Scholarly article on topic 'Influence of age, season, body condition and geographical area on concentrations of chlorinated and brominated contaminants in wild mink (Neovison vison) in Sweden'

Influence of age, season, body condition and geographical area on concentrations of chlorinated and brominated contaminants in wild mink (Neovison vison) in Sweden Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Chemosphere
OECD Field of science
Keywords
{Mink / PCB / "Brominated flame retardants (PBDE)" / "Methoxylated PBDE" / "Age-dependent accumulation" / "Seasonal variation"}

Abstract of research paper on Biological sciences, author of scientific article — Sara Persson, Anna Rotander, Bert van Bavel, Björn Brunström, Britt-Marie Bäcklin, et al.

Abstract The wild mink has gained acceptance as a sentinel species in environmental monitoring. However, only limited data are available in the literature on factors driving variability in concentrations of organic pollutants in this species. This study characterizes the differences in contaminant concentrations in subcutaneous fat of male mink from four different areas in Sweden and demonstrates how age, season and body condition influence concentrations of polychlorinated biphenyl (PCB) congeners, polybrominated diphenyl ether (PBDE) congeners (including methoxylated forms, MeO-PBDEs), as well as the pesticides dichlorodiphenyldichloroethylene (DDE), chlordane and hexachlorobenzene (HCB). The data were statistically treated using multiple regression and principal component analysis. The ∑PCB concentration and concentrations of PCB congeners 138, 156, 157, 180, 170/190, 189, 194, 206, 209 as well as PBDE 153/154 varied with age. Season had an influence on ∑PCB, PBDE 47 and PBDE 153/154 concentrations, as well as concentrations of most PCB congeners, with the exception of PCB 101, 110, 141 and 182/187. Lean mink had higher concentrations of most PCBs and PBDEs than mink with larger fat depots. The analyzed pesticides (DDE, oxychlordane, HCB) showed no systematic variation with season, age or body condition. The concentrations of MeO-PBDEs were generally low and 6MeO-PBDE 47 was the most commonly detected MeO-PBDE in mink from marine, brackish and freshwater areas. The results indicate that age, season and body condition are factors that may influence the concentrations of PCBs and PBDEs, and it is thus recommended to take these factors into account when analyzing mink exposure data.

Academic research paper on topic "Influence of age, season, body condition and geographical area on concentrations of chlorinated and brominated contaminants in wild mink (Neovison vison) in Sweden"

Contents lists available at SciVerse ScienceDirect

Chemosphere

journal homepage: www.elsevier.com/locate/chemosphere

Influence of age, season, body condition and geographical area on concentrations of chlorinated and brominated contaminants in wild mink (Neovison vison) in Sweden

Sara Persson a, Anna Rotanderb, Bert van Bavelb, Björn Brunströmc, Britt-Marie Bäcklind, Ulf Magnusson

a Division of Reproduction, Swedish University of Agricultural Sciences, P.O. Box 7054, SE-750 07 Uppsala, Sweden b MTM Research Centre, Örebro University, Fakultetsgatan 1, Örebro, Sweden

c Department of Environmental Toxicology, Uppsala University, Norbyvägen 18A, SE-75236 Uppsala, Sweden d Department of Contaminant Research, Swedish Museum of Natural History, P.O. Box 50007, SE-10405 Stockholm, Sweden

HIGHLIGHTS

" Concentrations of persistent organic pollutants in wild mink were analyzed. " Using multivariate methods, we identified sources of variation in the concentrations. " Age, season, body condition and geographical area caused significant variation. " It is recommended to take this into account when monitoring contaminants in mink.

ARTICLE INFO

Article history:

Received 30 December 2011 Received in revised form 17 July 2012 Accepted 10 September 2012 Available online 7 November 2012

Keywords:

Brominated flame retardants (PBDE) Methoxylated PBDE Age-dependent accumulation Seasonal variation

ABSTRACT

The wild mink has gained acceptance as a sentinel species in environmental monitoring. However, only limited data are available in the literature on factors driving variability in concentrations of organic pollutants in this species. This study characterizes the differences in contaminant concentrations in subcutaneous fat of male mink from four different areas in Sweden and demonstrates how age, season and body condition influence concentrations of polychlorinated biphenyl (PCB) congeners, polybrominated diphenyl ether (PBDE) congeners (including methoxylated forms, MeO-PBDEs), as well as the pesticides dichlorodiphenyldichloroethylene (DDE), chlordane and hexachlorobenzene (HCB). The data were statistically treated using multiple regression and principal component analysis. The J^PCB concentration and concentrations of PCB congeners 138,156,157,180,170/190,189,194, 206,209 as well as PBDE 153/154 varied with age. Season had an influence on X1PCB, PBDE 47 and PBDE 153/154 concentrations, as well as concentrations of most PCB congeners, with the exception of PCB 101, 110, 141 and 182/187. Lean mink had higher concentrations of most PCBs and PBDEs than mink with larger fat depots. The analyzed pesticides (DDE, oxychlordane, HCB) showed no systematic variation with season, age or body condition. The concentrations of MeO-PBDEs were generally low and 6MeO-PBDE 47 was the most commonly detected MeO-PBDE in mink from marine, brackish and freshwater areas. The results indicate that age, season and body condition are factors that may influence the concentrations of PCBs and PBDEs, and it is thus recommended to take these factors into account when analyzing mink exposure data.

© 2012 Elsevier Ltd. All rights reserved.

1. Introduction

The increasing concern about endocrine disruptors and possible effects of mixtures of contaminants on humans and wildlife emphasize the need for environmental monitoring and the use of sentinel species at a high trophic level. The American mink is a

* Corresponding author. Tel.: +46 18 67 23 24; fax: +46 18 67 35 45. E-mail address: ulf.magnusson@slu.se (U. Magnusson).

