Document 12070946

advertisement
Environmental Toxicology and Chemistry, Vol. 27, No. 10, pp. 2076–2087, 2008
䉷 2008 SETAC
Printed in the USA
0730-7268/08 $12.00 ⫹ .00
EXPOSURE AND EFFECTS ASSESSMENT OF RESIDENT MINK (MUSTELA VISON)
EXPOSED TO POLYCHLORINATED DIBENZOFURANS AND OTHER DIOXIN-LIKE
COMPOUNDS IN THE TITTABAWASSEE RIVER BASIN, MIDLAND, MICHIGAN, USA
MATTHEW J. ZWIERNIK,*† DENISE P. KAY,‡ JEREMY MOORE,† KERRIE J. BECKETT,§ JONG SEONG KHIM,㛳
JOHN L. NEWSTED,‡ SHAUN A. ROARK,‡ and JOHN P. GIESY†㛳#
†Department of Zoology, National Food Safety and Toxicology Center, 224 National Food Safety and Toxicology Building,
Michigan State University, East Lansing, Michigan 48824, USA
‡ENTRIX, 4295 Okemos Road, Okemos, Michigan 48864, USA
§Woodlot Alternatives, 122 Main Street, Topsham, Maine 04086, USA
㛳Department of Biomedical Veterinary Sciences and Toxicology Centre, University of Saskatchewan,
Saskatoon, Saskatchewan S7J 5B3, Canada
#Department of Biology and Chemistry, City University of Hong Kong, Kowloon, Hong Kong, Special Administrative Region, China
( Received 7 September 2007; Accepted 12 March 2008)
Abstract—Historically, sediments and floodplain soils of the Tittabawassee River (TR; MI, USA) have been contaminated with
polychlorinated dibenzofurans (PCDFs), polychlorinated dibenzo-p-dioxins (PCDDs), and polychlorinated biphenyls (PCBs). Median
concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalents (TEQs) based on 2006 World Health Organization tetrachlorodibenzo-p-dioxin toxic equivalency factors (TEFs) in the diet of mink (Mustela vison) ranged from 6.8 ⫻ 10⫺1 ng TEQ/kg wet
weight upstream of the primary source of PCDF to 3.1 ⫻ 101 ng TEQ/kg wet weight downstream. Estimates of toxicity reference
values (TRVs) derived from laboratory studies with individual PCDDs/PCDFs and PCB congeners or mixtures of those congeners,
as well as application of TEFs, were compared to site-specific measures of mink exposure. Hazard quotients based on exposures
expressed as concentrations of TEQs in the 95th percentile of the mink diet or liver and the no-observable-adverse-effect TRVs
were determined to be 1.7 and 8.6, respectively. The resident mink survey, however, including number of mink present, morphological
measures, sex ratios, population age structure, and gross and histological tissue examination, indicated no observable adverse effects.
This resulted for multiple reasons: First, the exposure estimate was conservative, and second, the predominantly PCDF congener
mixture present in the TR appeared to be less potent than predicted from TEQs based on dose–response comparisons. Given this,
there appears to be great uncertainty in comparing the measured concentrations of TEQs at this site to TRVs derived from different
congeners or congener mixtures. Based on the lack of negative outcomes for any measurement endpoints examined, including jaw
lesions, a sentinel indicator of possible adverse effects, and direct measures of effects on individual mink and their population, it
was concluded that current concentrations of PCDDs/PCDFs were not causing adverse effects on resident mink of the TR.
Keywords—Mink
Jaw lesions
Polychlorinated dibenzofurans
2,3,7,8-Tetrachlorodibenzo-p-dioxin equivalents
Hazard assessment
oratory to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and structurally related compounds have exhibited responses including
reproductive and developmental toxicity, such as decreased litter
size, growth of kits, reduced body weight, jaw lesions, and mortality [4–7]. Based on the measured concentrations of TCDD
equivalents (TEQs) in some dietary items and the results of controlled laboratory studies with related chemicals, the mink population of the TR would have been expected to be affected at
the population level. Yet, during preliminary work, mink frequently were sighted, and evidence in the form of tracks and scat
indicates a substantial presence of mink on the TR. The present
study was conducted in an effort to determine the reason for this
apparent contradiction.
Site-specific information, including dietary composition,
concentrations of PCDDs/PCDFs in mink tissues and dietary
items, mink abundance, and measures of mink individual
health, were used in a multiple-lines-of-evidence approach to
evaluate the potential of adverse effects to wild mink chronically exposed to these compounds in the TR basin. The potential for effects was predicted using a hazard assessment
approach in which concentrations of TEQs in the diets and
livers of mink from the TR were compared to toxicity reference
values (TRVs). Measures of individual health, including body
condition, age, body weight, and gross morphology, were re-
INTRODUCTION
The Tittabawassee River (TR) in central Michigan (USA)
is the largest tributary of the Saginaw River, draining 5,426
km2 of mostly woodlands and agricultural lands. The TR and
its tributaries, the Chippewa and Pine rivers, contribute nearly
50% of the total water discharge of the Saginaw River (Fig.
1). Midland (MI, USA) is a major industrial and population
center on the TR, where significant concentrations of polychlorinated dibenzofurans (PCDFs) and polychlorinated dibenzo-p-dioxins (PCDDs) have been found in sediments and
floodplain soils, including some of the greatest concentrations
reported [1]. The Michigan Department of Public Health has
issued fish advisories based on elevated concentrations of
PCDFs, PCDDs, and polychlorinated biphenyls (PCBs) in tissues of fish collected downstream of Midland.
Mink (Mustela vison) are top mammalian predators, which
consume great amounts of fish and are among the more sensitive
species to the effects of PCDDs, PCDFs, and PCBs [2]. Thus,
mink often are considered as a surrogate or sentinel species in
ecological hazard assessments for river systems contaminated
with these classes of chemicals [2,3]. Mink exposed in the lab* To whom correspondence may be addressed
(zwiernik@msu.edu).
Published on the Web 4/24/2008.
2076
Multiple-lines-of-evidence assessment for mink
Environ. Toxicol. Chem. 27, 2008
2077
Fig. 1. Tittabawassee River (MI, USA) study and reference areas. 1 ⫽ Tittabawassee River (reference); 2 ⫽ Chippewa River (reference); 3 ⫽
Pine River (reference); 4 ⫽ Tittabawassee River (study area).
corded for 48 field-collected mink. Tissue weights were measured for and histological examination performed on select
tissues, including liver, brain, kidney, and jaw, with particular
care given to evaluating the incidence of mandibular and maxillary squamous cell proliferation, which has been reported to
be a sensitive response of mink to aryl hydrocarbon receptor
(AhR)–active compounds that occur at exposures less than
those required to cause effects on survival or growth of adults
or kits [7]. The results of the predicted hazard for TR mink
were then compared to the direct field measures of individual
health and population dynamics, and the results are discussed
relative to known uncertainties.
MATERIALS AND METHODS
Study area, habitat, and abundance
The study area was located on the portion of the TR from
the Dow Dam in Midland to the confluence of the Shiawassee
River just upstream of Saginaw (MI, USA) (Fig. 1). Portions
of the lower Chippewa and Pine rivers were used as reference
areas. The quality of mink (M. vison) habitat was evaluated
based on the habitat suitability index protocols developed and
validated for the midwestern United States by the U.S. Fish
and Wildlife Service [8]. Habitat suitability indices were de-
termined within a 200-m segment that was randomly selected
on each side of the river and centered approximately every
500 m at the top, middle, and bottom of the defined, 1.0-km
reaches within the study and reference areas. The presence,
relative abundance, and estimates of population were based
on visual observations of mink, trapping data, and track surveys. Relative abundance was based on trapping data quantified in terms of success per unit effort. Estimates of population were determined by multiple years of track survey data.
