Toxicology 213 (2005) 177193

Young Investigators Award Lecture

Molecular responses to xenoestrogens: Mechanistic insights from toxicogenomics

Jonathan G. Moggs
Syngenta Central Toxicology Laboratory, Alderley Park, Cheshire SK10 4TJ, UK Available online 5 July 2005

Abstract The xenoestrogen group of endocrine disruptors has the potential to cause reproductive and developmental effects through stimulation or disruption of sex steroid nuclear receptor signalling pathways. A more detailed understanding of the ways in which xenoestrogens interact with biological systems at the molecular level will provide a mechanistic basis for improved safety assessment. The recent sequencing of mammalian genomes has driven the development of toxicogenomic technologies, including microarray based gene expression proling, which allow the expression levels of thousands of genes to be measured simultaneously. Since the cellular responses to xenoestrogens are predominantly mediated by estrogen receptors, which function as ligand-activated transcription factors to regulate gene expression, the application of toxicogenomics has great potential for providing insights into the molecular mechanisms of xenoestrogen action. A major challenge in applying toxicogenomics to the eld of endocrine disruption is the need to dene how xenoestrogen-induced changes in gene expression relate to conventional physiological and toxicological endpoints. Gene Ontology Mapping, Pathway Mapping and Phenotypic Anchoring of xenoestrogen-induced gene expression changes to cellular pathways and processes represent key steps in dening these relationships. Mechanistic insights into how xenoestrogens target specic genes and into the functional signicance of xenoestrogen-induced alterations in gene expression can be further enhanced by combining transcript proling with transgenic animal models or cell-based systems in which the estrogen receptor signalling pathways have been modied experimentally. This review illustrates how these toxicogenomic approaches are providing an unprecedented amount of mechanistic information on the molecular responses to xenoestrogens and how they are likely to impact on hazard and risk assessment. 2005 Published by Elsevier Ireland Ltd.
Keywords: Toxicogenomics; Endocrine disruption; Xenoestrogens; Gene expression

Contents
1. 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecular mechanisms of xenoestrogen signalling in mammals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 179

Tel.: +44 1625 519315; fax: +44 1625 590 996. E-mail address: jonathan.moggs@syngenta.com.

0300-483X/$ see front matter 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.tox.2005.05.020

178

J.G. Moggs / Toxicology 213 (2005) 177193

3.

4. 5.

6. 7.

Application of toxicogenomics to understanding mechanisms of estrogen action in vivo: using the rodent uterotrophic assay as a model experimental system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Holistic identication of E2-responsive genes in vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Gene ontology and pathway mapping of E2-responsive gene functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Phenotypic anchoring of gene expression changes during E2-induced uterine growth . . . . . . . . . . . . . . . . . . . . . Comparative transcript proling of physiological, synthetic and plant-derived estrogens . . . . . . . . . . . . . . . . . . . . . . . . . Understanding mechanisms of ER-mediated gene regulation: application of toxicogenomics to biological systems in which the ER signalling has been disrupted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Transgenic animals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Cell-based model systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Toxicogenomic analysis of xenoestrogen-induced carcinogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary and perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180 180 181 181 185 187 187 188 188 189 189 189

1. Introduction The sex steroid hormone estrogen regulates development, growth and differentiation in multiple tissues, including the reproductive tract, mammary gland, bone and cardiovascular system (Dechering et al., 2000; Nilsson and Gustafsson, 2002). Synthetic estrogens are also widely used in the treatment of several diseases, including breast cancer and osteoporosis (Riggs and Hartmann, 2003), and are components of contraceptive pills and post-menopausal hormone replacement therapies. In addition, there is concern over the potential health and environmental risks associated with exposure of humans and wildlife to xenoestrogens (Safe et al., 2002; Singleton and Khan, 2003). Understanding the molecular mechanisms of estrogen and xenoestrogen action will, therefore, be of considerable utility in many areas. The sequencing of mammalian genomes has driven the development of new -omic technologies that are capable of dening, in a holistic manner, the individual genes, proteins and biological pathways that mediate adverse responses to drugs and chemicals. Among these technologies is microarray based gene expression proling (transcript proling), which allows the expression levels of thousands of genes to be measured simultaneously. The application of transcript proling to toxicology has been termed toxicogenomics and can be used in either mechanistic or predictive modes of analysis. In the mechanistic mode, transcript proling is used to implicate specic genes or biological pathways in the mechanism of

action of a toxicant. Alternatively, this technology can be utilized in a predictive context, where the mode of action of a novel toxicant may be identied by comparing the expression pattern it elicits with established expression signatures of reference toxicants. The molecular and cellular responses to xenoestrogens are predominantly mediated by estrogen receptors, which function as ligand-activated transcription factors to regulate gene expression. Thus, the application of toxicogenomics has great potential for providing insights into the molecular mechanisms of xenoestrogen action and also for identifying gene expression signatures that are characteristic of chemicals with estrogenic activity (Naciff and Daston, 2004; Deavall et al., 2005). These genomic approaches promise to provide a more detailed understanding of the ways in which xenoestrogens interact with biological systems, including how they might induce adverse health effects. A major challenge in applying toxicogenomics to the eld of endocrine disruption is the need to dene how xenoestrogen-induced changes in gene expression relate to conventional physiologically and toxicologically relevant endpoints. Gene expression proling has the potential to reveal, holistically, the molecular pathways and cellular processes that mediate the adverse responses to xenoestrogens. However, the initial output of toxicogenomics experiments generally consists of very large lists of genes. In order to be able to interpret these large-scale and multivariate molecular expression data, novel bioinformatic methods must be applied to place gene expression

J.G. Moggs / Toxicology 213 (2005) 177193

179

changes in the context of underlying biological processes and responses, so that the molecular pathways affected can be identied. One rapidly emerging approach for translating complex gene expression proling data is the application of gene ontology (GO) and pathway mapping tools. The GO consortium (Ashburner et al., 2000; Harris et al., 2004) has developed a controlled vocabulary that describes the biological processes, cellular components and molecular functions associated with a particular gene product. Importantly, this controlled vocabulary represents a machine-readable code for the diverse descriptions used by biologists to annotate individual genes. Complex xenoestrogen-induced gene expression proles can thus be interrogated in silico using a database of GO terms to elucidate the predominant gene functions and biological pathways that are affected by xenoestrogen exposure in a given experimental system. Furthermore, the functional interplay between multiple xenoestrogen-induced genes can be visualized rapidly using biological pathway maps that are available from a variety of online bioinformatic resources (e.g., the Kyoto Encyclopedia of Genes and Genomes (KEGG; http://www.genome.ad.jp/kegg) and the Gene Map Annotator and Pathway Proler (GenMAPP; http://www.genmapp.org; Dahlquist et al., 2002)). The application of these GO and pathway mapping tools to toxicogenomics data is dramatically improving the interpretation of xenoestrogen-induced gene expression proles in terms of the molecular pathways and cellular processes that are affected, and provides a key step in being able to associate gene expression changes with alterations in conventional indicators of phenotypic and toxicologic changes (e.g., histological and clinical chemistry markers). This process, termed Phenotypic Anchoring is very important for dening genes that are linked to mechanisms of toxicity and adverse effects (Tennant, 2002; Waters et al., 2003; Paules, 2003). This review illustrates how toxicogenomics approaches are providing an unprecedented amount of mechanistic information on the molecular responses to xenoestrogens and are likely to impact on hazard and risk assessment by providing mechanism-based biomarkers for xenoestrogen exposure, as well as short-term biomarkers for long-term endpoints, such as xenoestrogen-induced carcinogenesis.

