AN ABSTRACT OF THE DISSERTATION OF Kathleen A. Patnode for the degree of Doctor of Philosophy in Toxicology presented on May 4, 1995. Title: Alterations in Mink Reproductive Physiology following Exposure to Coplanar and Noncoplanar Polychlorinated Biphenyls (PCBs). Abstract approved: Redacted for Privacy Lawrence R. Curtis Female mink (Mustela vison) are highly sensitive to polychlorinated biphenyl (PCB)-induced reproductive impairment. The goal of this research was to develop a better understanding of the mechanism of PCB-induced effects on female mink, mink embryos, and surviving mink kits. Anestrous, juvenile female mink and pregnant, adult mink were exposed to 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB), a coplanar PCB, or 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB), a noncoplanar PCB. During anestrus or at the onset of implantation, female mink received intraperitoneal injections of 0.4 or 0.8 mg 3HCB/kg, 20 mg 2HCB/kg, corn oil (CO), or no treatment. Anestrous females were treated with 1713-estradiol for 24 h prior to analysis. In the second experiment, COexposed animals received empty silastic implants, while 3HCB-treated mink received either empty or progesterone (P)-filled implants. Weights were similar between treatment groups in the first experiment, but in the second study, 3HCB-treated females lost as much as 10 % of their body weight. Hepatic not 2HCB. Both congeners impaired 170-estradiol-stimulated up-regulation ofuterine nuclear estrogen receptors (ER) in anestrous mink. Embryotoxicity and reduced embryo growth were observed 14 days after exposure to 0.4 mg 3HCB/kg > 0.8 mg 3HCB/kg > 20 mg 2HCB/kg. In pregnant mink, all 3 HCB treatments decreased progesterone receptor (PR) affinity. ER concentration and cytosolic PR total receptor number (Rt) were significantly increased by 20 mg 2HCB/kg > 0.8 mg 3HCB/kg. No significant difference was observed in either the total number or affinity of PR between in vitro systems with or without 3HCB. From day 10 forward, implantation sites from 3HCB-treated mink contained necrotic tissue from resorbing embryos. Progesterone treatment did not prevent resorption, but increased uterine prolactin receptor concentration. Length of rib, ulna, tibia, and fibula and diameter of humerus were found to be significantly reduced in day 1 kits exposed to 2HCB in ?hero. We conclude that PCBs are antiestrogenic in anestrous females and antiprogestational in pregnant mink, but reduced PR affinity is not the proximate cause of embryo resorption. In addition, 2HCB produces estrogenic changes in bones of mink embryos. Alterations in Mink Reproductive Physiology following Exposure to Coplanar and Noncoplanar Polychlorinated Biphenyls (PCBs). by Kathleen A. Patnode A DISSERTATION submitted to Oregon State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy Completed May 4, 1995 Commencement June 1996 Doctor of Philosophy dissertation of Kathleen A. Patnode presented on May 4, 1995 APPROVED: Redacted for Privacy Major Professor, representing Toxicology Program Redacted for Privacy Chair of Toxicology Program Redacted for Privacy Dean of Graduate School I understand that my dissertation will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my dissertation to any reader upon request. Redacted for Privacy DEDICATION This dissertation is dedicated to my parents, Richard and Eleanor, and to my husband, Lou Reynolds. My parents instilled in me a love of learning and the work ethic needed to achieve farsighted goals. In my mind, these tools are the most important gifts a child can receive. My husband provided tireless support during my degree program. Without his encouragement, many of the professional and personal hurdles would have been insurmountable. ACKNOWLEDGMENTS This research was assisted by many willing and capable associates. Dr. Larry Curtis, my major professor, collaborated in all parts of this work and I benefited greatly from his experience. Dr. Fred Stormshak shared with me his knowledge of endocrinology and mink physiology and provided invaluable guidance in the research. Dr. Tony Frank, Colorado State University, and Dr. Jack Rose, Idaho State University, provided advice on experimental design and conducted analyses that were critical to my research. The assistance of Dr. Ov Slayden of Oregon Regional Primate Research Center was integral in modifying the Center's steroid receptor assay for use in mink. This thesis and the experiments described within were also improved by the recommendations of Drs. Tom Murray, Hillary Carpenter, Dan Selivonchick, and Paul Reno. Many others assisted in various aspects of the work. Ron Scott and Cliff Thompson of OSU's Experimental Fur Farm cared for and handled animals and cheerfully adjusted their schedules to accommodate my needs. Assistance with tissue collection and assays was enthusiastically provided by Beth Siddens, Carol Forsyth, and Lou Reynolds. Much of this work could not have been completed were it not for the cooperative attitude within the Toxicology Program. I am particularly grateful for access to equipment and expertise in the laboratories of Drs. Donald Buhler, Dan Selivonchick, and David Williams. Don Griffin and Gene Johnson shared their expertise in contaminant analysis and gave me access to and assistance with analytical equipment. Funding for this research was generously provided by grants from The Mink Farmers Research Foundation and National Institute of Environmental Health Sciences grants ES07060 and ES005543. I performed this research as a predoctoral trainee under a program of the National Institute of Environmental Health Sciences. CONTRIBUTION OF AUTHORS Dr. Larry Curtis was involved in the design, analysis, and writing of each manuscript. Dr. Tony Frank performed histological evaluations and analyzed the resulting data at his laboratory at Colorado State University for the second manuscript. Dr. Jack Rose conducted the prolactin receptor assays for the second manuscript at his laboratory at Idaho State University. TABLE OF CONTENTS Page INTRODUCTION Polychlorinated Biphenyl (PCB) Toxicity Nonortho-substititued congeners Monoortho-substituted congeners Diortho- substituted congeners Metabolites Reproductive Toxicity 1 1 3 4 4 5 7 Mink Reproductive Physiology 11 PCB-Induced Reproductive Toxicity in Mink 15 2,2',4,4',5,5'- and 3,3',4,4',5,5'-HEXACHLOROBIPHENYL ALTERATION OF UTERINE PROGESTERONE AND ESTROGEN RECEPTORS COINCIDES WITH EMBRYOTOXICITY IN MINK (Muste la visor) 18 Abstract 19 Introduction 20 Methods 22 Anestrous mink Pregnant mink Statistical Analysis Results Anestrous mink Pregnant mink Discussion 3,3',4,4',5,5'-HEXACHLOROBIPHENYL (3HCB)-INDUCED CHANGES IN UTERINE PHYSIOLOGY IN MINK (MUSTELA VISON) 23 25 26 27 27 27 40 46 Abstract 47 Introduction 48 TABLE OF CONTENTS (Continued) Page Methods 49 Results 52 Discussion 54 IN UTERO EXPOSURE TO 2, 4, 5, 2', 4',5'-HEXACHLOROBIPHENYL (2HCB) PRODUCES SKELETAL CHANGES IN NEONATAL MINK (MUSTELA VISON) 59 Abstract 60 Introduction 60 Methods 62 Results 64 Discussion 68 CONCLUSIONS 71 BIBLIOGRAPHY 76 APPENDICES 89 Appendix A Appendix B Embryo Resorption and Growth Impairment in Mouse Strains with Normal and Reduced Aryl Hydrocarbon Receptor Concentrations 90 Nonesterified Fatty Acid Concentrations in Plasma and Uterine Tissue of Pregnant and Nonpregnant Female Mink Following Exposure to 3,3',4,4',5,5'­ Hexachlorobiphenyl 97 LIST OF FIGURES Page Figure Figure 1-1 Figure 1-2 Figure 1-3 Figure 2-1 Figure 3-1 Figure 3-2 Saturation binding curves for [31-1]-R5020 specifically bound to progesterone receptors at 14 days for pregnant mink treated with 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg or 20 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB)/kg ip at implantation 36 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB) concentrations (Y±SE) in liver over time in anestrous and pregnant female mink exposed to 20 mg 2HCB/kg. n=3 per treatment-time group 38 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) concentrations (k+SE) in liver over time in anestrous mink exposed to 0.4 mg 3HCB/kg and pregnant female mink exposed to 0.4 or 0.8 mg 3HCB/kg 39 Uterine prolactin receptor concentrations in pregnant and nonpregnant mink exposed to corn oil (CO), 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) at 0.4 mg/kg bw with or without progesterone (P)-filled silastic implants 55 Humerus length by sex for day 1 neonatal mink from females treated with corn oil (CO) or 2,4,5,2',4',5'­ hexachlorobiphenyl (2HCB) ip at time of implantation 66 Humerus diameter by sex for day 1 neonatal mink from females treated with corn oil (CO) or 2,4,5,2',4',5'­ hexachlorobiphenyl (2HCB) ip at time of implantation 67 LIST OF TABLES Page Table Table 1-1 Table 1-2 Table 1-3 Table 1-4 Table 1-5 Table 1-6 Table 2-1 Table 3-1 Uterine nuclear estrogen receptor (ERn) in fmol/Kg DNA ±SE) in anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'- hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg 29 Uterine cytosolic estrogen receptor (ERc) in fmol/[tg DNA ±SE) in anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'- hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg 30 Total cytochrome P450 (P450) in nmol/mg protein (.5? ±SE) in livers from anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'- hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg 31 Ethoxyresorufin-O-deethylase (EROD) activity in nmol/min/mg ±SE) in livers from anestrous and pregnant mink over protein time following exposure to 20 mg 2,2',4,4',5,5'-hexachloro­ biphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachloro­ biphenyl (3HCB)/kg 32 Numbers of implantation sites (IP) and embryos and embryo characteristics (f ±SE) in mink at 14 days after ip injection on March 30 with 20 mg/kg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB) or 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) at 0.4 or 0.8 mg/kg 33 Uterine progesterone receptor concentrations (Rt) in fmoUpg DNA ±SE) and dissociation constants (Kd) in nmol (.5-c ±SE) from pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5' ­ hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5' hexachlorobiphenyl (3HCB)/kg 35 Histological evaluation of implantation sites from pregnant mink treated with 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) or corn oil (CO) with empty or progesterone (P)-filled silastic implants 53 Bone length (cm) and diameter (cm) on body weight (g) basis (mean ± SE) for day 1 mink of females exposed to corn oil (control) or 2,4,5,2',4',5'-hexachlorobiphenyl (2HCB) ip at time of implantation 65 LIST OF APPENDIX FIGURES Figure Figure A-1 Figure A-2 Page Percent of litter resorbed in female mice with normal (AhR+) and reduced (AhR-) aryl hydrocarbon receptors exposed to 80 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg, 300 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB)/kg, or corn oil (CO) via ip injection on gestation day 5 92 Embryo weights from female mice with normal (AhR+) and reduced (AhR -) aryl hydrocarbon receptors exposed to 80 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg, 300 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2FICB)/kg, or corn oil (CO) via ip injection on gestation day 5 94 LIST OF APPENDIX TABLES Table Table B-1 Table B-2 Page Nonesterified fatty acid (NEFA) concentrations in plasma of pregnant and nonpregnant female mink treated with 0.4 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) /kg or corn oil (CO) via ip injection at the putative time of implantation 101 Nonesterified fatty acid (NEFA) concentrations in uterine tissue of pregnant and nonpregnant female mink treated with 0.4 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) /kg or corn oil (CO) via ip injection at the putative time of implantation 102 Alterations in Mink Reproductive Physiology following Exposure to Coplanar and Noncoplanar Polychlorinated Biphenyls (PCBs) INTRODUCTION Polychlorinated biphenyls (PCBs) are ubiquitous contaminants in aquatic ecosystems of the world. While manufacture of these compounds has been banned in many countries, they continue to enter the environment and will remain a contaminant problem for many more decades. The impact of environmental exposure to PCBs on mammalian populations needs to be investigated. This objective is best accomplished by using a sensitive model species in which the mechanisms of PCB toxicity are well understood. The mink (Mustela vison) is a carnivorous mammal in the family Mustelidae. Although mink are not a traditional laboratory animal, they are highly valued as a wild and domestic fur-bearing species and are unique among North American mammals in their top predator position in both aquatic and terrestrial food webs. These factors have resulted in substantial information regarding their reproductive physiology, contaminant exposure, and toxicological responses, particularly responses to PCBs. Polychlorinated Biphenyl (PCB) Toxicity PCBs constitute a major group of environmental contaminants. The characteristics of chemical stability, lipophilicity, and inflammability made PCBs valuable as plasticizers, dielectric fluids, hydraulic lubricants, and flame retardants. They have been, and continue 2 to be, released into the environment as a result of improper disposal, leaching, or atmospheric deposition. Those same properties that made PCBs valuable to industry are responsible for their persistence in abiotic compartments such as sediments and their bioaccumulationlbiomagnification in aquatic food chains (Safe 1994). Aside from occupational and accidental exposures, trophic transfer of these compounds through aquatic food chains is the primary route of exposure for wildlife, livestock, and humans. Although manufacture of PCBs has been banned in the United States for more than 20 years, these contaminants continue to occur in environmental samples and animal tissues (McFarland and Clarke 1989). With 209 possible congeners (Ballschmiter and Zell 1980), these extracts contain complex PCB mixtures that differ significantly from the commercial mixture due to selective degradation and absorption (Safe 1994). Exposure of animals to commercial PCB mixtures produces a broad spectrum of effects including mortality, inhibition of body weight gain or body weight loss, porphyria, immunotoxicity, hepatotoxicity, neurotoxicity, thymus atrophy, dermal toxicity, carcinogenicity, endocrine disruption, and reproductive toxicity (Safe 1994). This pleiotropic response pattern is logical given the understanding that the toxicity of commercial PCB mixtures is the summation of effects produced by numerous, individual congeners. Investigations of PCBs, individually and in groups, indicate that the congeners fall into the following 3 structural categories with characteristic of toxic effects. 3 Nonortho-substititued congeners Chlorine located at both para positions (4 and 4') and at least one meta position (3, 3', 5,or 5') are conformationally unrestricted to the extent that the biphenyl rings commonly lay in a single plane. Studies of structure-activity relationships in birds, mammals, and fish have determined that these PCBs are structurally and toxicologically similar to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). These planar congeners induce hepatic cytochrome P450 isozymes CYP IA1 and CYP1A2 (Brunstrom 1992), glutathione S-transferases (Aoki et al. 1992), and epoxide hydrolase (Ahotupa 1981); produce thymic atrophy (Brunstrom et al. 1991), hepatomegaly (Aulerich et al. 1987), and fatty livers (Borlakoglu et al. 1990); and cause porphyria (Sinclair et al. 1990), reproductive impairment (Aulerich et al. 1987), developmental malformations (Marks et al. 1980, 1981), wasting syndrome (Leece et al. 1985), and immunosuppression (Mayura et al. 1993). As with TCDD, these effects are thought to be mediated by high affinity, low capacity binding to the cytosolic aryl hydrocarbon receptor (AhR) (Kafafi et al. 1993). The mechanism of AhR action is believed to parallel that of other cytosolic transcription factors. Upon binding of the ligand, the cytosolic AhR/PCB complex undergoes transformation, translocation to the nucleus and DNA binding to xenobiotic response elements (XREs) resulting in increased mRNA synthesis (Whitlock 1993). These steps have been clearly demonstrated for CYP1A1 induction (Whitlock 1993) and XREs have been identified upstream of DNA coding regions for human and mouse estrogen receptor (ER) (White and Gasiewicz 1993). However, the application of this mode of action to all toxic effects of TCDD-like PCB congeners is based on indirect evidence using 4 models with high and low AhR concentrations and in vitro structure-activity relationships (Safe 1994). While these congeners are the most potent AhR agonists, only congener 3,3',4,4'-tetrachlorobiphenyl occurs in large enough quantities in commercial mixtures and environmental samples to contribute substantially to dioxin-like PCB toxicity (Leonards et al. 1995). Monoortho-substituted congeners The steric hindrance of a single ortho chlorine on either biphenyl ring (2 or 2') decreases the probability that the PCB will be in the planar conformation. As a result, monoortho PCBs have lower binding affinities for the AhR than do the nonortho congeners (Kafafi et al. 1993). They still exert TCDD-like effects such as hepatic CYP1A1, CYP1A2, and epoxide hydrolase induction (Parkinson et al. 1983); liver toxicity and thymic atrophy (Brunstrom et al. 1991); reduced body weight gain (Leece et al. 1985); immunosuppression (Davis and Safe 1990); and reproductive impairment (Kihlstrom et al. 1992). Since they also induce hepatic CYP2B1, CYP2B2, and CYP2A1, these monoortho PCBs are considered to be mixed inducers (Safe 1994). In contrast to the nonorotho PCBs, these congeners often comprise a large proportion of commercial and environmental mixtures. Thus, they are considered to contribute substantially to TCDD-like toxicity of PCB mixtures (Leonards et al. 1995). Diortho-substituted congeners The steric hindrance of diortho substitution (2 and 2') virtually eliminates planar conformation of the biphenyl rings. These congeners have negligible binding affinities for 5 the AhR (Kafafi et al. 1993). They do not induce CYP1A 1 or CYP1A2 (Brunstrom et al. 1991), but are even more potent inducers of CYP2B1 and CYP2B2 than the monoortho PCBs (Parkinson et al. 1983). They produce hepatomegaly and promote liver tumor growth (Buchmann et al. 1991). Along with some of the monoortho congeners, they reduce dopamine concentrations and are neurotoxic (Shain et al. 1991). Some of these congeners are estrogenic (Jansen et al. 1993) and cause embryotoxicity without inducing malformations (Marks and Staples 1980). Furthermore, PCB 153 partially antagonizes TCDD-mediated cleft palate formation in prenatally-exposed mice (Biegel et al. 1989). Concentrations of diortho PCBs in commercial mixtures and environmental samples are varied. Concentrations of the lower chlorinated congeners are commonly low, while the more highly chlorinated ones can predominate the total PCB mixture (McFarland and Clark 1989, Mason and Ratford 1994, Leonards et al. 1994). However, their role in environmental toxicity of PCBs is not known because their toxicity has not been as thoroughly studied as that of TCDD-like congeners. Metabolites Like toxicity, the rate of metabolism and metabolite structure depend upon the chlorine-substitution pattern of the individual congener. In general, lower chlorinated congeners with at least 2 adjacent unsubstituted carbons are more readily metabolized than more completely substituted PCBs (Borlakoglu et al. 1990). Metabolism also varies with species as demonstrated by the much higher rate of 2,2',4,4',5,5'-hexachlorobiphenyl metabolism in dog as compared to rodents (Duignan et al. 1988). 