To focus on resident breeding adults and to minimize the influence of trapping adults during the previous winter, surveys
of mink tracks and scat were preformed in mid-October of
2005 and 2006 after dispersion of kits and before breeding
activity [9,10]. The river was divided into 1.0-km intervals,
and results of population surveys are reported on a per-kilometer basis. Tracks or scat indicated the presence or absence
of mink in each survey area. Photographic records were made
of all tracks and scat along with Global Positioning System
coordinates as well as time and date. A 100-m buffer zone at
each end of the 1.0-km interval was excluded from potential
survey site selection to decrease the chance of counting a single
track set as a positive track sighting in multiple intervals. Track
survey data are reported as both percentage occurrences for
2078
Environ. Toxicol. Chem. 27, 2008
M.J. Zwiernik et al.
comparisons of relative abundance as well as an estimate of
the number of mink per kilometer. It is reported that mink
movement within a single activity period generally is less than
300 m [10]. Given the survey method and the conservative
assumption that an individual mink would leave 300 m of
shoreline tracks in any activity period, it is possible to determine the relationship between the numbers of track sets (i.e.,
mink) present within 1 km of shoreline and the frequency of
positive track surveys (Eqn. 1):
mink
⫽ 0.1515e 0.0318(% positive track surveys)
km shoreline
(1)
Mink trapping and collection of mink prey items
Forty-eight wild mink (22 from the study area and 26 from
the reference areas) were collected throughout the TR drainage
basin for assessments of stressor exposure, dietary composition, and overall condition. All mink and mink prey items were
collected from within the 100-year floodplain. Animals collected in the present study were handled in accordance with
guidelines established by the Michigan State University Institutional Animal Care and Use Committee. All prey items,
with the exception of muskrats (Ondatra zibethicus), were
collected at six predetermined biological sampling areas
(BSAs). Each BSA included two 30- ⫻ 30-m grids proximal
to the river bank, one for terrestrial sampling and one for
aquatic sampling. Four study area BSAs were downstream of
Midland, and two reference area BSAs were upstream. Study
area BSA locations were selected based on maximal exposure
potential given the available data and were located at the Caldwell boat launch, Tittabawassee Township Park, Freeland Festival Park, and Imerman Park, which were 5.3, 10, 10.9, and
17.7 km, respectively, downstream of the Dow Dam in Midland. The reference area BSAs were located on the TR at the
city of Sanford (MI, USA), located 15 km upstream of Midland, and on an upstream tributary just downstream of the
confluence of the Pine and Chippewa rivers at the Chippewa
Nature Center in Midland. Reference area locations were selected based on congruency with study area habitat and available contaminant data.
Mink and muskrats were caught using the same trap sets
during the winters of 2003 to 2005 and were processed individually. Electrofishing and seining were used to collect composite samples of fish species preyed on by mink at each BSA.
Small mammals were collected with Sherman and pitfall traps.
Crayfish were collected with seines and modified minnow
traps. Fish, small mammals, and crayfish were collected in
September 2003. The latter two dietary items were sampled
again in May and June of 2004. Three species of frogs—
northern leopard (Rana pipiens), green (Lithobates clamitans), and gray tree frogs (Hyla versicolor)—were sampled
by hand catching in June 2005. Excluding forage fish composites, whole bodies of all prey items were analyzed individually. The gastrointestinal (GI) tracts of small mammals
were rinsed out, and pelts of muskrats were removed, before
whole-body analysis.
Mink necropsy
Necropsies of mink were conducted by a board-certified
veterinary pathologist at the Michigan State University Diagnostic Center for Population and Animal Health (Lansing,
MI, USA) to determine overall condition, nutritional status,
and presence of gross abnormalities. Gender, body mass, and
body length, both including and excluding the tail, were determined for each mink. An upper and lower premolar tooth
was used to age each mink by microscopic analysis of the
tooth’s cementum annuli [11]. Livers were removed and the
total mass determined, and livers were then apportioned for
analysis. Two grams of liver were placed in a 10% formalin–
saline solution (10% formalin in 0.9% sodium chloride) for
subsequent histological examination. The remaining liver tissue was divided into three 10-g aliquots, placed in I-Chem威
jars (I-Chem, New Castle, DE, USA), and frozen at ⫺20⬚C
for storage before quantification of PCDDs, PCDFs, and PCBs.
Kidneys, GI tracts, and brains were removed and their masses
determined. Contents of GI tracts were preserved for prey item
identification. The baculum (os penis) was removed from
males, and the uterus, including the horns and ovaries, was
removed from females and preserved for subsequent examination. Uteri were examined grossly and histologically for
implantation sites based on placental scaring using a Prussian
blue histochemical technique. Signs of early embryonic death
or late fetal death also were noted. In addition, ovaries were
examined histologically for signs of ovulation and degree of
follicular development. Heads were decalcified in Decal II
solution (SurgiPath, Richmond, IL USA), and upper and lower
jaws were examined histologically for mandibular and maxillary squamous epithelial cell proliferation as described previously [7]. Lesions were graded as mild, moderate, or severe
based on the number and size of foci of squamous cell proliferation in the maxilla and mandible [7].
Mink diets
Composition of the diet as estimated from contents of the
GI tract were as described previously [12]. Prey items were
identified to the lowest practical taxonomic classification and
were grouped by species or genus. Mean values for occurrence,
excluding plant material, were converted to biomass based on
the site-specific mean weights for collected individual prey
items (small mammals, shrew, crayfish, and frogs) or by comparisons to site-specific individuals (fish scales to fish scales)
when possible.
Mink tissue concentrations
Concentrations of the seventeen 2,3,7,8-substituted PCDD
and PCDF congeners as well as the AhR-active PCB congeners
(PCBs 105, 114, 118, 123, 126, 156, 157, 167, 169, 189, 77,
and 81) were measured in the diet and livers of 48 mink from
within the TR basin. Individual congeners were identified and
quantified by use of U.S. Environmental Protection Agency
(EPA) methods 8290 and 1668 [13,14]. Sample matrix- and
congener-specific limits of quantification were based on threefold the signal to noise ratios and were never greater than 1.0
ng/kg wet weight. When a congener was not detected at the
limits of quantification, a proxy value of half the limit of
quantification was used [12]. Surrogate values of zero or the
full limit of quantification resulted in less than 0.5% variation
in the calculation of concentrations of TEQs in liver tissue or
dietary items. In addition to the individual congeners, the total
concentrations of TEQs were calculated as the sum of the
product of the concentration of each congener and the toxicity
equivalency factor (TEF) assigned by the World Health Organization (WHO) [15] (Eqn. 2):
TEQ ⫽
冘 [(Congener · TEF ) ⫹ · · · ⫹ (Congener · TEF )]
i→n
i
i
n
n
(2)
Environ. Toxicol. Chem. 27, 2008
Multiple-lines-of-evidence assessment for mink
2079
Table 1. Dietary and tissue-based toxicity reference values for mink normalized to the toxic equivalents (TEQs) of 2,3,7,8-tetrachlorodibenzop-dioxina
Diet concentration
(ng TEQ/kg food wet wt)
Housatonic River feeding study
Saginaw River feeding study
Furan feeding study
Dietary doseb
(ng TEQ/kg body wt/d)
Liver
(ng TEQ/kg liver, wet wt)
NOAEC
LOAEC
NOAEL
LOAEL
NOAEC
LOAEC
12
57
26
50
NA
242
1.6
7.2
3.3
6.4
NA
31
50
78
NA
190
NA
NA
a
TEQs calculated based on World Health Organization 2006 toxic equivalency factors [15]. Toxicity reference values are described in terms of
the no-observable-adverse-effect concentration (NOAEC) and the lowest-observable-adverse-effect concentration (LOAEC). The actual concentration at which effects may occur lies between these values. NA ⫽ not available.
b Dietary dose calculated based on 1 kg of body weight and a food consumption rate of 128 g/d [20].
where i → n is the concentration of the individual WHOidentified dioxin-like compound and the associated WHOidentified TEF.
There can be non-AhR-mediated effects of some of these
congeners, but it has been suggested that at the concentrations
and relative proportions of congeners generally observed at
contaminated sites, the critical mechanism (i.e., that which
occurs at the least concentration of the entire mixture) will be
the AhR-mediated effects [16,17]. Thus, to be conservative
and to account for the variations in relative potencies and
occurrences of congeners, the TEQ approach was used to compare concentrations of the PCDDs, PCDFs, and PCBs among
matrices and locations and as well as to TRVs.
Estimates of dietary exposure
Estimates of daily contaminant exposure experienced by
mink were calculated using a modification of the generalized
exposure model presented by the U.S. EPA [18] (Eqn. 3):
[冘 C · (P · NIR)]· AUF
Estimates of potential effects
n
ADDpot ⫽
i⫽1
i
i
crystalball.com). This procedure, which is conceptually simple
but computationally intense, involves repeatedly calculating
the dietary concentration component of Equation 3, each time
using a randomly sampled (with replacement) dietary item
concentration from the data set for each dietary category. Thus,
a distribution composed of more than 100,000 dietary concentrations was created, and the median and upper 95th percentile of the distribution were used to estimate the central
tendency and upper end, respectively, of the daily dose using
Equation 3. Monte Carlo simulations often are conducted in
probabilistic risk assessments, but it is import to emphasize
that the method used here is not a probabilistic approach.