2. Molecular mechanisms of xenoestrogen signalling in mammals Estrogen signalling in mammalian cells is mediated primarily at the molecular level by two members of the nuclear receptor superfamily: estrogen receptor alpha (ER ) and estrogen receptor beta (ER ). ER and ER function as ligand-activated transcription factors, in conjunction with numerous coregulatory proteins, in order to activate or repress the transcription of ERresponsive genes (Moggs and Orphanides, 2001; Hall et al., 2001; McKenna and OMalley, 2002; Tremblay and Giguere, 2002; Moggs et al., 2003; Metivier et al., 2003). The tissue-specic effects of estrogenic chemicals are mediated by cellular variations in the expression of each ER subtype, together with the tissue-specic expression of coregulatory proteins (Shang and Brown, 2002). Xenoestrogens encompass a diverse group of natural and synthetic chemicals that can mimic the actions of physiological estrogens in mammals via interaction with estrogen receptors. Importantly, in vitro studies have demonstrated that ER and ER display signicant differences in binding afnity, coregulator recruitment and transcriptional activation in response to some natural and synthetic estrogenic chemicals (Kuiper et al., 1997; Kuiper et al., 1998; Nishikawa et al., 1999; Routledge et al., 2000; An et al., 2001; Mueller et al., 2004). The classical mechanism of estrogen receptor action involves estrogenic ligands diffusing into the cell nucleus and binding to ER, inducing the receptor to dimerise and bind to specic estrogen response elements (EREs) in the promoter regions of target genes, resulting in the regulation of their expression. ER has also been shown regulate the transcription of some genes independently of an ERE by interacting with DNA-bound AP-1 transcription factors (Kushner et al., 2000). However, several additional mechanisms of ER-mediated gene regulation are now well established. Both ER subtypes possess two distinct transactivation domains, termed AF-1 and AF-2. The AF-2 region mediates ligand-dependent transcriptional activation whilst the AF-1 domain can inuence gene expression in a ligand-independent manner. The regulation of ER AF-1 activation occurs, at least in part, through the reversible phosphorylation of conserved serine residues that results from growth factor-activated protein kinase signalling cascades. AF-1 and AF-2

180

J.G. Moggs / Toxicology 213 (2005) 177193

can also interact synergistically to inuence promoter activity (Metivier et al., 2001) and thus there is potential for considerable interplay between ER ligand binding and growth factor-activated signalling pathways to inuence gene expression responses to xenoestrogens. ER signalling can be further complicated by the rapid extranuclear or non-genomic responses to estrogenic chemicals (Falkenstein et al., 2000; Moggs et al., 2003; Bulayeva and Watson, 2004; Revankar et al., 2005). At present, relatively little is known about the relative contributions of these diverse receptor-mediated gene regulatory mechanisms to the pleiotropic and tissuespecic responses to either physiological estrogens or xenoestrogens that are observed in vivo. Toxicogenomic approaches provide an opportunity for more holistic studies of the transcriptional responses to estrogenic chemicals that should enhance our understanding of the genes targeted by xenoestrogens, the molecular mechanisms through which they are regulated, and how they function to elicit the pleiotropic cellular responses to xenoestrogen exposure.

3. Application of toxicogenomics to understanding mechanisms of estrogen action in vivo: using the rodent uterotrophic assay as a model experimental system Despite signicant progress in elucidating the molecular mechanisms by which estrogens regulate gene expression via estrogen receptors (ERs) in vitro, little is known about how they coordinate tissue growth and differentiation at the molecular level. The response of the rodent uterus to estrogens has provided researchers with a tractable model system for investigating these mechanisms and the physiological response of the uterus to exogenous E2 in this system has been documented in detail (Clark and Mani, 1994). The initial biological responses of the uterus to E2 are rapid (<6 h), and include the uptake of uid resulting from hyperemia and vasodilation of uterine capillaries. This phenomenon is termed water imbibition and leads to swelling of the uterus that results in a marked decompaction of stromal cells. Another early event is an increase in overall levels of mRNA and protein synthesis. The net result of these early E2 responses is to increase the availability of metabolites

and macromolecules required for subsequent uterine growth. The uterus then enters a proliferative phase that is responsible, at least in part, for the large increase in uterine weight that occurs 1630 h after E2 exposure. Later responses include alterations in the surface of the luminal epithelia in preparation for embryo implantation and mimic, to an extent, the changes in uterine physiology that accompany the onset of puberty. The immature rodent uterus is primed to respond to elevations in endogenous E2 levels that occur during puberty and this sensitivity of the immature uterus (and similarly of uteri in ovariectomized adults) to estrogens has been widely exploited as a toxicological test, the rodent uterotrophic assay, for measuring the estrogenic activity of chemicals as a function of increased uterine weight (Owens and Ashby, 2002). While the uterotrophic events described above have been well characterized at the physiological level, little is known about how E2, acting through the nuclear receptors ER and ER , coordinates at the molecular level the wide range of cellular processes involved. The recent application of gene expression proling to this model system (Watanabe et al., 2002, 2003a,b; Naciff et al., 2002; Naciff et al., 2003; Hewitt et al., 2003; Fertuck et al., 2003; Frasor et al., 2003b; Moggs et al., 2004a,b) is now providing a wealth of detail and novel mechanistic insights into the molecular responses to physiological estrogens and xenoestrogens in vivo (described below). 3.1. Holistic identication of E2-responsive genes in vivo The application of gene expression proling to the rodent uterotrophic assay, using microarrays that provide genome-wide coverage of the tens of thousands of known genes in mammals, is a powerful approach for dening holistically the genes targeted in vivo by estrogens. We have used a multi-step method to analyse gene expression proles from seven different time points during the uterotrophic response to the physiological estrogen 17 -estradiol E2 (Fig. 1; Moggs et al., 2004a). First, data were ltered and subjected to statistical analyses to identify the 3538 genes with altered expression in E2 -treated mice. In this particular study, each gene was subjected to a mixed-model ANOVA (Churchill, 2004) allowing for treatment, time and treatment by time interaction as xed effects and replicate study as

J.G. Moggs / Toxicology 213 (2005) 177193

181

random effect, although numerous alternative statistical approaches are currently available for microarray data analysis, the selection of which is strongly dependent on experimental design parameters. Unsupervised hierarchical clustering was then used to group these genes into temporally co-regulated clusters (Slonim, 2002), revealing a complex multi-stage transcriptional response to E2 in the uterus (gene clusters AI; Fig. 1). A similar approach has been used to dene the temporal patterns of gene expression during the uterotrophic response of immature, ovariectomized mice following exposure to E2 (Watanabe et al., 2003b) and 17 ethynylestradiol (Fertuck et al., 2003). The identication of co-regulated E2-responsive genes within these complex gene expression proles using pattern visualisation tools such as gene trees or heat maps generated by hierarchical clustering greatly facilitates the subsequent identication of genes exhibiting common molecular functions, particularly when used in conjunction with the GO and pathway mapping tools described in Section 3.2. 3.2. Gene ontology and pathway mapping of E2-responsive gene functions The application of gene ontology (GO) mapping and pathway analysis tools to toxicogenomic data is emerging as a powerful approach for interpreting complex gene expression proles in the context of the underlying biological pathways and processes affected (Section 1). In order to gain an overview of the predominant molecular functions and biological pathways that are regulated at the transcriptional level during the staged uterotrophic response to E2 , we interrogated the 3538 E2-responsive genes using GO and pathway mapping approaches (Moggs et al., 2004a; Currie et al., 2005). An unsupervised GO mapping analysis of the 3538 E2-responsive genes in the mouse uterus using the GOStat gene ontology mining tool (Beibarth and Speed, 2004) revealed that E2 predominantly targets genes involved in protein metabolism, cell cycle, cell proliferation, DNA replication, RNA metabolism, mRNA transcription, and blood vessel development (Table 1). Independent mapping of the same 3538 E2responsive genes to pre-dened pathway maps using MAPPnder (Doniger et al., 2003; Table 1) provided complementary insights into the major E2-regulated pathways associated with uterine growth and differen-

tiation (e.g., DNA replication and purine metabolism). The combination of GO and Pathway Mapping tools therefore provides a powerful approach for identifying predominant E2-regulated pathways amongst large scale and multivariate gene expression proles. Importantly, we gained further insights into the functional relationships between these pathways during the uterotrophic response by using supervised gene ontology-driven clustering in which co-regulated E2responsive genes (as dened by hierarchical clustering; Section 3.1) were interrogated using GO mapping. This approach allowed us to identify gene functions that were predominant in each co-regulated cluster in Fig. 1, and revealed that E2 induces uterine growth and maturation by regulating successively the activities of different biological pathways (Fig. 2A). These observations demonstrate that GO and pathway mapping represent powerful approaches for rapidly translating complex toxicogenomic data in order to reveal common target gene functions and interrelationships at the biological pathway level. 3.3. Phenotypic anchoring of gene expression changes during E2-induced uterine growth A major challenge in the interpretation of toxicogenomic data is the need to dene how chemically induced changes in gene expression relate to conventional physiological and toxicological endpoints (Tennant, 2002; Waters et al., 2003; Paules, 2003). The extensive knowledge of E2-induced physiological changes in the rodent uterus provides an opportunity to dene the relationships between E2-induced changes in gene expression and alterations in uterine physiology. Importantly, GO and pathway mapping of toxicantinduced gene expression changes to cellular pathways and processes represents a key step in this process. We have shown that E2 induces a tightly coordinated transcriptional program that regulates successive and interlinked cellular processes during the uterotrophic response (Sections 3.1 and 3.2). By comparing these changes in gene expression to alterations in uterine weight and histology, we have been able to identify classes of genes that drive specic histological changes in the uterus, including uid uptake and coordinated cell division (Fig. 2A; Moggs et al., 2004a). This is exemplied by the coordinate E2-dependent induction of chromosome replication genes between 8 and 24 h