0045-6535/$ - see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chemosphere.2012.09.060

semi-aquatic species that feeds on fish, birds, rodents and frogs (Gerell, 1967). It has been proposed as a suitable sentinel species for monitoring exposure and effects of chemical contaminants in the environment (Basu et al., 2007; Persson et al., 2012). The mink is particularly suitable as a sentinel in Sweden, where it is an invasive species and therefore is legally hunted throughout the year in unlimited quantities, which facilitates the possibility of systematic sampling and monitoring over time. Also, monitoring to compare pollutant concentrations in different geographical areas is feasible

possible as the mink is distributed all over the country of Sweden and thus found in several different environments: along marine and brackish water coastlines as well as at inland streams, rivers and lakes. The home range for each mink is relatively small, on average 2.5 km along a shore-line (Gerell, 1970), suggesting that the contaminant concentration in a specimen reflects the local contamination levels.

Variation in wildlife contaminant concentrations and patterns is not only attributable to temporal and spatial differences, but also to biological factors such as metabolism and reproductive state (Bleavins et al., 1981; Yordy et al., 2010; Bytingsvik et al., 2012). It has been shown in several species that the body burden of contaminants varies due to factors such as age, sex, and diet (Addison and Smith, 1974; Borga et al., 2004; McKinney, 2011). Knowledge about factors underlying differences in body burden between individuals may be important in order to identify subgroups within populations at particular risk of adverse health effects due to chemical exposure. Such knowledge could also be used to standardize environmental monitoring and thereby reduce variation.

For the persistent organic pollutants included in the Stockholm convention, concentrations in marine wildlife are well documented in remote areas such as the Arctic (AMAP, 1998, 2004), as well as more densely populated areas such as the Baltic region (Jensen et al., 1969; Andersson and Wartanian, 1992; Bergek et al., 1992; Blomkvist et al., 1992; Koistinen et al., 1997; Roos et al., 1998). Polybrominated diphenyl ethers (PBDEs) have recently been added to the Stockholm convention and their concentrations are well documented in several species in the marine environment (de Wit et al., 2006; Law et al., 2006). In terrestrial and semi-aquatic or semi-terrestrial wildlife on the other hand, sparse data are available on persistent organic pollutants. Even less is known about concentrations of methoxylated PBDEs (MeO-PBDEs) and the origin of these compounds is debated (Teuten et al., 2005).

In the present study we examined the possible influence of age, season and body condition on the concentrations of various environmental contaminants in wild mink. Male mink from four populations occupying ecologically different habitats in Sweden were analyzed for polychlorinated biphenyls (PCBs), PBDEs and pesticides.

2. Materials and methods

2.1. Sampling

One hundred and one male wild mink were killed by local hunt-

ers from August to the end of April from 2004 to 2009, in four areas in Sweden: the Koster Islands in Skagerrak (K; n = 26), the Marsta inland region (M; n = 25), the Gavle Baltic Coast (G; n = 25), and

the inland of Northern Sweden (N; n = 25). The areas are characterized as follows: K is a sea water-environment about 8 km off the Swedish coast in the North Sea; M is an industrial and agricultural fresh water environment next to a town with 25 000 inhabitants, a large international airport and the training camp of the Swedish Rescue Services Agency; G is a brackish water costal environment

at the Baltic sea near two towns (70000 and 12000 inhabitants)

and with moderate industrial activities 10-20 km away; and N is

a sparsely populated inland environment with few industries and low agricultural activity.

Free-ranging mink were killed by a trap or shot and then frozen

at approximately -20 °C within 24 h. They were transported to the necropsy facilities at the National Veterinary Institute in Uppsala

and thawed just before necropsy, which was performed as previ-

ously described (Persson et al., 2011). The age in years of each mink was determined by tooth cementum analysis (Matson, 1981) at Matson's Laboratory (Milltown, Montana, USA). Assuming the

mink were born at the 1st of May, they were categorized into three age groups; juvenile (0-12 months old, n = 51), 1 year old (1324 months, n = 33) and two or more years old (older than 24 months, n = 17). Hours of daylight at the specific capture date and site for each mink was used to construct three seasonal groups; autumn (17-9 h of daylight before winter solstice, n = 42), winter (<9 h daylight, n = 29) and spring (9-17 h daylight after winter solstice, n = 30). In the necropsy procedure, the subcutaneous fat between the hind legs on the ventral part of the abdomen was excised and weighed. The body condition was set as the weight of the subcutaneous fat in relation to total body weight (g subcutaneous fat/kg body weight). Due to the lipophilic nature of the chemicals that were to be analyzed, body fat was saved (and refrozen) for chemical analysis.

2.2. Sample preparation and analytical determination

Detailed information on chemicals used in sample preparation and reference standards can be found in the Supporting Information. The sample preparation was performed using open column extraction followed by instrumental analysis using GC-ECNI-MS, as previously documented (Covaci et al., 2007; Rotander et al., 2012). Subcutaneous fat (3-5 g) was homogenized in a mortar with anhydrous sodium sulfate (5 * fat weight), and approximately 8 g of the homogenate was transferred to glass columns (18 mm in diameter). The 13C-labeled internal standards were added and the lipids were eluted with hexane/dichloromethane (1:1, v/v). After solvent evaporation using low-pressure rotary evaporation, the lipid contents were determined gravimetrically. Sample clean up was performed using a multi-layer silica column (18 mm diameter) containing KOH silica gel, neutral activated silica, 40% H2SO4 silica gel, 20% H2SO4 silica gel, neutral activated silica gel and activated Na2SO4. The analytes were eluted with hexane. Prior to instrumental analysis, a mix of three 13C-labeled PCBs used as recovery standards was added. PCBs, pesticides, PBDEs and MeO-PBDEs were analyzed by an Agilent 6890 GC coupled to a low-resolution mass spectrometer (Agilent 5975). PBDEs and MeO-PBDEs were analyzed in the negative chemical ionization mode, monitoring m/z 79 and m/z 81, and analysis of PCBs and pesticides was performed in electron impact selective ion recording mode monitoring the two most abundant ions of the molecular chlorine cluster. Splitless injection was used to inject 1 il of the final extract and the column used for separation of analytes was a 30 m SGE DB-5 (0.25 mm i.d., 25 im film thickness). For analysis of PBDEs and MeO-PBDEs, the temperature program was set to initial 180 °C for 2 min, ramped 15 °C/min to 205 °C, 2 °C/min to 251 °C and 6 °C/min to 325 °C. For analysis of PCBs and pesticides, the temperature program was set to initial 180 °C for 2 min, ramped 15 °C/ min to 205 °C, 2 °C/min to 230 °C and 5 °C/min to 325 °C.