6 Several metabolic pathways exist for PCBs. The only known direct metabolism results in the formation of phenols which may then be further metabolized to catechols and phenol conjugates (Safe 1994). All other metabolic pathways proceed through an arene oxide intermediate which is highly reactive and can form protein, RNA, and DNA adducts (Schnellman et al. 1985). The arene oxide can also be reduced to phenol. Thus, phenols can be produced by 2 distinct pathways and have been documented to occur in animal tissues (Bergman et al. 1994a). The intermediate can also be acted upon by epoxide hydrolase to form dihydrodiols and subsequently by dehydrogenase to yield catechols. The reactive nature of catechols is likely to be responsible for their absence in tissue samples. Alternatively, the arene oxide can be conjugated to glutathione by glutathione S­ transferases. These conjugates can then be eliminated in the feces. However, another pathway for these conjugates can lead to the formation of methylsulfones (Bergman et al. 1978). Cleavage by intestinal microbial lyases produces thiols. If these thiols are methylated, their lipophilicity increases facilitating their resorption. Subsequent Soxidation of the resorbed methyl metabolites yields methylsulfones. These metabolites have been documented in human (Haraguchi et al. 1984) and wild mammal tissues (Bergman et al. 1994b). PCB metabolites exert several unique toxic effects compared to the parent compounds. PCB phenols inhibit mitochondrial oxidative phosphorylation (Narasimhan et al. 1991) and certain P450 activities (Schmoldt et al. 1977), as well as bind to various hormone receptors and carrier proteins (Korach et al. 1988, Rickenbacher et al. 1986, 7 Brouwer et al. 1990). Methylsulfones bind with high affinity to several hormone (Gillner et al. 1988, Norlund-Moller et al. 1990) and fatty acid binding proteins (Larsen et al. 1992) and inhibit P4501A1 activity (Kiyohara et al. 1992). These actions have not been thoroughly investigated, so the role that metabolites play in PCB toxicity is unknown. Reproductive toxicity Current knowledge indicates that PCB-induced reproductive toxicity is likely to occur via multiple mechanisms. Structurally diverse congeners from all 3 categories (non-, mono-, and diortho-substituted) impair reproduction at environmentally-relevant exposures. Multiple reproductive tissues are targets for PCBs and diverse alterations including endocrine disruption, malformations, and embryo mortality occur (Peterson et al. 1993). Furthermore, species differences in dose/response relationships for acute toxicity and liver enzyme induction do not correlate with sensitivity to reproductive impairment (Lubet et al. 1990, Olson et al. 1990). As mediators of gestation, hormones are likely targets for contaminant-induced reproductive impairment, but effects vary with species, sex, age, and tissue. Some congeners rapidly accumulate and remain concentrated in steroidogenic corpora lutea (CLs) and adrenals of pregnant mice (Brandt 1975). Barsotti et al. (1979) demonstrated that TCDD-treated female rhesus monkeys had suppressed progesterone (P) and estrogen (E) plasma concentrations throughout the menstrual cycle. In pregnant rats, TCDD had no effects on plasma E despite decreased hepatic E metabolism (Shiverick and Muther 1983). Steroidogenesis was inhibited by TCDD in rat testes via decreased mobilization of cholesterol by P450scc (Moore et al. 1991). Aroclor 1254 interferes with P metabolism in 8 male guinea pig adrenals, but not testes (Goldman and Yawetz 1990). Serum PRL concentration was transiently reduced (Russell et al. 1988) and PRL/PRLR function suppressed (Jones et al. 1987) following TCDD exposure in male rats. The mechanism by which TCDD induces these hormone changes is unknown. Alternatively, reproductive toxicity may involve changes in hormone receptor function. TCDD caused dose-dependent decreases in uterine ER and progesterone (PR) in immature rats (Romkes and Safe 1988) and mice (De Vito et al. 1992a), but failed to down-regulate uterine ER concentrations in mature mice even at 3 times the effective dose in immature animals (De Vito et al. 1992b). In addition to age-specificity, TCDD-induced changes in receptors vary between species. Acutely toxic doses of TCDD in prepubertal guinea pigs produced no change in uterine ER, while hamster ER concentrations were elevated (Hruska and Olson 1989). Gestation-specific effects of TCDD or PCBs on uterine steroid receptors have not been described; therefore, their potential role in embryotoxicity and teratogenicity is unknown. The mechanism of PCB-induced changes in receptors is also unknown. Recent identification of an AhR response element on human and mouse DNA upstream from the ER coding region suggests that TCDD-like PCBs down-regulate ER concentrations by blocking transcription (White and Gasiewicz 1993). However, Nelson et al. (1989) proposed that TCDD modifies the steroid binding properties of cytosolic glucocorticoid receptors without altering quantities of immunoreactive receptor protein. While binding of TCDD to ER or PR is not known to be competitive in rodent uteri (Romkes and Safe 9 1988), PCBs and their metabolites bind to steroid binding proteins in both the rat testes (Soderkvist et al. 1986) and the rabbit uterus (Gillner et al. 1988). PCB metabolites may cause other toxic effects. Alterations in vitamin A and thyroid hormone metabolism, PCB binding to lung and adrenal proteins, and altered energy metabolism occur only after increased PCB metabolite formation (Brouwer 1991). While some congeners are metabolized at a slower rate than others (Millis et al. 1985), limited metabolism of even the most highly chlorinated PCBs occurs via direct hydroxylation, chlorine-shift hydroxylation, dechlorination, and sulfation in rodents (Kato et al. 1980, Kurachi 1983, Duignan et al. 1988). Differences in PCB metabolite formation, excretion, and tissue disposition also exist between pregnant and nonpregnant animals (Darnerud et al. 1986). Hydroxy- and methylsulfonyl-PCB metabolites are enriched in uterine luminal fluid in early gestation, yolk sac placenta in midgestation, and fetuses in late gestation in mice (Brandt et al. 1982). Some hydroxy-PCB metabolites bind to murine ER; but effective concentrations range from 42 to 5000 times that of E (Korach et al. 1988). However, steroid receptors may possess species-specific binding characteristics as demonstrated for hamster PR (Gray and Leavitt 1987) or alterations in receptor number and affinity in response to different physiological states (Chilton and Daniel 1987, Kawaguchi et al. 1991). In addition, methylsulfonyl metabolite binding kinetics have not been evaluated. Although TCDD and other AhR ligands are considered antiestrogens, some PCBs exert estrogenic effects on the rat uterus (Gellert 1978). PCB mixtures, diortho congeners, and some hydroxy metabolites of diortho congeners increase uterine weight 10 (Jansen et al. 1993, Soontornchat et al. 1994). Estrogenic effects of PCB congeners in female rats are hypothesized to be caused by metabolites based on the following observations: 1) efficacy increases with multiple doses and with doses that maximally induce pentoxyresorufin-O-deethylase activity (Li et al. 1994), 2) differences in congener efficacy parallel relative metabolism rates (Ecobichon and MacKenzie 1974), and 3) the structural similarity between hydroxy metabolites and E (Korach et al. 1988). Alternatively, PCBs may indirectly affect reproduction via their influence on other endogenous substances. PCB exposure causes significant weight loss and mobilization of fat stores in mink (Gillette et al. 1987). In anestrous mink, lipoprotein lipase activity increases and circulating triacylglycerol levels decrease (Brewster and Matsumura 1989). These changes may expose uterine ER and PR to elevated concentrations of nonesterified fatty acids (NEFAs) which are known inhibitors of steroid receptor binding (Kato 1989, Viscardi and Max 1993). Direct binding at a site distinct from the hormone and heat shock protein binding sites is the proposed mechanism of reduced receptor binding by NEFAs. This mechanism of action is compatible with observations that the quantity of immunoreactive receptor protein is unchanged by TCDD exposure (Nelson et al. 1989). In addition to regulating female reproductive tissues, hormones control growth and differentiation in developing embryos and neonates. ER has been documented in bone cell cultures (Eriksen et al. 1988), and intestine (Prince 1994). Both prenatal and neonatal exposure of mice to E results in shorter femurs without changes in body weight or reproductive tract development indicating that bone development is highly sensitive to estrogenic substances (Migliaccio et al. 1992). Research indicates that E complexes with 11 ERs on bone cells to decrease longitudinal and radial growth of the tibia resulting in shorter and narrower bones in female rodents compared to males (Turner et al. 1994). PCBs effects on growth and development are structure-dependent. PCB congeners that bind to AhR produce TCDD-like malformations such as cleft palate and hydronephrosis. However, some diortho congeners are embryotoxic without producing cleft palate or hydronephrosis in surviving embryos (Marks and Staples 1980). First-born children of PCB-exposed mothers where significantly shorter than siblings and children from unexposed mothers (Guo et al. 1994) and had reduced head size and neurodevelopmental deficits (Jacobson et al. 1990). Mink Reproductive Physiology Mink normally mate during a 4-5 week period in late winter producing just one litter per year (Enders 1952). Mating induces ovulation with approximately 10 eggs released per mating (Adams 1973). Fertilization occurs in the oviduct and transport to the uterus requires approximately 8 days (Hansson 1947). Once in the uterus, the blastocysts enter an obligate, embryonic diapause that varies in duration from 5-55 days depending on the date of mating (Enders 1952). During diapause, cell division ceases, and cell growth is minimal. The blastocysts become equally distributed throughout the bicornate uterus, but do not attach to the uterine wall (Enders 1952). During embryonic diapause, CLs formed after ovulation remain small and relatively inactive and systemic concentrations of P are low compared with those during postimplantation pregnancy (Moller 1973, Pilbeam et al. 1979, Stoufflet et al. 1989). 12 Systemic concentrations of E increase during the breeding season, decline to below detection levels during pregnancy, and remain low during anestrus (Pilbeam et al. 1979, Slayden and Stormshak 1992). Uterine concentrations of ER and PR are elevated during embryonic diapause and at the time of implantation, but are low during anestrus (Slayden 1990, Slayden and Stormshak 1992). Treatment of anestrous mink with E induces an increase in the uterine concentrations of ER and PR, as well as increasing uterine metabolic activity (Slayden and Stormshak 1992). Embryonic diapause in mink appears to be controlled by changes in photoperiod and shortly after the vernal equinox (March 21), blastocyst development resumes and embryos implant into the endometrium (Mead and Wright 1983). Implantation in carnivores involves the trophoblast adhesion to the uterine wall followed by penetration of the endometrium (Enders 1976). Cells hypertrophy and smooth and rough endoplasmic recticulum proliferate as the endometrium enters the secretory phase (Schlafke et al. 1981). A significant increase in mucopolysaccharide occurs (Murphy and James 1974). A uteroglobin-like protein occurs in mink uterine fluid during gestation (Daniel 1968). The remaining secretory products in mink uterine fluid are unknown. Progestational support is critical to the maintenance of the developing embryos. CLs undergo pronounced morphological and physiological changes after the vernal equinox (Mead 1981). Increasing P concentrations, which coincide with the time of renewed embryo development, reach a peak at mid gestation and then decline gradually in late gestation (Stoufflet et al., 1989). Reductions in P concentrations during gestation by ovariectomy (Moller 1974) or antiprogesterone antibody exposure (Rider and Heap 1986) 13 arrest development of embryos. In contrast, E peaks during the mating season, then declines to below detectable levels during gestation (Stoufflet et al. 1989). Thus, E appears to have very little regulatory influence in P-dominated mink gestation. E and P are believed to exert their actions at target tissues by binding to nuclear receptors (Truss and Beato 1993). Occupied receptors undergo conformational changes, release of companion proteins, complexing with transcription factors, and ultimately, binding to hormone response elements of DNA to initiate transcription of hormonespecific mRNA. Concentrations of steroid receptors are used to assess the responsiveness of a tissue to steroid exposure (Tate and Jordan 1983). Unoccupied receptors are quantified by exposing tissues to increasing quantities of [3H]- labeled ligand in the presence or absence of non-labeled ligand. Dissociation from endogenous ligand is prerequisite to quantifying occupied receptors. Commonly, this is accomplished by manipulating the assay temperature to increase dissociation or using a synthetic ligand with greater affinity to displace the endogenous compound. Using these techniques, it has been shown that a finite population of receptors exists at a given time in a specific tissue type, but this population exceeds the number of receptors necessary to produce a hormone-induced biological response. Receptor binding of steroid follows Michaelis-Menten saturation kinetics. Affinity and specificity of receptors are high to facilitate binding of a specific hormone at physiological concentrations. P and E binding to their receptors stimulate a variety of responses. These responses vary with target tissues as a result of tissue-specific cofactors required for 14 binding to response elements on DNA (Truss and Beato 1993). E/ER responses in the uterus include increased glucose metabolism, protein and lipid biosynthesis, and increased activity of RNA polymerase, cellular hypertrophy and hyperplasia (Tate and Jordan 1983). Included in E/ER-induced protein biosynthesis is the upregulation of ER and PR. Both receptors, maintained at low constitutive levels, are synthesized in response to E resulting in an increased responsiveness to subsequent E or P exposures (Kraus and Katzenellenbogen 1993). P in the absence of E inhibits recruitment of both ER and PR (Hsueh et al 1975). Thus, tissue responsiveness to E and P is linked to E-priming of receptors (Kraus and Katzenellenbogen 1993). This priming occurs as part of the normal reproductive cycle. With E concentrations peaking during estrus, ER and PR concentrations are elevated preparing the uterus for control by P produced by the CL during gestation. P stimulates differentiation of target cells thus lowering its responsiveness to subsequent exposures of both E and P. Mink uterine PR may also be under the direct control of prolactin (PRL). With increasing daylength following the vernal equinox, melatonin levels decline and circulating PRL concentration increase (Martinet and Main 1985, Slayden 1990). Increasing PRL is prerequisite for activation of the CL (Papke et al. 1980), for the cessation of diapause, and, subsequently, for embryo implantation (Murphy et al. 1981). At this time, a significant increase in PR occurs without any change in ER concentration in the uterus (Slayden 1990). Prolactin receptors (PRLR) occur in the mink ovary (Rose et al. 1986) and uterus (Rose et al. 1983). Thus, PRL may be directly responsible for increased P 15 synthesis by the CL and PR upregulation in the uterus which both coincide with the end of the embryonic diapause. Although the duration of diapause varies widely with date of mating, the interval from implantation to parturition is approximately 30 days (Enders 1952). While diapause synchronizes birth with optimum conditions for survival, these arrested embryos are incredibly sensitive to changes in the uterine environment and as many as 60% of them can die (Enders 1952). Thus, litter size varies from 2 to 10 kits (avg = 4), which are born in late April or early May. Kits are altricial and are unable to maintain their own body temperature until 35 days of age (Travis and Schaible 1961). Cold temperatures are stressful and can reduce survival in combination with other stressors (Wren et al. 1987a). Kits are weaned in early summer at which time adult females become anestrus until the onset of the next breeding season (Enders 1952). Kits born in May of one year are sexually mature by February of the following year at approximately 10 months of age (Enders 1952). PCB-Induced Reproductive Toxicity in Mink Mink are an excellent model for PCB-induced reproductive toxicity because of their high sensitivity and ecological equivalence to humans as high trophic level consumers in both aquatic and terrestrial ecosystems. In the 1960s, ranch mink experienced reproductive failure when fed diets containing fish byproducts from the Great Lakes ecosystem. Effects ranged from no live young to reduced size, growth, and survival of kits (Aulerich and Ringer 1977) Ovulation, fertilization, and implantation were not 16 impaired, but gestation was arrested prior to term (Platonow and Karstad 1973). As a result of an intensive research effort, PCBs were identified to be the causative agents (Aulerich et al. 1973). Experiments comparing mink to ferrets (Bleavins et al. 1980, Shull et al. 1982) and rats (Gillette et al. 1987) have clearly demonstrated the exceptional sensitivity of mink to PCBs. Recent investigations have shown that reduced litter size occurs as low as 656 pg TCDD equivalents/g liver (Heaton et al. 1991) which is significantly lower than most laboratory and domestic species (Golub et al. 1991). In addition to sensitivity, several other differences in the responses of mink to PCB exposure have been noted. Liver enlargement does not occur at exposures which produce complete reproductive failure (Aulerich et al. 1973). Toxicity in mink is not related to hepatic P450 CYP1A1 or CYP1A2 induction as is seen in rats (Gillette et al. 1987). PCB exposure which caused complete reproductive failure induced EROD activity only 2.2-fold (Brunstrom et al. 1991). P450 isozymes in the 2B family are either unchanged or inhibited by diortho congener exposure (Shull et al. 1982). Mink also appear to have a significantly different metabolic capacity in comparison to other domestic animals (Platonow and Karstad 1973). Highly chlorinated diortho congeners do not have the long half lives in fat deposits observed for rodents (Hornshaw et al. 1983). Both penta- and hexa-PCB metabolites occur in mink experimentally exposed to a 2-4 ortho-substituted PCB fraction (Bergman et al. 1992), as well as wild mink (Bergman et al. 1994b) which feed on aquatic organisms with PCB burdens dominated by highly chlorinated congeners (Maack and Sonzogni 1988, McFarland and Clarke 1989). 17 Wild mink are opportunistic predators consuming prey from both highly contaminated aquatic and relatively uncontaminated terrestrial food webs. As a result of their feeding habits, wild mink contaminant exposures are not as high as those in ranch mink fed PCBs in their diet. However, PCB exposure in prey items is more toxic to mink than exposure to equivalent concentrations of commercial PCB mixtures (Aulerich et al. 1986) due to biomagnification of the more highly chlorinated congeners. In addition, other environmental stresses to which wild mink are exposed can exacerbate the toxicity of PCBs. Contaminant toxicity is further exacerbated in adults by exposure to cold temperatures (Wren et al. 1987a). Simultaneous exposure to mercury, which is also widespread in aquatic ecosystems, synergizes the toxic effects of PCBs on kit growth and survival (Wren et al. 1987b). 