- Top of page
The authors evaluated effects of feeding ranch mink (Mustela vison) diets containing polychlorinated biphenyl (PCB)–contaminated fish from the upper Hudson River (New York, USA) on adult and offspring organ mass and pathology. Diets contained 2.5 to 20% Hudson River fish, providing 0.72 to 6.1 µg ΣPCBs/g feed (4.8–38 pg toxic equivalents [TEQWHO 2005]/g feed). Absolute thyroid and adrenal gland masses were increased in adult females and 31-week-old juveniles, respectively, and absolute liver and heart masses were decreased in six-week-old kits exposed to dietary PCBs. Dietary concentrations of 0.72 µg ΣPCBs/g feed (4.8 pg TEQWHO 2005/g feed) and greater induced mandibular and maxillary squamous epithelial proliferation in adult animals. The dietary concentration of ΣPCBs predicted to result in 20% incidence of the jaw lesion (EC20) was 2.3 µg ΣPCBs/g feed (15 pg TEQWHO 2005/g feed), and the hepatic concentration was 2.8 µg ΣPCBs/g liver (89 pg TEQWHO 2005/g liver). The EC20 values were greater than the dietary and hepatic concentrations predicted to result in a 20% increase in kit mortality (LC20) at six weeks of age (0.34 µg ΣPCBs/g feed or 2.6 pg TEQWHO 2005/g feed and 0.80 µg ΣPCBs/g liver or 13 pg TEQWHO 2005/g liver). However, the EC20 values reflect exposure of adults to PCBs for approximately six months, and the LC20 values reflect exposure of offspring from conception onward. Environ. Toxicol. Chem. 2013;32:794–801. © 2013 SETAC
- Top of page
The Hudson River flows 507 km from its source in the Adirondack Mountains to its confluence with New York Harbor at the Battery in Manhattan, New York, USA. The river is contaminated with polychlorinated biphenyls (PCBs) from Fort Edward, New York, to New York City, a distance of approximately 285 km. The source of the PCBs is attributed to two plants in Fort Edward and Hudson Falls, New York, that manufactured electrical capacitors containing PCBs. It is estimated that in the course of manufacturing activities, almost 600 metric tons of PCBs were discharged into the Hudson River between the 1940s and 1977, accounting for 50% of the PCB contamination in New York Harbor 1, 2.
A number of mammalian species, including the mink (Mustela vison), depend on the Hudson River and its floodplain for habitat, food, and breeding sites. As a piscivorous mammal, the mink is at risk from exposure to PCBs and related contaminants that have entered the aquatic environment and bioconcentrated within the food chain. Reports spanning the past 20 years have indicated that mink collected within one home range (6 km) of the upper Hudson River contained concentrations of PCBs in their fat and livers comparable to concentrations causing reproductive impairment in controlled studies with ranch mink and that tissue concentrations of PCBs have not decreased appreciably 3, 4.
Because of the mink's sensitivity to PCBs 5–7 and its designation as a sentinel wildlife species 8, a study was designed to evaluate the health effects of feeding farm-raised mink diets containing PCB-contaminated fish collected from the Hudson River. One objective, which is addressed in a companion article [9; this issue] was to assess effects on adult reproductive performance (percentage of females whelping, gestation length, percentage of stillbirths, and live kits whelped) and offspring growth and mortality through 31 weeks of age. A second objective, which is addressed in the present article, was to determine if exposure to PCB-contaminated fish would affect mass and morphology of specific organs and tissues in adult mink and their offspring. Of particular interest was a jaw lesion characterized as mandibular and maxillary squamous epithelial proliferation. This lesion has been described in a number of controlled studies involving exposure of mink to specific 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)–like congeners including 3,3′,4,4′,5-pentachlorobiphenyl (PCB 126 using International Union of Pure and Applied Chemistry nomenclature) 10–12 and environmentally derived mixtures of PCBs, polychlorinated dibenzo-p-dioxins (PCDDs), and polychlorinated dibenzofurans (PCDFs) through incorporation of contaminated fish into the diet 13, 14. In addition, the lesion has been reported in wild mink collected from sites contaminated with PCBs, PCDDs, and PCDFs 15, 16. As the jaw lesion appears to be associated with exposure to TCDD-like chemicals, can be induced in a laboratory setting in a sentinel species, and is known to occur in the wild, there is interest in mandibular and maxillary squamous epithelial proliferation as a potential indicator of exposure and adverse effects.
This article presents the effects of feeding diets containing PCB-contaminated fish from the Hudson River on the mass and morphology of specific organs and tissues, including the mandible and maxilla, of adult farm-raised mink and their offspring in terms of dietary and hepatic concentrations of ΣPCBs and World Health Organization (WHO) TCDD toxic equivalents (TEQWHO 2005).
