Environ. Sci. Technol. 2005, 39, 6016-6020
Hair as an Indicator of Endogenous Tissue Levels of Brominated Flame Retardants in Mammals H E L G A D ’ H A V EÄ , * , † A D R I A N C O V A C I , ‡ JAN SCHEIRS,§ PAUL SCHEPENS,‡ RON VERHAGEN,§ AND WIM DE COEN† Ecophysiology, Biochemistry and Toxicology Group and Evolutionary Biology Group, Department of Biology, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, and Department of Pharmaceutical Sciences, Toxicological Center, University of Antwerp, Universiteitsplein 1, B-2610 Wilrijk, Belgium
Few data are available on brominated flame retardants (BFRs) in terrestrial mammalian wildlife. Moreover, the use of hair in nondestructive monitoring of BFRs in mammals or humans has not been investigated. In the present study, concentrations of polybrominated diphenyl ethers (PBDEs) and brominated biphenyl 153 (BB 153) were analyzed in tissues of the European hedgehog Erinaceus europaeus. Road kills and carcasses from wildlife rescue centers were used to investigate relationships between concentrations of BFRs in hair and internal tissues, BFR tissue distribution (hair, liver, kidney, muscle, and adipose tissue), and PBDE congener tissue pattern dissimilarities. Liver concentrations of PBDEs and BB 153 were in the ranges 1-1178 and 0-2.5 ng/g of liver wet weight, respectively. PBDEs were predominant in adipose tissue and liver, while accumulation of BB 153 was tissue independent. The less persistent compound BDE 99 was more dominant in hair than in internal tissues. We observed positive relationships between BFR levels in hair and internal tissues for sum PBDEs and BDE 47 (0.37 < r < 0.78). The present study demonstrated that hair is a suitable indicator of PBDE exposure in terrestrial mammals which can be used in nondestructive monitoring schemes.
Introduction Polybrominated diphenyl ethers (PBDEs) and polybrominated biphenyls (PBBs), two classes of brominated flame retardants (BFRs), have been widely used in plastics, electronic equipment, textiles (not PBBs), and other materials for more than 30 years (1). Unfortunately, they have become widespread in the environment in the past few decades (1) and have been shown to exert toxic effects on man and wildlife (2). Most available data on BFRs are related to fish (3-5), birds (6, 7), and marine mammals (8, 9), while information on the distribution of PBDEs and PBBs in mammalian terrestrial wildlife is very limited (10). Measurements of persistent organic pollutants (POPs), such as polychlorinated biphenyls (PCBs) and PBDEs, are classically performed on organs of sacrificed individuals (11). * Corresponding author phone: ++32 3 265 33 42; fax: ++32 3 265 34 97; e-mail:
[email protected]. † Ecophysiology, Biochemistry and Toxicology Group. ‡ Department of Pharmaceutical Sciences. § Evolutionary Biology Group. 6016
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
However, increasing ethical concernsespecially for mammalian speciessgave rise to the need for nondestructive methods. Nondestructive methods impose minimal stress to individuals and allow investigation of pollution in endangered species or threatened populations. Additionally, they permit long-term follow-up of the same populations and individuals (12, 13). The use of hair analysis for investigation of POPs in humans and mammals has been limited (14, 15). Yet, few studies showed that significant relationships may exist between concentrations of these pollutants in hair and internal tissues (14, 16), indicating the promising role of hair as an indicator of endogenous tissue concentrations. To our knowledge, there are no available data on the levels of BFRs in human or mammalian hair. The main objective of this study was to investigate the use of hair of the European hedgehog Erinaceus europaeus as a nondestructive biological monitoring tool of contamination with BFRs in terrestrial ecosystems. This study is the first to investigate the relationships between concentrations of BFRs in hair and internal tissues and may provide a basis for the use of hair of hedgehogs, and possibly other mammalian wildlife and humans, in nondestructive monitoring studies of BFRs. Tissue distribution and tissue PBDE congener patterns were also investigated to better understand the presence or lack of relationships between BFR levels in hair and internal tissues. Analyses were performed on hedgehog road kills and carcasses from wildlife rescue centers all over Flanders, Belgium, and as such the available number of organisms represents the regional pollution level of Flanders.