18 2,2',4,4',5,5'- and 3,3',4,4',5,5'-HEXACHLOROBIPHENYL ALERATION OF UTERINE PROGESTERONE AND ESTROGEN RECEPTORS COINCIDES WITH EMBRYOTOXICITY IN MINK (Mustela visor) Kathleen A. Patnode and Larry R. Curtis Published in Toxicology and Applied Pharmacology volume 127, pages 9-18, 1994 copyright Academic Press, Inc. 19 Abstract Female mink (Mustela visor) are highly sensitive to organochlorine (0C)-induced reproductive impairment. However, mechanisms of this reproductive toxicity are unknown. We have investigated the possible role of steroid receptors in embryotoxicity and reduced neonate weights. Anestrous, juvenile female mink and pregnant adult mink were exposed to 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB), a coplanar polychlorinated biphenyl (PCB) or 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB), a noncoplanar PCB congener. Both congeners impaired 1713-estradiol-stimulated (24 hr after ip administration of 1001Ag E2f3 ip) up-regulation of uterine nuclear estrogen receptors (ERn) in anestrous mink. Embryotoxicity and reduced embryo growth were first observed 14 days after exposure to 0.4 mg 3HCB/kg > 0.8 mg 3HCB/kg > 20 mg 2HCB/kg. In pregnant mink, all 3 HCB treatments significantly increased progesterone receptor dissociation constants (PR Kd). ER concentration and PR total receptor number (Rt) were increased by 20 mg 2HCB/kg > 0.8 mg 3HCB/kg, but were unaffected by 0.4 mg 3HCB/kg. Serum E213 was below assay detection limits. Progesterone (P) concentrations were increased by 2HCB, decreased by 0.8 mg 3HCB/kg, and unchanged by 0.4 mg 3HCB/kg. Hepatic cytochrome P450 (P450) was induced 1.8-fold in anestrous and 2.2- fold in pregnant mink by 3HCB. Ethoxyresorufin-O-deethylase (EROD) was induced 13-fold and 4-fold in anestrous and pregnant mink, respectively. 2HCB exposure resulted in decreased P450 concentration in anestrous juveniles, but had no effect on P450 during gestation or EROD activity at any time. We propose that embryotoxicity and retarded embryo growth result from impairment of PR function and that differences in the efficacy of HCB treatments are a result of their dose-dependent, partial estrogenic actions which increase PR Rt via upregulation of ER. 20 Introduction Persistent organochlorines (OCs) tend to accumulate in animal tissues and constitute a major class of environmental contaminants. Reproductive function can be impaired by environmental exposure to these structurally-diverse OCs (Tanabe, 1988). Mink are an appropriate mammalian model for 0C-induced reproductive toxicity because of their high sensitivity and ecological equivalence to humans as high trophic level consumers in both aquatic and terrestrial ecosystems. Since the early 1960s, impaired reproduction has been reported in ranch mink fed diets including marine and freshwater fish byproducts. Researchers have documented effects of these 0C-contaminated fish ranging from decreased kit growth to complete reproductive failure (Sundqvist et al., 1989). Reduced litter size occurs as low as 656 pg 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) equivalents/g liver (Heaton et al., 1991). Species differences in sensitivity to reproductive dysfunction are not correlated with dose/response relationships for acute toxicity and cytochrome P450 lA 1 induction. Despite 80- and 40-fold differences in LD50 values and ethoxyresorufin-O-deethylase (EROD) induction response, respectively, between hamsters and rats (Lubet et al., 1990), reproductive toxicity occurs at similar doses (Olson et al., 1990). Mink responsiveness to TCDD isostereomers corresponds most closely to that of the guinea pig. Both species have high susceptibility to acute (Hochstein et al., 1988; Kociba and Schwetz, 1982) and reproductive toxicity (Kihlstrom et al., 1992; Olson and McGarrigle, 1992), yet exhibit low hepatic enzyme induction (Gillette et al., 1987; Holcomb et al., 1988). 21 As mediators of gestation, hormones are likely targets for contaminant-induced reproductive impairment; however, correlations have not been documented. Aulerich et a/. (1987) observed no changes in plasma concentrations of 1713-estradiol (E213) or progesterone (P) in 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)-exposed mink. In pregnant rats, TCDD had no effects on plasma E20 despite decreased hepatic E213 metabolism (Shiverick and Muther, 1983). Alternatively, reproductive toxicity may involve changes in steroid hormone receptor concentration or ligand affinity. TCDD down-regulation of uterine estrogen receptor (ER) concentration has been documented in immature Long-Evans rats (Romkes et al., 1987) and immature CD1 mice (DeVito et al., 1992a). However, uterine ER concentrations in ovariectomized, immature CD1 mice (DeVito et al., 1992a) and mature CD1 mice (DeVito et al., 1992b) are unaffected by TCDD exposure. Hruska and Olson (1989) found that TCDD treatment increased maximum ER binding in uteri of immature hamsters and Sprague-Dawley rats, but not in uteri of immature guinea pigs. TCDDinduced down-regulation of ER is thought be responsible for decreases in progesterone receptor (PR) concentration (Romkes and Safe, 1988). Lundholm (1988) documented decreased PR binding in rabbit endometrium following polychlorinated biphenyl (PCB) exposure, but did not distinguish between changes in PR concentration and affinity. Effects of TCDD or PCBs on mink uterine steroid receptors have not been described; therefore, their potential role in 0C-induced reproductive failure is unknown. The objectives of our research were to identify possible HCB-induced changes in uterine steroid receptors in mink and investigate their involvement in the course of events 22 leading to fetal resorption. Experiments were performed on both anestrous juvenile and pregnant adult mink to facilitate comparison of HCB uptake, hepatic cytochrome P450 enzyme induction, and uterine steroid horomone receptor responsiveness at different physiological states. Both coplanar and noncoplanar congeners were used to indirectly investigate AhR involvement in the mechanism of action. Methods 2,2',4,4',5,5'-Hexachlorobiphenyl (2HCB) was chosen as the noncoplanar congener because it is a major contaminant in wild mink (Larsson et al., 1990) and humans (Bordet et al., 1993). 3HCB was selected for its equivalence as a HCB with a coplanar structure. We based our 3HCB doses on total doses which produced embryo resorption in feeding studies (Aulerich et al., 1985). 2HCB dose was set at 50 times the 3HCB dose to reflect approximate ratios of these congeners in wild mink (Larsson et al., 1990). HCB congeners (>99% purity) were obtained from AccuStandard, New Haven,CT. Moxestrol (R2858) and promegestone (R5020) from Dupont-NEN (Boston, MA) were the synthetic ligands used in ER and PR binding assays, respectively. Mink were housed individually in wire cages in open-sided sheds under natural changes in photoperiod and temperature. Anestrous mink were fed approximately 125 g of growth ration with 22% fat and 38% protein once daily. Following mating, females were fed approximately 125 glday of a gestation ration containing 20% fat and 45% protein. Water was provided ad libitum. 23 Anestrous mink In this experiment a combination of HCB and estrogen treatments were used: 1) to approximate uterine estrus physiology, and 2) to test HCBs ability to inhibit estrogen stimulation. Anestrous, immature female mink were injected (ip) with either 0.4 mg 3HCB/kg (n=9), 20.0 mg 2HCB/kg (n=9), or corn oil (vehicle; n=18). Treatments were scheduled so that livers and uteri were collected within a 6-day period from mink at 6 hr, 3 days, or 14 days after HCB exposure. Twenty-four hours prior to tissue collection, HCBtreated animals received 100 lig E213 injections sc and corn oil treated animals were injected with E213 (n=9) or corn oil (n=9). Steroid receptor binding assays were performed in triplicate after Slayden et al. (1993). Briefly, uteri were homogenized on ice in 10 volumes (w/v) of cytosolic buffer (10 nM Tris-HCI, pH 7.4, 1.5 nM EDTA, 1 nM dithiothreitol, 10% (v/v) glycerol, and 25 mM Na2Mo04) We separated uterine cytosolic and nuclear fractions by differential centrifugation (1000 x g for 10 min) of homogenates. Nuclear pellets were washed 3 times with cytosolic buffer and then resuspended in nuclear buffer (10 nM Tris-HCI, pH 7.4, 1.5 nM EDTA, 1 nM dithiothreitol, 0.8 M KC1) and incubated for 1 hr. Samples (300 IA1) of nuclear fraction were frozen for DNA determination after Burton (1956). Nuclear and cytosolic fractions were purified by ultracentrifugation (100,000 x g for 1 hr). To determine total binding, we incubated nuclear supernatant for 30 min at 37°C and then both nuclear and cytosolic supernatants for 18 hr at 4°C with 10 nM [3H] -R2858 which was determined to be a saturating concentration in preliminary binding assays (data not shown). Replicate samples including 200 fold concentrations of unlabelled 82858 were 24 used to determine nonspecific binding. We separated receptor bound ligand on minicolumns of Sephadex LH-20 (1.5 ml packed volume) with 0.7 ml of the appropriate buffer and quantified ER by liquid scintillation counting. Hepatic microsomes were prepared from 5 g liver after Guengerich (1989). Total cytochrome P450 (P450) enzymes were quantified following Omura and Sato (1964) and EROD activity was determined after Prough et al. (1978). Protein concentration in microsomes was determined according to Bradford (1976). From remaining liver tissue, we extracted and purified HCB residues according to Gundersen and Pearson (1992) with the following modifications. Briefly, frozen tissue was homogenized with a Sorvall Omni Mixer. We dried 1 g subsamples of whole liver homogenate to a powder by macerating each with 7 g of anhydrous Na2SO4 with a mortar and pestle. Powdered samples were extracted in 100 ml redistilled acetone for 8 hr using a Soxhlet apparatus. We removed acetone by vacuum evaporation using a Biichi Rotavapor-RE and redissolved the lipid mixture in 5 ml redistilled hexane. We quantitatively transferred the solutions to 25 g partially-deactivated alumina columns for liquid chromatography cleanup. HCBs were eluted with 75 ml hexane and elutant was reduced to a known volume of less than 5 ml to concentrate HCBs. We determined concentrations by gas chromatography-mass spectrometry analysis on a Saturn II ion trap detector with a Varian 3400 GC equipped with a 30 m by 0.25 mm db5 capillary column. Helium at 30 cm/sec served as the carrier gas. We used a splitless injection with injector and detector temperatures of 200 and 350 °C, respectively. The column program was 80 °C for 1 min, 3 min rise to 200 °C, 8 min rise to 280 °C, and 280 °C for 8 additional min. 25 Ten percent of the samples were spiked with 2HCB or 3HCB to determine percent recovery for each congener. Pregnant mink Our objective in this experiment was to investigate the role of P and PR, the dominant steroid complex in mink gestation, in HCB-induced embryotoxicity. E213 and ER were quantified in light of changes observed in anestrous mink On March 30, average date of implantation (Stoufflet et al., 1989), pregnant female mink received either 0.4 or 0.8 mg 3HCB/kg, 20 mg 2HCB/kg, or corn oil (vehicle) via ip injection. At 3, 7, or 14 days after HCB exposure, blood samples were collected and uteri, ovaries, and livers were removed. Implantation sites were removed from uterine horns, measured, and weighed. Embryos were removed and weighed. We measured crown-rump length and head length with a micrometer. Corpora lutea on each ovary were counted. ERn and ERc concentrations, P450 concentrations, EROD activities, and liver HCB concentrations were determined as in anestrous animals. Blood samples were allowed to clot overnight at 4°C and sera were obtained by centrifugation (500 x g for 15 min). Steroids were extracted from 1 ml of serum using 1 ml reversed phase C18 silica gel columns (J.T. Baker Inc., Phillipsburg, NJ) and eluted with 5 ml of methanol. For 10% of the samples, replicates were spiked with E213 or P to determine the percent recovery for each steroid. We determined concentrations of E213 and P using enzyme immunoassay kits according to manufacturer's protocol (Caymen Chemical Co., Ann Arbor, MI). 26 Since it was unknown if PR binding parameters varied during gestation in mink, saturation binding analyses with R5020 were conducted on each sample. We incubated supernatant with 0.13 nM to 16.67 nM [3H] -R5020 for total binding determinations. Replicate samples with 200 fold concentration of unlabelled R5020 were used to estimate nonspecific binding. Uterine PRn were not measurable due to high concentrations of endogenous P. Otherwise, receptor assays were conducted as described for ER for anestrous mink. Statistical analysis We analyzed log transformed data for hepatic P450 and EROD, uterine ER, and serum E213 and P and normal data for HCB liver concentrations with 2-way analysis of variance (ANOVA). For ANOVAs with significant F values, Duncan's multiple comparisons (DMC) were used to test for differences among means. We used 2-way ANOVA and DMC nested on litter (Healy, 1972) to identify differences in embryo weights, lengths, and head sizes. A contingency table was constructed to compare numbers of resorbed and unresorbed embryos. Paired comparisons were made with Fisher exact tests (Seigel, 1956). PRc total receptor concentration (Rt) and PRc dissociation constant (Kd) were estimated using Marguardt-Levenberg non-linear least squares curve fitting algorithm (Probst and Hermkes, 1991) where specific binding = (total ligand added) (Rt) / (total ligand added + Kd). Log transformed estimates were then compared across treatments using 2-way ANOVA and DMC. 27 Results Anestrous mink After 24 hr, exogenous E213 alone increased nuclear estrogen receptor (ERn) concentrations (Table 1). Concentrations were comparable to those previously reported for E213-treated anestrous mink (Slayden and Stormshak, 1992). Treatment with HCBs significantly impaired E2f3's effect on ERn at 6 hr and 3 days, but not at 14 days after exposure. Cytosolic estrogen receptor (ERc) concentrations were unchanged by E2f3 (Table 2) suggesting the dose and/or exposure time were insufficient to induce ER synthesis. ERc concentrations were increased 29 to 42 % by HCB treatment after 6 hr and 3 days, but differences were not significant. E213 treatment alone had no significant effect on total cytochrome P450 concentrations (Table 3). P450 concentrations were reduced by 2HCB; reductions of 40 and 48% at 6 hr and 14 days, respectively were significant. Treatment with 0.4 mg 3HCB/kg also suppressed P450 initially, but induced a 1.5-fold increase in concentrations at 14 days. E2f1 and 2HCB treatments had no significant effect on EROD activity (Table 4). In 3HCB-treated mink, EROD activity was significantly induced 13-fold at 14 days. Pregnant mink Reproductive impairment was not evident after 3 or 7 days, but was observed 14 days after treatment with both congeners. Implantation sites did not differ significantly in number across treatments (Table 5), yet they were 80%, 70%, and 50% smaller in weight in 0.4 mg 3HCB/kg, 0.8 mg 3HCB/kg, and 2HCB-treated animals, respectively than in 28 corn oil treated mink (data not shown). Only amorphous necrotic tissue masses were present within implantation sites of 0.4 mg 3HCB/kg treated mink (Table 5). Significant increases in numbers of resorbed embryos also occurred in 0.8 mg 3HCB/kg and 2HCB­ treated animals. Embryo weight, crown-rump length, and head length were significantly reduced by both congeners with 3HCB causing greater growth impairment. 29 Table 1-1. Uterine nuclear estrogen receptor (ERn) in fmol/pg DNA (7 ±SE) in anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'­ hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg. TIME AFTER HCB EXPOSURE 6 hr 3 days 7 days 14 days corn oil 0.447±0.029 0.410±0.092 NE,a 0.448±0.033 corn oilb 1.799±0.187* 1.922±0.147* ND 1.676±0.233* 1.248±0.077** 1.114±0.191** ND 1.388±0.115* 1.350±0.194** 1.176±0.088** ND 1.352±0.070* corn oil ND 0.128±0.014 0.108±0.016 0.049±0.010 2HCB ND 0.150±0.031 0.053±0.009 0.104±0.021 0.4 mg/kg ND 0.158±0.024 0.099±0.022 0.044±0.008 0.8 mg/kg ND 0.103±0.016 0.075±0.007 0.072±0.013 TREATMENT Anestrous Mink 2HCBb 3HCBb 0.4 mg/kg Pregnant Mink 3HCB a Parameter was not determined (ND) for this time point b 100 pg estrogen injected sc 24 hr prior to tissue collection * differs from corn oil (p<0.05) ** differs from corn oil and corn oil/estrogen (p<0.05) 30 Table 1-2. Uterine cytosolic estrogen receptor (ERc) in fmol/p,g DNA CV ±SE) in anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'­ hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg. TIME AFTER HCB EXPOSURE 6 hr 3 days 7 days 14 days corn oil 30.100±0.841 22.277±1.573 Npa 25.919±1.938 corn oilb 31.087±0.708 26.647±3.106 ND 21.535±2.548 2HCBb 40.079±4.436 36.259±3.414 ND 28.604±2.510 41.345±1.644 37.834±4.714 ND 22.967±1.958 corn oil ND 4.526±0.496 2.992±0.642 1.268±0.227 2HCB ND 5.748±0.369 2.119±0.193 4.169±0.405* 0.4 mg/kg ND 3.103±0.632 3.410±0.853 1.290±0.118 0.8 mg/kg ND 3.040±0.148 1.660±0.204 1.779±0.081* TREATMENT Anestrous Mink 3HCBb 0.4 mg/kg Pregnant Mink 3HCB a Parameter was not determined (ND) for this time point b 100 pg estrogen injected sc 24 hr prior to tissue collection * differs from corn oil (p<0.05) 31 Table 1-3. Total cytochrome P450 (P450) in nmol/mg protein (.)7 ±SE) in livers from anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'­ hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg. TIME AFTER HCB EXPOSURE TREATMENT 6 hr 3 days 7 days 14 days corn oil 0.342±0.046 0.315±0.028 NDa 0.227±0.019 corn oil b 0.304±0.050 0.272±0.021 ND 0.254±0.017 2HCBb 0.156±0.007* 0.213+0.016 ND 0.137±0.035* 0.152±0.023* 0.311±0.030 ND 0.399±0.036* corn oil ND 0.389±0.101 0.269±0.013 0.259±0.034 2HCB ND 0.301±0.043 0.315±0.056 0.281±0.022 0.4 mg/kg ND 0.455±0.077 0.559±0.053* 0.581+0.097* 0.8 mg/kg ND 0.658±0.118 0.527±0.042* 0.667±0.125* Anestrous Mink 3HCBb 0.4 mg/kg Pregnant Mink 3HCB a Parameter was not determined (ND) for this time point b 100 pg estrogen injected sc 24 hr prior to tissue collection * differs from corn oil (p<0.05) 32 Table 1-4. Ethoxyresorufin-O-deethylase (EROD) activity in nmoUmin/mg protein (.f ± SE) in livers from anestrous and pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'- hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5' ­ hexachlorobiphenyl (3HCB)/kg. TIME AFTER HCB EXPOSURE 6 hr 3 days 7 days 14 days corn oil 0.089±0.008 0.085±0.011 ND" 0.115±0.021 corn oil° 0.062±0.008 0.075±0.026 ND 0.082±0.034 2HCB° 0.057±0.034 0.357±0.124 ND 0.115±0.030 0.163±0.067 0.486±0.167 ND 1.262±0.167* corn oil ND 0.016±0.001 0.012±0.003 0.008±0.005 2HCB ND 0.018±0.001 0.023±0.005 0.011±0.004 0.4 mg/kg ND 0.020±0.001 0.020±0.003 0.030±0.006* 0.8 mg/kg ND 0.023±0.005 0.024±0.003* 0.018±0.001 TREATMENT Anestrous Mink 3HCB° 0.4 mg/kg Pregnant Mink 3HCB a Parameter was not determined (ND) for this time point 100 [ig estrogen injected sc 24 hr prior to tissue collection * differs from corn oil (p<0.05) 33 Table 1-5. Numbers of implantation sites (IP) and embryos and embryo characteristics ±SE) in mink at 14 days after ip injection on March 30 with 20 mg/kg 2,2',4,4',5,5'­ hexachlorobiphenyl (2HCB) or 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) at 0.4 or 0.8 mg/kg. EMBRYO TREATMENT corn oil* 2HCB* #IP SITES EMBRYOS Weight(mg) 23" 21" 406.80± 69.19" 14.25± 1.68a 5.79± 101.62± 6.29+ 22.88° 1.096 2.63+ 0.476 0.00+ 0.00c 6.05+ 2.31c 0.00+ 0.00c 25" # 166 3HCB* 0.4mg/kg 22a Oc 0.8mg/kg 27" 7d * CR Length(mm) 1.21+ 0.41c n =3 per treatment group abcd dissimilar letters within columns denote differences (p<0.05) Head Length(mm) 0.62a 0.00+ 0.00c 0.57± 0.21c 34 PRc Kd and PRc Rt estimates in corn oil treated animals varied during gestation reaching a peak at 7 days (Table 6). PRc Kds were increased as early as 3 days after HCB exposure. However, these increases were not significant until 14 days after HCB treatment when PRc Kd in corn oil treated mink declined. Significant increases were seen in Rt estimates at 14 days for 2HCB and 0.8, but not 0.4 mg 3HCB/kg (Fig 1). It is unlikely that PR Kd changes reflect alterations in PR Rt as increases in Kd occur at times and HCB exposures when Rt is uneffected. ERn and ERc concentrations in pregnant mink were markedly reduced compared to those in anestrous mink (Tables 1 and 2). In pregnant mink, uterine ERn concentrations were unchanged by either congener. During gestation, ERc was increased significantly at 14 days by 2HCB and 0.8 mg 3HCB/kg exposure, while 0.4 mg 3HCB/kg had no effect. In corn oil treated animals, P concentrations (ng/ml) rose from 15.21 ± 2.01 at 3 days to a peak of 24.05 ± 2.67 at 7 days before falling to 10.23 ± 2.51 at 14 days. E213 was not measurable at any time. These determinations were comparable to those previously reported in pregnant mink (Stoufflet et al., 1989). HCB treatment did not produce significant differences in P concentrations at any particular sampling time. However, when data were pooled across time, 2HCB significantly elevated (19.24 ng/ml ± 1.56) and 0.8 mg 3HCB/kg significantly depressed (8.00 ± 0.90) serum P concentrations, while 0.4 mg 3HCB/kg-treated mink (13.73 ± 1.74) had P concentrations similar to corn oil treated animals (14.27 ± 1.85). High individual animal variability, which may obscure 35 Table 1-6. Uterine progesterone receptor concentrations (Rt) in frnol/pg DNA (-.f ±SE) and equilibrium dissociation constants (Kd) in nM ±SE) from pregnant mink over time following exposure to 20 mg 2,2',4,4',5,5'- hexachlorobiphenyl (2HCB)/kg or 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg. TIME AFTER HCB EXPOSURE (days) PARAMETER Rt TREATMENT 3 7 14 corn oil 1.472±0.360 1.703±0.982 0.761±0.257 2HCB 2.162±0.465 1.258±0.328 2.779±0.211* 0.4 mg/kg 1.548±0.133 1.324±0.495 0.960±0.060 0.8 mg/kg 1.341±0.166 1.071±0.053 1.509±0.113* corn oil 1.935±0.068 2.926±0.254 0.981±0.243 2HCB 2.578±0.359 2.682±0.591 1.973±0.019* 0.4 mg/kg 4.475±1.491 4.400±1.564 3.587±0.518* 0.8 mg/kg 4.005±1.383 3.550 ±0.664 2.617±0.489* 3HCB Kd 3HCB * differs from corn oil (p.0.05) 4 3 2 1 0 I 0 33 c14 6 9 12 total [3H] -R5020 added (nM) s0.4hcb14 © s0.8hcb14 15 A 18 s2Ohcb14 Figure 1-1. Saturation binding curves for [3H] -R5020 specifically bound to progesterone receptors at 14 days for pregnant mink treated with 0.4 or 0.8 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg or 20 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB)/kg ip at implantation. n=3 per treatment group. Total and nonspecific curves omitted for clarity. 