Chinese pond heron is a piscivorous bird feeding primarily on fish (95%) and aquatic insects, while the spotted dove and Chinese francolin are terrestrial phytophagous birds feeding on corn, fruit, leaves, and weed (21). The stable carbon isotope ratio of aquatic plants is substantially higher than that of terrestrial plants (22). Therefore, birds feeding on aquatic food have a higher δ13C than birds feeding on terrestrial food. White-breasted waterhen, common snipe, and slaty-breasted rail inhabit all types of freshwater wetlands, feeding on a wide range of foods with both aquatic and terrestrial origin. White-breasted waterhen mainly eats seeds, insects, and small fish, and it often forages above ground, in low bushes, and in small trees. Common snipe mainly eats insects, earthworms, and plant material. Slaty-breasted rail feeds mainly on crab, shrimp, and insects (21). The δ13C values in these three bird species span a wide range from -27.1‰ to -19.4‰ (Figure 1). Levels of TBBPA and HBCD. The statistics for the levels of HBCD and TBBPA are summarized in Table 1. TBBPA was detected in all muscle tissues of the birds at concentrations ranging from 9.0 to 1482 ng/g lipid weight (lw), with the highest levels found in white-breasted waterhen. Few studies have reported on the concentrations of TBBPA in bird tissues. Morris et al. (23) reported TBBPA in the eggs of common tern collected from western Scheldt with concentrations of