182

J.G. Moggs / Toxicology 213 (2005) 177193

J.G. Moggs / Toxicology 213 (2005) 177193

183

Table 1 GO and Pathway Mapping provide complementary insights into the predominant E2-regulated pathways amongst 3538 E2-responsive genes in the immature mouse uterotrophic assay GO Ids GO:0043170 GO:0019538 GO:0043039 GO:0006418 GO:0043038 GO:0006400 GO:0009059 GO:0043037 GO:0009057 GO:0006412 GO:0016126 GO:0007049 GO:0008283 GO:0000067 GO:0006260 GO:0000074 GO:0019752 GO:0006082 GO:0006520 GO:0006631 GO:0009308 GO:0016125 GO:0008203 GO:0019219 GO:0045449 GO:0006355 GO:0009451 GO:0016070 GO:0006350 GO:0006351 GO:0016049 GO:0008361 GO:0001568 GO:0006259 GO annotation Macromolecule metabolism Protein metabolism tRNA aminoacylation tRNA aminoacylation for protein translation Amino acid activation tRNA modication Macromolecule biosynthesis Translation Macromolecule catabolism Protein biosynthesis Sterol biosynthesis Cell cycle Cell proliferation DNA replication and chromosome cycle DNA replication Regulation of cell cycle Carboxylic acid metabolism Organic acid metabolism Amino acid metabolism Fatty acid metabolism Amine metabolism Sterol metabolism Cholesterol metabolism Regulation of nucleic acid metabolism Regulation of transcription Regulation of transcription, DNA-dependent RNA modication RNA metabolism Transcription Transcription, DNA-dependent Cell growth Regulation of cell size Blood vessel development DNA metabolism p-Value 3.04e20 4.87e17 7.9e07 7.9e07 7.9e07 3.04e06 6.61e06 1.62e05 5.34e05 5.34e05 6e05 1.31e18 4.87e17 2.21e11 9.28e10 8.51e08 7.28e13 7.28e13 8.58e09 7.35e05 1.26e08 5.95e08 7.9e07 6.73e07 6.73e07 2.44e06 6.73e07 9.23e06 1.08e06 2.26e06 4.89e06 2.06e05 1.77e05 6.15e05 GenMAPP (Z-score/p-value)

Mm tRNA Synthetases (2.676/0.014)

Mm DNA replication (2.622/0.017)

Mm Glycine serine and threonine metabolism(2.366/0.034) Mm Arginine and proline metabolism(2.48/0.018)

Mm Purine metabolism (1.426/0.184)

GO analysis was performed using GOStat (http://gostat.wehi.edu.au; Beibbarth and Speed, 2004). Shown are GO terms that are signicantly over-represented (p < 0.0001, using the Benjamini and Hochberg false discovery rate multiple testing correction). GenMAPP analysis was performed using MAPPFinder (Doniger et al., 2003). Shown are GenMAPPs that are signicantly over-represented (positive Z-score) and have p-values < 0.2. Adapted from Moggs et al. (2004a) and Currie et al. (2005).

Fig. 1. Experimental strategy for holistic identication of E2-responsive genes during rodent uterotrophic response. E2-responsive genes were identied by microarray analysis using ANOVA of three independent biological replicate studies in which seven different time points were analysed for E2-treated animals together with equivalent time points for vehicle-treated animals. Hierarchical clustering of 3538 E2-responsive genes revealed that the immature mouse uterus undergoes a staged transcriptional response to E2. The color bar scale indicates the mean fold change of E2-induced gene expression relative to time-matched AO-treated control samples. The predominant molecular functions of temporally co-regulated genes (clusters AI) were interrogated using Gene Ontology and Pathway Mapping tools and then compared with alterations in uterine physiology in order to facilitate phenotypic anchoring of toxicogenomic data. Adapted from Moggs et al. (2004a).

184

J.G. Moggs / Toxicology 213 (2005) 177193

Fig. 2. Phenotypic anchoring of E2-induced changes in gene expression. (A) GO mapping of co-regulated E2-responsive genes shows that E2induced uterine growth is associated with the successive regulation of genes with distinct molecular functions. (B) Coordinated expression of genes involved in chromosome replication and cell division. See Fig. 1 for color bar scale of altered gene expression. (C) The coordinated regulation of genes controlling chromosome replicationand cell division coincides with uterine histological changes associated with cell proliferation. E2 induces the thickening of luminal (LE) and glandular epithelium (GE) and an increased number of mitotic cells (indicated by arrowheads) between 8 and 24 h. Longitudinal 0.3 m thick parafn sections of uteri were stained with hematoxylin and eosin. Scale bar, 50 m. Quantitative mitotic index data were derived from four animals per group. Adapted from Moggs et al. (2004a).

J.G. Moggs / Toxicology 213 (2005) 177193

185

(Fig. 2B), the timing of which tightly correlates with histological alterations (e.g., increased epithelial cell height and mitotic index) that accompany increased cell proliferation (Fig. 2C). Together these data dene, at an unprecedented level of detail, the mechanism by which an estrogen induces organ growth and tissue maturation, and demonstrate that comparison of temporal changes in gene expression and conventional toxicology endpoints can facilitate the phenotypic anchoring of toxicogenomic data. To date, most gene expression proling studies of the rodent uterotrophic response to estrogenic chemicals represent the combined response of all cells in the uterus, and thus provide an integrated view of the transcriptional networks operating within the tissue as a whole. However, since the uterotrophic response is known to involve complex stromal-epithelial signalling events (Cooke et al., 1997), a limitation of the studies described above is that it is likely that many genes are only induced or repressed by estrogens in a fraction of uterine cells, and further analysis of the uterine cell-type specic expression patterns of genes, using immunohistochemistry or in situ hybridization, will be required for these relationships to be dened. The application of transcript proling to specic cell populations isolated by laser capture dissection (Ho Hong et al., 2004) promises to provide a more holistic view of cell-type specic transcriptional responses to estrogens within the uterus and additional target tissues. In addition to exploring the molecular mechanisms underlying physiological responses of the rodent uterus to exogenous estrogens, gene expression proling has also been used recently to dene the molecular basis for endogenous estrogen signalling during normal uterine development and physiological functions, including the aging (Khalyfa et al., 2003), estrous cycle (Tan et al., 2003) and embryo implantation (Reese et al., 2001; Bethin et al., 2003; Dey et al., 2004). Importantly, these studies reveal that there is considerable overlap between the genes regulated by exogenous estrogen exposure in the rodent uterotrophic assay and genes that are differentially regulated in response to endogenous hormones, thus lending support to the physiological relevance of many of the novel E2-responsive genes identied via genomic analyses of the rodent uterotrophic model system. Increasing attention is also being paid to the use of gene expression changes in the uterus as surrogate

biomarkers for short-term rodent estrogenicity assays (Owens and Ashby, 2002). Toxicogenomic analysis of E2-responsive target genes in vivo has revealed a large number of novel candidate marker genes for estrogenic chemicals that elicit the uterotrophic response (see Section 4). Importantly, the grouping of these marker genes into common molecular functions or biological pathways using Gene Ontology and Pathway Mapping facilitates their utility as mechanism-based markers in which the altered expression of multiple genes in an E2-induced pathway is more likely to provide a robust measure of estrogenic activity than an individual gene expression change.