37 differences at individual sampling times, is most likely due to imperfect synchronization of gestation. Constitutive P450 levels were not affected by gestation (Table 3) as previously demonstrated in rats and rabbits (Neale and Parke, 1973). P450 induction was 2.1 and 2.2-fold for 0.4 mg 3HCB/kg and 2.0 and 2.6-fold for 0.8 mg 3HCB/kg at 7 and 14 days, respectively. 2HCB treatment had no effect on P450 concentrations. Pregnancy did suppress constitutive EROD activity (Table 4). Similar reductions in hepatic microsomal xenobiotic metabolism during gestation have been reported for rats (Symons et al., 1982) and mice (Sinjari et al., 1993). EROD activity did not differ significantly at any time in 2HCB-treated mink. Maximum induction was 3.8- and 2.0-fold after 0.4 and 0.8 mg 3HCB/kg, respectively in pregnant mink. For both congeners, disposition differed between pregnant and anestrous mink (Figs 2 and 3). Hepatic 2HCB concentrations were approximately 7- and 5-fold greater in pregnant than anestrous mink at 3 and 14 days, respectively (Fig 2). Treatment of pregnant mink with 0.4 mg 3HCB/kg resulted in approximately 5-fold higher liver concentrations at 3 days than in anestrous females (Fig 3). This difference was no longer evident after 14 days of exposure. Doubling the dose of 3HCB in pregnant mink produced significantly higher liver concentrations only at 3 days. While ratios of 20 mg 2HCB/kg and 0.4 mg 3HCB/kg doses were 50:1, maximum ratios in liver were 65:1 for anestrous females and 20:1 for pregnant mink suggesting greater 3HCB accumulation during pregnancy. Time course patterns indicate a slower 3HCB accumulation relative to 2HCB in anestrous, but not pregnant animals. Background concentrations of 2HCB were 0.006378 [tg/ g liver, while 3HCB was undetectable in corn oil-treated animals (detection 25 0.125 Loanestrous 0 pregnant 20 - 0.100 15- - 0.075 10- 0.050 0.025 0.000 4 8 12 16 exposure time (days) Figure 1-2. 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB) concentrations (k±SE) in liver over time in anestrous and pregnant female mink exposed to 20 mg 2HCB/kg. n=3 per treatment-time group. * Significantly differs from anestrous mink (p<0.05). 2.4 0.6 0 anestrous 2.0 ­ 0pregnant A pregnant 0.5 0.04 mg 3HCB/kg 0.08 mg 3HCB/kg 1.6 ­ 0.4 1.2 ­ 0.3 0.8 ­ 0.2 0.4 ­ 0.1 0.0 0.0 0 4 8 12 16 exposure time (days) Figure 1-3. 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) concentrations (k±SE) in liver over time in anestrous mink exposed to 0.4 mg 3HCB/kg and pregnant female mink exposed to 0.4 or 0.8 mg 3HCB/kg. n=3 per treatment-time group. abc Dissimilar letters at exposure times denote differences (p<0.05). 40 limit of 0.0001 p.g). Percent recovery did not differ between congeners with an average and standard error of 72.8 ± 2.3 %. Discussion Embryo resorption at 14 days after implantation without reductions in number of implantation sites confirms evidence present at parturition (Aulerich et al., 1985; Backlin and Bergman, 1992) that PCB-induced embryotoxicity occurs following implantation. Impaired growth in surviving embryos is undoubtedly responsible for reductions in neonate weights from females fed PCBs (Aulerich et al., 1985; Kihlstrom et al., 1992). At the time of these toxicities, the mink uterine environment is controlled by P (Sundqvist et al., 1988). Reductions in functional P concentration by ovariectomy or P-antibody treatment produce this same spectrum of effects in mink and other mustelids (Moller, 1974; Rider and Heap, 1986). Yet, our results agree with earlier findings (Aulerich et al., 1985) that HCB-induced embryotoxicity is not correlated with changes in serum P. Embryo resorption and growth impairment can also occur when P binding to PR is impaired by antiprogestins. All 3 HCB treatments resulted in significant decreases in PRc affinity for ligand. While binding of TCDD to PR is not known to be competitive in immature rodent uteri (Romkes and Safe 1988), certain PCBs and PCB methyl-sulfonyl metabolites bind to P binding proteins in the rat lung (Lund et al., 1985) and the rabbit uterus (Gillner et a/., 1988). Methyl-sulfonyl metabolites have recently been documented to occur in pregnant mink exposed to coplanar, noncoplanar, and mixed PCB fractions (Bergman et al., 1992). Competitive PR binding assays with 2HCB, 3HCB, and tissue 41 specific metabolites of both congeners in mink are necessary to determine if these compounds are responsible for the observed decrease in PR binding affinity during gestation. Quantitation of cell-specific PR expression products will establish whether or not alterations in PR Kd are sufficient in magnitude to impair P/PR function. Alternatively, PCBs may act indirectly on the PR via their influence on other endogenous substances. Madej et al. (1992) found that urinary cortisol was significantly higher in PCB-treated pregnant mink beginning at implantation and continuing for 25 days. Cortisol causes increased dissociation rates of R5020 in non-estrogen primed, uterine tissue from rats (Moguilewsky and Deraedt, 1981). They propose that cortisol binds to sites other than the ligand binding site on PR and induces an allosteric change which reduces the receptor's affinity for its ligand. Replicating this work in mink would determine if elevated cortisol levels are responsible for altered PR binding affinity. While all 3 treatments decreased PR affinity, embryotoxicity varied with congener and dose. This observation suggests that a differential effect on some other reproductive end point also influences embryotoxicity. In anestrous, juvenile mink when uterine ER concentrations are high, both HCBs impaired E213-induced up-regulation of ERn in agreement with findings in immature rodents (Romkes et al., 1987; DeVito et at., 1992a). However, during mink gestaton when E213 concentrations are negligible and uterine ER and PR concentrations are declining, ERc and PR Rt were increased at 14 days after 2HCB and 0.8 mg 3HCB/kg, but not 0.4 mg 3HCB/kg. This suggests that both HCBs exert estrogenic effects which are capable of attenuating the normal physiological decline in uterine ER and PR. 42 Some PCBs and PCB metabolites exert estrogenic effects on mammalian uteri in the absence of exogenous E213 treatment (Jansen et al., 1993). In vitro binding assays indicate that hydroxy metabolites of some PCBs can bind to mouse uterine ER (Korach et al., 1988). Although their affinities are at least 40-fold lower than E2f3 (Korach et al., 1988), their occupancy of uterine ER may reach effective proportions during gestation in mink when endogenous estrogens are suppressed (Stoufflet et al., 1989). 2HCB, which induced the largest increase in uterine ERc and PRc, caused fewer resorptions than either 3HCB dose; 0.8 mg 3HCB/kg produced intermediate alterations in ER and PR concentrations and embryotoxicity; while 0.4 mg 3HCB/kg had no effect on ER or PR concentrations while producing complete litter failure. Based on these observations, HCB-induced embryotoxicity in the mink appears to be inversely related to the partial estrogenic effects of HCBs or their metabolites. HCB impaired P binding to uterine PR could lead to reduced synthesis of proteins essential to embryo growth. As plasma P concentrations rise , uterine protein secretions in mink increase (Daniel and Krishnan, 1969). One of these P/PR expression products is similar to rabbit uteroglobin (Daniel 1968), which increases cell division in cultured mink embryos (Daniel and Krishnan 1969). Reductions in this or other P/PR induced embryonic growth regulators could be responsible for HCB-induced growth impairment. Severe growth impairment may result in embryo inviability and activate maternal resorption. Alternatively, embryo resorption may also be triggered by antiprogestin treatment independent of growth impairment. Szekeres-Bartho et al. (1990) identified gestation­ 43 specific lymphocyte PR in the mouse through which P induces synthesis of an immunosuppressive protein involved in preventing the female's immune system from recognizing embryos as foreign bodies. The presence of gestation-specific PRs on mink lymphocytes must be substantiated and the effects of in vivo HCB exposure on PR kinetics by cell type determined. 2HCB had not previously been shown to cause reproductive failure in mink. Although we derived our doses from earlier studies, our experimental design may have produced different patterns of disposition and target organ concentrations. Contaminant exposure in feeding studies may be lower than computed total dose due to variables such as feed consumption (Auletta, 1988) and absorption (Houston et al., 1974). In addition, Spindler-Vomachka and Vodicnik (1984) demonstrated differences in 2HCB distribution among lipoprotein fractions with reproductive state. Thus, in longterm reproductive studies where dosing spans more than one reproductive state, lower target organ concentrations can result because of differential distribution during anestrus and gestation periods. This hypothesis is supported by 7- and 6-fold greater 2HCB concentrations in livers of pregnant mink than anestrous juveniles. Further support comes from teratogenic evaluations in mice in which 2HCB exposure during gestation was embryotoxic without producing malformations (Marks et al., 1980). Single gavage exposures to 2HCB in mice decreased fetal weights (Morrissey et al., 1992) and increased fetal resorptions (Biegel et al., 1989) while simultaneously antagonizing TCDD-induced fetal cleft palate. These results with 2HCB and the work of Kihlstrom et al., (1992) with a 2-4 ortho PCB fraction in mink demonstrate that embryotoxicity of PCBs is not mechanistically limited to 44 coplanar PCB actions and argue for the inclusion of all PCB congeners in reproductive risk assessments. Mink possess unique hepatic P450 characteristics. Previous studies have demonstrated relative resistance of mink to prototypic inducers (Shull et al., 1983), PCB mixtures (Shull et al., 1982), and planar and noncoplanar tetrachlorobiphenyl congeners (Gillette et al., 1987). Sensitivity to inducers of EROD activity appears to be greatest in kits (30-fold), moderate in juveniles (13-fold), and minimal in pregnant and postpartum adults (2-4-fold) (Brunstrom et al., 1991; this study). These characteristics are in contrast to those of mice where EROD induction potential increases with age and does not differ between pregnant and nonpregnant adult females (Sinjari et al., 1993). These interspecies differences are likely due to a combination of species-specific physiological parameters and incidental induction in all mink as a result of contaminated fish and meat byproducts in rations. Most effects of TCDD and coplanar PCBs are thought to be mediated by AhR. Induction of P450 1A1 is the best understood action of OC/AhR activation and correlation with P450 induction is taken to be indicative of AhR-mediated toxicosis (Safe, 1990). With less than a 4-fold EROD induction, 3HCB exposure caused complete reproductive failure. 2HCB caused no P450 or EROD induction in agreement with findings that 2HCB has virtually no affinity for AhR (Biegel et al., 1988), yet 2HCB increased embryo resorption and retarded embryo growth. All 3 treatments produced equivalent reductions in PR affinity. KihlstrOm and coworkers (Kihlstrom et al., 1992; BrunstrOm 1992) demonstrated increased embryotoxicity without EROD induction in 45 pregnant mink fed a 2-4 ortho PCB fraction. Collectively, these results indicate that reproductive dysfunction in mink is not correlated with P4501A1 induction. Events leading to changes in ER and PR must be elucidated to determine the mechanism of HCBinduced embryotoxicity and to identify appropriate biomarkers for this mode of action. 46 3,3',4,4',5,5'41EXACHLOROBIPHENYL (3HCB)-INDUCED CHANGES IN UTERINE PHYSIOLOGY IN MINK (MUSTELA VISON) K. A. Patnode, L. R. Curtis, W.J. Rose, and A.A. Frank Submitted to Fundamental and Applied Toxicology (1995) 47 Abstract Embryo resorption and coincident reductions in uterine cytosolic progesterone receptor (PRc) affinity occurred between days 7 and 14 following maternal exposure to 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) (Patnode and Curtis, 1994). Our objectives were to determine if 3HCB competitively binds to uterine PRc from pregnant mink, evaluate the histological changes in mink implantation sites, and test the efficacy of exogenous progesterone (P) treatment in inducing uterine prolactin receptors (PRLR) and overcoming 3HCB-induced embryo resorption. At the onset of implantation, mated female mink received 0.4 mg/kg 3HCB ip, corn oil (CO) ip, or no treatment. CO-exposed animals received empty silastic implants, while 3HCB-treated mink received either empty or P-filled implants. At 8, 10, and 12 days following exposure, implantation sites from treated mink were histologically evaluated and uterine PRLR concentrations were determined. Uteri from untreated mink at day 10 were used to evaluate the in vitro binding capacity of 3HCB to PRc. No significant difference was observed in either the total number or affinity of PRc between in vitro systems with or without 3HCB. From day 10 forward, implantation sites from 3HCB-treated mink contained necrotic tissue from resorbing embryos, while uteri from CO treatment group had no necrosis. Exogenous P treatment was ineffective in preventing resorption, but produced significant increases in uterine PRLR concentration. We conclude from these investigations that decreased affinity of the PRc was not the proximate cause of 3HCB-induced embryo resorption. 48 Introduction Previous work has demonstrated that embryo resorption and reduced uterine cytosolic progesterone receptor (PRc) affinity occur between days 7 and 14 following maternal exposure to 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) (Patnode and Curtis 1994). Progesterone action at uterine myometrial and stromal cells is critical to normal reproduction. Rising concentrations of P coincide with reactivation of embryonic growth and embryo implantation in mink (Stoufflet et al. 1989). Reductions in P concentrations by either ovariectomy (Moller 1974) or antiprogesterone antibody (Rider and Heap 1986) cause embryonic death in mustelids. With only a small fraction of circulating progesterone free (Louw et al. 1992), the modest changes we observed in PRc affinity may be sufficient to impair P/PR function leading to embryo resorption. The mechanism of PCB-induced changes in receptor affinity is unknown. While binding of TCDD to PR is not known to be competitive in rodent uteri (Romkes and Safe 1988), PCBs and their metabolites bind to steroid binding proteins in both the rat testes (Soderkvist et al. 1986) and the rabbit uterus (Gillner et al. 1988). In vitro assays to evaluate potential binding of 3HCB to uterine PR are lacking. In combination with P/PR, PRL complexing with prolactin receptor (PRLR) regulates gestation in the mink (Rose et al. 1983). Suppression of circulating P concentrations in mink at midgestation reduces uterine PRLR (Rose and Stormshak 1993). In ferrets, progesterone also induces increases in uterine PRLR in the presence of estrogen (Rose et al. 1993). In ovariectomized rabbits, progesterone treatment increases prolactin receptor (PRLR) concentrations nearly 5-fold (Chilton et al. 1988). These 49 studies suggest that uterine PRLR is an expression product of P binding to uterine PR. Thus, analysis of uterine PRLR concentrations at the time of embryo resorption could be used to determine if depressed PR affinity results in reductions in an expression product critical to sustaining gestation. Failure to maintain a progestational environment during early to mid gestation leads to embryo resorption in mink and other mammals. Histological evaluation of implantation sites has been used to determine the onset of embryo resorption (Duclos et al. 1994). Identification of the onset is necessary to elucidate the steps in 3HCB-induction of the resorption process. The overall goal of this research was to develop a better understanding of the mechanism of 3HCB-induced embryo resorption. Our first objective was to determine if 3HCB competitively binds to the cytosolic progesterone receptor of uteri from pregnant mink. A second objective was to evaluate the histological changes in mink implantation sites. Finally, we wanted to test the efficacy of exogenous P treatment in inducing PRLR concentrations and in overcoming 3HCB-induced embryo resorption. Methods Mink were housed individually in wire cages in open-sided sheds under natural changes in photoperiod and temperature. Anestrous mink were fed approximately 125 g of growth ration with 22% fat and 38% protein once daily. Following mating, females were fed approximately 125 g/day of a gestation ration containing 20% fat and 45% 50 protein. Water was provided ad libitum. Mink were weighed at the beginning and end of each treatment period. At the onset of implantation (March 30), 27 mated female mink received 0.4 mg/kg 3HCB (n = 18) or corn oil (CO; n = 9) via ip injection. Silastic implants (30mm x 0.5mm) were injected sc at the nape of the neck using a modified syringe. CO-treated animals received empty implants, while 3HCB-treated mink received either empty (n = 9) or progesterone-filled (n = 9) implants. At 8, 10, and 12 days following exposure, blood samples were collected. Animals were killed by CO2 asphyxiation and uteri were removed. Implantation sites were dissected free and fixed in neutral buffered formalin for histological evaluation. The remaining 3HCB-exposed uterine tissue and uteri from untreated mink were frozen in liquid nitrogen for PRLR and PR analyses, respectively. Histological evaluations were conducted on randomly selected implantation sites from each animal. Serial sections were generated to assure that a mid-implantation site section was evaluated. The following grading system was used to evaluate each site: GRADE DEFINITION 1 2 3 implantation site present; normal endometrium (minor superficial glandular dilation & modest mucous