Resampling was performed only on real, measured data to
estimate the daily dose rather than on a continuous distribution
inferred from the measured data or a range of exposure parameters, as performed in a typical probabilistic approach.
Thus, the approach employed here remains free from assumptions about the distribution of the data.
(3)
where ADDpot ⫽ potential average daily dose (ng/kg body
wt/d), NIR⫽ body weight–normalized food ingestion rate (kg/
kg body wt/d), Ci ⫽ concentration of TEQs in dietary item i
(ng/kg), Pi ⫽proportion of diet represented by dietary item i,
and AUF ⫽ area use factor (unitless; foraging area/site area).
To estimate the maximum possible exposure, it was assumed that the mink captured all prey items from within the
TR and adjacent floodplain and, thus, that AUF ⫽ 1. Dietary
exposure was estimated using the site-specific dietary composition according to U.S. EPA guidance, which recommends
describing the variability of an exposure profile using a measure of ‘‘central tendency,’’ the mean or median, and a measure
of the ‘‘high end,’’ represented by estimates expected to fall
between the 90th and 99.9th percentiles of the exposure distribution [19].
The distribution of concentrations of congeners for some
of the dietary items from the TR was right-skewed, violating
the assumption of normality implicit in using the mean to
describe central tendency. In other cases, data appeared to
approximate either a normal or possibly uniform distribution,
but in most cases, the sample size was not large enough to
accurately determine the data distribution. Therefore, to estimate the central tendency and an upper-percentile exposure
concentration without making potentially faulty assumptions
about normality, a Monte Carlo (or resampling) approach was
employed using Crystal Ball威 software (Oracle Crystal Ball
Global Business Unit, Denver, CO, USA; http://www.
The potential for adverse effects was determined by comparing either the concentration of TEQs in liver or diet to their
respective TRV. Hazard quotients (HQs) were calculated based
on concentrations of TEQs in the diet (Eqn. 4) or liver (Eqn.
5):
HQ ⫽
ADDpot (ng TEQ/kg/d)
dietary TRV
(4)
HQ ⫽
ng TEQ/kg
tissue residue TRV
(5)
The TRVs used in the present study were derived from a
rigorous assessment of the available literature regarding the
effects of dioxin-like compounds on mink [17]. All the studies
used in deriving the TRVs were conducted on mink [4,20–
22]. The studies available in the literature were evaluated based
on criteria such as duration, sensitivity, and ecological relevance of the endpoints as well as similarity of the constituents
in the study relative to those comprising the TEQs observed
in mink from the TR. The three studies that were deemed to
be most appropriate were chronic, controlled laboratory exposures in which mink were fed chow or fish from the Housatonic or Saginaw rivers that contained known concentrations
of TEQs. The TRVs derived from these studies ranged from
1.2 ⫻ 101 to 5.7 ⫻ 101 ng TEQ/kg wet weight (diet) and from
5.0 ⫻ 101 to 2.4 ⫻ 102 ng TEQ/kg wet weight (diet) for the
no-observable-adverse-effect level (NOAEL) and lowest-observable-adverse effect level (LOAEL), respectively (Table 1).
Toxicity reference values based on concentrations in the liver
2080
Environ. Toxicol. Chem. 27, 2008
M.J. Zwiernik et al.
Table 2. Measures of habitat suitability, relative abundance, and
population density for the Tittabawassee River (MI, USA; study area)
and upstream tributaries (reference areas)a
Reference areas
Chippewa
River
Habitat suitability
(100 ⫽ optimal)
Relative abundance
(track/200 m/km
shoreline)
Relative abundance
trapping success
(mink/1,000 trap-nights)
Population density
(mink/km shoreline)
a
71
Pine
River
Study area
Tittabawassee
River
67
transformed before conducting statistical analyses. The statistical significance of differences between sampling locations
was evaluated with Tukey’s honestly significant difference test.
When only comparing two groups of data, a Student’s t test
with the Satterthwaite approximation was used to evaluate the
significance of observed differences. Differences were considered to be statistically significant at p ⬍ 0.05.
RESULTS
57
Habitat suitability and relative abundance
0.44
0.76
0.45
4.1
8.5
5.0
0.87
2.1
0.88
Data are presented as the mean values for three sampling years (2003,
2004, and 2005).
were 5.0 ⫻ 101 to 7.8 ⫻ 101 and 1.9 ⫻ 102 ng TEQ/kg wet
weight based on the no-observable-adverse effect concentration (NOAEC) and the lowest-observable-adverse-effect concentration (LOAEC), respectively.
Statistical analyses
All statistical analyses were performed using SAS威 software (Release 9.0; SAS Institute, Inc., Cary, NC, USA). Before
the use of any parametric statistical methods, normality was
evaluated using the Shapiro–Wilks test, whereas Levene’s test
was used to examine the assumption of variance homogeneity.
When data were not normally distributed, data were log-transformed before statistical analysis. Masses of individual organs
were expressed as absolute values in terms of grams and as
relative masses and expressed as a percentage of brain weight.
When evaluating data from three or more locations, analysis
of variance (general linear model procedure) was used to test
for differences between mink morphological endpoints, including absolute and relative organ weights, whole-body
weights, body length, and age, as well as for total TEQ measures related to PCDDs, PCDFs, and PCBs. Relative organ
weight data as well as any other proportional data were arcsine
The habitat suitability index measures indicated that suitable habitat existed to support mink throughout the TR basin.
Habitat occupancy was then confirmed through direct visual
confirmation and by measures of relative abundance that were
not different between study and reference areas. Habitat suitability in the reference areas was slightly greater than that in
the study area, but this resulted largely from a 3.5-km stretch
of the study area containing urban development associated with
the city of Freeland (MI, USA) (Table 2). Mink sign surveys
found no statistical difference ( p ⱕ 0.05) in the percentage of
transects containing mink sign either between years or between
study and reference areas (Table 2). Similarly, no difference
in trapping success was found based on unit effort between
years or between study and reference areas.
Measures of population status and reproductive potential
The age and sex demographics of trapped mink were not
different between study and reference areas (Table 3) or between years and were indicative of a stable, light to moderately
harvested population [23]. Similarly, the reproductive potential
of trapped mink in terms of embryo fertilization and uterine
implantation rates also were not different between reference
and study areas (Table 3). A greater number of male mink
were captured per unit effort. This is consistent with commercial trapping efforts using the same methods in early winter. The increased incidence of captured males versus females
is normal and likely an artifact of the sampling method, because males tend to cover greater areas in search of food and
mates, thereby increasing their potential exposure to points of
capture. The mean age and the juvenile to adult ratio were not
different between sites. Reproductive potential was deemed to
Table 3. Measures of condition of individual adult wild mink trapped in the Tittabawassee River basin (MI, USA)a
Males
Reference (n ⫽ 18) Study (n ⫽ 15)
Body mass (g)
Body length (cm)
Liver mass (g)
Brain wt (g)
Ratio of liver to brain mass
Age (years)
Nutritional statusd
Baculum length (mm)e
Placental scarsf
Ovary weight (g)
Uterus weight (g)
Sex ratio (M:F)
910
58.8
50.6
8.29
6.14
1.8
2.9
42.6
⫾ 33
⫾ 0.61
⫾ 2.3
⫾ 0.14
⫾ 0.31
⫾ 0.8
⫾ 0.08
⫾ 0.86
NA
NA
NA
2.25:1
⫾ 56
⫾ 1.1
⫾ 4.7
⫾ 0.20c
⫾ 0.53c
⫾ 1.1
⫾ 0.10
⫾ 0.77
NA
NA
NA
(reference area)
962
58.7
55.7
8.05
7.21
2.2
3.0
42.7
Females
pb
Reference (n ⫽ 8)
0.414
0.922
0.348
0.329
0.090
0.734
0.380
0.952
NA
NA
NA
⫾ 26
⫾ 1.0
⫾ 1.1
⫾ 0.15
⫾ 0.22
⫾ 0.7
⫾ 0.25
NA
3.9 ⫾ 0.83
0.248 ⫾ 0.04
0.111 ⫾ 0.03
481
47.9
27.2
6.34
4.3
2.3
2.3
Study (n ⫽ 7)
⫾ 9.9
⫾ 0.39
⫾ 1.7
⫾ 0.22
⫾ 0.35
⫾ 0.9
⫾ 0.18
NA
4.3 ⫾ 1.1
0.325 ⫾ 0.05
0.152 ⫾ 0.04
2.41:1 (study area)
517
50.8
31.6
6.31
5.06
2.7
2.7
pb
0.235
0.023
0.069
0.924
0.103
0.327
0.160
NA
0.775
0.008
0.076
All measures are presented as the mean ⫾ standard error. Numbers in parentheses (n) for each location are the sample size. NA ⫽ not applicable.