2.3. Quality assurance

PBDE congeners 28, 47, 66, 100 and 99 and MeO-PBDEs 2MeO-PBDE 68, 6MeO-PBDE 47, 5MeO-PBDE 47 and 4MeO-PBDE 49 were quantified against 13C-labeled PBDE 77. PBDE congeners 85, 154, 153, 138, 183 and MeO-PBDEs 5MeO-PBDE 100, 4MeO-PBDE 103, 5MeO-PBDE 99 and 4MeO-PBDE 101 were quantified against 13C-labeled PBDE 139. Quantification was carried out using the internal standard method using 13C-labeled internal standards. PCBs were quantified against 13C-labeled analogues and pesticides were quantified against 13C-PCB 52, 70 or 101. The identification of individual congeners was based on accurate isotope ratio and retention time, and detection limits were set to three times the signal to noise. Quantification was carried out using the internal standard method. The average PCB recoveries ranged between 68% and 107% and the average PBDE 77 and PBDE 139 recoveries were

107% and 98%, respectively. With every batch of six to nine samples extracted, an extraction blank was also prepared and analyzed and instrumental blanks of toluene were monitored. No blank concentrations exceeded 10% of the concentrations measured in the samples.

2.4. Statistical analysis

Effects of age, season, body condition and geographical area on the chemical concentration were analyzed by least squares using the general linear model (GLM) procedure of SAS (SAS Institute Inc., Cary, NC, USA, version 9.02.01). Normality of data distribution was assessed using the univariate procedure of SAS and diagnostic plots generated by the GLM procedure. All dependent data were log-transformed to improve normal distribution before further analysis. The model used for analyzing the effects was:

Y = 1 + AGE + SEASON + BODYCOND + AREA + ERROR

where Y is an observed contaminant concentration; 1 is the population mean for the concentration; AGE is a fixed effect due to age category of the mink; SEASON is a fixed effect due to season of sampling; BODYCOND is a fixed effect due to the body condition of the mink; AREA is a fixed effect due to area of sampling; ERROR is a random residual error term. The variables year (of capture) and sample batch, as well as the effects of biologically relevant interactions between all variables were tested but none were included in the final model due to insignificance or small sample size. Pairwise comparisons of least square means were calculated by t-test. P-values less than 0.05 were considered significant. Also, an additional model was made for comparing seasonal variations in body condition in the four different areas. The model used was:

Y = 1 + SEASON + AREA + SEASON * AREA + ERROR

In this model, body condition is the dependent variable (Y) and SEASON * AREA is the interaction between the area and season.

A principal component analysis (PCA) on log-transformed contaminant data was also performed, using the SIMCA P + software (Umetrics, Umea, Sweden, version 12.0.1). In addition to the concentrations of contaminants, the variables area, season, age and body condition were included in the model. The data were centered and scaled (to variance 1) prior to modeling and the value of explained variation (R2) was calculated. Determination of the number of significant components and an estimate of the predictive ability of the model (Q2) were performed by cross validation. When analyzing biological data, models with R2 values >0.7 and Q2 values >0.4 are considered to be acceptable (Lundstedt et al., 1998).

A one-year-old mink caught in spring in the M region showed extremely high concentrations of most PCB congeners (sum 124000 ng/g fat, 10.2 standard deviations above mean) and was therefore excluded from all calculations. For samples in which a contaminant was not detected, half of the limit of detection was used in the calculations. Contaminants were excluded from the GLM analysis, the descriptive statistics and the PCA if <33% of the samples had detectable levels of residues. Also, contaminants that were found in very low concentrations (mean < 1 ng/g/fat and no individual concentration higher than 8 ng/g fat) were excluded from the GLM analysis and the PCA. Contaminants that were included are marked with an asterisk (*) in Table 1.

Unless indicated otherwise, the anti-logarithm of the least square means are reported. They are calculated from the multiple regression model and are therefore adjusted for the terms in the model. For arithmetic means, see Supporting Information.

Table 1

Concentrations of total PCBs and congeners, total PBDEs and congeners, Me-O-PBDEs and pesticides in subcutaneous fat from wild mink in Sweden (ng/g lipid weight).3

