Research Levels and Temporal Trends (1988-1999) of Polybrominated Diphenyl Ethers in Beluga Whales (Delphinapterus leucas) from the St. Lawrence Estuary, Canada M I C H E L L E B E U F , * ,† B R U N O G O U T E U X , ‡ LENA MEASURES,† AND STEVE TROTTIER† Department of Fisheries and Oceans, Maurice Lamontagne Institute, P.O. Box 1000, Mont-Joli, Quebec, Canada G5H 3Z4, and Institut des Sciences de la Mer, Universite´ du Que´bec a` Rimouski, 310 Alle´e des Ursulines, Rimouski, Quebec, Canada G5L 3A1
Polybrominated diphenyl ethers (PBDEs) were determined in blubber samples of 54 stranded adult beluga whales (Delphinapterus leucas) collected between 1988 and 1999 in the St. Lawrence Estuary (SLE), Quebec, Canada. Summed concentrations of 10 PBDE congeners (∑PBDEs) measured in beluga samples varied between 20 and almost 1000 ng/g wet weight. According to the PBDE concentrations in marine mammals reported in the scientific literature, SLE belugas appear to be relatively lightly contaminated. Only a few predominant congeners (namely, PBDE-47, -99, and -100) represent on average more than 75% of ∑PBDEs in SLE belugas. The accumulation of ∑PBDEs in both male and female belugas showed significant exponential increase throughout the 1988-1999 time period. The time necessary for beluga to double their blubber concentration of the most prevalent PBDE congeners was no longer than 3 years. The PBDE temporal changes reported in this study are generally faster but in agreement with the trend observed in other organisms collected in Canada, such as lake trout (Salvelinus namaycush) from the Great Lakes, ringed seal (Phoca hispida), and beluga whale from the Canadian Arctic. Some changes in the pattern of PBDEs in belugas were also observed during the time period investigated. The recent and important increase of PBDE levels in SLE belugas could explain the unexpected lack of statistical difference in PBDE contamination between males and females. This suggests that to date PBDEs tend to be accumulated by both male and female belugas, masking the elimination of PBDEs by females through post-natal transfer to their offspring. This study confirms that the growing use of PBDEs as flame retardants has resulted in rising contamination of Canadian aquatic environments. Additional studies are needed to assess the toxicological implications of the PBDE tissue levels found in SLE belugas. * Corresponding author phone: (418)775-0690; fax: (418)775-0718; e-mail:
[email protected]. † Maurice Lamontagne Institute. ‡ Universite ´ du Que´bec a` Rimouski. 10.1021/es035187j CCC: $27.50 Published on Web 04/29/2004
2004 American Chemical Society
Introduction Polybrominated diphenyl ethers (PBDEs) are used as flame retardants to reduce the risk of fire in a wide range of manufactured products such as polyurethane foams, computers, furniture, and automobiles (1-3). These compounds are typically produced in three commercial formulations of distinct degree of bromination (i.e., penta-PBDE, octa-PBDE, and deca-PBDE) (3). The penta-PBDE mixture consists mainly of a tetra (PBDE-47) and two penta bromo-substituted congeners (PBDE-99 and -100), and to a lesser extent of some tri- (PBDE-28) and hexa bromo-substituted congeners (PBDE153 and -154). The octa-PBDE mixture consists mainly of a hepta bromo-substituted congener (PBDE-183), of PBDE153, and to a lesser extent of some octa- and nona-PBDEs. The deca-PBDE mixture contains mainly the deca bromosubstituted congener PBDE-209 with only traces of nonaPBDEs. PBDEs are emitted into the environment through several sources such as during production, use, and recycling of products containing PBDEs as well as disposal in domestic waste, landfills, and incineration (4, 5). PBDEs have been reported in a variety of biotic and abiotic environmental matrixes such as air, biota, human milk, and plasma as well as suspended particles and sediments (4, 6, 7). The most predominant PBDE congeners found in biological matrixes contain 4-5 bromine atoms and possess physicochemical properties similar to those of persistent organic pollutants (POPs) of concern such as polychlorinated biphenyls (PCBs) and p,p′-dichlorodiphenyltrichloroethane (DDT) (4, 6, 8, 9). These bioaccumulated