p values from Student’s t test with Scatterthwaite approximation to account for variance heterogeneity.
c n ⫽ 14 (brain from one mink not suitable for analysis).
d Average nutritional status value scale: 1 ⫽ poor; 4 ⫽ very good.
e n ⫽ 13 (baculum from two mink lost during pelting process).
f Average number of scars per female mink.
a
b
Environ. Toxicol. Chem. 27, 2008
Multiple-lines-of-evidence assessment for mink
Table 4. Dietary dioxin, furan, and polychlorinated biphenyl (PCB)
concentrations in terms of 2,3,7,8-tetrachlorodibenzo- p -dioxin
equivalents (TEQs) for mink from the reference and study areas in
the Tittabawassee River (MI, USA)a
Diet concn. (ng TEQ/kg food wet wt)b
Total TEQs
⌺Doxin TEQs
⌺Furan TEQs
⌺Nn-ortho PCB TEQs
⌺Mno-ortho PCB TEQs
Reference area
Study area
95th
Median percentile
95th
Median percentile
0.679
0.35
0.120
0.200
0.02
0.879
0.519
0.181
0.203
0.0163
30.8
4.44
21.7
2.47
0.233
56.8
5.63
44.5
18.0
1.72
a
TEQs calculated based on World Health Organization 2006 toxic
equivalency factors [15].
b Median and 95th percentile dietary concentrations were estimated
by repeatedly resampling the measured concentrations in each dietary category and weighting in proportion to the dietary composition
(Eqn. 3) to estimate the distribution of potential dietary concentrations.
be normal, because no statistical difference was found in the
number of fetus implant points between sites (placental scarring is indicative of the most recent reproductive cycle).
PCDDs, PCDFs, PCBs, and TEQs in diet
Based on contents of the mink GI tracts, the diet of mink
trapped from the TR basin (converted from frequency of occurrence to kg wet wt) consisted primarily of fish (52%), followed by muskrats (19%), vegetation (9%), small mammals
(8%), crayfish (8%), and amphibians (4%). Small mammals
were further subdivided into shrew and nonshrew species. The
subdivision was a response to the great disparity in measured
contaminant body burdens of the two mammal categories and
uncertainty associated with the site-specific dietary composition. Two approaches were used to conservatively estimate the
proportions of shrews in the diets of resident mink. The first
approach was based on previous mink dietary studies for the
region, and the second was to assume that mink would con-
2081
sume shrews in proportion to their availability. For the two
regional mink dietary composition studies, shrews were identified in 0 of 1,028 digestive tracts [24] and in 5 of 201 digestive tracts [25], resulting in a wet-weight percentage contribution by the mammalian portion of the diet of 0% and less
than 3%, respectively. In a 43-year study of small-mammal
population dynamics in Ontario (Canada), the shrew relative
abundance was estimated to be 6% of the total diet [26]. The
relative abundance study did not include mammals larger than
a vole, so the percentage likely was an overestimate. When
converted to wet weight based on prey mean mass and percentage consumption, these studies resulted in shrew mammalian contribution of 0, 3, and 6%, respectively; we selected
6% as a conservative estimate of shrews consumed as part of
the mammalian dietary component. The estimated median and
95th percentile concentrations of TEQs in the diet of mink in
the study area were approximately 45- and 65-fold greater,
respectively, than those of the reference area (Table 4). The
TEQ concentrations for dietary items ranged from 1.4 ⫻ 10⫺1
to 2.7 ⫻ 103 and from 1.4 ⫻ 10⫺1 to 1.4 ⫻ 101 ng TEQ/kg
wet weight in the study and reference areas, respectively (Table
5). The estimated median ADDpot based on the site-specific
prey item analysis and dietary composition, calculated using
the food consumption rate (0.177 kg/d) and body weight (0.8
kg) recommended for wild mink in the Great Lakes region,
was approximately 45-fold greater in the study area (7 ng TEQ/
kg/d) than in the reference area (0.15 ng TEQ/kg/d) [27].
PCDDs, PCDFs, PCBs, and TEQs in liver
Polychlorinated dibenzo-p-dioxins, PCDFs, and PCBs were
detected in livers of all mink collected from the river basin
regardless of location (Table 6). Concentrations of PCDDs,
PCDFs, and total TEQs were significantly greater in the livers
of mink from the study area than those from the upstream
reference areas ( p ⬍ 0.001). Most of the difference resulted
from 2,3,4,7,8-pentchlorodibenzofuran, which accounted for
56% of the total TEQs and occurred at concentrations approximately 86-fold greater in livers of mink from the study
area. Polychlorinated dibenzo-p-dioxins, PCDFs, non-ortho
Table 5. Range of dietary item concentrations (ng 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalents [TEQs]/kg wet wt) for dioxins and furans used
in mink exposure estimatesa
Reference area
⌺Dioxin and furan TEQs
Forage fish
Muskrat
Terrestrial vegetation
Crayfish
Shrew
Other small mammal
Amphibian
⌺Polychlorinated biphenyl TEQs
Forage fish
Muskrat
Terrestrial vegetation
Crayfish
Shrew
Other small mammal
Amphibian
a
Study area
n
Min.
Max.
n
2
9
9
5
14
41
29
0.431
0.0881
0.194
0.140
0.205
0.180
0.166
0.436
0.206
1.42
1.38
14.2
6.36
2.13
7
13
25
15
35
128
106
2
2
0
0
0
1
0
0.311
0.0677
0.375
0.0798
0.118
0.118
7
8
62
3
2
5
4
Min.
5.11
1.97
0.143
3.15
6.92
1.15
0.649
2.10
0.119
0.00693
0.706
0.0654
0.290
0.494
Max.
73.5
11.4
2.62
75.4
2738
242
348
37.4
3.54
0.00803
0.951
5.71
1.59
0.894
TEQs calculated based on World Health Organization 2006 toxic equivalency factors [15]. Samples were collected from one reference and four
study biological sampling areas in the Tittabawassee River basin (MI, USA).
2082
Environ. Toxicol. Chem. 27, 2008
M.J. Zwiernik et al.
Table 6. Concentrations of dioxins, furans, and polychlorinated
biphenyls (PCBs) in terms of 2,3,7,8 tetrachlorodibenzo-p-dioxin
equivalents (TEQs) in mink liver collected from the reference and
study areas in the Tittabawassee River (MI, USA)a
Reference area
Study area
Class
n Concentration
n Concentration
⌺Dioxin
⌺Furan
⌺Non-ortho PCBs
⌺Mono-ortho
PCBs
Total TEQs
26
26
25
4.2 ⫾ 0.49
5.6 ⫾ 1.1
8.4 ⫾ 1.6
22
22
22
21 ⫾ 1.9
290 ⫾ 71
80 ⫾ 21
⬍0.0001
0.0007
0.003
25
25
2.0 ⫾ 0.36
20 ⫾ 3.2
22
22
5.8 ⫾ 1.1
400 ⫾ 74
0.003
⬍0.0001
pb
a
TEQs calculated based on World Health Organization 2006 toxic
equivalency factors [15]. Liver concentration data are presented as
the mean and standard error. Concentration data are given as ng
TEQ/kg wet weight. Data represent the results from both male and
female mink.
b p values from Student’s t tests with a Scatterthwaite approximation
to account for potential variance heterogeneity.
PCBs, and mono-ortho PCBs contributed an average of 9, 63,
27, and 0%, respectively, of the total concentrations of TEQs
in the livers of mink from the study area, whereas the relative
contributions were 37, 22, 37, and 3%, respectively, for mink
from the reference area.
