1999 Progress Report: Species-Specific Endocrine Disruption: PCB- and PAH-Induced Estrogenic Effects

EPA Grant Number: R826301
Title: Species-Specific Endocrine Disruption: PCB- and PAH-Induced Estrogenic Effects
Investigators: Zacharewski, Timothy
Institution: Michigan State University
EPA Project Officer: Carleton, James N
Project Period: January 1, 1998 through December 31, 2000
Project Period Covered by this Report: January 1, 1998 through December 31, 1999
Project Amount: $282,998
RFA: Endocrine Disruptors (1997) RFA Text |  Recipients Lists
Research Category: Economics and Decision Sciences , Endocrine Disruptors , Health , Safer Chemicals


This research aims to examine the alleged estrogen receptor-mediated activities of selected environmentally relevant compounds using a combination of in vitro and in vivo assays. These studies are being performed in a number of species, including fish, frogs, mice, and birds, to test the following general hypothesis: rodents are not appropriate surrogates for identifying and assessing the risks of alleged environmental estrogens to human and wildlife health due to a lack of significant amino acid sequence homology between species in the estrogen receptor (ER) ligand binding domains.

Progress Summary:

Species Comparison of ER Binding. Bacterially expressed fusion proteins were used to investigate potential differences in ligand preference and binding affinities of ERs from five different species using a semi-high throughput competitive binding assay. Fusion proteins consisted of glutathione-S-transferase (GST) linked to the ER ligand binding domain (LBD) of human (hER- ), mouse (mER- ), chicken (cER), green anole lizard (aER), and rainbow trout (rtER). Initially, the ability of 44 polychlorinated biphenyls (PCBs), 9 hydroxylated PCBs (HO-PCBs), and 8 Aroclor PCB mixtures to compete with [3H]17 -estradiol (E2) for binding to the hER- , aER, and rtER fusion proteins was examined. Of the 44 PCBs tested, only PCBs 104, 184, and 188 effectively competed with [3H]E2 for binding to the rtER protein, with binding affinities several orders of magnitude lower than that of E2. In contrast, these same congeners only caused a 30 percent displacement of [3H]E2 from the hER- and aER proteins. Several additional PCB congeners were found to bind to the rtER fusion protein, although the degree of interaction varied among congeners. Differences in binding were further explored by examining the ability of 35 natural and synthetic chemicals to compete with [3H]E2 for binding to each of the 5 GST-ER fusion proteins. Although ERs exhibited similar binding preferences and binding affinities for many of the chemicals examined, several differences in absolute and relative binding affinities were observed. For example, -zearalenol ( -ZEA) consistently bound with greater affinity to the rtER than E2, in contrast to results obtained with ERs from the other species. Similarly, coumestrol, genistein, and naringenin bound with approximately 10-fold higher affinity to aER than to the other ER fusion proteins. These results demonstrate that ERs from different species exhibit differential ligand preferences and relative binding affinities for estrogenic substances and PCBs, which may be due to the species-specific variations in amino acid sequence within the ER LBD. Rainbow Trout Studies. We have investigated the ability of genistein, 4-tert-octyl phenol (4t-OP), diethylstilbestrol (DES), and -ZEA to induce vitellogenin and eggshell protein levels in rainbow trout (Oncorhynchus mykiss). Both processes are physiologically controlled by estrogens, but also can be induced by other estrogenic compounds such as certain industrial compounds, phytoestrogens, and naturally occurring compounds. First, zona radiata proteins from rainbow trout were cloned (GenBank accession #AF185274). In addition, a method for quantifying eggshell protein and vitellogenin polymerase chain reaction (PCR) products using fluorescent probes (TaqMan assay) was developed. In three different experiments, rainbow trout were treated with genistein, 4t-OP (10, 50, and 75 mg/kg body weight), DES (0.1, 1.0, and 10 mg/kg), or -ZEA (0.1, 1.0, and 10 mg/kg). After 10 days of exposure, the fish were sacrificed and liver samples were analyzed for possible induction of vitellogenin and eggshell proteins using reverse transcription-PCR (RT-PCR) in combination with the TaqMan assay. Also, relative amounts of vitellogenin and eggshell proteins in blood plasma were analyzed by ELISA. Results from the TaqMan assay showed a dose-dependent induction of both genes after treatment with all compounds. ELISA results confirmed the dose-dependent induction of both proteins in blood plasma, except in the -ZEA experiment, in which no vitellogenin was detected in the two lowest dose groups. Frog Studies. In a competitive binding assay using African clawed toad (Xenopus laevis) ER (xERdef), all of the pesticides that were tested as well as butyl benzyl phthalate and dibutyl benzyl phthalate were found not to bind significantly to the xER. By contrast, Bisphenol A and 4t-OP bound weakly, and hydroxyphenyltrichloroethane (HPTE) and the pharmaceutical estrogens displayed a high affinity similar to that of E2. The ability of E2 to induce Xenopus ER-regulated gene expression was assessed in cultured MCF-7 human breast cancer cells transiently transfected with an xER-regulated luciferase reporter gene. Treatment of the cells with E2 was found to cause a 30- to 50-fold increase in xER-mediated reporter gene expression, and other chemicals are being evaluated in this system. In the in vivo studies, vitellogenesis in male frogs was used as a marker to assess potential estrogenic effects of genistein, 4t-octylphenol, and -ZEA. Adult male frogs were injected intraperitoneally for 3 consecutive days with genistein, -ZEA, or 4t-OP in corn oil. E2 at 1 mg/kg body weight was used as positive control. Animals were sacrificed on day 12, when vitellogenin mRNA and protein are known to be maximally induced by E2. Livers were removed, and hepatic vitellogenin mRNA levels were quantified using RT-PCR. Slight vitellogenin mRNA induction was observed at a cumulative dose of 0.15 mg/kg E2, while a maximal response was achieved at a cumulative dose of 3.0 mg/kg. In experiments with 4t-OP, a cumulative dose of 300 mg/kg was found to induce vitellogenin mRNA levels, while 30 and 150 mg/kg did not produce a detectable effect on vitellogenin mRNA, and 750 and 1,500 mg/kg were found to induce toxic effects. Mouse Studies. The ability of selected polycyclic aromatic hydrocarbons (PAHs) to bind to both the (LBD region) and the (full-length receptor) isoforms of the human ER was evaluated using a competitive binding assay. In addition, the ability of these compounds to induce human ER - or mouse ER -mediated reporter gene expression in transiently transfected MCF-7 cells was assessed. When it was found that certain PAHs induced reporter gene expression though none bound to either receptor isoform, the possibility that metabolites of these compounds were responsible for the ER-mediated action was investigated. Benzo[a]pyrene (B[a]P) was chosen as a model compound, and 10 B[a]P metabolites were tested. Certain monohydroxylated metabolites of B[a]P were found to bind and cause ER-mediated gene expression, showing a strong preference for the isoform of the receptor. When either these metabolites or the parent B[a]P compound was administered to immature ovariectomized mice, no increase in two common markers of estrogen action, uterine weight, and uterine lactoferrin mRNA expression was observed.