congeners are mostly contained in the penta-PBDE, but recent studies have suggested that some of these congeners may result from the debromination of higher brominated congeners (10, 11). Some of the lower brominated PBDE congeners generally found in biota and humans have the propensity to disrupt thyroid hormones, cause neurobehavioral deficits, and possibly cause cancer (12, 13). Unfortunately, available toxicological evidence is still limited, and additional studies are needed to adequately address the potential environmental risks of the PBDEs (13, 14). Recent studies have demonstrated continuously rising levels of PBDEs in biological samples from Canada. Increasing trends of PBDEs have been reported in human milk (15), marine mammals (16, 17), bird eggs (18), and fish (19). These trends are in agreement with the worldwide continuously increasing demand for the penta-PBDE mixture since the 1980s (16). In 1999, the penta-PBDE mixture was primarily used (>97%) in North America where no regulation is in place to control the use of these compounds (3, 20). Beluga whales (Delphinapterus leucas) inhabiting the St. Lawrence Estuary (SLE) have been identified as a species particularly propitious to bioaccumulate POPs (21-25). The prominent POP contamination reported in SLE belugas has been attributed to their high trophic position, their relatively long life span, and their habitat, which is located downstream of the St. Lawrence River and Lake Ontario, a highly populated and industrialized area (21, 23, 26, 27). The SLE beluga population has declined by about 80-90% of its estimated size of 5000 animals in 1885, in part due to hunting pressure. Although the SLE beluga was given the status of endangered species by the committee on the status of endangered wildlife in Canada (28) and hunting was prohibited in 1979, the population is considered stable at about 1000 individuals (29). The presence of elevated concentrations of POPs in belugas could have played a role in limiting the growth rate VOL. 38, NO. 11, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2971
of the population (30-33). Despite the fact that levels of PCBs and other POPs have been stable or even slightly decreasing since the late 1980s (24, 34), no evidence of a clear recovery of the SLE beluga population has been reported (29). Given that high PBDE concentrations are generally found in marine mammals located at the top of the food chain (35), it became imperative to investigate the presence of these emerging, bioaccumulative, and possibly deleterious chemicals in the SLE beluga population. The primary objective of this study was to report concentrations and patterns of PBDEs in blubber samples of beluga whales from the St. Lawrence Estuary. Samples collected between 1988 and 1999 were analyzed in order to assess temporal trends of PBDEs in SLE belugas during that time period.
Materials and Methods Sampling. Blubber samples were obtained from stranded adult male (n ) 28) and female (n ) 26) belugas found on the shores of the SLE between 1988 and 1999 (Table 1). Most whale carcasses were in good or fair (codes 2 or 3) condition according to the classification of Geraci and Lounsbury (36). The sex of the animals was recorded, and for most of them length was measured. The age of belugas was determined by counting the number of growth layer groups in sectioned teeth (37). Whenever possible, blubber samples were taken at 60-70% of body length from the nose, approximately midway between the spinal column and the mid-lateral region of the beluga. Most blubber samples collected prior to 1997 were initially separated in three layers; identified as close to the skin (upper), middle, and close to the muscle (lower); placed in individual solvent-rinsed glass jars; and stored at -20 °C. In an effort to represent the entire depth of the blubber, layers were combined in equal quantities and homogenized before chemical analysis. For animals sampled in 1997 and after, a block of blubber extending from the skin to the muscle was collected, wrapped in solventrinsed aluminum foil, placed in a sealed plastic bag, and stored at -20 °C. A subsample representative of the entire depth of the blubber layer was taken from the