Nb Meanc ± SDc Medianc Ranged

PCB 28* 82 6.4 ±19.8 1.0 0.23-123

PCB 52* 78 10.0 ±34.0 1.0 0.14-211

PCB 47/48* 88 89.5 ± 320 4.4 0.7-1976

PCB 74* 96 40.1 ± 95.9 8.6 0.6-491

PCB 66* 94 14.1 ± 35.8 3.5 0.3-217

PCB 101* 99 31.7 ±87.5 4.8 0.1-518

PCB 99/113* 100 342 ±816 86.3 2.0-5083

PCB 110* 92 39.3 ± 336 1.1 0.1-3362

PCB 118* 100 401 ± 908 131 6.7-6290

PCB 114/122* 100 7.1 ± 14.4 2.5 0.2-97.3

PCB 105* 100 122 ±239 42.0 1.9-1724

PCB 153* 100 1452 ±2149 591 19.6-10839

PCB 141* 85 2.4 ± 4.0 1.0 0.1-21.4

PCB 138* 100 1345 ±2134 532 22.2-13012

PCB 156* 100 246 ± 381 73.1 3.1-1761

PCB 157* 100 39.3 ± 63.4 12.9 0.2-371

PCB 182/187* 100 63.5 ± 99.0 28.0 1.9-655

PCB 180* 100 1555 ±2625 445 24.7-17277

PCB 183* 100 74.2 ±104 34.0 1.5-566

PCB 170/190* 100 933 ±1566 231 13.4-9079

PCB 189* 99 35.6 ± 54.8 11.4 0.5-339

PCB 194* 100 368 ± 809 70.7 4.2-5185

PCB 206* 100 69.2 ±158.4 15.7 0.7-1047

PCB 209* 100 29.0 ± 53.9 6.8 0.4-319

pPCB* 7316 ±11459 2366 108-55635

PBDE 28 51 0.007-1.1

PBDE 47* 100 19.2 ±30.1 11.0 0.5-245

PBDE 66 31 0.01-0.5

PBDE 85 14 0.008-1.2

PBDE 99* 100 3.9 ± 5.2 2.1 0.1-30.2

PBDE 100* 98 5.8 ± 8.8 2.3 0.2-62.8

PBDE 138 26 0.008-6.1

PBDE 153/154* 100 12.3 ±17.7 6.4 0.8-98.5

PBDE 183 49 0.2 ± 0.4 0.1 0.01-2.0

PBDE 209e 22 0.8 ±1.3 0.5 0.1-7.0

pPBDE 42.5 ± 53.8 25.8 2.0-390

2MeO-PBDE 68f 29 1.6 ±2.4 0.3 0.3-14.2

6MeO-PBDE 47* 67 2.1 ±4.6 0.3 0.03-33.4

4MeO-PBDE 49 1 0.06

5MeO-PBDE 100 1 0.03

4MeO-PBDE 103 1 0.06

5MeO-PBDE 99 3 0.05-0.15

HCB* 99 22.0 ± 47.8 10.6 2.7-439

Cis-Heptachlorepoxide 91 1.3 ±1.6 0.7 0.1-8.3

Cis-chlordane 2 0.3-0.4

Trans-chlordane 32 0.06-0.8

Oxychlordane* 100 21.5 ±27.9 11.6 0.5-156

Trans-nonachlordane 82 1.1 ±1.5 0.4 0.06-6.7

Cis-nonachlordane 36 0.05-1.8

p,p'-DDE* 100 476 ±1097 182 8.8-9708

a 5MeO-PBDE 47,4MeO-PBDE 101 and o,p-DDE were not detected in any sample. b Number of samples with detectable residues (of total 100 samples analyzed). c Arithmetic mean, standard deviation and median for contaminants in which >66% of the samples had detectable residues. Lowest detectable limit was halved and then used for calculations. d Only detectable values are reported. e 39 selected samples were analyzed.

f 39 samples were excluded due to co-elution with internal standard. * Contaminants that were included in the multiple regression model and the PCA.

3. Results

Mink characteristics are summarized in Table 2. The body condition of mink caught in winter was significantly higher (p < 0.05) than that of mink caught in autumn or spring (data not shown). Body condition was a significant covariate for the sum of the PCB concentrations (J^PCB, the sum of 29 congeners) and all PCB congeners in the model except PCB 28, 52, 47/48, 74, 66,101 and 110. It was also significant for the sum of PBDE concentrations (^PBDE,

Table 2

Sample data and characteristics for male mink from four different areas in Sweden.3

Area Age (years) Body weight (kg) Body lengthb (cm) Weight of subcutaneous fatc (g) Capture years

G(n = 25) Mean ± SDd 1.1 ±1.6 1.1 ±0.2 40.5 ±1.9 16.1 ±7.9

Range 0-5 0.81-1.45 37-45 2.7-34 2007-2008

K (n = 26) Mean ± SD 1.0 ±1.1 1.1 ±0.2 41.2 ±2.5 17.3 ±9.6

Range 0-4 0.4-1.5 33-44 0.4-37 2006-2009

M (n = 24) Mean ± SD 0.5 ±0.7 1.0 ±0.1 41.4 ±1.7 9.6 ± 4.9

Range 0-2 0.8-1.2 37.5-44 2.4-20 2004-2008

N (n = 25) Mean ± SD 0.6 ± 0.8 1.0 ±0.2 40.5 ± 2.0 11.9 ±7.1

Range 0-3 0.6-1.3 35-44 3.2-29 2006-2009

a The Gavle Baltic Coast (G), the Koster Islands in Skagerrak (K), the Marsta inland region (M), and the inland of Northern Sweden (N). b The length from nose to base of tail.

c The subcutaneous fat between the hind legs on the ventral part of the abdomen. d Arithmetic mean and standard deviation.

the sum of 11 congeners) and all PBDE congeners included in the model. The relationship was negative; lean mink had higher concentrations of these contaminants than mink with larger fat depots.

The PCA model (R2 = 0.72 and Q2 = 0.57) had four significant components. The scores and loadings plots of component 1 versus 2 explained 58% of the variation in the data (Fig. 1). Corresponding plots of component 3 versus 4, the descriptive data for the components and the R2 and Q2 calculated for each variable are found in the Supporting Information.

and the N area (p < 0.001-0.05), and between the M and the K area (p < 0.01-0.05 but only for congeners 28, 47/48 and 52).

The higher chlorinated congeners 138, 153, 180 and 170/190 constituted 73% of the pPCB. As shown in Table S3, PCB 180 was the predominant congener in the G area (25% of PCB), followed by congeners 153 (20%), 138 (19%) and 170/190 (15%). The M and N areas showed a similar PCB congener profile as the G area. In the K area the pattern was different, as 153 was the most abundant congener (27%), followed by PCB 138 (22%), PCB 180 (20%) and PCB 170/190 (11%).

3.1. PCBs

Age influenced the ]TPCB concentrations significantly (Table 3), but when comparing age groups with least square means, this was only true for congeners 138, 156, 157, 180, 170/190, 189, 194, 206 and 209 (Table S4). For these congeners and PCB, mink that were 2 years or older had significantly higher concentration than the juveniles and the one-year-old mink (p < 0.0001-0.05). There was no significant difference between juveniles and the one-year-olds for any congener.

Season significantly influenced the concentrations of most PCB congeners, with the exception of PCB 101, 110, 141 and 182/187 (Table S2). PCB and congeners that showed seasonal variation were present at significantly higher concentrations during spring than in autumn (p = 0.0001-p < 0.05; Table S5). In addition, pPCB and congeners 52, 47/48, 156, 157, 180, 183, 170/190, 189, 194, 206, 209 showed a significantly lower concentration in mink caught during winter compared to spring (p = 0.01-0.05). There were no significant differences in concentration between autumn and winter for any congener.