4. Comparative transcript proling of physiological, synthetic and plant-derived estrogens Comparative transcript proling represents a powerful method for identifying gene expression signatures or ngerprints that are characteristic of chemicals with estrogenic activity (Naciff and Daston, 2004; Deavall et al., 2005), and can potentially provide molecular markers for the xenoestrogen class of endocrine disruptors. Furthermore, this approach has great potential for investigating possible mechanistic differences in the cellular responses to different classes of xenoestrogens. The molecular events revealed by global gene expression proling for physiological estrogens (described in Section 3) provides a detailed mechanistic basis for understanding how other estrogenic chemicals induce their effects. Thus, having dened a reference transcriptional programme for the potent physiological estrogen, E2 (Moggs et al., 2004a), we investigated, using global gene expression proling, whether different classes of xenoestrogens, including the phytoestrogen genistein (GEN) and the synthetic estrogen diethylstilbestrol (DES), exert mechanistically distinct molecular effects in the rodent uterus (Moggs et al., 2004b). The selection of these three classes of estrogenic chemicals was based driven in part by the apparent paradox that the exposure of humans to g/kg levels of synthetic environmental estrogens is assumed to be hazardous, whilst exposure to mg/kg levels of phytoestrogens in foods and dietary supplements is frequently proposed to be life enhanc-

186

J.G. Moggs / Toxicology 213 (2005) 177193

ing. Thus an important question in human hazard and risk assessment is whether estrogens derived from different sources (e.g., synthetic versus plant-derived) are toxicologically similar (i.e., induce their estrogenic effects through similar mechanisms) or intrinsically different. The immature mouse uterus was selected for this comparative analysis because it is a major estrogen-responsive organ and forms the basis for a reference assay of estrogenic activity (Owens and Ashby, 2002), including xenoestrogen-induced carcinogenesis (Newbold et al., 2001). Several lines of evidence suggest possible differences in the molecular mechanisms induced by different xenoestrogens. Firstly, ER and ER display signicant differences in binding afnity for different estrogens (Kuiper et al., 1998). For example, the physiological estrogen 17 -estradiol (E2 ) and the synthetic estrogen diethylstilbestrol (DES) bind with a similar afnity to ER and ER , while the phytoestrogen genistein (GEN) binds with 20-fold higher afnity to ER than to ER (Kuiper et al., 1998). Secondly, ER and ER expression levels vary amongst different cell types and tissues. Since the uterus expresses both ER subtypes (Weihua et al., 2000) it might be expected to exhibit distinct molecular responses to ER-subtype selective ligands such as GEN when compared to E2. Thirdly, a variety of non-hormonal properties have been reported for GEN, including inhibition of tyrosine kinases (Akiyama et al., 1987), nitric oxide synthase (Duarte et al., 1997) and topoisomerase II (Okura et al., 1988) and also decreases in calcium-channel activity (Potier and Rovira, 1999), lipid peroxidation (Arora et al., 1998) and diacylglycerol synthesis (Dean et al., 1989). Non-hormonal properties of DES include the induction of aneuploidy in mammalian cells (Aardema et al., 1998) and binding to rat liver DNA (Williams et al., 1993). Despite these possible mechanistic differences between E2, DES and GEN, comparative transcript proling, under conditions where each chemical produced an equivalent gravimetric and histological uterotrophic effect, revealed that DES and GEN are capable of modulating the full spectrum of biological processes associated with the action of E2 in the rodent uterus (Moggs et al., 2004b). Thus, in the rodent uterus, a synthetic estrogen and a phytoestrogen operate via a molecular mechanism analogous to that of the physiological estrogen, E2. Together, these data indicate that estrogens of differing origin may share the potential for both bene-

cial and adverse health effects, thus highlighting the need for an holistic approach to hazard assessment that takes into account combined exposures to physiological, synthetic and plant-derived estrogens. It is noteworthy that previous studies have revealed both similarities and differences between transcriptional responses induced at a single time-point after exposure to E2 and DES in the uteri of immature ovariectomized mice (Watanabe et al., 2003a), and after exposure to either GEN, bisphenol A or 17 -ethynl-estradiol in the reproductive tract of intact adult rats (Naciff et al., 2002). However, the xenoestrogen-specic gene expression changes documented in these studies most probably arises from dose-dependent variations in the magnitude and kinetics of gene expression, rather than from the operation of distinct mechanisms of estrogenic action. The high degree of similarity of transcriptional responses to three different classes of estrogenic chemicals (Moggs et al., 2004b) raises the question of whether additional mitogenic stimuli drive similar transcriptional responses in the rodent uterus. In this respect, a recent comparative evaluation of global gene expression changes in the mouse uterus in response to epidermal growth factor, IGF-I and E2, revealed considerable overlap in the transcriptional responses to both growth factors and E2 (Hewitt et al., 2005). However, there were also signicant differences in the response of some genes to IGF-I or epidermal growth factor versus E2, and importantly, many of the changes in gene expression observed after exposure to growth factors were not dependent on the presence of functional ER , indicating that the uterotrophic response involves E2-specic ER -mediated responses together with responses that result from cross-talk between growth factor and ER signalling pathways. Another important question is whether tissuespecic differences exist in the mechanism of action of distinct classes of estrogenic chemicals. Comparative transcript proling of E2-responsive genes in rodent liver and uterus reveals some apparent tissue-specic differences between E2-responsive genes (Boverhof et al., 2004). Furthermore, the xenoestrogen nonylphenol has been reported to cause very similar effects to E2 on gene expression in uterine but not in liver tissue, indicating that tissue-specic effects may exist for some classes of xenoestrogen in certain tissues (Watanabe et al., 2004). However, it will be important to determine

J.G. Moggs / Toxicology 213 (2005) 177193

187

whether these apparent tissue-specic effects represent mechanistically distinct responses to physiological estrogens and xenoestrogens or whether they arise due to differential pharmacokinetics or potency of estrogenic chemicals in different tissues. In addition to the rodent uterotrophic assay, a variety of in vitro mammalian model systems have also been employed to elucidate gene expression proles or signatures that are characteristic of chemicals with estrogenic activity (Charpentier et al., 2000; Soulez and Parker, 2001; Lobenhofer et al., 2002; Frasor et al., 2003a; Hodges et al., 2003; Levenson et al., 2002; DeNardo et al., 2005; Inoue et al., 2002; Terasaka et al., 2004; Ise et al., 2005; Wang et al., 2004). Whilst these approaches are valuable for gaining insights into the relative similarity of transcriptional responses induced by a range of estrogenic chemicals, apparent xenoestrogenspecic target genes (Ramanathan and Gray, 2003; Terasaka et al., 2004; Ise et al., 2005) may arise in part to differences in timing and/or magnitude of the transcriptional response in the experimental system employed. Comparison of gene expression changes induced by equipotent doses of estrogenic chemicals for a dened in vitro biological endpoint (e.g., cell proliferation) would strengthen the case for there being mechanistic differences (e.g., off-target effects) in the cellular responses to different estrogenic chemicals.

5.1. Transgenic animals As described in Section 2, E2 can regulate transcription through a combination of at least two distinct signalling pathways: (i) via activation of the nuclear transcription factors ER and ER and (ii) via extranuclear or non-genomic signalling events. The transcriptional responses to xenoestrogens that have been dened using toxicogenomic approaches are thus likely to involve a combination of direct gene regulation by nuclear ERs and indirect gene regulation via extranuclear signalling pathways. However, the relative contributions of these complex ER-mediated signalling pathways are not well understood. Importantly, recent transcript proling studies using ovariectomized ER knockout mice revealed a predominant role for ER in the regulation of estrogen-responsive genes in the uterus (Watanabe et al., 2002, 2003b; Hewitt et al., 2003), consistent with the observation that only a partial uterotrophic response occurs in ER knockout mice (Lubahn et al., 1993). Nevertheless, it is noteworthy that both ER subtypes are expressed at comparable levels in the immature mouse uterus and that ER has been proposed to antagonize the proliferative functions of ER (Weihua et al., 2000). Consistent with this notion, recent uterine studies using ER subtype-selective agonist ligands indicate that whilst ER is the major regulator of estrogen function in the uterus, ER does exert effects on some uterine markers of estrogen action and can modulate ER activity in a response-specic and dose-dependent manner (Frasor et al., 2003a). Therefore, whilst the majority of E2-responsive genes identied in uterine toxicogenomics studies to date are likely to be regulated by ER , the identication of the direct gene targets for each ER subtype will ultimately require the development of methods for measuring the occupancy of receptor subtypes at promoters in vivo. For example, chromatin immunoprecipitation assays (Orlando, 2000) could, in principle, be used to map the endogenous genomic DNA binding sites for each ER subtype in uterine tissue. Another key area in which transgenic animal models are aiding our understanding of the molecular responses to estrogens is the development of luciferase-coupled ERE reporter gene mice for the in vivo imaging of transcriptionally active estrogen receptors (Ciana et al., 2003). In this model system, non-reproductive tissues, such as bone and brain, exhibit peak transcriptional activity of estrogen recep-

5. Understanding mechanisms of ER-mediated gene regulation: application of toxicogenomics to biological systems in which the ER signalling has been disrupted The application of gene expression proling to biological systems in which key ER signalling components have been disrupted is a powerful approach for understanding mechanisms of ER-mediated transcriptional regulation. Disruption strategies may include the alteration of signalling pathways and receptor expression via the generation of transgenic animals, genetically modied cell lines and through the use of small molecule chemical inhibitors. These approaches can provide mechanistic insights into how xenoestrogens target specic genes and into the functional significance of xenoestrogen-induced alterations in gene expression.