Individual condition
Overall, all mink appeared to be in good condition (n ⫽
48) (Table 3). Mink from the reference area (n ⫽ 26) were
described as having adequate body condition, whereas mink
from the study area (n ⫽ 22) had a slightly greater nutritional
status based on pathologist interpretation of adipose tissue as
a percentage of body mass, overall muscle tone, skin and coat
condition, and overall tissue color and condition. This difference in body condition was noted especially for study area
males (n ⫽ 15), which were reported as extremely large and
muscular compared to reference animals. Although qualitatively noted, differences in quantified measures of adult condition were not statistically significant when study and reference areas were compared. No significant differences between sites were found for any of the following parameters:
Absolute masses of whole bodies, brains, and livers, and relative masses of the three organs normalized to brain weight
and expressed as percentages (Table 3). Slight but statistically
significant increases, however, in body length without tails and
uterine weight in females from the study area were found
compared to the reference area mink ( p ⬍0.05) (Table 3). No
significant differences were observed for body length in male
mink, nor was a statistically significant difference found in
baculum length between the study and reference areas. Additionally, gross and histological examination of organs found
no abnormalities in liver, a tissue that is known to be sensitive
to exposure to AhR-active compounds. Remarkable kidney
tissue was noted both grossly and histologically for number
of mink from both the reference area (2/26) and the study area
(5/22) and was described as minimal to moderate multifocal
interstitial infiltration of lymphocytes and plasma cells, which
is suggestive of past bacterial infection. An additional abnormality that was observed histologically in 1 of the 26 mink
collected from the reference area (RA-020) was mild squamous
epithelial proliferation of the right and left maxillae. This condition was not identified in any of the 22 study area mink that
had liver tissue TEQ concentrations that were between 2- to
Fig. 2. Mink liver and dietary exposure hazard quotients for the study
area on the Tittabawassee River (MI, USA) based on the sum of
dioxins, furans, and polychlorinated biphenyls in terms of 2,3,7,8tetrachlorodibenzo-p-dioxin equivalents. LOAEC ⫽ lowest-observable-adverse-effect concentration; NOAEC ⫽ no-observable-adverseeffect concentration.
35-fold greater. No other changes of diagnostic significance
were found in any of the other mink. Neither the kidney nor
the jaw abnormalities were determined to be contaminant related.
Hazard estimates
Potential for adverse effects was evaluated by use of a HQ
approach. Hazard quotients were developed based on the concentrations of PCDDs/PCDFs expressed as concentrations of
TEQs in the diet (ADDpot) or in livers of mink trapped from
the TR. The exposure estimates were divided by TRVs, as
either NOAELs or LOAELs, and expressed as HQs. The TRVs
were derived from the results of controlled laboratory studies
in which mink were fed individual congeners or mixtures of
AhR congeners. The 5th percentile, median, and 95th percentile were used to represent a range of concentrations of TEQs
in the liver and diet expressed as the ADDpot. Hazard quotients
for the reference area were all less than 1.0 for both dietary
exposure and liver residues. The range of dietary exposure–
based HQs in the study area was from less than 1.0 for the
5th percentile (LOAEL) to 1.7 for the 95th percentile
(NOAEL). The median study area dietary exposure HQs were
less than 1.0 when calculated for both the NOAEL and
LOAEL. Liver residue–based HQs in the study area ranged
from less than 1.0 for the 5th percentile (LOAEC) to 8.6 for
the 95th percentile (NOAEC). The liver residue–based HQs
were 1.2 and 3.0 when the median concentrations were compared to the LOAEC and NOAEC, respectively (Fig. 2).
DISCUSSION
Multiple lines of evidence
A multiple-lines-of-evidence approach for ecological risk
assessment reduces uncertainty and provides the most accurate
evaluation of potential hazard [28,29]. This approach can include various types of data (chemistry, bioassays, laboratory,
and field measures) that are integrated into a conclusion that
best describes the entire data set. The desired result is more
Multiple-lines-of-evidence assessment for mink
informed decisions. Streamlined assessments for mink often
are inferred from limited data for a single line of evidence.
For example, concentrations of contaminants in sediments and/
or fish may be compared to literature-derived toxicity values
to determine the potential for effects. This relatively simplistic
approach requires several assumptions with considerable uncertainty. Generally, the results of such an exercise are useful
for screening out the possibility of risk but likely will not
provide sufficient information or certainty for risk management
decisions [30]. Site-specific empirical information can more
directly measure cause–effect relationships; however, uncertainties also are associated with these data sets. For instance,
in the present study, estimates for the relative abundance of
mink may be inferred from two apparently unrelated lines of
evidence (scat and track surveys, as compared to trapping
success per unit effort). At first glance, the similarity between
the two measures in terms of relative numbers in reference
versus study areas would add strength to the conclusion that
mink abundance does not appear to be different between the
reference and study areas. In fact, both measures, despite their
appearance of independence, likely underestimated mink abundance in the study area. This is because the water level in the
river fluctuates by 10 to 75 cm between approximately 8 AM
and approximately 5 PM as the result of hydroelectric generation, and this fluctuation likely eliminated near-shore tracks
and scat. Similarly, there may be a bias in trapping success
per unit effort, because this same daily fluctuation in water
level made it difficult to maintain trap effectiveness over the
duration of the set. The hydroelectric dams had little or no
influence on water levels in the reference areas. Thus, the
multiple-lines-of-evidence approach can be a useful tool, but
spatial, temporal, and analytical uncertainties still need to be
addressed.
In the present study, extensive site-specific data were used
in a multiple-lines-of-evidence approach to evaluate the potential for effects on wild mink chronically exposed to PCDFs,
PCDDs, and PCBs in the TR basin. The approach considered
the strengths and weaknesses of various measurement endpoints as well as the nature of the uncertainty associated with
each to evaluate the overall potential for effects. Site-specific
lines of evidence included dietary and tissue-based exposure
assessments as well as assessment of the current status of the
resident mink on both an individual and a population level.
Direct measures of conditions of individual mink
The primary contaminants of concern in the TR—PCDDs,
PCDFs, and some PCB congeners—can act through a common
mechanism of action, mediated by the AhR. Effects mediated
through this mechanism include chronic, sublethal effects on
mink [16,17], such as body condition, nutritional status, placental scaring, age, body mass, body length, liver mass, brain
mass, ratio of liver to brain mass, and histological examinations of select tissues, including kidney, liver, lymph, brain,
and jaw. In addition, some PCDDs, PCDFs, and PCB congeners have been found to affect the immune system of mammals, which could result in greater susceptibility to disease
[31]. Abnormal values for these parameters can indicate poor
condition, reduced growth, reduced survival of adults and kits,
and reduced fertility and fecundity. All these parameters are
ecologically relevant endpoints that could translate into reduced fitness and affect the robustness or sustainability of the
population, which is the assessment endpoint of interest [2,4–
6]. The fact that no statistically significant or toxicologically
Environ. Toxicol. Chem. 27, 2008
2083
adverse difference was found in any of the parameters measured between the study and reference areas, and that these
parameters were similar to those otherwise reported for wild
mink [18,32], indicates that the threshold for effects in individuals had not been reached.
Baculum length is not affected by AhR-mediated mechanisms, but it has been reported to potentially be affected by
exposure to some types of organochlorines [33]. Thus, even
if differences in exposure to other classes of compounds existed between the study and reference areas, it was not reflected
in differences of baculum length.
Absolute and relative organ masses
One effect of TCDD and structurally similar congeners of
PCDDs, PCDFs, and PCBs observed in mink is enlargement
of the liver, which has been suggested to result from adaptive
hypertrophy associated with lipidosis, and fatty infiltration [34]
and induction of mixed-function monooxygenase enzymes,
such as cytochrome P4501A [35]. No differences were found
in liver mass, either absolute or relative (to percentage of brain
mass), between mink from the study and reference areas observed in the present study. This finding was contrary to those
of previously conducted laboratory studies, which identified
changes in absolute and relative liver mass for mink exposed
to TEQ-based doses less than or equal those of the TR mink
[34]. In one study of particular relevance, mink were fed fish
collected just downstream of the TR study area in Saginaw
Bay (Lake Huron, MI, USA). As would be expected, the physical connection of these water bodies resulted in mink exposure
(fish) of very similar contaminant makeup. The notable exception was a shift in relative TEQ contribution from PCDFs
in TR fish and dietary items to PCBs further downstream in
Saginaw River and Saginaw Bay. The most dramatic shift in
the percentage contribution to TEQs were for 2,3,7,8-tetrachlorodibenzofuran (TCDF), which declined from 11 to 2.5%
of the total TEQs, and for PCB 126, which increased from 1
to 9%. For this Saginaw River–based laboratory study, the
relative masses of liver and spleen, expressed as a percentage
of brain mass, were greater in exposed than in unexposed mink.