Materials and Methods Sample Collection. Hedgehog road traffic victims and carcasses from wildlife rescue centers (Vogelbescherming Vlaanderen) were collected from Flanders, Belgium (n ) 42), and The Netherlands (n ) 2), during 2002 and 2003. Hedgehogs were stored in labeled plastic bags at -20 °C until dissection. Labels contained detailed information on the origin of the hedgehog, numbers of days spent in the wildlife center, and reason for rehabilitation. Individuals that spent more than one week in a wildlife center and individuals that showed severe wounds or illnesses were excluded from the analyses. Hair was cut with stainless scissors. The hedgehogs were dissected, and liver, kidney, muscle, and subcutaneous adipose tissue were removed. The majority of the hedgehogs did not possess sufficient fat that could easily become separated from muscle tissue. As a result, adipose tissue was sampled only from six individuals. Sample Preparation and Analysis. The PBDE congeners 28, 47, 99, 100, 153, 154, and 183 and BB 153 were investigated. PBDE and BB 153 standards were available from Wellington Laboratories (Guelph, ON, Canada) and Dr. Ehrenstorfer Laboratories (Augsburg, Germany), respectively. All solvents used were of pesticide-grade purity (Merck, Darmstadt, Germany). The acidified silica was prepared by dropwise addition of 27 mL of concentrated sulfuric acid (95-97%) to 50 g of silica gel and continuous stirring to ensure homogeneity. To remove external contamination (e.g., fine soil particles, dust), individual hair samples were incubated in Milli-Q water (Millipore, Brussels, Belgium) in a shaking incubator (1 h, 40 °C), and afterward rinsed with Milli-Q water (17). Samples were air-dried on absorbent paper and cut into pieces of ∼2 mm. The extraction procedure was similar to the method described for the determination of persistent organochlorine pollutants in human hair (17). Between 0.3 and 0.5 g of hair 10.1021/es0507259 CCC: $30.25
2005 American Chemical Society Published on Web 07/06/2005
sample was accurately weighed, spiked with internal standard (1 ng of BB 103), and incubated overnight at 40 °C with 4 mL of hydrochloric acid (4 M) and 3 mL of hexane/dichloromethane (4:1, v/v). Extraction of analytes from the incubation medium was done by a liquid-liquid extraction (LLE) procedure with 2 × 4 mL of hexane/dichloromethane (4:1, v/v). The organic layer was purified on a cartridge filled with 2 g of acidified silica. After loading, the cartridges were eluted with 4 mL of hexane. The final eluate was concentrated to approximately 80 µL. The sample preparation procedure for internal tissues was similar to the procedure described in detail by Voorspoels et al. (18). Between 1 and 3 g of tissue was mixed with Na2SO4, 4 ng of internal standard BB 103 was added, and the mixture was extracted for 2 h with 75 mL of hexane/acetone (3:1, v/v) into a hot Soxhlet manifold. The lipid content was determined gravimetrically on an aliquot of the extract (1 h, 105 °C), while the rest of the extract was cleaned up on a column filled with ∼8 g of acidified silica and eluted with 15 mL of hexane and 10 mL of dichloromethane. The eluate was concentrated to 100 µL under a gentle nitrogen stream, and transferred to an injection vial. PBDEs and BB 153 were determined on an Agilent (Palo Alto, CA) gas chromatograph coupled with a mass spectrometer operated in electron-capture negative ionization mode and equipped with a 25 m × 0.22 mm × 0.25 µm HT-8 capillary column (SGE, Zulte, Belgium). Instrumental operating conditions and quality control were presented in detail elsewhere (18, 19). The mass spectrometer was operated in selected ion-monitoring mode (m/z ) 79 and 81). Methane was used as the moderating gas, and the ion source, quadrupole, and interface temperatures were 250, 150, and 300 °C, respectively. Dwell times were set at 30 ms. Limits of quantification (LOQs) for the analyzed compounds in internal tissues and hair were 0.1 ng/g of wet weight and 0.05 ng/g of hair. The quality assurance was done as described by Voorspoels et al. (18) and Gill et al. (20). Statistical Analyses. A one-way analysis of variance (ANOVA; with the post hoc Tukey multiple comparisons test) was performed to test for differences in concentrations of PBDEs and BB 153 among tissues, using the Proc Mixed module in SAS (SAS Institute, Cary, NC, 1999). The dependent variable was always log(x + 1) transformed to meet the assumptions of analysis of variance. The degrees of freedom of the fixed effects F-test were adjusted for statistical dependence using the Kenward-Rogers method (21), while variance components were estimated by restricted maximum likelihood (22). Regression analyses of log(x + 1) transformed variables were used to investigate the relationships between wet weight (ww) concentrations of organic compounds in hair and internal tissues (SPSS, Chicago, IL, 2002). Inspection of the residuals of the regression analyses did not reveal deficiencies in any of the models. No regression analyses were performed for adipose tissue, because of the small sample size. Linear regressions provided the best fit for all compounds. Pearson correlations were used to correlate log(x + 1) transformed concentrations of liver, kidney, and muscle tissues. Regressions and correlations were only performed if more than 10 samples were above the detection limit for both of the investigated tissues. We used a principal component analysis (PCA) to evaluate PBDE pattern differences among the various tissue types for the different PBDE compounds. For the latter reason, the proportion (%) of each PBDE compound was calculated relative to the total concentration of PBDEs (Statsoft, Tulsa, OK, 1994). Proportion data were arcsine transformed. Factor extraction was performed by maximum likelihood, followed by calculation of the factor loadings (varimax rotation). The individual factor scores were calculated and subsequently
plotted. Each individual tissue is represented as a point in multivariate space (SPSS).