188

J.G. Moggs / Toxicology 213 (2005) 177193

tors at diestrus (i.e., when endogenous estrogen levels are at their lowest), thus emphasizing the importance of hormone-independent activation of estrogen receptors in vivo. The combination of transcript proling with model systems in which ligand-independent activities of ERs have been perturbed should facilitate more detailed mechanistic analyses of tissue-specic transcriptional responses mediated by ERs. 5.2. Cell-based model systems Immortalised cell lines are highly suited for detailed mechanistic studies of estrogen action due to them consisting of relatively homogenous cell populations and being experimentally amenable to manipulation of specic estrogen signalling pathways using a wide range of approaches including genetic modication, siRNAs and small molecule inhibitors. Transcript proling of cell-based model systems thus provides a powerful approach for elucidating the mechanism by which genes are regulated by estrogenic chemicals. For example, gene expression proling has been used to reveal fundamental differences in the molecular responses of human MCF-7 breast cancer cells to the selective estrogen receptor modulators (SERMs) tamoxifen and raloxifene or ICI 182,780 (Hodges et al., 2003; Frasor et al., 2004; Kian Tee et al., 2003). Cell line models have also been employed to dissect the relative contributions of ER and ER to estrogen-dependent transcriptional regulation. For example, gene expression prolling of U2OS human osteosarcoma cells stably transfected with either ER or ER , revealed both common as well as distinct target genes for these two ERs (Kian Tee et al., 2003; Stossi et al., 2004). It is noteworthy that for a subset of genes regulated equally by E2 via either ER or ER , the phytoestrogen genistein preferentially stimulated gene expression via ER (Stossi et al., 2004), consistent with the preferential binding of ER by genistein in vitro (Kuiper et al., 1998). The transition from identifying E2-responsive target genes to understanding their regulatory mechanisms can also be facilitated by employing in silico genomics approaches. ERs recognize 15-bp palindromic estrogen response elements (EREs) with high afnity in vitro (Klinge, 2001) and an in silico genome-wide screen for such high-afnity EREs resulted in the identication of approximately 70,000 ERE motifs in the human and mouse genomes (Bourdeau et al., 2004; Lin et al.,

2004). Importantly, the functional relevance of many of these EREs was subsequently conrmed by biochemical (chromatin immunoprecipitation) assays for ER -binding in human breast cancer cell lines. We have recently also used global gene expression proling to investigate the mechanistic basis for an aberrant proliferative response to E2 that is observed in ER-negative breast cancer cells that re-express ER (Moggs et al., 2005). Estrogen receptor (ER)-negative breast carcinomas do not respond to hormone therapy, making their effective treatment very difcult, and the re-expression of ER in ER-negative MDA-MB-231 breast cancer cells has been explored as a model system, in which hormone-dependent responses can be restored (Garcia et al., 1992; Levenson and Jordan, 1994; Lazennec and Katzenellenbogen, 1999). Paradoxically, in contrast to the mitogenic activity of E2 in ER-positive breast cancer cells, E2 suppresses proliferation in ER-negative breast cancer cells in which ER has been re-expressed. Transcript proling reveals that the E2-induced suppression of proliferation in these cells is associated with the down-regulation of genes encoding positive cell cycle regulators, chromosome replication proteins and an anti-apoptosis factor. In parallel, E2 up-regulated the expression of the negative cell cycle regulators and tumour-suppressor genes. Furthermore, the expression of several of these genes is regulated in the opposite direction by E2 compared with their regulation in ER-positive MCF-7 cells. Thus, transcript proling reveals that the anti-proliferative effect of estrogen in breast cancer cells that re-express ERalpha is mediated by aberrant regulation of cell cycle genes.

6. Toxicogenomic analysis of xenoestrogen-induced carcinogenesis Diethylstilbestrol (DES) has been investigated extensively both as an estrogen and as a carcinogen and the development of uterine carcinomas within a year of briey exposing neonatal mice to DES has long been established (Newbold et al., 1990). Additionally, perturbations of uterine morphology, such as increased or atypical hyperplasia, are observed prior to tumour formation. The development of DES-induced tumours is dependent on functional ER (Couse and Korach, 2004) and can be inhibited by prepubescent

J.G. Moggs / Toxicology 213 (2005) 177193

189

ovariectomy, implying that it is the combination of neonatal exposure to DES and subsequent pubertal exposure to endogenous estrogens which is necessary for the induction of the tumours (Newbold et al., 1990). However, the early cellular and molecular mechanisms underlying this transformation event are not well understood. A combination of global gene expression analysis and a knockout mouse model for the homeobox gene MSX2 have recently been used to delineate genetic pathways affected by DES and have revealed that neonatal DES treatment alters uterine cell fate, particularly in the luminal epithelium, by inducing genes associated with abnormal differentiation (Huang et al., 2005). Interestingly, neonatal exposure to DES has been associated with persistent changes in gene expression, through presumed epigenetic mechanisms involving altered DNA methylation status of gene regulatory regions (Li et al., 2003). The perturbation of epigenetic status is emerging as an important mechanism for toxicant action (Bombail et al., 2004) and toxicogenomics provides a new opportunity to explore, in a holistic manner, persistent changes in gene expression, which may be associated with epigenetic changes during the onset of DES-induced carcinogenesis (Moore et al., 2005), and may also provide short-term molecular markers for xenoestrogen-induced carcinogenesis.

the tissue- and species-specic molecular responses to xenoestrogens.

Acknowledgements I would like to thank all my colleagues at Syngenta CTL who have been involved in many of the studies described in this manuscript, in particular H. Tinwell, F.L. Lim, R. Currie, D. Moore, I. Pate, A. Soames, R. Stuckey, T. Murphy, T. Spurway, D. Deavall, J. Edmunds, K. Antrobus, I. Kimber, J. Ashby and G. Orphanides.

References
Aardema, M.J., Albertini, S., Arni, P., Henderson, L.M., KirschVolders, M., Mackay, J.M., Sarrif, A.M., Stringer, D.A., Taalman, R.D., 1998. Aneuploidy: a report of an ECETOC task force. Mutat. Res. 410, 379. Akiyama, T., Ishida, J., Nakagawa, S., Ogawara, H., Watanabe, S., Itoh, N., Shibuya, M., Fukami, Y., 1987. Genistein, a specic inhibitor of tyrosine-specic protein kinases. J. Biol. Chem. 262, 55925595. Arora, A., Nair, M.G., Strasburg, G.M., 1998. Antioxidant activities of isoavones and their biological metabolites in a liposomal system. Arch. Biochem. Biophys. 356, 133141. An, J.P., Tzagarakis-Foster, C., Scharschmidt, T.C., Lomri, N., Leitman, D.C., 2001. Estrogen receptor -selective transcriptional activity and recruitment of coregulators by phytoestrogens. J. Biol. Chem. 276, 1780817814. Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, J.M., Davis, A.P., Dolinski, K., Dwight, S.S., Eppig, J.T., Harris, M.A., Hill, D.P., Issel-Tarver, L., Kasarskis, A., Lewis, S., Matese, J.C., Richardson, J.E., Ringwald, M., Rubin, G.M., Sherlock, G., 2000. Gene ontology: tool for the unication of biology. The gene ontology consortium. Nat. Genet. 25, 2529. Beibarth, T., Speed, T.P., 2004. GOstat: nd statistically overrepresented Gene Ontologies within a group of genes. Bioinformatics 20, 14641465. Bethin, K.E., Nagai, Y., Sladek, R., Asada, M., Sadovsky, Y., Hudson, T.J., Muglia, L.J., 2003. Microarray analysis of uterine gene expression in mouse and human pregnancy. Mol. Endocrinol. 17, 14541469. Bombail, V., Moggs, J.G., Orphanides, G., 2004. Perturbation of epigenetic status by toxicants. Toxicol. Lett. 149, 5158. Bourdeau, V., Deschenes, J., Metivier, R., Nagai, Y., Nguyen, D., Bretschneider, N., Gannon, F., White, J.H., Mader, S., 2004. Genome-wide identication of high-afnity estrogen response elements in human and mouse. Mol. Endocrinol. 18, 14111427. Boverhof, D.R., Fertuck, K.C., Burgoon, L.D., Eckel, J.E., Gennings, C., Zacharewski, T.R., 2004. Temporal- and dose-dependent hepatic gene expression changes in immature ovariectomized