Increases in both the absolute and relative liver mass were
observed in male mink fed a diet composed of similar Saginaw
Bay fish for 18 months [36]. The absence of these responses
in the TR mink, despite calculated TEQ exposure concentrations of equal or greater value, suggests that the different patterns of PCDDs/PCDFs and PCB congeners appear to result
in different potencies based on dose response. Thus, even for
these closely related exposures, the WHO TEF system does
not appear to accurately predict effects.
Body mass, length, and age
Exposure to congeners that are structurally similar to TCDD
has been reported to cause a dose-dependent decrease in body
mass [22,31]. Concentrations of TEQs in the liver of mink
from the TR exceed the concentration that has been associated
with weight loss (wasting syndrome) [34]. Because the mean
body mass of mink in the study area actually was greater than
that in the reference area, we found no indication of wasting
syndrome in the wild populations of mink on the TR. Again,
this suggests that the difference in congener pattern making
up the total concentration of TEQs does not cause the same
magnitude of response as that predicted from studies with
mixtures comprised of different PCDDs/PCDFs and PCB congeners.
2084
Environ. Toxicol. Chem. 27, 2008
M.J. Zwiernik et al.
Histology
Based on results reported in the literature [7,34,37], the
concentrations of TEQs observed in the livers and diet of mink
from the TR should have resulted in dose-dependent histological changes in liver and jaw tissues, but no such TEQ-related
pathologies were observed. Adult female mink fed diets that
included Saginaw Bay carp had enlarged and diffusely yellow
livers [34]. Histologically, the livers of mink fed Saginaw Bay
carp expressed various degrees of congestion, hepatocellular
fatty changes, and scattered aggregates of lymphocytes [34].
Other studies have reported gross and cellular hepatic changes,
including hepatic lipidosis, in mink exposed to PCBs [38–43].
None of these previously identified contaminant-related tissue
variations were seen in adult TR basin mink.
Abnormal kidney tissue was noted in 2 of 26 reference
mink and in 5 of 22 study area TR basin mink. The tissue
change was described as commonly occurring, minimal to
moderate multifocal interstitial infiltration of lymphocytes and
plasma cells, which is suggestive of previous bacterial infection. This abnormality is relatively common in both wild and
ranch mink, and based on histological examination and dose–
response considerations, it did not appear to be contaminant
related.
An even more sensitive histological endpoint related to
AhR-active compounds is proliferation of mandibular and
maxillary squamous epithelial cells. This histological response
can be identified before other types of responses that may affect
survival and, thus, can be thought of as a biomarker or indicator of exposure. Mink exposed to PCB 126 and TCDD at
dietary concentrations of 2.4 ⫻ 104 and 2.4 ⫻ 103 ng/g feed
wet weight, respectively, developed proliferation of mandibular and maxillary squamous epithelia [44–46]. These results
were confirmed by another feeding study in which mink were
fed diets containing PCB 126 [7], in which concentrations as
low as 2.4 ⫻ 102 ng/kg feed wet weight (equivalent to 2.4 ⫻
101 ng TEQ/kg feed wet wt), caused mandibular and maxillary
squamous epithelia that was clinically detectable by loose and
missing teeth [7].
During laboratory exposure to PCB 126, the incidence of
the lesion at 24 ng TEQ/kg feed was greater than 50%. Only
one individual of the 48 mink from the TR watershed that were
examined exhibited the lesion, however, and this individual
was from the upstream reference area. Mink trapped in the
study area were exposed to approximately 31 ng TEQ/kg wet
weight in their median diet, which is similar to the concentration of PCB 126 that causes the lesion. Because no lesions
were observed in mink from the study area, the mixture of
AhR-active congeners comprising the TEQs appears to be less
potent than would be predicted based on the TEQs. This is
consistent with the fact TCDF, which composes approximately
31% of the total predicted TEQs in the diet of mink inhabiting
the study area, did not, under laboratory conditions, cause
proliferation of mandibular and maxillary squamous epithelia
at TEQ concentrations 100-fold less than the concentration of
PCB 126. Furthermore, proliferation of squamous epithelia in
the mandible and maxilla was detected in wild mink collected
from the Kalamazoo River (MI, USA) Superfund site, where
PCBs account for most of the TEQs [7]. An estimate of total
PCBs consumed by mink inhabiting the Kalamazoo River
ranged from 8.0 ⫻ 10⫺2 to 0.41 mg/kg body mass/d, corresponding to TEQ concentrations of 0.7 to 2.0 ng/kg body
weight/d, respectively [12].
Fig. 3. Polychlorinated dibenzofurans and polychlorinated dibenzop-dioxins congener distributions comprising mammalian 2,3,7,8-tetrachlorodibenzo-p-dioxin equivalents (TEQs) in mink liver and sitespecific diet. BAF ⫽ bioaccumulation factor; TR ⫽ Tittabawasee
River; WHO ⫽ World Health Organization. ⫽ other congeners; m
⫽ 1,2,3,4,7,8-hexachlorodibenzofuran; □ ⫽ 2,3,4,7,8-pentachlorodibenzofuran; u ⫽ 2,3,7,8-tetrachlorodibenzofuran; 䡵 ⫽ 1,2,3,7,8pentachlorodibenzodioxin; v ⫽ 2,3,7,8-tetrachlorodibenzo-p-dioxin.
The jaw lesion results provide valuable insight regarding
both the toxic potency of the TR exposure mixture and the
conditions of individual resident mink. Jaw lesions are a very
sensitive endpoint, and their occurrence at the histological level of identification does not adversely effect survival. Thus,
the absence of this diagnostic jaw lesion provides additional
confidence in the lines of evidence associated with direct measures for the condition of individual mink. The absence of jaw
lesions and of any responses directly measured and associated
with the condition of individual mink suggests that the fitness
of mink is not being adversely affected. Additionally, the absence of a response for any of these direct measures of conditions of individual mink, including jaw lesions, at TEQ concentrations in excess of those at which a response would be
predicted based on other AhR-active mixtures suggests that
the mixture to which TR mink are exposed (primarily PCDFs)
appears to be less potent than predicted. Thus, for mink from
the TR, even with the more detailed site-specific information
that is reported here, use of the TEQ approach with comparison
to available TRVs overestimates the potential for adverse effects.
Dietary exposure
The dietary items consumed by mink on the TR were similar to those reported elsewhere for Michigan rivers [12,25,47].
Fish accounted for the greatest proportion of the site-specific
diet. The proportion of fish in the TR mink diet was slightly
greater than those reported by Millsap et al. [12] and Sealander
[25] but slightly less than that reported by Alexander [47].
The estimated TR dietary composition is supported by the fact
that the congener pattern in the site-specific diet matches what
can be calculated for dietary exposure based on the residues
measured in resident mink livers and published bioaccumulation factors (Fig. 3).
Multiple-lines-of-evidence assessment for mink
Exposure hazard assessment
Hazard quotients based on concentrations of TEQs in both
the diet and liver of resident TR basin mink were all between
0.0 and 0.5 in the reference. In the study area, hazard quotients
were all less than 10, ranging from 9.5 ⫻ 10⫺1 to 1.7 for dietary
exposure and from 3.0 to 8.7 for the tissue-based assessment.
The large range of HQs is an artifact of the attempt to capture
the uncertainties associated with both the exposure concentrations and the toxic potency of the contaminant mixture. The
greatest HQs resulted from a comparison of the 95th percentile
of the exposure estimate to a no-observed-adverse-effect TRV.
The HQs in the lower range were based on the 5th percentile
of dietary exposure estimates as compared to LOAELs.
Direct measures of population status
Relative abundance based on surveys and trapping success
was not different between exposed and unexposed populations
and correlated well with habitat suitability. The predicted population density for the study area was approximately 1.7 mink/
stream km, or three to six mink/breeding territory based on a
defended male territory of 2 to 4.5 river km [10]. Great uncertainty is involved in population estimates, but it appears
that the TR basin downstream of Midland exceeds the reported
carrying capacity of 0.6 mink/km based on a similar Michigan
stream with excellent habitat and sufficient food [18,48]. This
conclusion is supported by population demographics, which
are consistent with a stable, lightly harvested mink population
[23] and are inconsistent with a mink population experiencing
adverse effects from contaminants [49,50]. In all, each of these
data sets indicates no population-level effects of the PCDDs/
PCDFs.