Least square mean of PCB was significantly higher in the G area (9995 ng/g lipid weight (lw), p <0.0001-0.001) than in the K, M and N areas (3194, 3058 and 1534 ng/g lw, respectively; Table S3). The PCA showed that individual mink with the highest concentrations of PCBs came from the M and G areas whereas mink from the N and K areas in general contained relatively low concentrations. It can be seen in the scores and loadings plot (Fig. 1) that mink from the M area showed a large variation in contaminant burden, and tended to have high concentrations of the low-chlorinated congeners 28, 47/48, 52, 66, and 74. Some of the mink from the M area were located outside the Hotelling T2 ellipse (0.95). These results were confirmed when least square means were compared, but significant differences were only found between the M area

3.2. Pesticides

The analyzed pesticides showed no systematic variation due to season, age or body condition. Oxychlordane was predominant among the chlordane components analyzed. Other chlordane components were found in lower concentrations (Table 1). In the PCA, the pesticides co-varied with brominated compounds and were mainly found in high concentrations in mink from the G area. The higher concentrations of pesticides in the G area were confirmed by least square mean comparisons; significantly higher concentrations of HCB, oxychlordane and p,p'-DDE than in mink from the K, M and N areas (p <0.001-0.01) were found (Table S7). Mink from the M area had significantly lower concentrations of oxychlordane than mink from the K and N areas (p <0.0001).

3.3. PBDEs

Age did not influence PBDE 153/154 in the overall model, but the least square mean comparisons showed that the concentrations were higher in the 2 years and older mink than in juvenile mink (p < 0.01; data not shown). Also, there was a seasonal influence on PBDE 153/154 (Table 3), with higher concentrations during spring than in winter (p < 0.05; data not shown). In addition, least square means of PBDE 47 showed higher concentrations during spring than in autumn (p < 0.05). However, neither PBDE concentrations nor concentrations of congeners 99 and 100 varied with season or age.

Congeners 28, 66, 85, 138, and 183 were found in low concentrations (Table 1). PBDE 47 was the most abundant congener (45% of total PBDEs), followed by PBDE 153/154 (29%), PBDE 100 (14%), and PBDE 99 (9%). This congener profile was true for all areas and all age classes, except in mink that were 2 years and old-

Fig. 1. (A) Principal component analysis (PCA) scores plot of components 1 and 2, with individual mink from the Gâvle Baltic Coast (G), the Koster Islands in Skagerrak (K), the Marsta inland region (M), and the inland of Northern Sweden (N). (B) The corresponding loadings plot of components 1 and 2 showing PCBs, PBDEs and pesticides in mink adipose tissue.

er. In this group, PBDE 153/154 was as predominant as PBDE 47 (40.6% and 40.3% of total PBDEs, respectively).

The PCA (Fig. 1) shows that high concentrations of PBDEs were particularly found in mink from the G area, but also in some mink from the K area. Least square mean for ]TPBDE concentrations was 45.7 ng/g lw in the G area, which was significantly (p < 0.001-0.05) higher than in the other three areas (15.3-27.3 ng/g lw; Table S6). Concentrations of PBDE congeners 47, 99 and 100, but not 153/ 154, were significantly influenced by area (p < 0.01; Table S2).

6MeO-PBDE 47 was detected in 67% of the mink in this study. 2MeO-PBDE 68 was detected in 29 mink and its concentration was below the detection limit in 32 mink samples (Table 1). All other methoxylated PBDE compounds were found in only a few samples and in very low concentrations, i.e. three- to ten-fold lower than the concentrations of PBDEs. No effect of season, age or body condition was found for 6MeO-PBDE 47. As shown in Table S6, mink from the G area had higher (p < 0.0001) concentrations of 6MeO-PBDE 47 (3.6 ng/g lw) than those from the K (0.4 ng/g lw),

Table 3

Analysis of variance for concentrations of selected chemicals in subcutaneous fat from wild mink a.

Dependent variable R2b Source of variance (level of significance)

Agec Seasonc Body conditionc Areac

PCBs pPCB PCB#153 PCB#180

Pesticides p.p-DDE HCB

Oxychlordane

PBDEs pPBDE PBDE 153/154 6Me-O-PBDE 47

47 40 51

44 32 57

25 28 49

0.0399

0.0038

n.s. n.s. n.s.

n.s n.s. n.s.

0.0036 0.0204 0.0087

n.s. n.s n.s.

0.0453 n.s.

0.0005 0.0017 0.0002

n.s. n.s. n.s.

0.0015 0.0016 n.s.

<.0001 <.0001 <.0001

<.0001 <.0001 <.0001

0.0024

<.0001

a Full data set is available in the supporting information (Table S2). b Coefficient of determination for the statistical model (%), i.e. the proportion of variability that is accounted for by the statistical model. c The effect of age, season, body condition and area, respectively. d Not significant (p < 0.05).

M (0.2 ng/g lw) and N areas (0.3 ng/g lw). In the loadings plot from the PCA model, PBDE 47 and 6MeO-PBDE47 are located close to each other (Fig. 1). The correlation between these variables was 0.31 (R2).

4. Discussion

To our knowledge, this is the first study on wild mink to report influence of age on concentrations of specific PCB and PBDE congeners, and to report seasonal variations in contaminant concentrations.

There are few studies addressing the influence of age on contaminant concentrations in mink. In one study, PCB concentrations were significantly higher in adults than in juveniles in some sites, which is in line with the findings in this study (Martin et al., 2006). In another study, no age differences between mink <2 years old and mink p2 years old were found for ]TPCB or HCB, but concentrations were higher in juveniles than in adults for DDT and chlordane (Poole et al., 1995). However, as the authors stated, the statistical method that was used might not have revealed true age differences due to small sample size and spatial differences in concentrations. In contrast to those findings, but in line with the results in this study, earlier work stated that there was no age influence on ]TDDT in mink (Franson et al., 1974).

4.1. PCBs

Although PCBs are known to bioaccumulate, there were no significant differences in chemical concentrations between the juveniles and the one-year-olds. This could possibly be due to diet differences or a growth dilution effect. The significant age-related differences between mink that were 2 years or older and mink in the two younger age categories could, however, be assumed to be mainly due to bioaccumulation, since the congeners in question are highly chlorinated. Biotransformation rates of highly chlorinated congeners are generally lower than those of less chlorinated congeners, which have been shown in, for example, otters (Leonards et al., 1997; Giesy and Kannan, 1998) and other mustelids (Leonards et al., 1998).