Results and Discussion Tissue Concentrations. All PBDE congeners and BB 153 were detected in hair, liver, kidney, muscle, and adipose tissue of hedgehogs. Concentrations of sum PBDEs, individual PBDE congeners, and BB 153, together with the mean lipid percentages of the tissues, are provided on a ww basis in Table 1. Hedgehogs showed considerably higher PBDE levels (mean 653 ng/g of muscle lipid weight (lw), min-max 7-24212 ng/g of muscle lw, seven congeners) than moose Alces alces (mean 1.7 ng/g of muscle lw, three congeners) and rabbits Oryctolagus cuniculus (mean < 2 ng/g of muscle lw; three congeners) sampled 10 years ago in Sweden (10). The relatively high PBDE levels in Flemish hedgehogs compared to other terrestrial wildlife probably result from their higher (insectivorous) trophic position in the food chain. To our knowledge no data are available for BB 153 in terrestrial mammals. Tissue Distribution and PBDE Congener Pattern Analysis. Our mixed model analyses indicated that the concentrations (ww) of PBDEs were significantly different between some, but not all, selected tissues (Table 1). Residue levels of PBDEs were significantly highest in liver and adipose tissue. Similar to PCBs, BFRs are highly lipophilic pollutants and preferably accumulate in fatty tissues (9). This may pose a threat to hibernating animals, such as hedgehogs, since the mobilization of lipids during starvation causes accumulated contaminants to reenter the animal’s circulation and become distributed toward target organs (23, 24). Hedgehogs are especially at risk for these seasonal potentially toxic effects of lipophilic persistent pollutants, since they may lose up to 37% of their body mass during hibernation (25). The PBDE pattern in all hedgehog tissues, except hair, was dominated by the tetrabromodiphenyl ether 47, followed by BDEs 153 and 99 in that order (Figure 1). The observation that BDE 153 belongs to the three most important PBDE congeners is different from findings in other terrestrial mammals (rabbits, moose, and reindeer Rangifer tarandus) (10) sampled 10 years ago and marine mammals (9), where the dominant congeners were BDEs 47, 99, and 100 in that order. In some predatory birds on the other hand, BDE 153 was the predominant congener (7, 9). These pattern differences are probably explained by species-specific differences in PBDE metabolism and accumulation (9, 26) and food pattern dissimilarities (9). Hedgehogs, being insectivorous, are positioned on a higher trophic level than herbivorous moose, reindeer, and rabbits, and therefore, PBDE food patterns and metabolization may differ. An additional factor which may contribute to the different PBDE patterns is the present use of different PBDE technical mixtures compared to those used 15-20 years ago. Only the deca-BDE is now allowed for use in Europe, while the penta- and octa-BDE mixtures are banned (27). PBDE tissue pattern analysis by means of PCA analysis (Figure 2) resulted in a first axis that explained 21% of the pattern variation and that was characterized by higher proportions of BDE 99 in the negative direction. The second axis explained 27% of the variation and was characterized by high proportions of BDE 153 and BDE 47 in the negative and positive directions, respectively. Liver and fat had more positive loadings along the second axis than kidney and muscle, which indicated relatively higher proportions of BDE 47 and lower proportions of BDE 153 in these tissues. Hair had more negative factor loadings along the first axis than the other tissues due to relatively higher BDE 99 concentrations. A possible explanation may be found in the fact that BDE 99 is directly and unaltered transported from the VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6017
TABLE 1. Concentrations of PBDEs and BB 153 in Hair and Internal Tissues of Hedgehogs (ng/g of wet weight)a mean (median) lipid % PBDEs BDE 28 BDE 47 BDE 99 BDE 100 BDE 153 BDE 154 BDE 183 BB 153
mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max mean ( SE median min-max
hair, n ) 32 n.a.
liver, n ) 43 3.46 (3.11)
kidney, n ) 44 3.71 (3.29)
muscle, n ) 44 3.07 (1.93)
fat, n ) 6 71.23 (75.68)
1.9 ( 0.3 a 1.5 0.8-11.0