7. Summary and perspectives Toxicogenomics approaches are providing an unprecedented amount of mechanistic information on the molecular responses to xenoestrogens and are likely to impact on hazard and risk assessment by providing mechanism-based biomarkers for xenoestrogen exposure, as well as short-term biomarkers for long-term endpoints, such as xenoestrogen-induced carcinogenesis. In addition to the ongoing challenge of dening the relationships between xenoestrogen gene expression changes and conventional physiological and toxicological endpoints, the advent of systems toxicology promises to provide complementary mechanistic information to gene expression proling data via the integration of genomics, proteomics and metabolomics databases (Waters and Fostel, 2004). The development of novel bioinformatic tools for interrogation and modelling of combined -omic datasets should ultimately lead to an improved understanding of

190

J.G. Moggs / Toxicology 213 (2005) 177193 Duarte, J., Ocete, M.A., Perez-Vizcaino, F., Zarzuelo, A., Tamargo, J., 1997. Effect of tyrosine kinase and tyrosine phosphatase inhibitors on aortic contraction and induction of nitric oxide synthase. Eur. J. Pharmacol. 338, 2533. Falkenstein, E., Tillmann, H.C., Christ, M., Feuring, M., Wehling, M., 2000. Multiple actions of steroid hormonesa focus on rapid, nongenomic effects. Pharmacol. Rev. 52, 513556. Fertuck, K.C., Eckel, J.E., Gennings, C., Zacharewski, T.R., 2003. Identication of temporal patterns of gene expression in the uteri of immature, ovariectomized mice following exposure to ethynylestradiol. Physiol. Genomics 15, 127141. Harris, M.A., Clark, J., Ireland, A., Lomax, J., Ashburner, M., Foulger, R., Eilbeck, K., Lewis, S., Marshall, B., Mungall, C., Richter, J., Rubin, G.M., Blake, J.A., Bult, C., Dolan, M., Drabkin, H., Eppig, J.T., Hill, D.P., Ni, L., Ringwald, M., Balakrishnan, R., Cherry, J.M., Christie, K.R., Costanzo, M.C., Dwight, S.S., Engel, S., Fisk, D.G., Hirschman, J.E., Hong, E.L., Nash, R.S., Sethuraman, A., Theesfeld, C.L., Botstein, D., Dolinski, K., Feierbach, B., Berardini, T., Mundodi, S., Rhee, S.Y., Apweiler, R., Barrell, D., Camon, E., Dimmer, E., Lee, V., Chisholm, R., Gaudet, P., Kibbe, W., Kishore, R., Schwarz, E.M., Sternberg, P., Gwinn, M., Hannick, L., Wortman, J., Berriman, M., Wood, V., de la Cruz, N., Tonellato, P., Jaiswal, P., Seigfried, T., White, R., 2004. Gene ontology consortium. The gene ontology (GO) database and informatics resource. Nucl. Acids Res. 32, D258D261. Frasor, J., Barnett, D.H., Danes, J.M., Hess, R., Parlow, A.F., Katzenellenbogen, B.S., 2003a. Response-specic and ligand dose-dependent modulation of estrogen receptor (ER) alpha activity by ERbeta in the uterus. Endocrinology 144, 3159 3166. Frasor, J., Danes, J.M., Komm, B., Chang, K.C., Lyttle, C.R., Katzenellenbogen, B.S., 2003b. Proling of estrogen up- and down-regulated gene expression in human breast cancer cells: insights into gene networks and pathways underlying estrogenic control of proliferation and cell phenotype. Endocrinology 144, 45624574. Frasor, J., Stossi, F., Danes, J.M., Komm, B., Lyttle, C.R., Katzenellenbogen, B.S., 2004. Selective estrogen receptor modulators: discrimination of agonistic versus antagonistic activities by gene expression proling in breast cancer cells. Cancer Res. 64, 15221533. Garcia, M., Derocq, D., Freiss, G., Rochefort, H., 1992. Activation of estrogen receptor transfected into a receptor-negative breast cancer cell line decreases the metastatic and invasive potential of the cells. Proc. Natl. Acad. Sci. U.S.A. 89, 1153811542. Hall, J.M., Couse, J.F., Korach, K.S., 2001. The multifaceted mechanisms of estradiol and estrogen receptor signaling. J. Biol. Chem. 276, 3686936872. Hewitt, S.C., Deroo, B.J., Hansen, K., Collins, J., Grissom, S., Afshari, C.A., Korach, K.S., 2003. Estrogen receptor-dependent genomic responses in the uterus mirror the biphasic physiological response to estrogen. Mol. Endocrinol. 17, 20702083. Hewitt, S.C., Collins, J., Grissom, S., Deroo, B., Korach, K.S., 2005. Global uterine genomics in vivo: microarray evaluation of the estrogen receptor {alpha}-growth factor cross talk mechanism. Mol. Endocrinol. 19, 657668.

mice following exposure to ethynyl estradiol. Carcinogenesis 25, 12771291. Bulayeva, N.N., Watson, C.S., 2004. Xenoestrogen-induced ERK-1 and ERK-2 activation via multiple membrane-initiated signaling pathways. Environ. Health Perspect. 112, 14811487. Charpentier, A.H., Bednarek, A.K., Daniel, R.L., Hawkins, K.A., Lain, K.J., Gaddis, S., MacLeod, M.C., Aldaz, C.M., 2000. Effects of estrogen on global gene expression: identication of novel targets of estrogen action. Cancer Res. 60, 59775983. Churchill, G.A., 2004. Using ANOVA to analyze microarray data. Biotechniques 37, 173175. Ciana, P., Raviscioni, M., Mussi, P., Vegeto, E., Que, I., Parker, M.G., Lowik, C., Maggi, A., 2003. In vivo imaging of transcriptionally active estrogen receptors. Nat. Med. 9, 8286. Clark, J.H., Mani, S.K., 1994. Actions of ovarian steroid hormones. In: Knobil, E., Neill, J. (Eds.), The Physiology of Reproduction, vol. 1, 2nd ed. Raven Press, New York, pp. 10111059. Cooke, P.S., Buchanan, D.L., Young, P., Setiawan, T., Brody, J., Korach, K.S., Taylor, J., Lubahn, D.B., Cunha, G.R., 1997. Stromal estrogen receptors mediate mitogenic effects of estradiol on uterine epithelium. Proc. Natl. Acad. Sci. U.S.A. 94, 6535 6540. Couse, J.F., Korach, K.S., 2004. Estrogen receptor-alpha mediates the detrimental effects of neonatal diethylstilbestrol (DES) exposure in the murine reproductive tract. Toxicology 205, 5563. Currie, R.A., Orphanides, G., Moggs, J.G., 2005. Mapping the molecular responses to xenoestrogens through gene ontology and pathway analysis of toxicogenomic data. Reprod. Toxicol., in press. Dahlquist, K.D., Salomonis, N., Vranizan, K., Lawlor, S.C., Conklin, B.R., 2002. GenMAPP, a new tool for viewing and analyzing microarray data on biological pathways. Nat. Genet. 31, 1920. Dean, N.M., Kanemitsu, M., Boynton, A.L., 1989. Effects of the tyrosine-kinase inhibitor genistein on DNA synthesis and phospholipid-derived second messenger generation in mouse 10T1/2 broblasts and rat liver T51B cells. Biochem. Biophys. Res. Commun. 165, 795801. Deavall, D.G., Moggs, J.G., Orphanides, G., 2005. Toxicogenomics applied to endocrine disruption. In: Borlak, J. (Ed.), Handbook of Toxicogenomics: A Strategic View of Current Research and Applications, vol. 17. WileyVCH Verlag GmbH & Co., KGaA, Weinheim, pp. 415435. Dechering, K., Boersma, C., Mosselman, S., 2000. Estrogen receptors a and b: two receptors of a kind? Curr. Med. Chem. 7, 561576. DeNardo, D.G., Kim, H.T., Hilsenbeck, S., Cuba, V., Tsimelzon, A., Brown, P.H., 2005. Global gene expression analysis of estrogen receptor transcription factor cross talk in breast cancer: identication of estrogen-induced/activator protein-1-dependent genes. Mol. Endocrinol. 19, 362378. Dey, S.K., Lim, H., Das, S.K., Reese, J., Paria, B.C., Daikoku, T., Wang, H., 2004. Molecular cues to implantation. Endocr. Rev. 25, 341373. Doniger, S.W., Salomonis, N., Dahlquist, K.D., Vranizan, K., Lawlor, S.C., Conklin, B.R., 2003. MAPPFinder: using gene ontology and GenMAPP to create a global gene-expression prole from microarray data. Genome Biol. 4, R7.