Leaner mink had higher concentrations of PCBs than fat mink. A reason for this may be that rapid mobilization of fat reserves, for example during times of starvation, results in an increased concentration of PCBs in adipose tissues (Wania, 1999). This has been seen

in other high trophic level species such as polar bears (Polischuk et al., 2002).

In this study, the PCB concentrations were generally higher during spring than in autumn. Possibly, the seasonal variation in PCB concentrations is not solely driven by lipid dynamics. The variation could also be explained by seasonal changes in concentrations in the prey. For example, PCB concentrations in herring in the Baltic Sea have been shown to change seasonally, with higher concentrations in spring-summer than in autumn (Bignert et al., 2007). Another explanation could be seasonal shifts in food preferences, as the diet of the mink has been shown to be influenced by season (Gerell, 1967; Jedrzejewska et al., 2001).

The mean PCB concentration in area G in this study is similar to concentrations found in mink from both northern and southern Sweden almost 20 years before our mink were sampled (Larsson et al., 1990). In addition, the congener profiles found in areas G, M and N are similar to the profile found in mink in that study. Notably, the marine west coast area (K) showed a different profile with PCB 153 as the most abundant congener. In comparison to the other areas, the inland fresh water environment, area M, showed a relatively large variation in PCB concentrations, as well as in concentrations of the PBDEs and pesticides. This intraregional variation could be a result of the relatively small home range of the mink, causing exposure of some mink to point sources and other mink to average background exposure. Possible point sources could be the adjacent large international airport or the former training camp for fire-fighters. In the M area, the PCB congener profile indicated a more recent exposure than in the other areas, as the concentrations of the less chlorinated PCBs were found to be relatively high compared to the concentrations in the other areas.

4.2. Pesticides

The lack of age dependence of HCB concentrations seen in this study is in line with previous findings in mink (Poole et al., 1998). In other species, the results are contradictive; HCB concentrations were significantly higher in adult male otters (p2 years) than in juveniles (Grove and Henny, 2008), whereas results from polar bears indicate that age has no influence on HCB concentrations (Bernhoft et al., 1997). In the study on male otters by Grove and Henny (2008), a significant increase between juveniles and adults was also found for oxychlordane and DDE concentrations, which is contrary to the findings in the present study.

Concentrations of p,p'-DDE in the mink in this study were similar to those found in a previous study on mink from Sweden (Larsson et al., 1990). Mink from the Baltic coast area (G) had much higher concentrations of DDE, HCB and oxychlordane than mink from the other areas. Higher concentrations of HCB and DDE on the Baltic coast than on the west coast of Sweden is in line with findings in herring (Bignert et al., 2011).

4.3. PBDEs

PBDE 47 was the predominant congener in the mink of this study, a pattern which is also seen in marine mammals (Law et al., 2003). However, PBDE 153/154 had the second highest concentration and was even similar to PBDE 47 in concentration in the oldest mink. This is usually not the case in marine mammals, but in line with findings for polar bears (Muir et al., 2005). An additional finding in that study was that PBDE 153 was the most bioaccumu-lative congener, which is in accordance with observations on farmed mink (Zhang et al., 2008). This was also indicated by the findings in this study, as PBDE153/154 was the only congener to significantly increase with age. In other species, there seems to be a lack of influence of age on concentrations of PBDEs, for

example in polar bears (Muir et al., 2005), arctic fox (Fuglei et al., 2007) and river otters (Stansley et al., 2010).

Least square means showed seasonal variation in the concentrations of PBDE 153/154 and PBDE 47 in mink in this study. The sources of variation are probably similar to those of the PCBs (as discussed in Section 4.1). Season should therefore not be dismissed as an influencing factor when monitoring PBDEs in mink.

In the literature, there are limited data available on PBDE concentrations in wild mink. Liver concentrations ranging from 29.4 to 2890 ng/g lw (^PBDE), with a median of 136 ng/g lw, were reported for 20 mink in the Great Lakes area (Zhang et al., 2009), which is higher than the concentrations in our study.

The concentrations of MeO-PBDEs were generally low in the present study and the concentrations were below those found in fish from the southern Bothnian Sea (Kierkegaard et al., 2004). This indicates a relatively rapid metabolism of these compounds or it could possibly reflect the semi-terrestrial diet of the mink. The predominance of 6MeO-PBDE 47 is in line with previous findings in Baltic biota, such as herring, salmon and seal (Haglund et al., 1997). Mink from the Baltic coast area (G, with brackish water) had the highest concentrations of 6MeO-PBDE 47. However, some inland mink (13 from the N area and 15 from the M area) contained very low but detectable concentrations. This might indicate a freshwater source of exposure, even though the origin is unknown. The concentrations of methoxylated compounds were much lower than the concentrations of PBDEs, unlike the findings in other studies in Swedish inland and Baltic biota where the concentrations of MeO-PBDEs were similar to or exceeded the concentrations of PBDEs (Haglund et al., 1997; Asplund et al., 1999; Kierkegaard et al., 2004). The relatively weak correlation between 6MeO-PBDE 47 and the tentative parent compound PBDE 47 indicates that a metabolic hydroxylation and methoxylation in mink is unlikely.

5. Conclusions

In conclusion, mink from the four areas showed different contaminant patterns and there were also variations in concentrations of the various pollutants between individual mink within the areas. Both the PCA and the multiple regression model showed variation in contaminant burden connected to age, season and body condition, especially for the PCB and PBDE congeners. This information is, for example, important when using exposure data in assessment of contaminant-related health effects, in temporal and spatial trend studies, and for understanding trophic transfer of contaminants. The findings in this study suggest that it is not adequate to only distinguish between juveniles and adults, as the 2 years and older mink may be the ones with higher concentrations of certain contaminants. For monitoring of environmental contamination, it is recommended to sample 2 years and older mink during spring to include individuals with the highest body burden of PCBs and PBDEs.

Acknowledgements

The Environmental Monitoring program at the Swedish University of Agricultural Sciences is thanked for financial support. The mink hunters B. Almberg, G. Anttila, A. Degermark, B. Engstrom, M. Eriksson, W. Eriksson, T Hall, J. Karlsson, S. Karlsson, S-G. Lunne-ryd, M. Nilsson, B. Nyberg, A. Olofsson, E. Olsson, S. Sundin and S-A. Angwald are thanked for their work of providing mink carcasses for this study.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/jj.chemosphere. 2012.09.060.