Hazard to TR mink
Based on multiple lines of evidence, we find little present
potential for adverse effects to the resident mink of the TR
basin. Direct measures of effects on individuals and the population all suggest that resident mink are in good condition
and that the population is robust and sustainable. This was
true for all the measurement endpoints, including those that
have been shown under laboratory conditions to be affected
by exposure to various AhR agonists, such as PCDDs/PCDFs.
No biologically significant effects were observed for any of
the endpoints measured in the present study between reference
and study populations, and all parameters measured were within the range expected for a typical wild mink. This included
measurements made at the level of the individual, including
body mass, relative tissue masses, baculum length, gross pathology, liver and jaw histology, and fecundity, as measured
by placental scarring, as well as estimates of the population
density of mink inhabiting the study and reference areas.
Indirect measures for the condition of individual mink,
based on exposure calculations when compared to TRVs based
on non-site-specific contaminant mixtures, do suggest the possibility of, albeit small, adverse effects; however, none of the
responses for which the TRVs were established was seen
among the sampled mink. This included jaw lesions that should
occur at TEQ concentrations 10- to 100-fold less than measurement endpoints associated with possible adverse effects
on individual survival and population status. The absence of
a response for any of these direct measures for the condition
of individual mink, including jaw lesions, at TEQ concentrations in excess of those at which a response would be predicted
Environ. Toxicol. Chem. 27, 2008
2085
based on other AhR-active mixtures suggests that the mixture
to which TR mink are exposed (primarily PCDFs) appears to
be less potent than predicted based on TEFs. The exact reason
for this is unknown; however, a recent laboratory feeding study
suggests that TCDF, which is a major source of TR TEQs, is
less potent than predicted and is metabolized more rapidly
than other AhR-active compounds, such as TCDD and PCB
126 [51]. Thus, given this scenario, there appears to be significant uncertainty associated with the applicability of the
presently available TRVs, resulting in a likely overestimate of
hazard for TR mink. Even when HQs were calculated using
this likely overestimate of potency and assumptions leading
to maximal estimates of exposure (95th percentile), all the
resulting HQs were less than 2.0 and 9.0 based on estimates
of exposure in the diet or concentrations of TEQ measured in
livers, respectively. Hazard quotients calculated using less conservative estimates of exposure (e.g., median) but conservative
TRVs (based on the LOAEC or NOAEC) were all less than
1.0 when based on predicted dietary exposure and ranged between 1.2 and 3.0 when based on measured concentrations of
TEQs in liver. Taken together, predicted and directly measured
indicators of fitness of individual mink or robustness of their
population do not suggest that significant potential exists for
adverse effects because of exposure to PCDDs and PDCFs in
the environment along the TR. The population is robust and
either at or near carrying capacity.
Acknowledgement—The authors thank all staff and students of the
Michigan State University–Aquatic Toxicology Laboratory and field
crew—namely, M. ‘‘Never’’ Fales, D. Tazelaar, T. Fredricks, R. Seston, S. Coefield, M. Nadeau, S. Plautz, M. Barker, J. van Dam, L.
Williams, P. Bradley, M. Kramer, N. Ikeda, and M. Shotwell of ENTRIX. The authors also thank J. Dastyck and S. Kahl of the U.S. Fish
and Wildlife Service–Shiawassee National Wildlife Refuge and T.
Lennon and the Chippewa Nature Center for their expertise and collaboration. The authors recognize the Saginaw County Parks and Recreation Commission for access to Imerman Park and the Tittabawassee
Township Park Rangers for access to Tittabawassee Township Park
and Freeland Festival Park. Most importantly, the authors acknowledge the more than 50 cooperating landowners throughout the study
area who granted us access to their property, making our research
possible. Funding was provided through an unrestricted grant from
The Dow Chemical Company (Midland, MI, USA) to J.P. Giesy and
M.J. Zwiernik of Michigan State University.
REFERENCES
1. Hilscherova K, Kannan K, Nakata H, Yamashita N, Bradley P,
McCabe J, Taylor AB, Giesy JP. 2003. Polychlorinated dibenzop-dioxin and dibenzofuran concentration profiles in sediments and
flood-plain soils of the Tittabawassee River, Michigan. Environ
Sci Technol 37:468–474.
2. Basu N, Scheuhammer AM, Bursian SJ, Elliott J, Rouvinen-Watt
K, Chan HM. 2007. Mink as a sentinel species in environmental
health. Environ Res 103:130–144.
3. U.S. Environmental Protection Agency. 1993. Interim report on
data and methods for assessment of 2,3,7,8-tetrachlorodibenzop-dioxin risks to aquatic life and associated wildlife. EPA/600/
R-93/055. Office of Research and Development, Washington, DC.
4. Brunstrom B, Lund B-O, Bergman A, Asplund L, Athanassiadis
I, Athanasiadou M, Jensen S, Orberg J. 2001. Reproductive toxicity in mink (Mustela vison) chronically exposed to environmentally relevant polychlorinated biphenyl concentrations. Environ Toxicol Chem 20:2318–2327.
5. Bursian SJ, Beckett KJ, Yamini B, Martin PA, Kannan K, Shields
KL, Mohr FC. 2006. Assessment of effects in mink caused by
consumption of carp collected from the Saginaw River, Michigan,
USA. Arch Environ Contam Toxicol 50:614–623.
6. Leonards PEG, De Vries TH, Minnaard W, Stuijfzand S, De Voogt
P, Cofino WP, van Straalen NM, van Hattum B. 1995. Assessment
of experimental data on PCB-induced reproduction inhibition in
2086
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
Environ. Toxicol. Chem. 27, 2008
mink, based on an isomer- and congener-specific approach using
2,3,7,8-tetrachlorodibenzo-p-dioxin toxic equivalency. Environ
Toxicol Chem 14:639–652.
Beckett KJ, Millsap SD, Blankenship AL, Zwiernik MJ, Giesy
JP, Bursian SJ. 2005. Squamous epithelial lesion of the mandibles
and maxillae of wild mink (Mustela vison) naturally exposed to
polychlorinated biphenyls. Environ Toxicol Chem 24:674–677.
Allen AW. 1986. Habitat suitability index models: Mink, revised.
FWS/OBS-82/10.61. Revised report. U.S. Fish and Wildlife Service, Washington, DC.
Bonesi L, Macdonald DW. 2004. Evaluation of sign surveys as
a way to estimate the relative abundance of American mink (Mustela vison). J Zool 262:65–72.
Gerell R. 1970. Home ranges and movements of the mink Mustela
vison Schreber in southern Sweden. Oikos 21:160–173.
Matson GM. 1981. Workbook for Cementum Analysis. Milltown,
MT, USA.
Millsap SD, Blankenship AL, Bradley PW, Jones PD, Neigh A,
Park C, Strause KD, Zwiernik MJ, Giesy JP. 2004. Comparison
of risk assessment methodologies for exposure of mink to PCBs
on the Kalamazoo River, Michigan. Environ Sci Technol 38:
6451–6459.
U.S. Environmental Protection Agency. Method 1668: Revision
A: Chlorinated biphenyl congeners in water, soil, sediment, and
tissue by HRGC/HRMS. 1999. EPA 821/B-94/005.
U.S. Environmental Protection Agency. 1998. Polychlorinated dibenzodioxins (PCDDs) and polychlorinated dibenzofurans
(PCDFs) by high-resolution gas chromatography/high-resolution
mass spectrometry (HRGC/HRMS), Revision 1. Method 8290A.
SW-846. Washington, DC.
Van den Berg M, Birnbaum LS, Denison M, De Vito M, Farland
W, Feeley M, Fiedler H, Hakansson H, Hanberg A, Haws L, Rose
M, Safe S, Schrenk D, Tohyama C, Tritscher A, Tuomisto J,
Tysklind M, Walker N, Peterson RE. 2006. The 2005 World
Health Organization reevaluation of human and mammalian toxic
equivalency factors for dioxins and dioxin-like compounds. Toxicol Sci 93:223–241.
Kannan K, Blankenship AL, Jones PD, Giesy JP. 2000. Toxicity
reference values for the toxic effects of polychlorinated biphenyls
to aquatic mammals. Human and Ecological Risk Assessment 6:
181–201.
Blankenship AL, Kay DP, Zwiernik MJ, Holem RR, Newsted JL,
Hecker M, Giesy JP. 2008. Toxicity reference values for mink
exposed to 2,3,7,8-tetrachlodibenzo-p-dioxin (TCDD) equivalents (TEQs). Ecotoxicol Environ Saf 69:325–349.