- Top of page
The mink feeding study, approved by the Michigan State University Institutional Animal Care and Use Committee, has been previously described [9; this issue]. Briefly, 1,521 kg of carp (Cyprinus carpio) collected from three different locations in the upper Hudson River were ground and incorporated into a basal mink diet 9 at 0, 2.5, 5.0, 10, 15, and 20% of the total diet. The upper limit of 20% dietary Hudson River fish was based on the mean ΣPCB concentration in the collected fish (36 µg ΣPCBs/g) and the desire to not cause complete reproductive failure and/or adult mortality. Measured mean ΣPCB concentrations (± standard deviation) were 0.0074 (0.0016), 0.72 (0.12), 1.5 (0.21), 2.8 (0.38), 4.5 (0.49), and 6.1 (0.51) µg/g feed, respectively. The TEQWHO 2005 concentrations (± standard deviation), based on toxic equivalency factors (TEFsWHO 2005) 17, were 0.41 (0.022), 4.8 (0.15), 10 (0.074), 18 (0.13), 28 (0.12), and 38 (0.38) pg/g feed, respectively, as presented in Bursian et al. [9, this issue]. All concentrations are expressed on a wet weight basis unless otherwise specified.
A total of 75 first-year (virgin), natural dark, female mink and 30 first-year, natural dark, male mink from the Michigan State University Experimental Fur Farm herd were randomly assigned to the six treatment groups. The control and 4.5- and 6.1 µg ΣPCBs/g feed groups had 15 females and five males each. The 0.72-, 1.5-, and 2.8 µg ΣPCBs/g feed groups had 10 females and five males each. Mink were fed the experimental diets for approximately 170 d, beginning two months before breeding and continuing through gestation, lactation, and early kit growth. Those kits not necropsied were maintained on their respective treatment diets for an additional 170 d. Body weights were determined on a monthly basis until termination of the trial, when the juvenile mink were approximately 31 weeks old.
Adults and a subsample of kits were necropsied when kits were six weeks old, and surviving juveniles were necropsied at 31 weeks of age. The number of surviving kits and juveniles at key time points during the study and the number of kits and juveniles necropsied are presented in Table 1.
Table 1. Number of surviving kits/juveniles at key time points during the study
|Feed group based on dietary concentration||Whelped alive||Weaned (approximately 6 weeks old)||Alive and available for necropsy||Necropsied||Alive post necropsy period||Alive at start of growth trial||Alive at end of growth trial||Necropsied at end of growth trial|
|0.72 µg ΣPCBs/g feed||49||39||33||9||24||24||23||23|
|1.5 µg ΣPCBs/g feed||72||51||40||13||27||5||0||0|
|2.8 µg ΣPCBs/g feed||46||9||8||0||8||1||0||0|
|4.5 µg ΣPCBs/g feed||40||15||9||0||9||0||0||0|
|6.1 µg ΣPCBs/g feed||31||2a||2||0||2||0||0||0|
At time of necropsy, animals were weighed and euthanized with CO2, and specific tissues were collected for subsequent analyses. The liver, brain, heart, kidneys, spleen, thyroid gland, adrenal glands, and testes or uterus (adults only) were removed, weighed, and placed in buffered formalin for histological examination. A portion of the liver also was collected for determination of ΣPCB and TEQWHO 2005 concentrations, as described in Bursian et al. [9, this issue]. After removal of the brain, the mandible and maxilla were detached from the rest of the skull, placed in buffered formalin, and processed and evaluated as described in Beckett et al. 15. The primary pathologist as well as an independent pathologist conducted a blind review of the slides.
Statistical analyses of measurement end points were conducted using a generalized linear model framework 18, where the most appropriate class of linear models was selected based on classification of data type and correlation structure. A summary of end points classified by data type and analysis method is provided in Table 2. Continuous end points were analyzed by a linear regression model 19 when the end point was measured at the experimental unit level and the experimental units within a treatment group were expected to be independent (e.g., adult mink). Continuous end points having kits clustered within litters were analyzed by linear generalized estimating equation (GEE) regression. The GEE models are a contemporary extension of generalized linear models for clustered data that adjust for within-cluster correlation 20. Examples of clustered data in the present study were kit organ masses within litters. Binary end points measured on adults (e.g., presence of jaw lesions) were analyzed with logistic regression or, in the case of complete separation (i.e., for at least one dietary concentration end point values were the same), with Fisher's exact test 21. For histological end points having significant treatment effects, the effective concentration resulting in 20 (EC20) and 50% (EC50) prevalence of that pathology was estimated with logistic regression. Dose–response relationships were estimated for dietary and hepatic ΣPCBs as well as TEQWHO 2005. Binary end points measured on kits were analyzed with beta-binomial models or, in the case of complete separation, with Fisher's exact test. Beta-binomial models were used to account for the variation that occurred both between kits in the same litter (binomial variation) and between litters in the same treatment group (extrabinomial variation).
Table 2. Summary of study end points, data types, and statistical methods
|End point||Data type||Statistical methods|
|Adult female organ masses||Continuous||Linear regression/ANOVA|
|Six-week-old kit organ masses||Continuous||Linear generalized estimating equation regression|
|31-week-old juvenile organ masses||Continuous||Linear generalized estimating equation regression|
|Histopathology of adult organs and jaws||Binary||Logistic/Fisher's exact test|
|Histopathology of six-week-old kit organs and jaws||Binary||Logistic regression/Fisher's exact test|
|Histopathology of 31-week-old juvenile organs and jaws||Binary||Beta-binomial regression/Fisher's exact test|
All statistical analyses were conducted using R statistical software (http://www.r-project.org/), including the additional R packages geepack for GEE models and doBy for linear functions of estimated GEE regression parameters.