block of blubber, homogenized, and used for chemical analysis. However, only a part of the entire depth of the blubber layer, corresponding to a thickness of 3 cm, was sampled for five females and six males. For most of these animals, this blubber sample represents more than half of the entire depth of the blubber layer, which demonstrated no significant concentration bias due to stratification of contaminants in the blubber of belugas (38). Although two animals (DL-102-98 and DL-102-99) were collected outside the SLE, levels of PCBs and other POPs suggest that these individuals were members of the SLE population and not from the Arctic (M. Lebeuf, personal data). Chemical Analysis. Blubber samples (0.5-1 g wet weight) were chemically dried with sodium sulfate before being transferred to a glass column. A single 13C12 PCB (PCB-170) was added to the column before the extraction procedure. Lipids and lipophilic compounds were extracted from the sample with dichloromethane (DCM)-hexane (50:50 v/v). The extraction solution received a mixture of five 13C12 PBDEs (PBDE-47, -99, -100, -126, and -153) before being reduced in volume for purification. Lipids were removed from the extract by gel permeation chromatography. The lipid-free extracts were further cleaned up by elution through a two-layer column packed with 10 g of hydrated neutral silica (5% water) and 5 g of hydrated alumina (5% water). The first 15-mL hexane fraction was discarded, and the second 70 mL of 50% DCM in hexane fraction containing the PBDE congeners was collected. The final extracts were reduced in volume and spiked with an instrument performance solution containing 2972
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 11, 2004
TABLE 1. Sampling and Biological Characteristics of Individual Belugas Collected in the St. Lawrence Estuary during the 1988-99 Time Period and Concentrations of ∑PBDEs in Blubber Samples blubber thick∑PBDEs length ness lipids (ng/g wet sex (m) (cm) (%) weight)
collection datea
beluga no.
ageb (yr)
88-10-07 88-10-24 88-08-26 89-06-13 89-08-17 90-07-29 92-09-24 92-10-12 93-10-20 94-09-23 95-05-14 95-05-21 97-05-23 97-06-08 97-07-29 97-07-23 97-09-30 97-11-10 98-05-24 98-05-25 98-07-17 99-05-31 99-07-05 99-08-19 99-10-16 99-04-11
DL-09-88 DL-12-88 DL-139-88 DL-04-89 DL-06-89 DL-04-90 DL-05-92 DL-06-92 DL-08-93 DL-04-94 DL-01-95 DL-02-95 DL-02-97 DL-03-97 DL-06-97 DL-102-97 DL-105-97 DL-106-97 DL-03-98 DL-04-98 DL-06-98 DL-03-99 DL-05-99 DL-07-99 DL-09-99 DL-107-99
22+ 16 31+ 14+ 25+ 13 21+ 31.5+ 12.5 10 28+ 26+ 28.5 21+ 31+ 22.5+ 30+ 31+ 26+ 20.5+ 11 16 11 10 18+ 15
F F F F F F F F F F F F F F F F F F F F F F F F F F
3.7 3.5 3.7 3.4 3.6 3.3 3.4 3.9 3.6 3.4 3.7 3.3 3.4 3.6 3.9 3.4 3.9 3.4 3.8 3.6 2.9 3.7 3.6 3.4 3.9 3.6
ndc nd nd nd nd nd nd nd nd nd nd 6 5 10 8 6 7 7 10 5 10 10 6 7 7 8
76 94 80 94 94 94 95 90 94 97 94 95 93 91 90 98 94 90 85 90 96 95 88 88 95 93
17.2 18.0 22.5 29.3 37.5 31.9 109 110 111 196 210 297 644 400 325 331 289 421 526 821 330 287 935 576 614 440
88-08-13 89-04-26 89-05-30 90-04-08 90-06-16 90-07-19 93-04-18 93-05-20 93-06-18 94-10-28 94-10-31 94-04-18 95-07-01 97-11-26 97-05-16 97-09-12 98-05-23 98-08-08 98-09-29 98-04-12 98-04-22 98-10-12 99-04-08 99-06-07 99-08-03 99-04-06 99-08-18 99-11-22
DL-08-88 DL-01-89 DL-02-89 DL-01-90 DL-02-90 DL-03-90 DL-01-93 DL-02-93 DL-03-93 DL-05-94 DL-06-94 DL-101-94 DL-06-95 DL-07-97 DL-101-97 DL-103-97 DL-02-98 DL-07-98 DL-08-98 DL-09-98 DL-102-98 DL-105-98 DL-01-99 DL-04-99 DL-06-99 DL-102-99 DL-105-99 DL-108-99
20+ 27+ 26+ 18+ 23+ 19+ 11+ 25+ 13+ 25+ 27.5 29+ 25+ 26+ 16+ 23.5+ 12+ 17+ 22+ 18+ 22+ 24.5 15+ 17 14+ 19+ 9+ 9+
M M M M M M M M M M M M M M M M M M M M M M M M M M M M
4.0 4.0 4.2 4.2 4.2 4.4 4.2 4.0 nd 4.0 4.2 4.1 4.4 4.3 4.4 4.4 4.0 4.1 4.0 4.1 4.3 4.6 4.0 4.0 4.2 nd 3.8 3.7
nd nd nd nd nd nd nd nd nd nd nd nd 7 7 7 5 5 6 8 8 8 8 8 8 6 6 4 9
87 91 97 88 64 90 92 93 93 96 96 93 93 91 93 87 91 95 76 92 86 75 91 94 76 88 87 86
28.4 41.6 60.0 42.2 50.8 49.2 131 108 97.9 255 267 162 196 156 277 585 356 206 359 389 221 281 414 598 425 290 675 423
a Given as yy-mm-dd. b A plus sign (+) indicates that ages may be underestimated due to difficulty in reading worn dentine layers. c nd, not determined.