References

AMAP, 1998. AMAP assessment report: Arctic pollution issues. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway.

AMAP, 2004. AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway.

Addison, R., Smith, T., 1974. Organochlorine residue levels in Arctic ringed seals: variation with age and sex. Oikos, 335-337.

Andersson, O., Wartanian, A., 1992. Levels of polychlorinated camphenes (toxaphene), chlordane compounds and polybrominated diphenyl ethers in seals from Swedish waters. Ambio, 550-552.

Asplund, L., Athanasiadou, M., Sjodin, A., Bergman, A., Borjeson, H., 1999. Organohalogen substances in muscle, egg and blood from healthy Baltic salmon (Salmo salar) and Baltic salmon that produced offspring with the M74 syndrome. Ambio 28, 67-76.

Basu, N., Scheuhammer, A.M., Bursian, S.J., Elliott, J., Rouvinen-Watt, K., Chan, H.M., 2007. Mink as a sentinel species in environmental health. Environ. Res. 103, 130-144.

Bergek, S., Bergqvist, P.A., Hjelt, M., Olsson, M., Rappe, C., Roos, A., Zook, D., 1992. Concentrations of PCDDs and PCDFs in seals from Swedish waters. Ambio, 553556.

Bernhoft, A., Wiig, O., Skaare, J.U., 1997. Organochlorines in polar bears (Ursus maritimus) at Svalbard. Environ. Pollut. 95, 159-175.

Bignert, A., Nyberg, E., Sundqvist, K.L., Wiberg, K., 2007. Spatial variation in concentrations and patterns of the PCDD/F and dioxin-like-PCB content in herring from the northern Baltic Sea. J. Environ. Monit. 9, 550-556.

Bignert, A., Boalt, E., Danielsson, S., Hedman, J., Johansson, A.K., Miller, A., Nyberg, E., Berger, U., Borg, H., Eriksson, U., 2011. Comments concerning the national Swedish contaminant monitoring programme in marine biota, 2011. Annual report to the Swedish Environmental Protection Agency.

Bleavins, M., Aulerich, R., Ringer, R., 1981. Placental and mammary transfer of polychlorinated and polybrominated biphenyls in the mink and ferret. In: Avian and Mammalian Wildlife Toxicology. American Society for Testing and Materials, Philadelphia, PA, pp. 121-131.

Blomkvist, G., Roos, A., Jensen, S., Bignert, A., Olsson, M., 1992. Concentrations of DDT and PCB in seals from Swedish and Scottish waters. Ambio. Stockholm 21, 539-545.

Borga, K., Fisk, A.T., Hoekstra, P.F., Muir, D.C.G., 2004. Biological and chemical factors of importance in the bioaccumulation and trophic transfer of persistent organochlorine contaminants in arctic marine food webs. Environ. Toxicol. Chem. 23, 2367-2385.

Bytingsvik, J., Lie, E., Aars, J., Derocher, A.E., Wiig, 0., Jenssen, B.M., 2012. PCBs and OH-PCBs in polar bear mother-cub pairs: A comparative study based on plasma levels in 1998 and 2008. Sci. Total Environ. 1 (1).

Covaci, A., Voorspoels, S., Ramos, L., Neels, H., Blust, R., 2007. Recent developments in the analysis of brominated flame retardants and brominated natural compounds. J. Chromatogr. A. 1153,145-171.

de Wit, C.A., Alaee, M., Muir, D.C.G., 2006. Levels and trends of brominated flame retardants in the Arctic. Chemosphere 64, 209-233.

Franson, J.C., Dahm, P.A., Wing, L.D., 1974. Chlorinated hydrocarbon insecticide residues in adipose, liver, and brain samples from Iowa mink. Bull. Environ. Contam. Toxicol. 11, 379-385.

Fuglei, E., Bustnes, J.O., Hop, H., Mork, T., Bjornfoth, H., van Bavel, B., 2007. Environmental contaminants in arctic foxes (Alopex lagopus) in Svalbard: relationships with feeding ecology and body condition. Environ. Pollut. 146, 128-138.

Gerell, R., 1967. Food Selection in Relation to Habitat in Mink (Mustela vison Schreber) in Sweden. Oikos 18, 233-246.

Gerell, R., 1970. Home Ranges and Movements of the Mink (Mustela vison Shreber) in Southern Sweden. Oikos 21,160-173.

Giesy, J.P., Kannan, K., 1998. Dioxin-like and non-dioxin-like toxic effects of polychlorinated biphenyls (PCBs): implications for risk assessment. Crit. Rev. Toxicol. 28, 511-569.

Grove, R.A., Henny, C.J., 2008. Environmental contaminants in male river otters from Oregon and Washington, USA, 1994-1999. Environ. Monit. Assess. 145, 49-73.

Haglund, P.S., Zook, D.R., Buser, H.R., Hu, J.W., 1997. Identification and quantification of polybrominated diphenyl ethers and methoxy-polybrominated diphenyl ethers in Baltic biota. Environ. Sci. Technol. 31, 3281-3287.

Jedrzejewska, B., Sidorovich, V.E., Pikulik, M.M., Jedrzejewski, W., 2001. Feeding habits of the otter and the American mink in Bialowieza Primeval Forest (Poland) compared to other Eurasian populations. Ecography 24, 165-180.

Jensen, S., Johnels, A., Olsson, M., Otterlind, G., 1969. DDT and PCB in marine animals from Swedish waters. Nature 224, 247.

Kierkegaard, A., Bignert, A., Sellstrom, U., Olsson, M., Asplund, L., Jansson, B., de Wit, C.A., 2004. Polybrominated diphenyl ethers (PBDEs) and their methoxylated derivatives in pike from Swedish waters with emphasis on temporal trends, 1967-2000. Environ. Pollut. 130, 187-198.

Koistinen, J., Stenman, O., Haahti, H., Suonpera, M., Paasivirta, J., 1997. Polychlorinated diphenyl ethers, dibenzo-p-dioxins, dibenzofurans and biphenyls in seals and sediment from the Gulf of Finland. Chemosphere 35, 1249-1269.