accumulations); no necrosis implantation site present; hyperplastic endometrium (basilar glandular dilation & abundant mucous accumulations); ± necrosis no implantation site present Results of these histological evaluations were then analyzed using Fisher exact probability testing across treatment groups and gestation day. Uterine homogenates from 7 pregnant, untreated female mink at day 10 post- implantation were used to conduct in vitro binding assays. Aliquots of uterine 51 homogenate were incubated for 30 min. at 4°C in duplicate in uncoated or 3HCB-coated tubes at a final 3HCB concentration of 100pM. Steroid receptor binding assays were then performed in triplicate after Patnode and Curtis (1994). Briefly, uteri were homogenized on ice in 10 volumes (w/v) of cytosolic buffer. We separated uterine cytosolic and nuclear fractions by differential centrifugation of homogenates. Nuclear pellets were washed 3 times with cytosolic buffer and then resuspended in nuclear buffer and incubated for 1 hr. Samples (300 pl) of nuclear fraction were frozen for DNA determination after Burton (1956). Cytosolic fractions were purified by ultracentrifugation and saturation binding analyses with R5020 were conducted on each sample. Uterine PRn were not measurable due to high concentrations of endogenous P. We incubated supernatant with 0.13 nM to 16.67 nM [3H] -R5020 for 18 hr at 4°C to determine total binding. Replicate samples with 200-fold concentration of unlabelled R5020 were used to estimate nonspecific binding. We separated receptor bound ligand on minicolumns of Sephadex LH-20 (1.5 ml packed volume) with 0.7 ml of the appropriate buffer and quantified bound PRc by liquid scintillation counting. PRc total receptor concentration (Rt) and PRc dissociation constant (Kd) were estimated using MarguardtLevenberg non-linear least squares curve fitting algorithm (Probst and Hermkes, 1991) where specific binding = (total ligand added) (Rt) / (total ligand added + Kd). Log transformed estimates were then compared between treatments using students t-test. To investigate possible effects of 3HCB on PRLR concentration and the function of P/PR complexing in 3HCB-treated mink, PRLR point assays were performed on uterine samples on 8, 10 and 12 days following treatment in CO, 3HCB-treated, and 3HCB+P­ 52 treated pregnant mink after Rose et al. (1993). Briefly, uterine homogenate was centrifuged at 1500 x g for 10 min. The resulting supernatant was centrifuged at 15,000 x g for 20 min. This second supernatant was centrifuged at 50,000 x g for 4 hr. The resulting pellet was resuspended in buffer and stored at -80°C until assayed. Homogenates were incubated with a saturating concentration of125I-PRL (ovine) with or without a 500-fold excess of competing nonradioactive PRL (saturating concentration was previously determined by incubating a constant concentration of uterine microsomal protein with increasing concentrations of125I-PRL for 18 hr at 25°C). At the end of the incubation period, 3 ml of cold buffer (Tris -HCI 0.025M, CaC12 0.025M, 0.1% bovine serum albumin, pH 7.6) were added to each tube to terminate the reaction. Bound and free hormone were separated by centrifugation at 1140 x g for 30 min. Bound PRL hormone was then quantified in the pellets by scintillation counting. Results Mink treated with CO increased in weight as gestation progressed with percent weight changes from day 0 of 4.36±1.51 on day 8, 7.13±1.12 on day 10, and 8.67±0.92 on day 12. In contrast, 3HCB-treated animals gained weight only between days 0 and 8 (2.35±1.91). From day 0 to day 10, they lost 2.95±4.40 % of their body weight and from day 0 to day 12, the loss increased to 10.75±5.98 %. Body weight loss was not previously observed at this dose of 3HCB in pregnant mink (Patnode and Curtis 1994). Progesterone receptor concentration and dissociation constant from uteri of pregnant, untreated mink were 3.926±0.989 fmol/i.ig DNA and 3.992±0.827 nmol. 53 Addition of 1001AM 3HCB to the in vitro assay did not significantly change either PR parameter (PRc concentration = 2.367±0.732 fmol/pg DNA and 3.758±0.962 nmol). Histological evaluation of the implantation sites showed no detectable differences between corn oil- and 3HCB-treated animals through day 8 (Table 1). However, at day 10 and beyond, implantation sites from 3HCB-treated mink contained necrotic tissue from resorbing embryos, while CO-exposed uteri had no necrosis. In 3HCB exposed uteri, the endometrium was hyperplastic with basilar glandular dilation and abundant mucous accumulations compared to the endometrial tissue of CO mink. Implantation sites could not be detected macroscopically or histologically in two 3HCB-treated and five 3HCB+P­ treated mink. These animals were considered nonpregnant in all subsequent analyses. The proportion on nonpregnant females in the 3HCB+P treatment group was significantly less than for control and 3HCB groups (p<0.03). Table 2-1. Histological evaluation of implantation sites from pregnant mink treated with 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) or corn oil (CO) with empty or progesterone (P)-filled silastic implants. TREATMENT CO 3HCB on day 8 3HCB on days 10/12 3HCB+P NUMBER OF SITES Necrosis/Hyperplastic Normal Endometrium 8 3 0 0 * significant difference @ p<0.05 1 0 4* 2* 54 Results of PRLR binding assays are presented in Fig. 1. Data were pooled across time because we detected no significant differences between 8, 10 and 12 day exposures. 3HCB treatment did not decrease uterine PRLR concentration (Figure 1). Exogenous P treatment significantly increased in PRLR concentration in both pregnant and nonpregnant mink. Discussion In vitro assay data indicate that 3HCB-induced changes in PRc affinity previously observed (Patnode and Curtis, 1994) are not due to direct binding of the parent PCB to uterine PR. However, it does not eliminate the possibility that a 3HCB metabolite is a PR ligand. Hydroxy-PCBs compete with estrogen for the steroid binding site on the estrogen receptor (Korach et al. 1998) and methylsufones bind to progesterone-binding proteins in the rabbit uterus (Gillner et al. 1988) and rat lung (Andersson et al. 1991). Methylsulfones occur in ranch mink exposed to nonortho-substituted PCB fractions (Bergman et al. 1992), as well as in wild mink exposed to environmental mixtures of PCBs (Bergman et al. 1994). In vitro binding assays with identified 3HCB metabolites are needed to determine if competition with uterine PRc occurs in pregnant mink. Histological evaluation indicates that resorption is not microscopically detectable on day 8, but has progressed to a highly necrotic stage by day 10. This sequence is the same as that observed in resorption-prone crosses of mouse strains (Duclos et a. 1994). In these mice, nonspecific immune cells invade decidua on day 6, increase sharply on days 7 and 8, and plateau on days 8 to 10. Thus, the onset of resorption 0 27 b -r 24 E 21 0c 18 Om. 5 4.0 CIS 4 15 ON. a 12 a 9 9 T a 0 7 tr. 6 am. PREGNANT 4. a.4 g NONPREGNANT 3 CO 3HCB 3HCB+P 1 3HCB 3HCB+P Figure 2-1. Uterine prolactin receptor concentrations in pregnant and nonpregnant mink exposed to corn oil (CO), 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) at 0.4 mg/kg bw with or without progesterone (P)-filled silastic implants (different letters denote significant differences at p<0.05; numbers denote sample sizes). 56 precedes histological evidence first detectable in day 10 mouse embryos by as much as 4 days. Given the extent of necrosis in day 10 implantation sites of mink, we suspect that immune cells infiltrated the microscopically-normal day 8 decidua. This chronology indicates that the resorption process is triggered prior to the time when PRc affinity is significantly decreased (Patnode and Curtis 1994). As a means of evaluating the biological significance of reduced PRc affinity, we exposed 3HCB-treated mink to exogenous P. We hypothesized that boosting the circulating P concentration could overcome decreased PRc affinity to maintain a progestational environment. However, this treatment was ineffective at preventing embryo resorption. Similar results were observed in P treatment of pregnant rats exposed to an antiprogestational synthetic hormone where P reversed anti-P effects with the exception of embryo resorption (Kabir et al. 1992). These observations indicate that once the resorption process is triggered, P treatment is ineffective at halting its progress. The exogenous P treatment also assessed whether functional P/PR complexing did occur as measured by increases in uterine PRLR, a putative P/PR expression product. We found that PR was still capable of binding P and producing a statistically significant response. The magnitude of the increase in PRLR that we observed was comparable of that produced by a 10-fold decrease in P in mink early in gestation (Rose and Stormshak 1993). Greater than 5-fold increases in uterine PRLR are observed with similar P treatments in ovariectomized rabbits (Chilton et al. 1988). In addition to species-specific differences, mink may have control mechanisms for uterine PRLR during gestation which maintain their concentration within a critical range. 57 We conclude that decreased affinity of the PRc is not the proximate cause of 3HCB-induced embryo resorption. The observation that the resorption must be initiated prior to significant alterations in PRc affinity combined with P-induced upregulation of PRLR suggest that both PR and embryo resorption may be effects of an earlier 3HCB­ induced change. PRL, the other endocrine mediator of mink gestation, appears to be a susceptible target. Exposure to 3HCB (DeKrey et al. 1994) and TCDD (Jones et al. 1987, Moore et al. 1989) reduces circulating PRL concentration in rodents. Russell et al. (1988) identify increases in dopamine, the inhibitor of PRL synthesis, as the cause of reduced PRL secretion. Although PRL has not been measured in pregnant mink undergoing embryo resorption, Aulerich et al. (1985) demonstrated significant increases in dopamine concentrations in brain tissue of reproductively impaired, post-partum mink exposed to either Aroclor 1254 or 3HCB. In combination with P/PR, PRL/PRLR complexing mediates mammalian gestation. PRL regulates uterine PR in mink (Slayden and Stormshak 1993) and induces 5-fold increases in uterine estrogen receptor affinity in rabbits (Chilton and Daniel 1987). Thus, decreased PRc affinity may be due to PCB-induced reductions in PRL from high concentrations normally present in mink pregnancy (Rose et al. 1986). PRL is also known to stimulate immune function (Gala 1991). It may be the stimulus for mobilizing small lymphocytes which secrete a soluble factor that suppresses cytotoxic T lymphocyte generation and prevents embryo resorption (Daya et al. 1985). We are currently assessing 58 the concentrations of serum and uterine fluid PRL in CO and 31-ICB-treated pregnant mink to determine if alterations in this hormone trigger embryo resorption. 59 IN UTERO EXPOSURE TO 2, 4, 5, 2', 4',5'-HEXACHLOROBIPHENYL (2HCB) PRODUCES SKELETAL CHANGES IN NEONATAL MINK (MUSTELA VISON) K.A. Patnode and L.R. Curtis Submitted to Journal of Toxicology and Environmental Health (1995) 60 Abstract During mink gestation when circulating endogenous estrogens are below detectable concentrations, 2HCB produced estrogenic changes in uterine steroid receptors and retarded embryo growth (Patnode and Curtis 1994). This pattern of embryotoxicity differed from that which followed exposure to a coplanar PCB congener. Our objective was to determine if growth impairments observed in day 14 mink embryos exposed to 2HCB in zitero persisted in day 1 neonates and assessed potential effects of 2HCB on estrogen-sensitive bone growth. Female mink received intraperitoneal injections of 20 mg 2HCB/kg at the time of implantation. Day 1 neonates were examined for external, visceral, and skeletal abnormalities and bone measurements were performed. Weights were similar between treatment groups. Skeletal deformities were observed in 2HCB-exposed neonates, but not in the corn oil-treated group. No differences were detected in body length, head size, or urogenital-anal distance. Average length of rib, ulna, tibia, and fibula bones and average diameter of humerus were significantly reduced in kits exposed to 2HCB in utero (p<0.05). Malformations such as cleft palate and hydronephrosis were not detected and surviving neonates recovered from growth impairments observed in day 14 embryos exposed to 2HCB in utero. Skeletal defects indicated estrogenic effects of 2HCB persisted at birth and reflected significant changes in nonreproductive estrogen-sensitive tissues in mink. Introduction Polychlorinated biphenyls (PCBs) accumulate in animal tissues and constitute a major class of environmental contaminants known to cause reproductive impairment. 61 Mink are an appropriate mammalian model for PCB-induced reproductive toxicity because of their high sensitivity and ecological equivalence to humans as high trophic level consumers in both aquatic and terrestrial ecosystems. Researchers have documented effects of feeding mink PCB-contaminated fish ranging from decreased kit growth to complete reproductive failure (Sundqvist et al. 1989). Although reproductive toxicity in most species has been attributed to PCB congeners with high binding affinity to the aryl hydrocarbon receptor (AhR; Safe 1994), embryotoxicity and reduced embryo growth occur in mink exposed to 2-4 ortho-substituted PCB congeners (Kihlstrom et al. 1992; Patnode and Curtis 1994) with low affinity for the AhR (Kafafi et al. 1993). As regulators of uterine tissue, steroid hormones and their receptors are likely targets reproductive toxins. 2,3,7,8-Tetrachloro-dibenzo-p-dioxin (TCDD) exposure in immature rodents (Romkes et al. 1987; DeVito et al. 1992) and PCB treatment in anestrous mink antagonized estrogen-induced upregulation of uterine nuclear ER (Patnode and Curtis, 1994). In contrast, PCB mixtures (Aroclor 1242), noncoplanar congeners (2HCB and PCB 47), and some hydroxy metabolites of noncoplanar congeners increase uterine weight, albeit less potently than estradiol (Jansen et al. 1993; Soontornchat et al. 1994). During mink gestation when circulating endogenous estrogens are below detectable concentrations, 2HCB treatments at the time of implantation significantly increased estrogen receptor (ER) concentration and progesterone receptor (PR) total receptor number within 14 days of exposure (Patnode and Curtis, 1994). These observations suggest that noncoplanar PCBs or their metabolites exert both estrogenic and antiestrogenic effects in mink uterine tissue. 62 In addition to regulating female reproductive tissues, steroid hormones control growth and differentiation in developing embryos and neonates. ER has been detected in bone cell cultures (Eriksen et al. 1988), and intestine (Prince 1994). Research indicates that estrogens complex with ERs on bone cells to decrease longitudnal and radial growth of the tibia, resulting in shorter and narrower bones in female rodents compared to males (Turner et al. 1994). The objective of our research was to determine if growth impairments observed in day 14 mink embryos exposed to 2HCB in utero (Patnode and Curtis, 1994) persist in day 1 neonates. Since we had observed both estrogenic and antiestrogenic effects in mink reproductive tissues following exposure to 2HCB, we examined endpoints indicative of in utero exposure to estrogens and AhR agonists. Methods On March 30, the average date of implantation (Stoufflet et al. 1989), pregnant female mink received 20 mg 2HCB/kg (n = 3), or corn oil (CO, n = 6) via ip injection. The animals were housed individually in wire cages in open-sided sheds under natural changes in photoperiod and temperature. They were fed approximately 125 g gestation ration/day of a containing 20 % fat and 45% protein. Water was provided ad libitum. On the day of whelping, kits were asphyxiated with CO2 gas. Each kit was sexed and weighed. External examinations for gross deformities were performed on all kits. Following standard teratology guidelines (Taylor 1986), approximately half of the kits 63 from each litter were used for visceral examinations and the other half for skeletal examinations. Visceral exams were conducted using the sectioning technique described by Taylor (1986). Briefly, whole bodies were fixed in Bouin's solution until decalcification was complete (2 weeks). We then replaced Bouin's with 70% alcohol for 1 week. We began by examining the palate for closure and then sectioning the head into transverse slices 1-3 mm thick. Following a midline incision, orientation and position of internal organs was assessed. We compared urogenital and genital systems with the external sex determination. Heart and kidneys were removed, sectioned, and examined for abnormalities. Liver and lungs were removed and examined for abnormal lobulation. For skeletal assessments (Taylor 1986), whole bodies were fixed in 95% alcohol for at least 1 week followed by clearing with 2% KOH for 24 hours. After removing skin and internal organs, we stained calcified bone with 0.0025% Alizarin red S in 2% KOH. Kits were then exposed to increasing concentrations (30, 50 and 80%) of glycerin for 24 hours each and stored in 100% glycerin with thymol crystals until examinations were conducted. Presence of calcified bones and abnormal ossification or fusion patterns were determined. Skull length, width, and depth; body length; tail length; and urogenital-anal distance were measured with calipers. We then measured the length and diameter of all long bones and the 14th rib using a dissecting microscope with an ocular micrometer. A contingency table was constructed to compare numbers of deformed and normal embryos. Paired comparisons were made with Fisher exact tests (Seigel, 1956). Skeletal measurements were expressed on a per gram body weight ratio to control for differences 64 in kit body sizes. We tested for differences by treatment and sex using the analysis of variance option in Systat (version 5.03, 1993). If the sex-treatment interaction was significant, we examined the sex-specific differences graphically. Results External, visceral, and skeletal examinations identified 5 kits out of 14 from 2HCB-treated females with developmental abnormalities, while none were found in control kits. Deformities included missing digits, vestigial digits, an extra rib, and a missing sternum segment. Although no malformations were found in CO-exposed kits, we were unable to detect a significant difference between the 2 treatments. It is possible that small sample sizes limited the power of our analysis. Average long bone and rib measurements, expressed as a per gram body weight ratio, are presented in Table 1. Rib, ulna, tibia, and fibula bones were found to be significantly shorter in kits exposed to 2HCB in way) (p<0.05). Diameter was significantly reduced only in the humerus. Both diameter and length changes in humerus bones were sex-dependent. In female kits from 2HCB-treated females, humerus length (Fig. 1) and diameter (Fig. 2) were reduced. However, 2HCB irz utero exposure had no effect on length and may have slightly increased humerus diameter in male kits. No difference in neonatal weight was observed between treatment groups. 65 Table 3-1. Bone length (cm) and diameter (cm) on body weight (g) basis (mean ± SE) for day 1 mink of females exposed to corn oil (control) or 2,4,5,2',4',5'-hexachlorobiphenyl (2HCB) ip at time of implantation. BONE rib humerus radius ulna femur tibia fibula LENGTH/BODY WEIGHT RATIO Control (n=6) 2HCB (n=14) 1.007 ± 0.048 0.868 ± 0.042 a 1.030 ± 0.040 0.941 ± 0.036" 0.873 ± 0.025 0.825 ± 0.022 0.755 ± 0.025 0.676 ± 0.022a 0.742 ± 0.036 0.706 ± 0.031 0.857 ± 0.024 0.711 ± 0.022a 0.750 ± 0.025 0.649 ± 0.022a DIAMETER/BODY WEIGHT RATIO Control (n=6) 2HCB (n=14) 0.095 ± 0.011 0.096 ± 0.009 0.207 ± 0.009 0.180 ± 0.008ab 0.131 ± 0.006 0.137 ± 0.005 0.114 ± 0.007 0.121 ± 0.006 0.157 ± 0.009 0.143 ± 0.008 0.147 ± .0008 0.143 ± 0.007 0.063 ± 0.006 0.072 ± 0.005 a signicantly different from control at p_.5_0.05 treatment-sex interaction significant (see Figures 1 and 2) males females 1.3 1.2 T 1.1 4 12 1.0 2 0.9 2 0.8 0.7 CO 2HCB CO 2HCB Figure 3-1. Humerus length by sex for day 1 neonatal mink from females treated with corn oil (CO) or 2,4,5,2',4"5'-hexachlorobiphenyl (2HCB) ip at time of implantation. males females 0.26 T 0.24 4 12 0.22 0.20 0.18 2 0.16 0.14 2 0.12 CO 2HCB CO 2HCB Figure 3-2. Humerus diameter by sex for day l neonatal mink from females treated with corn oil (CO) or 2,4,5,2',4',5'-hexachlorobiphenyl (2HCB) ip at time of implantation. 