two additional 13C12 PCBs (PCB-111 and -189) for a final volume of 100 µL. Ten individual PBDEs for which authentic standards were available were measured in all blubber samples. Since the chemical structure of PBDEs resembles that of PCBs, the same numbering system was used. Concentrations were reported for one tri (PBDE-28), three tetra (PBDE-47, -49,
and -66), two penta (PBDE-99 and -100), three hexa (PBDE153, -154, and -155), and one hepta (PBDE-183) bromosubstituted congeners and as the sum of these 10 congeners (∑PBDEs). The quantification of PBDEs was performed using a ThermoQuest Trace GC gas chromatograph equipped with a Finnigan PolarisQ ion trap, a ThermoQuest PTV split/ splitless programmable injector (5-µL injection volume) operated in splitless mode, and a ThermoQuest AS2000 autosampler. The chromatographic separation of the contaminants was achieved using a 30-m DB-5MS column (0.25 mm i.d., 0.25 µm film thickness) with helium as the carrier gas. The temperature program for the injector was as follows: 90 °C during 0.3 min then raised to 310 °C at 14.5 °C/s and held for 40 min. The GC oven temperature was kept at 90 °C during 1.3 min, then raised to 160 °C at 30 °C/min, and raised again to 270 °C at 2 °C/min. Finally, the GC oven temperature was raised to 300 °C at 30 °C/min and held for 2 min. The ion source was operated in electron impact ionization mode, and the ion trap was operated in MS/MS mode. Concentrations of PBDE congeners were calculated using relative response factors determined from a threepoint calibration curve. PBDE congener concentrations were corrected on the basis of the recovery of the 13C12 labeled surrogate compounds, which ranges from 58 to 127%. The precision of the PBDE analytical method was assessed by repeated analysis (n ) 11) of a pilot whale (Globicephala sp.) blubber sample available as the standard reference material SRM 1945 (39). The coefficients of variation (CV) of replicate analysis varied between 9 and 29% and were inversely proportional to the concentration of individual PBDE congeners. For example, PBDE-47, -99, -100, -153, and -154, found at concentrations of 10 ng/g wet weight or higher in SRM 1945, showed CV lower than 15% in replicate analysis. The analytical accuracy of the method was validated through an international interlaboratory study on brominated flame retardants where seven congeners (PBDE-28, -47, -99, -100, -153, -154, and -183) were measured in several biological samples as well as in one solution of PBDEs in undisclosed concentrations (40). Our laboratory reported PBDE concentrations that were on average within 20% of the assigned values for each congener except for PBDE-153, where our results were positively biased in a range varying between 40 and 50% of the assigned values. Statistical Analysis. Statistical analyses were performed either on a spreadsheet using original formula (41) or conducted using Systat 10 (42). In all tests, statistical significance was set at R ) 0.05. Contaminant data were normalized to lipid content and used for statistical analyses after natural logarithmic transformation whereas biological parameters were not transformed. Differences of age, lipid content, length, blubber thickness, and contaminant level and pattern between female and male belugas were assessed by a single-factor analysis of variance (ANOVA), followed by a pairwise comparison test using the Bonferroni procedure (42). Multiple regression analyses, using backward elimination, were performed to assess (two-tailed t-test) the contribution of each independent variable such as age, lipid content, length and blubber thickness of belugas, and collection time in explaining PBDE concentrations in belugas. Time trends in contaminant levels were assessed by simple least-squared regression using linear model followed by ANOVA to confirm the linearity of the regression. Values of studentized residuals for linear regressions exceeding a threshold value of 3 were considered as outliers and removed from statistical analysis (42). Slopes of the linear regressions were compared using an analysis of covariance (ANCOVA) followed by a multiple comparison procedure using the mean critical values of Tukey and Newman-Keuls tests (43).