Future Activities:

Computational molecular modeling and ligand docking tools will be used to investigate the structural basis for these differences in binding preference, focusing on human and rainbow trout ERs. Three-dimensional models of the interactions between estrogenic chemicals and the human and rainbow trout ERE domains based on the crystal structure of the human ER E domain:E2 complex will be generated. Site-directed mutagenesis then will be performed on selected residues to verify whether these residues contribute to the observed differences in ligand preference and relative binding affinities between the two species. Vitellogenin mRNA induction experiments also will be performed to examine the possible estrogenic activity of genistein and -ZEA. In addition, manuscripts discussing the results of the rainbow trout and frog studies will be submitted for publication.

Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other project views: All 50 publications 11 publications in selected types All 10 journal articles
Type Citation Project Document Sources
Journal Article Matthews J, Zacharewski T. Differential binding affinities of PCBs, HO-PCBs, and aroclors with recombinant human, rainbow trout (Onchorhynkiss mykiss), and green anole (Anolis carolinensis) estrogen receptors, using a semi-high throughput competitive binding assay. Toxicological Sciences 2000;53(2):326-339. R826301 (1999)
R826301 (Final)
not available

Supplemental Keywords:

human health, metabolism, carcinogen, animal, mammalian, organism, cellular, diet, effluent, terrestrial, aquatic, estrogenic endocrine modulators., RFA, Health, Scientific Discipline, Toxics, Waste, Environmental Chemistry, Health Risk Assessment, Endocrine Disruptors - Environmental Exposure & Risk, HAPS, endocrine disruptors, chemical mixtures, Risk Assessments, Biochemistry, Children's Health, Endocrine Disruptors - Human Health, adverse outcomes, complex mixtures, endocrine disrupting chemicals, PCBs, fertility, industrial chemicals, PAH, animal models, carcinogens, human growth and development, cancer, human exposure, estrogen response, reproductive processes, estrogen receptors, biological effects

Relevant Websites:


Progress and Final Reports:

Original Abstract
  • 1998
  • Final Report