68 Discussion The objective of this study was to determine if 2HCB-induced growth impairments observed in day 14 mink embryos (Patnode and Curtis 1994) persisted to produce abormalities in neonates. Despite reduced weights in day 14 embryos, day 1 neonates did not weigh significantly less than control kits. Similarly, Aulerich et al. (1985) found no weight reduction in neonates of female mink chronically exposed to 2HCB in their feed. Differences in body length and head size between control and 2HCB exposed embryos no longer existed in day 1 neonatal mink. These data indicate that between day 14 of gestation and birth, surviving embryos are able to compensate for previous growth impairment. This may result from a growth advantage to surviving embryos when litter size is reduced by embryo resorption. In contrast to 100 % embryo resorption produced by a coplanar PCB (3, 3', 4, 4', 5, 5'-hexachlorobiphenyl), 2HCB exposure resulted in only 24% embryonic loss (Patnode and Curtis 1994). Given these observed potencies and the relative binding affinities of these 2 congeners to the AhR (Kafafi et al. 1993), it is possible that embryotoxicity of both PCBs could occur via mechanisms similar to TCDD. If 2HCB is acting through the AhR, malformations such as cleft palate and hydronephrosis should be present in surviving embryos. However, we did not observe either of these malformations in day 1 neonatal mink at a dose which causes embryo resorption. A similar pattern of embryotoxicity without cleft palate or hydronephrosis in late term fetuses was observed in CD-1 mice exposed to 2HCB in utero (Marks and Staples 1980). Collectively, these results suggest that the mechanism of 2HCB embryotoxicity is independent of the AhR. 69 Concurrent with embryo resorption, 2HCB exerted estrogenic effects on uterine ER and PR (Patnode and Curtis 1994). Both prenatal and neonatal exposure of mice to estrogens results in shorter femurs without changes in body weight or reproductive tract development indicating that bone development is highly sensitive to estrogenic substances (Migliaccio et al. 1992). We found that 2HCB exposure in Wen) resulted in shorter ulna, tibia, fibula, and rib bones independent of kit sex. We also observed a female-specific reduction in humerus length and diameter. All of the gross deformities we observed were also skeletal alterations. These data indicate that estrogenic effects of 2HCB are not limited to reproductive tissue, but include developing bone and potentially other estrogensensitive tissues. Similar effects on developing human embryos may explain the decreased height of first-born children of PCB-exposed Yu-Cheng mothers (Guo et al. 1994). The mechanism of 2HCB's estrogenicity is undetermined. Li et al. (1994) hypothesize that estrogenic effects of 2HCB in female rats is attributable to a metabolite based on observations that efficacy increases with multiple doses and with doses that maximally induce pentoxyresorufin-O-deethylase activity. This phenomenon is even more likely to occur in mink for several reasons and may account for the high sensitivity of mink. The mustelids, including the mink, are more closely related to the dog which possesses a P450 isozyme with a much greater capacity than the rodent isozyme for converting 2HCB to multiple metabolites (Duignan et al. 1988). In addition, both penta­ and hexa-PCB metabolites occur in mink experimentally exposed to a 2-4 ortho­ substituted PCB fraction (Bergman et al. 1992), as well as wild mink (Bergman et al. 1994) which are known to have high 2HCB tissue concentrations (Bergman et al. 1994). 70 Alternatively, the initial dose of 2HCB could be necessary to occupy binding sites of P450 isozymes and induce ER concentrations for interaction with a higher circulating concentration of 2HCB molecules. Binding of PCB congeners to the ER is thought to be limited to hydroxy metabolites due to presence of a phenolic ring (Korach et al. 1988). However, 2HCB has been shown to bind to a steroid binding protein in the rodent lung (Gillner et al. 1988). In vitro binding assays with parent and metabolites extracted from mink tissues are necessary to determine the mechanism of 2HCB estrogenicity in reproductive and other estrogen-sensitive tissues. 71 CONCLUSIONS Embryo resorption after implantation without reductions in number of implantation sites confirms evidence present at parturition (Aulerich et at, 1985; Backlin and Bergman, 1992) that PCB-induced embryotoxicity occurs following implantation. Histological evaluation indicates that the resorption process is not microscopically detectable on day 8, but has progressed to a highly necrotic stage by day 10. Given the extent of necrosis in day 10 implantation sites of mink, it is likely that immune cells have infiltrated day 8 decidua. All PCB treatments in these experiments significantly decreased cytosolic progesterone receptor (PR) affinity for ligand. However, in vitro assay data indicate that 3HCB-induced changes in PR affinity are not due to direct binding of the parent PCB to uterine PR. Metabolites occur in ranch mink exposed to PCB fractions (Bergman et al. 1992), as well as in wild mink exposed to environmental mixtures of PCBs (Bergman et al. 1994). Exogenous progesterone (P) treatment was ineffective at preventing embryo resorption. PR was still capable of binding P and producing a statistically significant response. The magnitude of the increase in prolactin receptor (PRLR) that we observed was comparable of that produced by a 10-fold decrease in P in mink early in gestation. Similar results were seen in P treatment of pregnant rats exposed to an antiprogestational synthetic hormone where P reversed anti-P effects with the exception of embryo 72 resorption (Kabir et al. 1992). These observations indicate that once the resorption process is triggered, P alone is ineffective at halting its progress. The observation that the resorption process must be initiated prior to significant alterations in PR affinity combined with P/PR responsiveness suggest that alterations in both PR and maternal immune cells may be effects of an earlier 3HCB-induced change. Prolactin (PRL), the other endocrine mediator of mink gestation, is a susceptible target. Exposure to 3HCB (DeKrey et al. 1994) reduces circulating PRL concentration in rodents. Although PRL has not been measured in pregnant mink undergoing embryo resorption, Aulerich et al. (1985) demonstrated significant increases in dopamine, an inhibitor of PRL release, in brain tissue of reproductively impaired, post-partum mink exposed to either Aroclor 1254 or 3HCB. In combination with P/PR, PRL/PRLR complexing mediates mammalian gestation. PRL regulates uterine PR in mink (Slayden and Stormshak 1993) and induces 5-fold increases in uterine estrogen receptor affinity in rabbits (Chilton and Daniel 1987). Thus, decreased PRc affinity may be due to PCB-induced reductions in PRL from high concentrations normally present in mink pregnancy (Rose et al. 1986). PRL is also known to stimulate immune function (Gala 1991). It may be the stimulus for mobilizing small lymphocytes which secrete a soluble factor that suppresses cytotoxic T lymphocyte generation and prevents embryo resorption (Daya et al. 1985). Induction of P450 1A1 is the best understood action of AhR activation and correlation with P450 induction is taken to be indicative of AhR-mediated toxicosis (Safe, 1990). Sensitivity to inducers of EROD activity appears to be greatest in kits (30-fold), 73 moderate in juveniles (13-fold), and minimal in pregnant and postpartum adults (2-4-fold) (Brunstrom et al., 1991; this study). With less than a 4-fold EROD induction, 3HCB exposure caused complete reproductive failure. 2HCB caused no P450 or EROD induction, yet increased embryo resorption and retarded embryo growth. Increased embryotoxicity without EROD induction occurred in pregnant mink fed a 2-4 ortho PCB fraction (Kihlstrom et al., 1992; Brunstrom 1992). Collectively, these results indicate that reproductive dysfunction in mink is far more sensitive than P4501A1 induction. This research was the first to demonstrate 2HCB-induced reproductive failure in mink. Embryos were resorbed and growth of surviving embryos was significantly impaired. Characteristic AhR malformations such as cleft palate and hydronephrosis were not present in neonatal mink at a 2HCB dose which causes embryo resorption. A PCB fraction of 2-4 ortho congeners was also embryotoxic in mink without causing malformations (KihIstrom et al. 1992). Collectively, these results suggest that the mechanism of 2HCB embryotoxicity is independent of the AhR and argue for the inclusion of all PCB congeners in reproductive risk assessments. Steroid receptor responses to PCB exposure vary with reproductive status in mink. In anestrous, juvenile mink when uterine ER concentrations are high, 2HCB antagonized E's effects on ER. However, during mink gestaton when E concentrations are negligible and uterine ER and PR concentrations are declining, ER and PR Rt were increased by 2HCB. In vitro binding assays indicate that hydroxy metabolites of some PCBs can bind to mouse uterine ER (Korach et al., 1988). Although their affinities are at least 40-fold lower than estradio1-1713 (Korach et al., 1988), their occupancy of uterine ER may reach 74 effective proportions during gestation in mink when endogenous estrogens are suppressed (Stoufflet et al., 1989). Thus, 2HCB is antiestrogenic when serum estrogen concentrations are elevated, yet is estrogenic during gestation. Despite reduced weights in day 14 embryos, day 1 neonates did not weigh significantly less than control kits. Similarly, Aulerich et al. (1985) found no weight reduction in neonates of female mink chronically exposed to 2HCB in their feed. Differences in body length and head size between control and 2HCB-exposed embryos no longer existed in neonatal mink. These data indicate that between day 14 of gestation and birth, surviving embryos compensate for previous growth impairment. 2HCB exposure in utero resulted in specific skeletal growth impairments. Ulna, tibia, fibula, and rib bones were shorter in 2HCB-exposed neonates independent of sex. We also observed a female-specific reduction in humerus length and diameter. Both prenatal and neonatal exposure of mice to estrogens results in shorter femurs without changes in body weight or reproductive tract development indicating that bone development is highly sensitive to estrogenic substances (Migliaccio et al. 1992). These data indicate that estrogenic effects of 2HCB are not limited to reproductive tissue, but include developing bone and potentially other estrogen-sensitive tissues. The mechanism of 2HCB's estrogenicity is undetermined. Li et al. (1994) hypothesize that estrogenic effects of 2HCB in female rats is attributable to a metabolite based on observations that efficacy increases with multiple doses and with doses that maximally induce pentoxyresorufin-O-deethylase activity. Alternatively, the initial dose of 2HCB could be necessary to occupy binding sites of P450 isozymes and induce ER 75 concentrations for interaction with a higher circulating concentration of 2HCB molecules. Binding of PCB congeners to the ER is thought to be limited to hydroxy metabolites due to presence of a phenolic ring (Korach et al. 1988). However, 2HCB has been shown to bind to a steroid binding protein in the rodent lung (Gillner et al. 1988). In conclusion, this research has identified critical timepoints in PCB-induced embryo resorption in mink. Changes in steroid receptors occur, but do not appear to the proximate cause of embryotoxicity. Both nonortho and diorth congeners are toxic to mink embryos and diortho congeners can exert both estrogenic and antiestrogenic effects. 76 BIBLIOGRAPHY Adams, C.E. 1973. The reproductive status of female mink, Mustela vison recorded as 'failed to mate'. J. Reprod. Fertil. 33:527. Ahotupa, M. 1981. Enhancement of epoxide-metabolizing enzyme activities by pure PCB isomers. Biochem. Pharmacol. 30:1866. Andersson 0., Nordlund-Moller L., Bronnegard M., Sirzea F., Ripe E., and Lund J. 1991. Purification and level of expression in bronchoalveolar lavage of a human polychlorinated biphenyl (PCB)-binding protein: evidence for a structural and functional kinship to the multihormonally regulated protein uteroglobin. Am. J. Respir. Cell. Mol. Biol. 5:6. Aoki, Y., Satoh, K., Sato, K. and Suzuki, K.T. 1992. Induction of glutathione S­ transferase P-form in primary cultured rat liver parenchymal cells by co-planar polychlorinated biphenyl congeners. Biochem. J. 281:539. Aulerich R. J., Bursian S. J., Evans M. G., Hochstein J. R., Koudele K. A., Olson B. A., and Napolitano A. C. 1987. Toxicity of 3,4,5,3',4',5'-hexachlorobiphenyl. Arch. Environ. Contam. Toxicol. 16:53. Aulerich R. J., Olson B. A., Ringer R. K., Bursian S. J., and Breslin W. J. 1985. Toxicological manifestations of 2,4,5,2',2',4',5'-hexachlorobiphenyl, 2,3,6,2',3',6'­ hexachlorobiphenyl, and 3,4,5,3',4',5'-hexachlorobiphenyl and Aroclor 1254 in mink. J. Toxicol. Environ. Health 15: 63. Aulerich, R.J. and Ringer, R.K. 1977. Current status of PCB toxicity to mink, and effect on their reproduction. Arch. Environ. Contam. Toxicol. 6:279. Aulerich, R.J., Ringer, R.K. and Iwamoto, S. 1973. Reproductive failure and mortality in mink fed on Great Lakes fish. J. Reprod. Fert. Suppl. 19:365. Aulerich, R.J., Ringer, R.K. and Safronoff, J. 1986. Assessment of primary vs. secondary toxicity of Aroclor 1254 to mink. Arch. Environ. Contam. Toxicol. 15:393. Auletta C. S. 1988. Acute systemic toxicity testing. Pages 43-71 IN Product Safety Evaluation Handbook (S. C. Gad, Ed.), Marcel Dekker, Inc., New York. Backlin B., and Bergman A. 1992. Morphological aspects on the reproductive organs in female mink (Mustela vison) exposed to polychlorinated biphenyls and fractions thereof. Ambio 21:596. Ballschmiter, K. and Zell, M. 1980. Analysis of polychlorinated biphenyls (PCB) by glass capillary gas chromatography: composition of technical Aroclor- and ClophenPCB mixtures. Fresenius Z. Analytical Chemistry 302:20. Barsotti, D.A., Abrahamson, L.J. and Allen, J.R. 1979. Hormonal alterations in female Rhesus monkeys fed a diet containing 2,3,7,8-tetrachlorodibenzo-p-dioxin. Bull. Environ. Contam. Toxicol. 21:463. 77 Bergman A., Athanasiadou M., Bergek S., Haraguchi K., Jensen S., and Wehler E. K. 1992. PCB and PCB methyl sulfones in mink treated with PCB and various PCB fractions. Ambio 21: 570. Bergman, A., Klasson-Wehler, E. and Kuroki, H. 1994a. Selective retention of hydroxylated PCB metabolites in blood. Environ. Health. Perspect. 102:464. Bergman A., Norstrom R. J., Haraguchi K., Kuroki H., and Be land P. 1994b PCB and DDE methyl sulfones in mammals from Canada and Sweden. Environ. Toxicol. Chem. 13: 121. . Bergman, A. and Wachtmeister, C.A. 1978. Synthesis of methylthio- and methylsulfonyl­ polychlorinated biphenyls via nucleophilic aromatic substitution of certain types of polychlorobiphenyls. Chemosphere 12:949. Biegel L., Harris M., Davis D., Rosengren R., Safe L., and Safe S. 1989. 2,2',4,4',5,5'­ hexachlorobiphenyl as a 2,3,7,8-tetrachlorodibenzo-p-dioxin antagonist in C57BL/6J mice. Toxicol. Appl. Pharmacol. 97:561. Bleavins, M.R., Aulerich, R.J. and Ringer, R.K. 1980. Polychlorinated biphenyls (Aroclors 1016 and 1242): effects on survival and reproduction in mink and ferrets. Arch. Environ. Contam. Toxicol. 9:627. Bordet F., Mallet J., Maurice L., Borrel S., and Venant A. 1993. Organochlorine pesticide and PCB congener content of French human milk. Bull. Environ. Contam. Toxicol. 50:425. Borlakoglu, J.T., Drew, M.G., Wilkins, J.P. and Dils, R.R. 1990. Effects of molecular substitution patterns on the cytochrome P-450-dependent metabolism of 2,2',3,5,5',6- and 2,2',3,4,4',6-hexachlorobiphenyl by rat liver microsomal monooxygenases. Biochim. Biophys. Acta 1036:167. Bradford M. M. 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248. Brandt, I. 1975. Distribution of 2,2',4,4',5,5'-hexachlorobiphenyl in mice and Chinese hamsters. Arch. Toxicol. 34:111. Brandt, I., Danrerud, P.O., Bergman, A. and Larsson, Y. 1982. Metabolism of 2,4',5­ trichlorobiphenyl: enrichment of hydroxylated and methyl sulphone metabolites in the uterine luminal fluid of pregnant mice. Chem. Biol. Interact. 40:45. Brewster, D.W. and Matsumura, F. 1989. Differential effect of 2,3,7,8­ tetrachlorodibenzo-p-dioxin on adipose tissue lipoprotein lipase activity in the guinea pig, rat, hamster, rabbit, and mink. Comp. Biochem. Physiol. C. 93:49. Brouwer, A. 1991. Role of biotransformation in PCB-induced alterations in vitamin A and thyroid hormone metabolism in laboratory and wildlife species. Biochem. Soc. Trans. 19:731. 78 Brouwer, A., Klasson-Wehler, E., Bokdam, M., Morse, D.C. and Traag, W.A. 1990. Competitive inhibition of thyroxin binding to transthyretin by monohydroxy metabolites of 3,4,3'4'-tetrachlorobiphenyl. Chemosphere 20:1257. Brunstrom B. 1992. Induction of cytochrome P-450-dependent enzyme activities in female mink (Mustela vison) and their kits by technical PCB preparations and fractions thereof. Ambio 21:585. Brunstrom B., Hakansson H., and Lundberg K. 1991. Effects of a technical PCB preparation and fractions thereof on ethoxyresorufin 0-deethylase activity, vitamin A levels and thymic development in the mink (Mustela vison). Pharmacol. Toxicol. 69:421. Buchmann, A., Ziegler, S., Wolf, A., Robertson, L.W., Durham, S.K. and Schwarz, M. 1991. Effects of polychlorinated biphenyls in rat liver: correlation between primary subcellular effects and promoting activity. Toxicol. Appl. Pharmacol. 111:454. Burton K. 1956. A study of the conditions and mechanism of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem. J. 62:315. Chilton B. S., and Daniel J. C. Jr. 1987. Differences in the rabbit uterine response to progesterone as influenced by growth hormone or prolactin. J. Reprod. Fertil. 79:581. Chilton B. S., Mani S. K., and Bullock D. W. 1988. Servomechanism of prolactin and progesterone in regulating uterine gene expression. Mol. Endocrinol. 2:1169. Daniel J. C. 1968. Comparison of electrophoretic patterns of uterine fluid from rabbits and mammals having delayed implantation. Comp. Biochem. Physiol. 24:297. Daniel J. C., and Krishnan R. S. 1969. Studies on the relationship between uterine fluid components and the diapausing state of blastocysts from mammals having delayed implantation. J. Exp. Zool. 172:267. Darnerud, P.O., Brandt, I., Klasson-Wehler, E., Bergman, A., D'Argy, R., Dencker, L. and Sperber, G.O. 1986. 3,3'4,4'-tetrachloro[14c]biphenyl in pregnant mice:enrichment of phenol and methyl sulphone metabolites in late gestational fetuses. Xenobiotica 16:295. Davis, D. and Safe, S. 1990. Immunosuppressive activities of polychlorinated biphenyls in C57BL/6N mice: structure-activity relationships as Ah receptor agonists and partial antagonists. Toxicology. 63:97. Daya S., Clark D. A., Devlin C., Jarrell J., and Chaput A. 1985. Suppressor cells in human decidua. Am. J. Obstet. Gynecol. 151:267. DeKrey G. K., Hollingshead N. C., Kerkvliet N. I., and Smith B. B. 1994. Suppression of prolactin and cytotoxic T lymphocyte activity in PCB-treated mice. Intnl. J. of Immunopharm. 16:251. 79 De Vito M. J., Thomas T., Martin E., Umbreit T. H., and Gallo M. A. 1992a Antiestrogenic action of 2,3,7,8-tetrachlorodibenzo-p-dioxin: tissue-specific regulation of estrogen receptor in CD1 mice. Toxicol. Appl. Pharmacol. 