Results and Discussion Age, Lipid Content, Length, and Blubber Thickness of Belugas. All beluga whales examined were at least 9+ years old (Table 1) and assumed to be sexually mature based on information from other beluga populations (27). The mean age ( 95% confidence interval (CI) of these animals was 20.8 ( 3.0 and 19.7 ( 2.3 yr for females and males, respectively. No significant difference in age (ANOVA; F ) 0.36; p ) 0.55) was detected between males and females. However, the age of the examined males significantly decreased as a function of sampling years as indicated by a linear regression analysis (r 2 ) 0.15; p ) 0.04) whereas no such a trend was observed for females (r 2 ) 0.01; p ) 0.65). Mean blubber lipid content (( 95% CI) was not significantly different (ANOVA; F ) 1.88; p ) 0.17) between male (89 ( 7%) and female belugas (92 ( 5%). Although data on blubber thickness are limited to animals collected in 1997 and after, no significant difference (ANOVA; F ) 0.94; p ) 0.34) was detected between males (6.5 ( 1.5 cm) and females (7.5 ( 1.9 cm). However, male belugas (4.2 ( 0.2 m) were significantly longer (ANOVA; F ) 100; p < 0.001) than females (3.5 ( 0.2 m). No linear relationship was detected between the lipid content (males, r 2 ) 0.11; p ) 0.09; females, r 2 ) 0.07; p ) 0.22) or the length of the animals (males, r 2 ) 0.01; p ) 0.56; females, r 2 ) 0.03; p ) 0.42) and the year of collection for both sexes. Concentrations and Patterns of PBDE Congeners in Blubber of Belugas. Summed concentrations of PBDE congeners measured in blubber of SLE belugas ranged from about 20 to almost 1000 ng/g wet weight (Table 1). The 10 PBDE congeners investigated were detected in nearly all blubber samples of SLE belugas (SI). Occasionally, PBDE-28, -49, -66, and -183 were below detection limits, particularly in animals collected in the 1988-1990 time period. In such a case, a value of 0.1 ng/g wet weight, which corresponds to half of the detection limit, was used in statistical analyses. PBDE levels on a lipid weight basis in SLE belugas were compared to those reported previously for belugas from the Arctic and other marine mammal species from North American and European coasts (Table 2). Complications in comparing PBDE data arise as the result of the rapid and considerable increase of PBDEs in biota during the last 20 yr (44). For this reason, only animals collected between 1996 and 2000 were considered. PBDE data must also be compared with caution due to differences in methods of presenting the results such as reporting the sum of a variable number of individual congeners or as homologue groups. In addition, biological parameters such as age, sex, diet, and nutritional condition of animals are known or expected to affect the level of contamination between animals, even from the same marine mammal species (45). To our knowledge, only two studies have reported PBDE contamination in belugas until now. Alaee and co-workers (46) measured PBDEs in male and female belugas from the Canadian Arctic at concentrations roughly 3 and 7 times lower than those in SLE belugas collected in the 1997-1999 time period, respectively. Unfortunately, the sampling time and age range of the Canadian Arctic belugas examined were not reported by the authors. Stern and Ikonomou (17) also reported PBDE concentrations in male belugas from the Canadian Arctic (southeast Baffin Island). The PBDE contamination in these animals collected in 1997, and of similar age range, was approximately 25 times lower than in male belugas from the SLE collected between 1997 and 1999. These results are in agreement with previous studies that reported elevated levels of POPs in SLE beluga tissues, generally 2030 fold higher than in other beluga populations from the Canadian Arctic (22, 47, 48). It also confirms that PBDEs, as most other POPs, are found at higher concentrations in VOL. 38, NO. 11, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2973