U.S. Environmental Protection Agency. 1993. Wildlife exposure factors handbook, Vol I and II. EPA/600/R-93/187b. Washington, DC.
U.S. Environmental Protection Agency. 1998. Guidelines for ecological risk assessment final. EPA/630/R-95/002F. Washington, DC.
Bursian SJ, Sharma C, Auman HJ, Yamini B, Mitchell RR, Orazio
C, Moore D, Sivirski S, Tillitt DE. 2006. Dietary exposure of
mink (Mustela vison) to fish from the Housatonic River, Berkshire
County, Massachusetts, USA: Effects on reproduction and kit
growth and survival. Environ Toxicol Chem 25:1533–1540.
Halbrook RS, Aulerich RJ, Bursian SJ, Lewis L. 1999. Ecological
risk assessment in a large river-reservoir: 8. Experimental study
of the effects of polychlorinated biphenyls on reproductive success in mink. Environ Toxicol Chem 18:649–654.
Heaton SN, Bursian SJ, Giesy JP, Tillitt DE, Render JA, Jones
PD, Verbrugge DA, Kubiak TJ, Aulerich RJ. 1995. Dietary exposure of mink to carp from Saginaw Bay, Michigan. 1. Effects
on reproduction and survival, and the potential risks to wild mink
populations. Arch Environ Contam Toxicol 28:334–343.
Whitman JS. 2003. Age structure differences in American mink,
Mustela vison, populations under varying harvest regimes. Can
Field Nat 117:35–38.
Korschgen LJ. 1958. December food habits of mink in Missouri.
J Mammal 39:521–527.
Sealander JA. 1943. Winter food habits of mink in southern Michigan. J Wildl Manag 7:411–417.
Fryxell JM, Falls JB, Falls EA, Brooks RJ. 1998. Long-term
dynamics of small-mammal populations in Ontario. Ecology 79:
213–225.
U.S. Environmental Protection Agency. 1995. Great lakes water
quality initiative criteria documents for the protection of wildlife:
DDT, mercury, 2,3,7,8-TCDD, PCBs. EPA-820-B-95-0083.
Washington, DC.
M.J. Zwiernik et al.
28. Fairbrother A. 2003. Lines of evidence in wildlife risk assessments. Human and Ecological Risk Assessment 9:1475–1491.
29. Menzie C, Henning MH, Cura J, Finkelstein K, Gentile J, Maughan J, Mitchell D, Petron S, Potocki B, Svirsky S, Tyler P. 1996.
Special report of the Massachusetts weight-of-evidence workgroup: A weight-of-evidence approach for evaluating ecological
risks. Human and Ecological Risk Assessment 2:277–304.
30. Hull RN, Swanson S. 2006. Sequential analysis of lines of evidence—An advanced weight-of-evidence approach for ecological
risk assessment. Integr Environ Assess Manag 2:302–311.
31. U.S. Environmental Protection Agency. 2000. Exposure and human health reassessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD) and related compounds. EPA/600/P-00/001 Ab-Ae.
Washington, DC.
32. Aulerich RJ, Powell DC, Bursian SJ. 1999. Handbook of Biological Data for Mink. Department of Animal Science, Michigan
State University, East Lansing, MI, USA.
33. Harding LE, Harris ML, Stephen CR, Elliott JE. 1999. Reproductive
and morphological condition of wild mink (Mustela vison) and river
otters (Lutra canadensis) in relation to chlorinated hydrocarbon contamination. Environ Health Perspect 107:141–147.
34. Heaton SN, Bursian SJ, Giesy JP, Tillitt DE, Render JA, Jones
PD, Verbrugge DA, Kubiak TJ, Aulerich RJ. 1995. Dietary exposure of mink to carp from Saginaw Bay, Michigan: 2. Hematology and liver pathology. Arch Environ Contam Toxicol 29:
411–417.
35. Shipp EB, Restum JC, Giesy JP, Bursian SJ, Aulerich RJ, Helferich WG. 1998. Multigenerational study of the effects of consumption of PCB-contaminated carp from Saginaw Bay, Lake
Huron, on mink. 2. Liver PCB concentration and induction of
hepatic cytochrome P-450 activity as a potential biomarker for
PCB exposure. J Toxicol Environ Health A 54:377–401.
36. Restum JC, Bursian SJ, Giesy JP, Render JA, Helferich WG,
Shipp EB, Verbrugge DA. 1998. Multigenerational study of the
effects of consumption of PCB-contaminated carp from Saginaw
Bay, Lake Huron, on mink. 1. Effects on mink reproduction, kit
growth and survival, and selected biological parameters. J Toxicol
Environ Health 54:343–375.
37. Bursian SJ, Sharma C, Aulerich RJ, Yamini B, Mitchell RR,
Beckett KJ, Orazio CE, Moore D, Svirsky S, Tillitt DE. 2006.
Dietary exposure of mink (Mustela vison) to fish from the Housatonic River, Berkshire County, MA, USA: Effects on organ
weights and histology and hepatic concentrations of polychlorinated biphenyls and 2,3,7,8-tetrachlorodibenzo-p-dioxin toxic
equivalents. Environ Toxicol Chem 25:1541–1550.
38. Aulerich RJ, Ringer RK, Seagran HL, Youatt WG. 1971. Effects
of feeding coho salmon and other Great Lakes fish on mink reproduction. Can J Zool 49:611–616.
39. Aulerich RJ, Ringer RK, Iwamoto S. 1973. Reproductive failure
and mortality in mink fed on Great Lakes fish. J Reprod Fertil
Suppl 19:365–376.
40. Aulerich RJ, Ringer RK. 1977. Current status of PCB toxicity to
mink, and effect on their reproduction. Arch Environ Contam
Toxicol 6:279–292.
41. Bergman A, Athanasiadou M, Bergek S, Haraguchi K, Jensen S,
Klasson-Wehler E. 1992. PCB and PCB methyl sulfones in mink
treated with PCB and various PCB fractions. Ambio 21:570–576.
42. Gillette DM, Corey RD, Helferich WG, Mcfarland JM, Lowenstine LJ, Moody DE, Hammock BD, Shull LR. 1987. Comparative
toxicology of tetrachlorobiphenyls in mink and rats. 1. Changes
in hepatic enzyme activity and smooth endoplasmic reticulum
volume. Fundam Appl Toxicol 8:5–14.
43. Platanow NS, Karstad LH. 1973. Dietary effects of polychlorinated biphenyls on mink. Can J Comp Med 37:391–400.
44. Render JA, Aulerich RJ, Bursian SJ, Nachreiner RF. 2000. Proliferation of maxillary and mandibular periodontal squamous cells
in mink fed 3,3⬘,4,4⬘,5-pentachlorobiphenyl (PCB 126). Journal
of Veterinary Diagnostic Investigation 12:477–479.
45. Render JA, Hochstein JR, Aulerich RJ, Bursian SJ. 2000. Proliferation of periodontal squamous epithelium in mink fed 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Vet Hum Toxicol 42:85–86.
46. Render JA, Bursian SJ, Rosenstein DS, Aulerich RJ. 2001. Squamous epithelial proliferation in the jaws of mink fed diets containing 3,3⬘,4,4⬘,5-pentachlorobiphenyl (PCB 126) or 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Vet Hum Toxicol 43:22–26.
47. Alexander GR. 1977. Food of vertebrate predators on trout waters
Multiple-lines-of-evidence assessment for mink
in north central lower Michigan. Michigan Academician 10:181–
195.
48. Marshall WH. 1936. A study of the winter activities of the mink.
J Mammal 17:382–392.
49. Aulerich RJ, Ringer RK, Iwamoto S. 1974. Effects of dietary
mercury on mink. Arch Environ Contam Toxicol 2:43–51.
50. O’Shea TJ, Kaiser TE, Askins GR, Chapman JA. 1981. Polychlorinated biphenyls in a wild mink population. Proceedings,
Environ. Toxicol. Chem. 27, 2008
2087
Worldwide Furbearer Conference, August 3–11, Frostburg, MD,
USA, pp 1746–1751.
51. Zwiernik MJ, Budinsky RA, Alyward L, Moore J, Kay DP, Newsted JL, Khim JS, Bursian SJ, Giesy JP. Pharmacokinetic evaluation of exposure to ecologically relevant concentrations of
2,3,7,8-TCDF and 2,3,4,7,8-PeCDF in mink (Mustela vison).
Toxicol Sci (in press).
Download