J.G. Moggs / Toxicology 213 (2005) 177193 Hodges, L.C., Cook, J.D., Lobenhofer, E.K., Li, L., Bennett, L., Bushel, P.R., Aldaz, C.M., Afshari, C.A., Walker, C.L., 2003. Tamoxifen functions as a molecular agonist inducing cell cycleassociated genes in breast cancer cells. Mol. Cancer Res. 1, 300311. Ho Hong, S., Young Nah, H., Yoon Lee, J., Chan Gye, M., Hoon Kim, C., Kyoo Kim, M., 2004. Analysis of estrogen-regulated genes in mouse uterus using cDNA microarray and laser capture microdissection. J. Endocrinol. 181, 157167. Huang, W.W., Yin, Y., Bi, Q., Chiang, T.C., Garner, N., Vuoristo, J., McLachlan, J.A., Ma, L., 2005. Developmental diethylstilbestrol exposure alters genetic pathways of uterine cytodifferentiation. Mol. Endocrinol. 19, 669682. Inoue, A., Yoshida, N., Omoto, Y., Oguchi, S., Yamori, T., Kiyama, R., Hayashi, S., 2002. Development of cDNA microarray for expression proling of estrogen-responsive genes. J. Mol. Endocrinol. 29, 175192. Ise, R., Han, D., Takahashi, Y., Terasaka, S., Inoue, A., Tanji, M., Kiyama, R., 2005. Expression proling of the estrogen responsive genes in response to phytoestrogens using a customized DNA microarray. FEBS Lett. 579, 17321740. Khalyfa, A., Klinge, C.M., Hall, W.C., Zhao, X., Miller, M.M., Wang, E., 2003. Transcription proling of estrogen target genes in young and old mouse uterus. Exp. Gerontol. 38, 10871099. Kian Tee, M., Rogatsky, I., Tzagarakis-Foster, C., Cvoro, A., An, J., Christy, R.J., Yamamoto, K.R., Leitman, D.C., 2003. Estradiol and selective estrogen receptor modulators differentially regulate target genes with estrogen receptors alpha and beta. Mol. Biol. Cell. 15, 12621272. Klinge, C.M., 2001. Estrogen receptor interaction with estrogen response elements. Nucl. Acids Res. 9, 29052919. Kuiper, G.G., Carlsson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S., Gustafsson, J.A., 1997. Comparison of the ligand binding specicity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 138, 863 870. Kuiper, G.G., Lemmen, J.G., Carlsson, B., Corton, J.C., Safe, S.H., van der Saag, P.T., van der Burg, B., Gustafsson, J.A., 1998. Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 139, 42524263. Kushner, P.J., Agard, D.A., Greene, G.L., Scanlan, T.S., Shiau, A.K., Uht, R.M., Webb, P., 2000. Estrogen receptor pathways to AP-1. J. Steroid Biochem. Mol. Biol. 74, 311317. Lazennec, G., Katzenellenbogen, B.S., 1999. Expression of human estrogen receptor using an efcient adenoviral gene delivery system is able to restore hormone-dependent features to estrogen receptor-negative breast carcinoma cells. Mol. Cell. Endocrinol. 149, 93105. Levenson, A.S., Jordan, C., 1994. Transfection of human estrogen receptor (ER) cDNA into ER-negative mammalian cell lines. J. Steroid Biochem. Mol. Biol. 51, 229239. Levenson, A.S., Svoboda, K.M., Pease, K.M., Kaiser, S.A., Chen, B., Simons, L.A., Jovanovic, B.D., Dyck, P.A., Jordan, V.C., 2002. Gene expression proles with activation of the estrogen receptor alpha-selective estrogen receptor modulator complex in breast cancer cells expressing wild-type estrogen receptor. Cancer Res. 62, 44194426.

191

Li, S., Hursting, S.D., Davis, B.J., McLachlan, J.A., Barrett, J.C., 2003. Environmental exposure, DNA methylation, and gene regulation: lessons from diethylstilbesterol-induced cancers. Ann. N. Y. Acad. Sci. 983, 161169. Lin, C.Y., Strom, A., Vega, V.B., Kong, S.L., Yeo, A.L., Thomsen, J.S., Chan, W.C., Doray, B., Bangarusamy, D.K., Ramasamy, A., Vergara, L.A., Tang, S., Chong, A., Bajic, V.B., Miller, L.D., Gustafsson, J.A., Liu, E.T., 2004. Discovery of estrogen receptor alpha target genes and response elements in breast tumor cells. Genome Biol. 5, R66. Lobenhofer, E.K., Bennett, L., Cable, P.L., Li, L., Bushel, P.R., Afshari, C.A., 2002. Regulation of DNA replication fork genes by 17beta-estradiol. Mol. Endocrinol. 16, 12151229. Lubahn, D.B., Moyer, J.S., Golding, T.S., Couse, J.F., Korach, K.S., Smithies, O., 1993. Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc. Natl. Acad. Sci. U.S.A. 90, 1116211166. Metivier, R., Penot, G., Flouriot, G., Pakdel, F., 2001. Synergism between ERalpha transactivation function 1 (AF-1) and AF2 mediated by steroid receptor coactivator protein-1: requirement for the AF-1 alpha-helical core and for a direct interaction between the N- and C-terminal domains. Mol. Endocrinol. 15, 19531970. Metivier, R., Penot, G., Hubner, M.R., Reid, G., Brand, H., Kos, M., Gannono, F., 2003. Estrogen receptor-alpha directs ordered, cyclical, and combinatorial recruitment of cofactors on a natural target promoter. Cell 115, 751763. McKenna, N.J., OMalley, B.W., 2002. Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108, 465474. Moggs, J.G., Orphanides, G., 2001. Estrogen receptors: orchestrators of pleiotropic cellular responses. EMBO Rep. 2, 775781. Moggs, J.G., Deavall, D.G., Orphanides, G., 2003. Sex steroids, ANGELS and osteoporosis. Bioessays 25, 195199. Moggs, J.G., Tinwell, H., Spurway, T., Chang, H.-S., Lim, F.L., Moore, D.J., Pate, I., Soames, A.R., Stuckey, R., Currie, R.A., Zhu, T., Kimber, I., Ashby, J., Orphanides, G., 2004a. Phenotypic anchoring of gene expression changes during estrogen-induced uterine growth. Environ. Health Perspect. 112, 15891606. Moggs, J.G., Ashby, J., Tinwell, H., Lim, F.L., Moore, D.J., Kimber, I., Orphanides, G., 2004b. The need to decide if all estrogens are intrinsically similar. Environ. Health Perspect. 112, 11371142. Moggs, J.G., Murphy, T.C., Lim, F.L., Moore, D.J., Stuckey, R., Antrobus, K., Kimber, I., Orphanides, G., 2005. Antiproliferative effect of estrogen in breast cancer cells that reexpress ERa is mediated by aberrant regulation of cell cycle genes. J. Mol. Endocrinol. 34, 535551. Moore, D.J., Lim, F.L., Tinwell, H., Kimber, I., Ashby, J., Orphanides, G., Moggs, J.G., 2005. Transcript proling of gene expression changes associated with an increased incidence of cancer after exposure of neonatal mice to xenoestrogens. Toxicol. Sci. 84, 1739, Toxicologist. Mueller, S.O., Simon, S., Chae, K., Metzler, M., Korach, K.S., 2004. Phytoestrogens and their human metabolites show distinct agonistic and antagonistic properties on estrogen receptor alpha (ERalpha) and ERbeta in human cells. Toxicol. Sci. 80, 1425.