TABLE 2. PBDE Concentrations (ng/g Lipid Weight) in Blubber of Marine Mammals from North American and European Coasts Collected in 1996 or after ∑PBDEs (ng/g lipid weight) species
location
time period (yr)
n
sex
beluga whale St. Lawrence Estuary 1997-1999 (Delphinapterus leucas) 1997-1999 Canadian Arctic
15 14
SE Baffin Island, Canadian Arctic long-finned pilot whale Vestmanna, (Globicephala melas) Faroe Islands
1997
17
M F M F M
1996 1996 1996 1996 1997 1997 1997 1997 1996-1999 1996-1999 1996 2000 2000 1996 2000 1997-1998 1998
8 (pooled) 13 (pooled) 19 (pooled) 4 (pooled) 3 3 3 3 22 38 8 8 8 9 5 2 1
M M F F M M F F M F M M M F F M F
Torshavn, Faroe Islands
harbor porpoise (Phocoena phocoena) ringed seal (Phoca hispida)
harbor seal (Phoca vitulina)
England and Wales Holman Island, Canadian Arctic
San Francisco Bay
a ∑10 congeners. b Di to hexa homologue groups. c Tri to hexa homologue groups. limits. g ∑13 congeners. h ∑5 congeners.
environmental matrixes collected near industrialized or human population centers. Mean concentrations of ∑PBDEs in SLE beluga whales collected between 1997 and 1999 reach 430 and 540 ng/g lipid weight in males and females, respectively. These values stand at the lower end of the range of levels reported in other marine mammals from North America and Europe (Table 2). For instance, PBDEs in long-finned pilot whales (Globicephala melas) from Vestmanna in the Faroe Islands, Denmark; in harbor porpoises (Phocoena phocoena) from England and Wales, U.K.; in ringed seals (Phoca hispida) from Holman Island, Northwest Territories, Canada; and in harbor seals (Phoca vitulina) from the San Francisco Bay, California, USA, range from 1000 to 5000 ng/g (16, 49-51). However, PBDE concentrations in long-finned pilot whales from Torshavn in the Faroe Islands are in the same range as those found in the SLE belugas (52). PBDE congener patterns, representing the percent composition of each individual congener relative to ∑PBDEs, were calculated for both males and females using all animals examined (Figure 1). Only a few PBDE congeners (namely, PBDE-47, -99, and -100) were dominant in belugas from the SLE with, on average, more than 75% of ∑PBDEs accounted for by the sum of these three congeners. These results are in agreement with the general PBDE congener pattern reported in biological samples, especially marine mammals (6, 7). Although some congeners were found in significantly higher proportion in females (PBDE-49, -66, and -99) and others in males (PBDE-155 and -183), the overall PBDE patterns were almost identical for both sexes (Figure 1). Very similar PCB or toxaphene congener patterns have also been reported in male and female belugas from a given population (23, 34, 53). Unexpectedly, there was no significant difference in PBDE levels between male and female SLE belugas collected during the 1997-1999 time period (ANOVA; F ) 2.157, p ) 0.154) (Table 2). Multiple regression analyses revealed that independent variables such as age, length, and blubber thickness of belugas have no significant effect (p > 0.05) on the magnitude of PBDE concentrations in both male and female 2974
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 11, 2004
age (yr) 9-26 10-31
mean 430a
540a 160b 81b ca. 8-ca. 20 ca. 16c adults young adults young adults juveniles adults juveniles