Larsson, P., Woin, P., Knulst, J., 1990. Differences in uptake of persistent pollutants for predators feeding in aquatic and terrestrial habitats. Holarct. Ecol. 13,149155.

Law, R.J., Alaee, M., Allchin, C.R., Boon, J.P., Lebeuf, M., Lepom, P., Stern, G.A., 2003. Levels and trends of polybrominated diphenylethers and other brominated flame retardants in wildlife. Environ. Int. 29, 757-770.

Law, R.J., Allchin, C.R., de Boer, J., Covaci, A., Herzke, D., Lepom, P., Morris, S., Tronczynski, J., de Wit, C.A., 2006. Levels and trends of brominated flame retardants in the European environment. Chemosphere 64,187-208.

Leonards, P.E.G., Broekhuizen, S., de Voogt, P., Van Straalen, N.M., Brinkman, U.A.T., Cofino, W.P., van Hattum, B., 1998. Studies of bioaccumulation and biotransformation of PCBs in mustelids based on concentration and congener patterns in predators and preys. Arch. Environ. Contam. Toxicol. 35, 654-665.

Leonards, P.E.G., Zierikzee, Y., Brinkman, U.A.T., Cofino, W.P., vanStraalen, N.M., vanHattum, B., 1997. The selective dietary accumulation of planar polychlorinated biphenyls in the otter (Lutra lutra). Environ. Toxicol. Chem. 16,1807-1815.

Lundstedt, T., Seifert, E., Abramo, L., Thelin, B., Nystrom, A., Pettersen, J., Bergman, R., 1998. Experimental design and optimization. Chemometrics Intellig. Lab. Syst. 42, 3-40.

Martin, P.A., McDaniel, T.V., Hunter, B., 2006. Temporal and spatial trends in chlorinated hydrocarbon concentrations of mink in Canadian Lakes Erie and St. Clair. Environ. Monit. Assess. 113, 245-263.

Matson, G.M., 1981. Workbook for Cementum Analysis. Matson's Laboratory, Milltown, Montana. 30 pp.

McKinney, M.A., Letcher, R.J., Aars, J., Born, E.W., Branigan, M., Dietz, R., Evans, T.J., Gabrielsen, G.W., Muir, D.C.G., Peacock, E., Sonne, C., 2011. Regional contamination versus regional dietary differences: understanding geographic variation in brominated and chlorinated contaminant levels in Polar bears. Environ. Sci. Technol. 45, 896-902.

Muir, D.C.G., Backus, S., Derocher, A.E., Dietz, R., Evans, T.J., Gabrielsen, G.W., Nagy, J., Norstrom, R.J., Sonne, C., Stirling, I., Taylor, M.K., Letcher, R.J., 2005. Brominated flame retardants in polar bears (Ursus maritimus) from Alaska, the Canadian Arctic, East Greenland, and Svalbard. Environ. Sci. Technol. 40, 449-455.

Persson, S., Brunstrom, B., Backlin, B.M., Kindahl, H., Magnusson, U., 2012. Wild mink (Neovison vison) as sentinels in environmental monitoring. Acta Vet. Scand. 54, 1-3.

Persson, S., Backlin, B.-M., Kindahl, H., Brunstrom, B., Magnusson, U., 2011. Influence of age, nutritional status and season on the reproductive system in wild male mink (Neovison vison). Eur. J. Wildl. Res., 1-7.

Polischuk, S.C., Norstrom, R.J., Ramsay, M.A., 2002. Body burdens and tissue concentrations of organochlorines in polar bears (Ursus maritimus) vary during seasonal fasts. Environ. Pollut. 118, 29-39.

Poole, K.G., Elkin, B.T., Bethke, R.W., 1995. Environmental contaminants in wild mink in the Northwest Territories. Canada. Sci. Total Environ. 160-61, 473-486.

Poole, K.G., Elkin, B.T., Bethke, R.W., 1998. Organochlorine and heavy metal contaminants in wild mink in western Northwest Territories, Canada. Arch. Environ. Contam. Toxicol. 34, 406-413.

Roos, A., Bergman, A., Greyerz, E., Olsson, M., 1998. Time trend studies on (sigma) DDT and PCB in juvenile greys seals (Halichoerus grypus), fish and guillemot eggs from the Baltic Sea. Organohalog. Compd. 39,109-112.

Rotander, A., van Bavel, B., Polder, A., Riget, F., Au3unsson, G.A., Gabrielsen, G.W., Vikingsson, G., Bloch, D., Dam, M., 2012. Polybrominated diphenyl ethers (PBDEs) in marine mammals from Arctic and North Atlantic regions, 19862009. Environ. Int. 40, 102-109.

Stansley, W., Velinsky, D., Thomas, R., 2010. Mercury and halogenated organic contaminants in river otters (Lontra canadensis) in New Jersey, USA. Environ. Toxicol. Chem. 29, 2235-2242.

Teuten, E.L., Xu, L., Reddy, C.M., 2005. Two abundant bioaccumulated halogenated compounds are natural products. Science 307, 917-920.

Wania, F., 1999. On the origin of elevated levels of persistent chemicals in the environment. Environ. Sci. Pollut. Res. 6, 11-19.

Yordy, J.E., Wells, R.S., Balmer, B.C., Schwacke, L.H., Rowles, T.K., Kucklick, J.R., 2010. Life history as a source of variation for persistent organic pollutant (POP) patterns in a community of common bottlenose dolphins (Tursiops truncatus) resident to Sarasota Bay, FL. Sci. Total Environ. 408, 2163-2172.

Zhang, S., Bursian, S., Martin, P.A., Chan, H.M., Martin, J.W., 2008. Dietary accumulation, disposition, and metabolism of technical pentabrominated diphenyl ether (dE-71) in pregnant mink (Mustela vison) and their offspring. Environ. Toxicol. Chem. 27,1184-1193.

Zhang, S., Bursian, S.J., Martin, P.A., Chan, H.M., Tomy, G., Palace, V.P., Mayne, G.J., Martin, J.W., 2009. Reproductive and developmental toxicity of a pentabrominated diphenylether mixture, DE-71, to ranch mink (Mustela vison) and hazard assessment for wild mink in the Great Lakes region. Toxicol. Sci. 110,107-116.