113: 284. . DeVito M. J., Thomas T., Umbreit T. H., and Gallo M. A. 1992b. Differential sensitivity to the antiestrogenic effects of TCDD between sexually immature and mature mice. Toxicologist 12:137. Duclos A. J., Pomerantz D. K., and Baines M. G. 1994. Relationship between decidual leukocyte infiltration and spontaneous abortion in a murine model of early fetal resorption. Cell. Immunol. 159:184. Duignan D. B., Sipes I. G., Ciaccio P. J., and Halpert J. R. 1988 The metabolism of xenobiotics and endogenous compounds by the constitutive dog liver cytochrome P450 PBD-2. Arch. Biochem. Biophys. 267: 294. . Ecobichon, D.J. and MacKenzie, D.O. 1974. The uterotrophic activity of commercial and isomerically-pure chlorobiphenyls in the rat. Res. Commun. Chem. Pathol. Pharmacol. 9:85. Enders, A.C. 1976. Anatomical aspects of implantation. J. Reprod. Fert. Suppl. 25:1. Enders, R.K. 1952. Reproduction in mink. Proc. Amer, Phil. Soc. 96:691. Eriksen E. F., Colvard D. S., Berg N. J., Graham M. L., Mann K. G., Spelberg T. C., and Riggs B. L. 1988 Evidence of estrogen receptor in normal human osteoblast-like cells. Science. 241: 84. . Gala R. R. 1991. Prolactin and growth hormone in the regulation of the immune system. Proc. Soc. Exp. Biol. Med. 198:513. Gellert, R.J. 1978. Uterotrophic activity of polychlorinated biphenyls (PCB) and induction of precocious reproductive aging in neonatally treated female rats. Environ. Res. 16:123. Gillette D. M., Corey R. D., Helferich W. G., McFarland J. M., Lowenstine L. J., Moody D. E., Hammock B. D., and Shull L. R. 1987. Comparative toxicology of tetrachlorobiphenyls in mink and rats. I. Changes in hepatic enzyme activity and smooth endoplasmic reticulum volume. Fundam. Appl. Toxicol. 8:5. Gillner M., Lund J., Cambillau C., Alexandersson M., Hurtig U., Bergman A., KlassonWehler E., and Gustafsson J. A. 1988. The binding of methylsulfonyl-polychloro­ biphenyls to uteroglobin. J. Steroid. Biochem. 31:27. Goldman, D. and Yawetz, A. 1990. The interference of aroclor 1254 with progesterone metabolism in guinea pig adrenal and testes microsomes. J. Biochem. Toxicol. 5:99. Golub, M.S., Donald, J.M. and Reyes, J.A. 1991. Reproductive toxicity of commercial PCB mixtures: LOAELs and NOAELs from animal studies. Environ. Health. Perspect. 94:245. 80 Gray, G.O. and Leavitt, W.W. 1987. RU486 is not an antiprogestin in the hamster. J. Steroid. Biochem. 28:493. Guengerich F. P. 1989. Analysis and characterization of enzymes. Pages 777-814 IN Principles and Methods of Toxicology (A. W. Hayes, Ed.), Raven Press, New York. Gundersen D. T., and Pearson W. D. 1992. Partitioning of PCBs in the muscle and reproductive tissues of paddlefish, Polyodon spathula, at the Falls of the Ohio River. Bull. Environ. Contam. Toxicol. 49:455. Guo Y. L., Lin C. J., Yao W. J., Ryan J. J., and Hsu C. C. 1994 Musculoskeletal changes in children prenatally exposed to polychlorinated biphenyls and related compounds (Yu-Cheng children). J. Toxicol. Environ. Health 41: 83. . Hansson, A. 1947. The physiology of reproduction in mink (Mustela vison Schreb.) with special reference to delayed implantation. Acta Zool. Stockhlm. 28:1. Haraguchi, H., Kuroki, H., Masuda, Y. and Shigematsu, N. 1984. Determination of methylthio and methylsulphone polychlorinated biphenyls in tissues of patients with "Yusho". Food. Chem. Toxicol. 22:283. Healy M. J. R. 1972. Animal litters as experimental units. Appl. Stat. 21:155. Heaton S. J., Aulerich R. J., Bursian S. J., Giesy J. P., Render J. A., Tillitt D. E., and Kubiak T. J. 1991. Reproductive performance of mink fed Saginaw Bay carp. Proceedings of the Annual Meeting of the Society of Environmental Toxicology and Chemistry :85. Hochstein J. R., Aulerich R. J., and Bursian S. J. 1988. Acute toxicity of 2,3,7,8­ tetrachlorodibenzo-p-dioxin to mink. Arch. Environ. Contam. Toxicol. 17:33. Holcomb M., Yao C., and Safe S. 1988. Biologic and toxic effects of polychlorinated dibenzo-p-dioxin and dibenzofuran congeners in the guinea pig: quantitative structure-activity relationships. Biochem. Pharmacol. 37:1535. Hornshaw, T.C., Aulerich, R.J. and Johnson, H.E. 1983. Feeding Great Lakes fish to mink: effects on mink and accumulation and elimination of PCBS by mink. J. Toxicol. Environ. Health 11:933. Houston J. B., Upshall D. G., and Bridges J. W. 1974. A Re-evaluation of the importance of partition coefficients in the gastrointestinal absorption of anutrients. J. Pharmacol. Exp. Ther. 189:244. Hruska R. E., and Olson J. R. 1989. Species differences in estrogen receptors and in the response to 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure. Toxicol. Lett. 48:289. Hsueh, A.J., Peck, E.J. and Clark, J.H. 1975. Progesterone antagonism of the oestrogen receptor and oestrogen-induced uterine growth. Nature. 254:337. Jacobson, J.L., Jacobson, S.W. and Humphrey, H.E. 1990. Effects of in utero exposure to polychlorinated biphenyls and related contaminants on cognitive functioning in young children. J. Pediatr. 116:38. 81 Jansen H. T., Cooke P. S., Porcelli J., Lui T. C., and Hansen L. G. 1993 Estrogenic and antiestrogenic actions of PCBs in the female rat: in vitro and in vivo studies. Reprod. Toxicol. 7: 1. . Jones M. K., Weisenburger W. P., Sipes I. G., and Russell D. H. 1987. Circadian alterations in prolactin, corticosterone, and thyroid hormone levels and downregulation of prolactin receptor activity by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 87:337. Kabir S. N., Ghosh S., Pal A. K., and Pakrashi A. 1992. Effect of RMI 12,936, an antiprogestational steroid on decidual cell reaction and embryos in rat. Indian J. Exp. Biol. 30:342. Kafafi S. A., Afeefy H. Y., Ali A. H., Said H. K., and Kafafi A. G. 1993 . Binding of polychlorinated biphenyls to the aryl hydrocarbon receptor. Environ. Health. Perspect. 101: 422. Kato, J. 1989. Arachidonic acid as a possible modulator of estrogen, progestin, androgen, and glucocorticoid receptors in the central and peripheral tissues. J. Steroid. Biochem. 34:219. Kato, S., McKinney, J.D. and Matthews, H.B. 1980. Metabolism of symmetrical hexachlorobiphenyl isomers in the rat. Toxicol. Appl. Pharmacol. 53:389. Kawaguchi, K., Fujii, S., Konishi, I., Iwai, T., Nanbu, Y., Nonogaki, H., Ishikawa, Y. and Mori, T. 1991. Immunohistochemical analysis of oestrogen receptors, progesterone receptors and Ki-67 in leiomyoma and myometrium during the menstrual cycle and pregnancy. Virchows. Arch. A. Pathol. Anat. Histopathol. 419:309. Kihlstrom J. E., Olsson M., Jensen S., Johansson A., Ahlbom J., and Bergman A. 1992 Effects of PCB and different fractions of PCB on the reproduction of the mink (Mustela vison). Ambio 21: 563. . Kiyohara, C., Omura, M., Hirohata, T. and Masuda, Y. 1992. Comparison of suppression of mutagenicity of benzo(a)pyrene among methylsulfonyl polychlorinated biphenyl isomers. Bull. Environ. Contam. Toxicol. 48:877. Kociba R. J., and Schwetz B. A. 1982. Toxicity of 2,3,7,8-TCDD. Drug. Metab. Rev. 13:387. Korach K. S., Sarver P., Chae K., McLachlan J. A., and McKinney J. D. 1988 Estrogen receptor-binding activity of polychlorinated hydroxybiphenyls: conformationally restricted structural probes. Mol. Pharmacol. 33: 120. . Kraus, W.L. and Katzenellenbogen, B.S. 1993. Regulation of progesterone receptor gene expression and growth in the rat uterus: modulation of estrogen actions by progesterone and sex steroid hormone antagonists. Endocrinology. 132:2371. Kurachi, M. 1983. A new sulfur-containing derivative and possibility of conjugate formation of PCBs in mice and rats. Agric. Biol. Chem. 47:1183. 82 Larsen, G.L., Huwe, J.K., Bergman, A., Klasson-Wehler, E. and Hargis, P. 1992. Methylsulfonyl metabolites of xenobiotics can serve as ligands for fatty acid binding proteins in chicken liver and intestinal mucosa. Chemosphere 25:1189. Larsson P., Woin P., and Knulst J. 1990. Differences in uptake of persistent pollutants for predators feeding in aquatic and terrestrial habitats. Holarctic Ecol. 13:149. Leece, B., Denomme, M.A., Towner, R., Li, S.M. and Safe, S. 1985. Polychlorinated biphenyls: correlation between in vivo and in vitro quantitative structure-activity relationships (QSARs). J. Toxicol. Environ. Health 16:379. Leonards, P.E., De Vries, T.H., Minnaard, W., Stuijfzand, S., De Voogt, P., Cofino, W.P., Van Straalen, N.M. and Van Hattum, B. 1995. Assessment of experimental data on PCB-induced reproduction inhibition in mink, based on an isomer- and congener-specific approach using 2,3,7,8-tetrachlorodibenzo-p-dioxin toxic equivalency. Environ. Toxicol. Chem. 14:639. Leonards, P.E., Van Hattum, B., Cofino, W.P. and Brinkman, U.A. 1994. Occurrence of non-ortho-, mono-ortho, and di-ortho-substituted PCB congeners in different organs and tissues of polecats (Mustela putorius L.) from the Netherlands. Environ. Toxicol. Chem. 13:129. Li M. H., Zhao Y. D., and Hansen L. G. 1994 Multiple dose toxicokinetic influence on the estrogenicity of 2,2',4,4',5,5'-hexachlorobiphenyl. Bull. Environ. Contam. Toxicol. 53: 583. . Louw A. I., van-Wyk V., and van-Aarde R. J. 1992. Properties of proteins binding plasma progesterone in pregnant Cape porcupines (Hystrix africaeaustralis). J. Reprod. Fertil. 96:117. Lubet R. A., Syi J. L., Nelson J. 0., and Nims R. W. 1990. Induction of hepatic cytochrome P-450 mediated alkoxyresorufin 0-dealkylase activities in different species by prototype P-450 inducers. Chem. Biol. Interact. 75:325. Lund J., Brandt I., Poellinger L., Bergman A., Klasson-Wehler E., and Gustafsson J. A. 1985. Target cells for the polychlorinated biphenyl metabolite 4,4'­ bis(methylsulfony1)-2,2',5,5'-tetrachlorobiphenyl. Characterization of high affinity binding in rat and mouse lung cytosol. Mol. Pharmacol. 27: 314. Lundholm C. E. 1988. The effects of DDE, PCB and chlordane on the binding of progesterone to its cytoplasmic receptor in the eggshell gland mucosa of birds and the endometrium of mammalian uterus. Comp. Biochem. Physiol. C. 89:361. Maack, L. and Sonzogni, W.C. 1988. Analysis of polychlorobiphenyl congeners in Wisconsin fish. Arch. Environ. Contam. Toxicol. 17:711. Madej A., Forsberg M., and Edqvist L. 1992. Urinary excretion of cortisol and oestrone sulfate in pregnant mink fed PCB and fractions of PCB. Ambio 21:582. 83 Marks, T.A., Kimmel, G.L. and Staples, R.E. 1981. Influence of symmetrical polychlorinated biphenyl isomers on embryo and fetal development in mice: I. teratogenicity of 3,3',4,4',5,5'-hexachlorobiphenyl. Toxicol. Appl. Pharmacol. 61:269. Marks T. A., and Staples R. E. 1980 Teratogenic evaluation of the symmetrical isomers of hexachlorobiphenyl (HCB) in the mouse. Teratol. 21: 54A. . Martinet, L. and Allain, D. 1985. Role of the pineal gland in the photoperiodic control of reproductive and non-reproductive functions in mink (Mustela vison). Ciba Found. Symp. 117:170. Mason, C.F. and Ratford, J.R. 1994. PCB congeners in tissues of European otter (Lutra Infra). Bull. Environ. Contam. Toxicol. 53:548. Mayura, K., Spainhour, C.B., Howie, L., Safe, S. and Phillips, T.D. 1993. Teratogenicity and immunotoxicity of 3,3',4,4',5-pentachlorobiphenyl in C57BL/6 mice. Toxicology. 77:123. McFarland, V.A. and Clarke, J.U. 1989. Environmental occurrence, abundance, and potential toxicity of polychlorinated biphenyl congeners: considerations for a congener-specific analysis. Environ. Health. Perspect. 81:225. Mead, R.A. 1981. Delayed implantation in the mustelidae with special emphasis on the spotted skunk. J. Reprod. Fert. Suppl. 29:11. Mead, R.A. and Wright, P.L. 1983. Reproductive cycles of mustelidae. Acta Zool. Fenn. 174:169. Migliaccio S., Newbold R. R., Bullock B. C., McLachlan J. A., and Korach K. 1992 Developmental exposure to estrogens induces persistent changes in skeletal tissue. Endocrinol. 130: 1756. . Mills, R.A., Millis, C.D., Dannan, G.A., Guengerich, F.P. and Aust, S.D. 1985. Studies on the structure-activity relationships for the metabolism of polybrominated biphenyls by rat liver microsomes. Toxicol. Appl. Pharmacol. 78:96. Moguilewsky M., and Deraedt R. 1981. Interrelations between glucocorticoid and progestin receptors. J. Steroid. Biochem. 15:329. Moller, O.M. 1973. The progesterone concentrations in the peripheral plasma of the mink (Mustela vison) during pregnancy. J. Endocrinol. 56:121. Moller 0. M. 1974. Plasma progesterone before and after ovariectomy in unmated and pregnant mink, Mustela vison. J. Reprod. Fertil. 37:367. Moore, R.W., Jefcoate, C.R. and Peterson, R.E. 1991. 2,3,7,8-Tetrachlorodibenzo-p­ dioxin inhibits steroidogenesis in the rat testis by inhibiting the mobilization of cholesterol to cytochrome P450scc. Toxicol. Appl. Pharmacol. 109:85. Morrissey R. E., Harris M. W., Diliberto J. J., and Birnbaum L. S. 1992. Limited PCB antagonism of TCDD-induced malformations in mice. Toxicol. Lett. 60:19. 84 Murphy, B.D., Concannon, P.W., Travis, H.F. and Hansel, W. 1981. Prolactin: the hypophyseal factor that terminates embryonic diapause in mink. Biol. Reprod. 25:487. Murphy, B.D. and James, D.A. 1974. Mucopolysaccharide histochemistry of the mink uterus during gestation. Can. J. Zool. 52:687. Narasimhan, T.R., Kim, H.L. and Safe, S. 1991. Effects of hydroxylated polychlorinated biphenyls on mouse liver mitochondrial oxidative phosphorylation. J. Biochem. Toxicol. 6:229. Neale M. G., and Parke D. V. 1973. Effects of pregnancy on the metabolism of drugs in the rat and rabbit. Biochem. Pharmacol. 22:1451. Nelson, K., Sunahara, G., Ryan, R. and Lucier, G. 1989. Decrease hepatic cytosolic glucocorticoid receptor binding upon treatment of rodents (mice and rats) with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Chemosphere 18:737. Nordlund-Moller, L., Andersson, 0., Ahlgren, R., Schilling, J., Gillner, M., Gustafsson, J.A. and Lund, J. 1990. Cloning, structure, and expression of a rat binding protein for polychlorinated biphenyls. Homology to the hormonally regulated progesterone-binding protein uteroglobin. J. Biol. Chem. 265:12690. Olson J. R., and McGarrigle B. P. 1992. Comparative developmental toxicity of 2,3,7,8­ tetrachlorodibenzo-p-dioxin (TCDD). Chemosphere 25:71. Olson J. R., McGarrigle B. P., Tonucci D. A., Schecter A., and Eichelberger H. 1990. Developmental toxicity of 2,3,7,8-TCDD in the rat and hamster. Chemosphere 20:7. Omura T., and Sato R. 1964. The carbon monoxide-binding pigment of liver microsomes. J. Biol. Chem. 239:2370. Papke, R.L., Concannon, P.W., Travis, H.F. and Hansel, W. 1980. Control of luteal function and implantation in the mink by prolactin. J. Animal Sci. 50:1 102. Parkinson, A., Safe, S., Robertson, L., Thomas, P.E., Ryan, D.E. and Levin, W. 1983. Immunochemical quantitation of cytochrome P450 isozymes and epoxide hydrolase in liver microsomes from polychlorinated and polybrominated biphenyls: a study of structure-activity relationships. J. Biol. Chem. 258:5967. Patnode K. A., and Curtis L. R. 1994. 2,2',4,4',5,5'- and 3,3',4,4',5,5'-hexachlorobiphenyl alteration of uterine progesterone and estrogen receptors coincides with embryotoxicity in mink (Mustela vison). Toxicol. Appl. Pharmacol. 127: 9. Peterson, R.E., Theobald, H.M. and Kimmel, G.L. 1993. Developmental and reproductive toxicity of dioxins and related compounds: cross-species comparison. Crit. Rev. Toxicol. 23:283. Pilbeam, T.E., Concannon, E.P. and Travis, H.F. 1979. The annual reproductive cycle of mink (Mustela vison). J. Animal Sci. 48:578. 85 Platonow, N.S. and Karstad, L.H. 1973. Dietary effects of polychlorinated biphenyls on mink. Can. J. Comp. Med. 37:391. Prince R. L. 1994 Counterpoint: estrogen effects on calcitropic hormones and calcium homeostasis. Endocrin. Rev. 15: 301. . Probst B., and Hermkes A. 1991. Quantitative characterization of hormone receptors by a nonlinear regression approach. J. Steroid. Biochem. Mol. Biol. 39:299. Prough R. A., Burke M. D., and Mayer R. T. 1978. Direct fluorimetric methods for measuring mixed-function oxidase activity. Methods Enzymol. 52:372. Rickenbacher, U., McKinney, J.D., Oatley, S.J. and Blake, C.C. 1986. Structurally specific binding of halogenated biphenyls to thyroxine transport protein. J. Med. Chem. 29:641. Rider V., and Heap R. B. 1986. Heterologous anti-progesterone monoclonal antibody arrests early embryonic development and implantation in the ferret (Mustela putorius). J. Reprod. Fertil. 76:459. Romkes M., Piskorska-Pliszczynska J., and Safe S. 1987 Effects of 2,3,7,8­ tetrachlorodibenzo-p-dioxin on hepatic and uterine estrogen receptor levels in rats. . Toxicol. Appl. Pharmacol. 87, 306. Romkes M., and Safe S. 1988. Comparative activities of 2,3,7,8-tetrachlorodibenzo-p­ dioxin and progesterone as antiestrogens in the female rat uterus. Toxicol. Appl. Pharmacol. 92:368. Rose J., Huang J. L., and Mead R. A. 1993. Role of ovarian steroids in development of uterine binding sites for prolactin in the ferret. Biol. Reprod. 48:1266. Rose J., Oldfield J. E., and Stormshak F. 1986. Changes in serum prolactin concentrations and ovarian prolactin receptors during embryonic diapause in mink. Biol. Reprod. 34:101. Rose J., and Stormshak F. 1993. 26th Annual Meeting of Society for the Study of Reproduction, Colorado State University, Fort Collins. p. 98.. Rose J., Stormshak F., Adair J., and Oldfield J. E. 1983. Prolactin binding sites in the uterus of the mink. Mol. Cell. Endocrinol. 31:131. Russell D. H., Buckley A. R., Shah G. N., Sipes I. G., Blask D. E., and Benson B. 1988. Hypothalamic site of action of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Toxicol. Appl. Pharmacol. 94:496. Safe S. 1990. Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and related compounds: environmental and mechanistic considerations which support the development of toxic equivalency factors (TEFs). Crit. Rev. Toxicol. 21:51. 86 Safe S. H. 1994 Polychlorinated biphenyls (PCBs): environmental impact, biochemical and toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24: . 87. Schlafke, S., Enders, A.C. and Given, R.L. 1981. Cytology of the endometrium of delayed and early implantation with special reference to mice and mustelids. J. Reprod. Fert. Suppl. 29:135. Schmoldt, A., Herzberg, W. and Benthe, H.F. 1977. On inhibition of microsomal drug metabolism by polychlorinated biphenyls (PCBs) and related phenolic compounds. Chem. Biol. Interact. 16:191. Schnellmann, R.G., Vickers, E.M. and Sipes, I.G. (1985). Metabolism and disposition of polychlorinated biphenyls. In Reviews in Biochemical Toxicology (E. Hodgson, J.R. Bend and R.M. Philpot, Eds.), p. 247. Elsevier, Amsterdam. Seigel S. 1956.. In Non-parametric Statistics for the Behavioral Sciences (S. Seigel, Ed.), pp. 96-104. McGraw-Hill, New York. Shain, W., Bush, B. and Seegal, R. 1991. Neurotoxicity of polychlorinated biphenyls: structure-activity relationship of individual congeners. Toxicol. Appl. Pharmacol. 111:33. Shiverick K. T., and Muther T. F. 1983. 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) effects on hepatic microsomal steroid metabolism and serum estradiol of pregnant rats. Biochem. Pharmacol. 32:991. Shull L. R., Bleavins M. R., Olson B. A., and Aulerich R. J. 1982. Polychlorinated biphenyls (Aroclors 1016 and 1242): effect on hepatic microsomal mixed function oxidases in mink and ferrets. Arch. Environ. Contam. Toxicol. 11:313. Shull L. R., Rush G. F., Olson B. A., Sleight S. D., Aulerich R. J., and Wisniewski J. A. 1983. Biological and induction effects of phenobarbital and 3-methylcholanthrene in mink (Mustela vison). Drug. Metab. Dispos. 11:441. Sinclair, P.R., Bement, W.J., Lambrecht, KW., Gorman, N. and Sinclair, J.F. 1990. Chlorinated biphenyls induce cytochrome P450IA2 and uroporphyrin accumulation in cultures of mouse hepatocytes. Arch. Biochem. Biophys. 281:225. Sinjari T., Tornwall U., and Darnerud P. 0. 1993. Induction of 7-ethoxyresorufin-O­ deethylase (EROD) activity in mice foetuses by the PCB-congener 3,3',4,4'­ tetrachlorobiphenyl. Xenobiotica. 23:107. Slayden, O.D. 1990. Endocrine regulation of uterine physiology in mink. Ph.D., Oregon State University, Corvallis. 180. Slayden 0. D., Hirst J. J., and Brenner R. M. 1993. Estrogen action in the reproductive tract of rhesus monkeys during antiprogestin treatment. Endocrinology. 132:1845. Slayden 0., and Stormshak F. 1992. Uterine metabolic activity and steroid receptor concentrations in response to suppressed secretion of PRL in anestrous mink. Gen. Comp. Endocrinol. 