192

J.G. Moggs / Toxicology 213 (2005) 177193 Riggs, B.L., Hartmann, L.C., 2003. Selective estrogen-receptor modulatorsmechanisms of action and application to clinical practice. N. Engl. J. Med. 348, 618629. Routledge, E.J., White, R., Parker, M.G., Sumpter, J.P., 2000. Differential effects of xenoestrogens on coactivator recruitment by estrogen receptor (ER) alpha and ERbeta. J. Biol. Chem. 275, 3598635993. Safe, S.H., Pallaroni, L., Yoon, K., Gaido, K., Ross, S., McDonnell, D., 2002. Problems for risk assessment of endocrine-active estrogenic compounds. Environ. Health Perspect. 110, 925929. Shang, Y., Brown, M., 2002. Molecular determinants for the tissue specicity of SERMs. Science 295, 24652468. Singleton, D.W., Khan, S.A., 2003. Xenoestrogen exposure and mechanisms of endocrine disruption. Front. Biosci. 8, s110s118. Slonim, D.K., 2002. From patterns to pathways: gene expression data analysis comes of age. Nat. Genet. 32 (Suppl.), 502508. Soulez, M., Parker, M.G., 2001. Identication of novel oestrogen receptor target genes in human ZR75-1 breast cancer cells by expression proling. J. Mol. Endocrinol. 27, 259274. Stossi, F., Barnett, D.H., Frasor, J., Komm, B., Lyttle, C.R., Katzenellenbogen, B.S., 2004. Transcriptional proling of estrogenregulated gene expression via estrogen receptor (ER) alpha or ERbeta in human osteosarcoma cells: distinct and common target genes for these receptors. Endocrinology 145, 34733486. Tan, Y.F., Li, F.X., Piao, Y.S., Sun, X.Y., Wang, Y.L., 2003. Global gene proling analysis of mouse uterus during the oestrous cycle. Reproduction 126, 171182. Tennant, R.W., 2002. The National Center for Toxicogenomics: using new technologies to inform mechanistic toxicology. Environ. Health Perspect. 110, 8. Terasaka, S., Aita, Y., Inoue, A., Hayashi, S., Nishigaki, M., Aoyagi, K., Sasaki, H., Wada-Kiyama, Y., Sakuma, Y., Akaba, S., Tanaka, J., Sone, H., Yonemoto, J., Tanji, M., Kiyama, R., 2004. Using a customized DNA microarray for expression proling of the estrogen-responsive genes to evaluate estrogen activity among natural estrogens and industrial chemicals. Environ. Health Perspect. 112, 773781. Tremblay, G.B., Giguere, V., 2002. Coregulators of estrogen receptor action. Crit. Rev. Eukaryot. Gene Exp. 12, 122. Wang, D.Y., Fulthorpe, R., Liss, S.N., Edwards, E.A., 2004. Identication of estrogen-responsive genes by complementary deoxyribonucleic acid microarray and characterization of a novel early estrogen-induced gene: EEIG1. Mol. Endocrinol. 18, 402 411. Watanabe, H., Suzuki, A., Mizutani, T., Khono, S., Lubahn, D.B., Handa, H., Iguchi, T., 2002. Genome-wide analysis of changes in early gene expression induced by oestrogen. Genes Cells 7, 497507. Watanabe, H., Suzuki, A., Kobayashi, M., Lubahn, D.B., Handa, H., Iguchi, T., 2003a. Similarities and differences in uterine gene expression patterns caused by treatment with physiological and non-physiological estrogens. J. Mol. Endocrinol. 3, 487497. Watanabe, H., Suzuki, A., Kobayashi, M., Takahashi, E., Itamoto, M., Lubahn, D.B., Handa, H., Iguchi, T., 2003b. Analysis of temporal changes in the expression of estrogen-regulated genes in the uterus. J. Mol. Endocrinol. 30, 347358.

Naciff, J.M., Jump, M.L., Torontali, S.M., Carr, G.J., Tiesman, J.P., Overmann, G.J., Daston, G.P., 2002. Gene expression prole induced by 17alpha-ethynyl estradiol, bisphenol A, and genistein in the developing female reproductive system of the rat. Toxicol. Sci. 68, 184199. Naciff, J.M., Overmann, G.J., Torontali, S.M., Carr, G.J., Tiesman, J.P., Richardson, B.D., Daston, G., 2003. Gene expression prole induced by 17 alpha-ethynyl estradiol in the prepubertal female reproductive system of the rat. Toxicol. Sci. 72, 314 330. Naciff, J.M., Daston, G.P., 2004. Toxicogenomic approach to endocrine disrupters: identication of a transcript prole characteristic of chemicals with estrogenic activity. Toxicol. Pathol. 32, 5970. Newbold, R.R., Bullock, B.C., McLachlan, J., 1990. Uterine adenocarcinoma in mice following developmental treatment with estrogens: a model for hormonal carcinogenesis. Cancer Res. 50, 76777681. Newbold, R.R., Padilla Banks, E., Bullock, B., Jefferson, W.N., 2001. Uterine adenocarcinoma in mice treated neonatally with genistein. Cancer Res. 61, 43254328. Nilsson, S., Gustafsson, J.-A., 2002. Estrogen receptor action. Crit. Rev. Biochem. Mol. Biol. 37, 128. Nishikawa, J., Saito, K., Goto, J., Dakeyama, F., Matsuo, M., Nishihara, T., 1999. New screening methods for chemicals with hormonal activities using interaction of nuclear hormone receptor with coactivator. Toxicol. Appl. Pharmacol. 154, 76 83. Okura, A., Arakawa, H., Oka, H., Yoshinari, T., Monden, Y., 1988. Effect of genistein on topoisomerase activity and on the growth of [Val 12]Ha-ras-transformed NIH 3T3 cells. Biochem. Biophys. Res. Commun. 157, 183189. Orlando, V., 2000. Mapping chromosomal proteins in vivo by formaldehyde-crosslinked-chromatin immunoprecipitation. Trends Biochem. Sci. 25, 99104. Owens, J.W., Ashby, J., 2002. Critical review and evaluation of the uterotrophic bioassay for the identication of possible estrogen agonists and antagonists: in support of the validation of the OECD uterotrophic protocols for the laboratory rodent. Organisation for Economic Co-operation and Development. Crit. Rev. Toxicol. 32, 445520. Paules, R., 2003. Phenotypic anchoring: linking cause and effect. Environ. Health Perspect. 111, A338A339. Potier, B., Rovira, C., 1999. Protein tyrosine kinase inhibitors reduce high-voltage activating calcium currents in CA1 pyramidal neurones from rat hippocampal slices. Brain Res. 816, 587597. Ramanathan, L., Gray, W.G., 2003. Identication and characterization of a phytoestrogen-specic gene from the MCF-7 human breast cancer cell. Toxicol. Appl. Pharmacol. 91, 107117. Reese, J., Das, S.K., Paria, B.C., Lim, H., Song, H., Matsumoto, H., Knudtson, K.L., DuBois, R.N., Dey, S.K., 2001. Global gene expression analysis to identify molecular markers of uterine receptivity and embryo implantation. J. Biol. Chem. 276, 4413744145. Revankar, C.M., Cimino, D.F., Sklar, L.A., Arterburn, J.B., Prossnitz, E.R., 2005. A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science 307, 16251630.

J.G. Moggs / Toxicology 213 (2005) 177193 Watanabe, H., Suzuki, A., Goto, M., Lubahn, D.B., Handa, H., Iguchi, T., 2004. Tissue-specic estrogenic and non-estrogenic effects of a xenoestrogen, nonylphenol. J. Mol. Endocrinol. 33, 243252. Waters, M.D., Olden, K., Tennant, R.W., 2003. Toxicogenomic approach for assessing toxicant-related disease. Mutat. Res. 544, 415424. Waters, M.D., Fostel, J.M., 2004. Toxicogenomics and systems toxicology: aims and prospects. Nat. Rev. Genet. 5, 936948.

193

Weihua, Z., Saji, S., Makinen, S., Cheng, G., Jensen, E.V., Warner, M., Gustafsson, J.A., 2000. Estrogen receptor (ER) beta, a modulator of ERalpha in the uterus. Proc. Natl. Acad. Sci. U.S.A. 97, 59365941. Williams, G.M., Iatropoulos, M., Cheung, R., Radi, L., Wang, C.X., 1993. Diethylstilbestrol liver carcinogenicity and modication of DNA in rats. Cancer Lett. 68, 193198.