88:307. 87 Soderkvist P., Poellinger L., and Gustafsson J. A. 1986. Carcinogen-binding proteins in the rat ventral prostate: specific and nonspecific high affinity binding sites for benzo(a)pyrene, 3-methylcholanthrene, and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Cancer Res. 46:651. Soontornchat, S., Li, M.H., Cooke, P.S., and Hansen, L.G. 1994. Toxicokinetic and toxicodynamic influences on endocrine disruption by polychlorinated biphenyls. Environ, Health Persp. 102::568. Spindler-Vomachka M., and Vodicnik M. J. 1984. Distribution of 2,4,5,2',4',5'­ hexachlorobiphenyl among lipoproteins during pregnancy and lactation in the rat. J. Pharmacol. Exp. Ther. 230:263. Stoufflet I., Mondain-Monval M., Simon P., and Martinet L. 1989. Patterns of plasma progesterone, androgen and oestrogen concentrations and in-vitro ovarian steroidogenesis during embryonic diapause and implantation in the mink (Mustela vison). J. Reprod. Fertil. 87: 209. Sundqvist C., Amador A. G., and Bartke A. 1989. Reproduction and fertility in the mink (Mustela vision). J. Reprod. Fertil. 85:413. Sundqvist C., Ellis L. C., and Bartke A. 1988. Reproductive endocrinology of the mink (Mustela vison). Endocrin. Rev. 9:247. Symons A. M., Turcan R. G., and Parke D. V. 1982. Hepatic microsomal drug metabolism in the pregnant rat. Xenobiotica 12:365. Szekeres-Bartho J., Chaouat G., and Kinsky R. 1990. A progesterone-induced blocking factor corrects high resorption rates in mice treated with antiprogesterone. Am. J. Obstet. Gynecol. 163:1320. Tanabe S. 1988. PCB problems in the future: foresight from current knowledge. Environ. Pollut. 50:5. Tate, A.C. and Jordan, V.C. 1983. The estrogen receptor. In Principles of Recepterology (M.K. Agarwal, Ed.), pp. 381-463. Walter de Gruyter & Co., Berlin. Taylor P. 1986 Practical Teratology. Academic Press, New York. . Travis, H.F. and Schaible, P.J. 1961. Fundamentals of mink ranching. Circ. Bull. 229, Coop. Ext. Serv., Michigan State Univ., E. Lansing, ML Truss, M. and Beato, M. 1993. Steroid hormone receptors: interaction with deoxyribonucleic acid and transcription factors. Endocrin. Rev. 14:459. Turner R. T., Riggs B. L., and Spelsberg T. C. 1994. Skeletal effects of estrogen. Endocrin. Rev. 15: 275. Viscardi, R.M. and Max, S.R. 1993. Unsaturated fatty acid modulation of glucocorticoid receptor binding in L2 cells. Steroids. 58:357. 88 White, T.E. and Gasiewicz, T.A. 1993. The human estrogen receptor structural gene contains a DNA sequence that binds activated mouse and human Ah receptors: a possible mechanism of estrogen receptor regulation by 2,3,7,8-tetrachlorodibenzo­ p-dioxin. Biochem. Biophys. Res. Commun. 193:956. Whitlock, J.P. 1993. Mechanistic aspects of dioxin action. Chem. Res. Toxicol. 6:754. Wren, C.D., Hunter, D.B., Leatherland, J.F. and Stokes, P.M. 1987a. The effects of polychlorinated biphenyls and methylmercury, singly and in combination, on mink. I: uptake and toxic responses. Arch. Environ. Contam. Toxicol. 16:441. Wren, C.D., Hunter, D.B., Leatherland, J.F. and Stokes, P.M. 1987b. The effects of polychlorinated biphenyls and methylmercury, singly and in combination on mink. II: Reproduction and kit development. Arch. Environ. Contam. Toxicol. 16:449. 89 APPENDICES 90 APPENDIX A Embryo Resorption and Growth Impairment in Mouse Strains with Normal and Reduced Aryl Hydrocarbon Receptor Concentrations K.A. Patnode and L.R. Curtis Submitted to Fundamental and Applied Toxicology (1995) 91 Summary Both coplanar and noncoplanar polychlorinated biphenyl (PCB) congeners induce embryo resorption in mink (Patnode and Curtis 1994). This observation suggests a common mechanism of action despite known differences in congener binding affinity to the aryl hydrocarbon receptor (AhR) (Kafafi et al. 1993). To distinguish between AhR­ dependent and AhR-independent mechanisms, we exposed C57 Black (AhR+) mice to 80 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg (n=10) or corn oil (CO; n=10) via ip injection at the time of implantation (gd 5). We also treated B6N:D2N (Ahd) N13F13:B6JN28F28F3 (AhR) mice with 80 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg (n=10), 300 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB) (n=10), or corn oil (CO; n=10) using the same protocol. After 8 days (gd 13), we asphyxiated mice with CO2, removed uteri, and dissected implantation sites. Resorption was documented macroscopically and surviving embryos were weighed. Fisher exact probability test was used to analyze the numbers of resorptions by treatment. Embryo weights were compared across treatments and strains using analysis of variance. The percent of litter resorbed was similar in CO-treated mice of both strains indicating no strain-specific differences in spontaneous resorption rates (Fig Al.). Treatment of AhR+ mice with 3HCB did not increase the proportion of resorptions. In contrast, AhR mice exposed to either 3HCB or 2HCB had increased resorptions compared to CO-treated animals. This finding suggests that AhR does not mediate embryotoxicity and may, in fact, be protective of embryos. Binding to AhR in non-target tissues may reduce the amount of PCB reaching the target tissue. 50 AhR+ AhR­ 40 b 30 20 10 ii CO 3HCB CO 3HCB 2HCB Figure A-1. Percent of litter resorbed in female mice with normal (AhR+) and reduced (AhR-) aryl hydrocarbon receptors exposed to 80 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg, 300 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB)/kg, or corn oil (CO) via ip injection on gestation day 5. Dissimilar letters denote significant differences (p<0.05). 93 Embryo weights were similar for CO-treated mice of both strains indicating similar rates of development (Fig. A2.). 3HCB treatment of AhR+ mice significantly reduced embryo weights. In AhR mice, 3HCB-exposed embryo weights were between those of CO-treated AhR and 3HCB-treated AhR+ and did not differ significantly from either treatment-strain group. The order of treatments from CO to 3HCB-treated AhR to 3HCB-treated AhR+ data suggest that embryo growth impairment may be a AhR­ mediated effect. The lack of impairment in 2HCB exposed mice may reflect an insufficient dose in a strain deficient in AhR. The combination of AhR-mediated growth impairment and AhR-independent embryo resorption are only compatible if the target tissues differ for these two reproductive effects. The embryo resorption process may be triggered within the hypothalamus. Jones et al. (1987) observed decreased serum prolactin concentrations and attributed these to changes in brain dopamine levels. Aulerich et al. (1985) documented decreased dopamine concentrations in mink that experienced PCB-induced embryo resorption. In contrast, growth impairment may be due to a direct action of PCBs on the uterus or embryos. PCBs and their metabolites concentrate in the uterus and embryos of pregnant mice (Darnerud et al. 1986, Lucier et al. 1978). Decreases in body weight in children exposed to PCBs in utero are associated with decreased capacity of epidermal growth factor to stimulate autophosphorylation of placental growth factor receptor (Sunahara et al. 1987). Glucocorticoid receptor binding capacity in fetal livers is also 300 AhR+ AhR­ 250 a 20 200 ab T 47 150 ab_ 100 be T 33 T 24 39 50 0 CO 3HCB CO 3HCB 2HCB Figure A-2. Embryo weights from female mice with normal (AhR+) and reduced (AhR-) aryl hydrocarbon receptors exposed to 80 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB)/kg, 300 mg 2,2',4,4',5,5'-hexachlorobiphenyl (2HCB)/kg, or corn oil (CO) via ip injection on gestation day 5. Dissimilar letters denote significant differences (p<0.05). 95 decreased by dioxin-like contaminants (Ryan et al. 1989) which may reduce nutrients available for embryo growth. These results must be interpreted with caution as timed mating of these mouse strains was problematic. We observed numerous matings which failed to produce a vaginal plug. In addition, these mice had a low conception rate. The staff veterinarian at a laboratory animal supplier indicated that C57 Black males experience poor fertility and in commercial operations only proven breeders are used for timed mating studies. These complications reduced our sample sizes to 6 and 5 for CO- and 3HCB-treated animals of each strain and to 4 for 2HCB-treated AhR mice. Due to the difficulties in accurately documenting gestation day 1 using vaginal plugs, it is possible that not all of these embryos are gestation day 13. We recommend repeating this experiment to increase the sample size of females and embryos accurately aged with the vaginal plug technique. 96 Literature Cited Aulerich, R.J., Olson, B.A., Ringer, R.K., Bursian, S.J. and Breslin, W.J. 1985. Toxicological manifestations of 2,4,5,2',2',4',5'-hexachlorobiphenyl, 2,3,6,2',3',6'­ hexachlorobiphenyl, and 3,4,5,31,4',5'-hexachlorobiphenyl and Aroclor 1254 in mink. J. Toxicol. Environ. Health 15:63. Darnerud, P.O., Brandt, I., Klasson-Wehler, E., Bergman, A., D'Argy, R., Dencker, L. and Sperber, G.O. 1986. 3,314,4'-tetrachloro[14C]biphenyl in pregnant mice: enrichment of phenol and methyl sulphone metabolites in late gestational fetuses. Xenobiotica 16:295. Jones, M.K., Weisenburger, W.P., Sipes, I.G. and Russell, D.H. 1987. Circadian alterations in prolactin, corticosterone, and thyroid hormone levels and downregulation of prolactin receptor activity by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 87:337. Kafafi, S.A., Afeefy, H.Y., Ali, A.H., Said, H.K. and Kafafi, A.G. 1993. Binding of polychlorinated biphenyls to the aryl hydrocarbon receptor. Environ. Health. Perspect. 101:422. Lucier, G.W., McDaniel, 0.S., Schiller, C.M. and Matthews, H.B. 1978. Structural requirements for the accumulation of chlorinated biphenyl metabolites in the fetal rat intestine. Drug. Metab. Dispos. 6:584. Patnode, K.A. and Curtis, L.R. 1994. 2,2',4,4',5,5'- and 3,3',4,4',5,5'-hexachlorobiphenyl alteration of uterine progesterone and estrogen receptors coincides with embryotoxicity in mink (Mustela vison). Toxicol. Appl. Pharmacol. 127:9. Ryan, R.P., Sunahara, G.I., Lucier, G.W., Birnbaum, L.S. and Nelson, K.G. 1989. Decreased ligand binding to the hepatic glucocorticoid and epidermal growth factor receptors after 2,3,4,7,8-pentachlorodibenzofuran and 1,2,3,4,7,8­ hexachlorodibenzofuran treatment of pregnant mice. Toxicol. Appl. Pharmacol. 98:454. Sunahara, G.I., Nelson, K.G., Wong, T.K. and Lucier, G.W. 1987. Decreased human birth weights after in utero exposure to PCBs and PCDFs are associated with decreased placental EGF-stimulated receptor autophosphorylation capacity. Mol. Pharmacol. 32:572. 97 APPENDIX B Nonesterified Fatty Acid Concentrations in Plasma and Uterine Tissue of Pregnant and Nonpregnant Female Mink Following Exposure to 3,3',4,4',5,5'-Hexachlorobiphenyl K.A. Patnode and L.R. Curtis Submitted to Lipids (1995) 98 Summary Both coplanar and noncoplanar polychlorinated biphenyl (PCB) congeners induce embryo resorption in mink (Patnode and Curtis 1994). In addition to potential direct actions of PCBs on hormones and their receptors, PCBs may act indirectly via their influence on other endogenous substances. Aroclor 1254 increases heptic membrane concentration of arachidonic acid (AA) (Borlakoglu et al. 1990) via induction of microsomal fatty acid desaturases (Borlakoglu et al. 1991) in rats and pigeons. Agents which block AA release from the membrane or subsequent steps in the cascade leading to prostaglandin synthesis reduce the toxicity of 2,3,7,8-tetrachloro-dibenzo-p-dioxin (TCDD) in rat livers (Al-Bayati and Stohs 1991) and chick hearts (Rifkind et al. 1984). AA and other nonesterified fatty acids (NEFAs) are modulators of numerous physiological processes and their liver concentrations are reduced during gestation (Feuer 1979). They inhibit progesterone receptor (PR), glucocorticoid receptor (GR), estrogen receptor (ER) binding by direct binding at .a site distinct from the hormone and heat shock protein binding sites (Kato et al. 1987, Kato 1989, Vallette et al. 1991). Low concentrations of saturated NEFAs also suppress lymphocyte proliferation (Buttke 1984). Increasing desaturation of NEFAs reverses the inhibition of DNA synthesis and leads to increased B and T lymphocyte populations. PCB exposure causes significant weight loss and mobilization of fat stores in mink (Gillette et al. 1987). In anestrous mink, adipose lipoprotein lipase activity increases and circulating triglyceride levels decrease (Brewster and Matsumura 1989). These PCBinduced changes in lipid metabolism could produce changes in NEFAs that may be the 99 proximate causes of embryo resorption in mink. Our objective was to determine if circulating NEFAs or uterine tissue NEFAs in pregnant mink were altered by exposure to 3,3 ',4,4' ,5,5 ' -hexachlorobiphenyl. On March 30, 1994, we injected 16 mated mink with 0.4 mg/kg 3HCB or corn oil ip. After 10 d, a 10 ml blood sample was collected via cardiac puncture and sera obtained by centrifugation. The uterus was removed and homogenized. Lipids were extracted from 5 ml serum and whole uterine homogenate after Bligh and Dyer (1959). Extracts were desiccated under nitrogen, weighed, dissolved in benzene, and stored at -20°C. Methyl esters were prepared by methanolysis (Selivonchick et al. 1977). NEFA composition was determined with a Varian aerograph 1200 series gas chromatograph. A 183-cm stainless steel column packed with GP 10% SP-2330 on 100/120 mesh Chromosorb W AW was used. Column temperatures were 160°C for the first 8 min., increased to 190°C in steps of 4°C per min., and then held at 190°C for 20 min. The flow rate of nitrogen was 30 ml/min. Injection port and detector temperatures were 250°C. Fatty acid esters were identified by comparison to standard mixtures of commercially available known methyl mixtures and quantified by a Hewlett-Packard 3380 integrator. Analysis of variance was used to detect differences in the relative percentages of NEFAs. Three mated females in each treatment group were not pregnant and were analyzed as separate treatment groups. Three plasma and six uterine samples were damaged and were not suitable for analysis. NEFA profiles for plasma and uterine tissue are presented in Tables B1 and B2, respectively. We did not detect significant differences between pregnant and nonpregnant 100 control animals. 3HCB treatment did not significantly alter plasma or uterine tissue NEFA concentrations in either pregnant or nonpregnant mink. Caution should be used in interpreting these results because small sample sizes have severely reduced the power of our statistical analysis. In light of the known changes in lipid metabolism in mink, this experiment should be repeated before concluding that NEFAs are not a causative factor in 3HCB-induced embryo resorption in mink. 101 Table B-1. Nonesterified fatty acid (NEFA) concentrations in plasma of pregnant and nonpregnant female mink treated with 0.4 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) /kg or corn oil (CO) via ip injection at the putative time of implantation. NEFA 14:0 16:0 16:10)7 18:0 18:10)9 18:1co7 18:2(1)6 18:3co6 18:3co3 20:3(06 20:4(06 20:5(o3 22:40)6 22:5co3 22:6(0 Non-Pregnant CO (n=2) 3HCB (n=3) 0.561±0.124 1.002±0.250 22.958±3.360 23.302±0.809 1.407±0.124 1.760±0.239 14.846±1.692 15.378±0.562 11.108±0.773 12.619±0.905 1.695±1.198 2.506±1.037 22.803±0.310 22.373±0.839 0.352±0.056 0.113±0.092 0.126±0.089 0.877±0.065 0.454±0.191 14.378±2.056 13.535±0.522 4.614±0.980 2.723±0.868 0.135±0.096 0.896±0.185 0.223±0.182 3.244±1.149 4.011±1.647 Pregnant CO (n=3) 3HCB (n=5) 0.705±0.107 1.062±0.382 20.642±0.363 22.514±2.669 1.581±0.127 1.946±0.612 12.684±0.580 15.947±1.892 12.649±0.619 11.936±1.184 3.569±0.111 3.063±0.747 23.710±0.702 17.689±1.992 0.436±0.052 2.270±1.767 0.313±0.050 0.155±0.138 0.902-1-0.017 1.274±0.344 13.744±0.821 12.081±2.296 4.764±0.306 4.228±0.851 0.230±0.096 0.435±0.312 0.808±0.046 0.794±0.178 3.264±0.004 4.606±1.000 102 Table B-2. Nonesterified fatty acid (NEFA) concentrations in uterine tissue of pregnant and nonpregnant female mink treated with 0.4 mg 3,3',4,4',5,5'-hexachlorobiphenyl (3HCB) /kg or corn oil (CO) via ip injection at the putative time of implantation. NEFA 14:0 16:0 16:10)7 18:0 18:1(1)9 18:16)7 18:2(1)6 18:30)6 18:30)3 20:36)6 20:46)6 20:5(03 22:46)6 22:50 22:6(1)3 Non-Pregnant CO (n=2) 3HCB (n=3) 3.069±0.014 3.749±0.463 18.531 ±0.166 19.097±1.535 5.966±0.013 7.724±0.687 7.812±0.089 7.382±1.294 30.611±0.951 33 236±0. 903 2.737±1.935 2.272±1.855 16.564±0.240 15.611±0.438 0.605±0.022 0.161±0.079 2.770±1.453 0.726±0.020 0.564±0.017 0.426±0.045 5.585±0.117 4.819±0.703 0.926±0.044 0.797±0.079 0.861±0.091 0.493±0.218 0.926±0.006 0.858±0.041 2.473±.0.187 2.651±0.227 . Pregnant CO (n=4) 3HCB (n=1) 2.563±0.211 1.650±0.000 19.944±0.576 19.322±0.000 5.976±0.701 3.842±0.000 8.312±0.489 13.945±0.000 30.772±1.745 20.748±0.000 3.589±1.185 8.870±0.000 15.884±0.451 12.517±0.000 0.247±0.075 0.049±.0.000 0.605±0.062 0.225±0.000 0.607±0.063 0.781±0.000 6.176±0.732 11.393±0.000 1.192±0.095 0.967±0.000 0.927±0.093 1.338±0.000 0.866±0.055 1.099±0.000 2.339±0.096 3.254±0.000 103 Literature Cited Al-Bayati, Z.A. and Stohs, S.J. 1991. The possible role of phospholipase A2 in hepatic microsomal lipid peroxidation induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Arch. Environ. Contam. Toxicol. 20:361. Bligh, E.G. and Dyer, W.J. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911. Borlakoglu, J.T., Edwards-Webb, J.D. and Dils, R.R. 1990. Polychlorinated biphenyls increase fatty acid desaturation in the proliferating endoplasmic reticulum of pigeon and rat livers. Eur. J. Biochem. 188:327. Borlakoglu, J.T., Edwards-Webb, J.D. and Dils, R.R. 1991. Evidence for the induction of fatty acid desaturation in proliferating hepatic endoplasmic reticulum in response to treatment with polychlorinated biphenyls. Are fatty acid desaturases cytochrome P-450-dependent monooxygenases? Int. J. Biochem. 23:925. Brewster, D.W. and Matsumura, F. 1989. Differential effect of 2,3,7,8­ tetrachlorodibenzo-p-dioxin on adipose tissue lipoprotein lipase activity in the guinea pig, rat, hamster, rabbit, and mink. Comp. Biochem. Physiol. C. 93:49. Buttke, T.M. 1984. Inhibition of lymphocyte proliferation by free fatty acids. Immunol. Today 53:235. Feuer, G. 1979. Action of pregnancy and various progesterones on hepatic microsomal activities. Drug. Metab. Rev. 9:147. Gillette, D.M., Corey, R.D., Helferich, W.G., McFarland, J.M., Lowenstine, L.J., Moody, D.E., Hammock, B.D. and Shull, L.R. 1987. Comparative toxicology of tetrachlorobiphenyls in mink and rats. I. Changes in hepatic enzyme activity and smooth endoplasmic reticulum volume. Fundam. Appl. Toxicol. 8:5. Kato, J. 1989. Arachidonic acid as a possible modulator of estrogen, progestin, androgen, and glucocorticoid receptors in the central and peripheral tissues. J. Steroid. Biochem. 34:219. Kato, J., Takano, A., Mitsuhashi, N., Koike, N., Yoshida, K. and Hirata, S. 1987. Modulation of brain progestin and glucocorticoid receptors by unsaturated fatty acid and phospholipid. J. Steroid. Biochem. 27:641. Patnode, K.A. and Curtis, L.R. 1994. 2,2',4,4',5,5'- and 3,3',4,4',5,5'-hexachlorobiphenyl alteration of uterine progesterone and estrogen receptors coincides with embryotoxicity in mink (Mustela vison). Toxicol. Appl. Pharmacol. 127:9. Rifkind, A.B., Hattori, Y., Levi, R., Hughes, M.J., Quilley, C. and Alonso, D.R. (1984). The Chick embryo as a model for PCB and dioxin toxicity: evidence of cardiotoxicity and increased prostaglandin synthesis. In Biological Mechanisms of Dioxin Toxicity (A. Poland and R. Kimbrough, Eds.), pp. 255-266. Cold Spring Harbor Press, Cold Spring Harbor. 104 Selivonchick, D.P., Johnston, P.V. and Roots, B.I. 1977. Acy and alkenyl group composition of brain subcellular fractions of goldfish (Carassius auratus L.) acclimated to different environmental temperatures. Neurochem. Res. 2:379. Vallette, G., Vanet, A., Sumida, C. and Nunez, E.A. 1991. Modulatory effects of unsaturated fatty acids on the binding of glucocorticoids to rat liver glucocorticoid receptors. Endocrinol. 129:1363.