ToxSci Advance Access originally published online on July 11, 2006
Toxicological Sciences 2006 93(2):298-310; doi:10.1093/toxsci/kfl057
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Expression Profiling of Endocrine-Disrupting Compounds Using a Customized Cyprinus carpio cDNA Microarray
* Department of Biology, Laboratory for Ecophysiology, Biochemistry and Toxicology, and Department of Mathematics and Computer Science, Intelligent Systems Laboratory, University of Antwerp, B-2020 Antwerp, Belgium; and Department of Biomedical Sciences, Laboratory for Molecular Genetics, University of Antwerp, B-2610 Antwerp (Wilrijk), Belgium
1 To whom correspondence should be addressed at Department of Biology Laboratory for Ecophysiology, Biochemistry and Toxicology, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, Belgium. Fax: +(32) 3-265 34 97. E-mail: lotte.moens{at}ua.ac.be.
Received April 25, 2006; accepted June 29, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Exposure to a variety of anthropogenic compounds has been shown to interfere with normal development, physiology, and reproduction in a wide range of organisms, both in laboratory studies and wildlife. We have developed a Cyprinus carpio cDNA microarray consisting of endocrine-related genes. In the current study, we investigated the applicability of this microarray (1) to study the molecular effects induced by exposure to a variety of endocrine-disrupting compounds (EDCs) in fish and (2) to discriminate the specific transcriptional profiles associated with these compounds. To that purpose, gene expression profiles were generated in livers of juvenile carp exposed to 14 Organization of Economical Cooperation Development (OECD)-recommended reference EDCs (17beta-estradiol, 17alpha-ethinylestradiol, 4-nonylphenol, bisphenol A, tamoxifen, 17alpha-methyltestosterone, 11-ketotestosterone, dibutyl phthalate, flutamide, vinclozolin, hydrocortisone, CuCl2, propylthiouracil, and a mixture of L-triiodothyronine and L-thyroxine). Our results show that, in addition to some expression similarities between analogous acting substances, each individual compound produced its own unique expression pattern on the array, distinct from the profiles generated by the other compounds. In addition, we were able to identify a minimal subset of genes, which also allowed to discriminate between the different compounds. Overall, our findings suggest that the developed cDNA array has great promise to screen new and existing chemicals on their endocrine-disruptive potential and to identify distinct classes of EDCs.
Key Words: endocrine disruption; gene expression profiling; microarray; fish; classification.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Today, ecotoxicologists worldwide are faced with the enormous challenge of evaluating the toxicity and mode of action of chemicals and waste discharges ending up in the (aquatic) environment. Especially, endocrine-disruptive effects are hard to characterize due to the intrinsic complex nature of the endocrine system and the limited fundamental insights in these pathways in aquatic organisms. As a result, a significant majority of the chemicals, which are now in commercial use (
87,000), have not been sufficiently tested for potential endocrine-disrupting effects (EDSTAC, 1998
). The development of proficient in vitro and in vivo screening tools is, thus, of vital importance to evaluate the adverse effects on a system as complex as the endocrine metabolism.
Thanks to recent advances in functional genomics, powerful platforms have become available for unraveling the spectrum of signaling pathways involved in the endocrine system. The emerging field of "toxicogenomics," which is an integration of genomic techniques as well as bioinformatics in the research area of toxicology, could be very helpful not only to identify mechanisms of action of endocrine-disrupting compounds (EDCs) but also to help predict potential actions of newly synthesized chemicals on the basis of similarities of expression profiles to known toxicants. Several recent publications have described the use of microarray analysis to identify gene expression changes associated with xenobiotic exposure (Terasaka et al., 2004; Van der Ven et al., 2005; Williams et al., 2003). In addition, research has shown that compounds associated with similar mechanisms of toxicity also yield similar gene expression profiles that are distinct from profiles generated by other classes of chemicals (Bartosiewicz et al., 2001; Hamadeh et al., 2002a,b; Thomas et al., 2001; Waring et al., 2001a,b). To date, gene expression profiling has widely been used for discriminating between various types of hepatotoxicants, but its application for characterizing different classes of endocrine disruptors is relatively new.
Especially in aquatic organisms, the use of molecular techniques for the evaluation of endocrine-disruptive effects is still in its infancy. Although several recent publications (Hoyt et al., 2003; Larkin et al., 2002, 2003) have used the array technology to study the effects of EDCs in fish, most of the traditional studies in this area have focused on expression changes of either individual gene transcripts (e.g., vitellogenin) or a very limited set of specific genes. Furthermore, the emphasis was mostly put on estrogenic responses, whereas interference of xenobiotics with other hormonal pathways—such as the androgenic, thyroidal, and corticosteroidal pathways—remains largely unexplored. This is not surprising, given the substantial difficulty in pinpointing the most relevant genes to consider in order to create a comprehensive screening method for endocrine-disruptive effects. If the various hormonal signaling pathways involved are not known or only poorly understood, which genes should then be studied? Ideally, one would like to analyze all genes that have the potential to be transcriptionally regulated during an endocrine-modulating response. A possible way to meet this ambitious goal to a certain extent is by constructing cDNA libraries enriched for genes that are regulated by a wide range of endocrine modulators. A few research groups, as well as ours, have made a start on this process, by using a variety of methods including Differential Display RT-PCR, shotgun sequencing of cDNAs from estradiol-treated fish, PCR amplification of target genes of the endocrine system (Bowman et al., 2002; Denslow et al., 2001; Larkin et al., 2002), and Suppression Subtractive Hybridization PCR (SSH) of hormone-responsive and gender-related genes (Moens et al., in press a,b). The resulting gene sets showed to be very useful for detecting gene expression changes in fish exposed to EDCs (Larkin et al., 2002, 2003; Moens et al., in press a) and to characterize molecular differences between male and female fish in several tissues (Moens et al., in press b). Until now they have hardly been applied for discriminating between different classes of EDCs.
In the present study, we used a previously constructed custom cDNA microarray, consisting of 960 hormone-responsive and gender-associated gene fragments from common carp (Cyprinus carpio), to analyze gene expression profiles generated in the liver of juvenile carp exposed to 14 OECD-recommended reference EDCs with various mechanisms of action. These compounds can roughly be divided into six main groups: estrogen agonists and antagonists, androgen agonists and antagonists, modulators of the cortisol pathway, and modulators of the thyroid hormone pathway.
The objectives of this study were to analyze changes in gene expression levels induced by exposure to these 14 model compounds and to determine if the resulting gene signatures allow a discrimination—or classification—of compound-associated profiles. This study demonstrates the potential of this specific cDNA array to screen chemicals with an unknown mode of action on their endocrine-disruptive potential.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Chemicals, exposure, and preparation of liver RNA.
Fish were treated with 14 OECD-recommended reference chemicals, with different mechanisms of action (Table 1). All chemicals were obtained from Sigma Aldrich (Bornem, Belgium), except 17alpha-methyltestosterone (MT) and 11-ketotestosterone (11KT), which were purchased from Steraloids Inc. (Newport, RI).
|
Juvenile fish were obtained from the University of Wageningen (the Netherlands). Fish were acclimated for 3 weeks in aerated OECD water prior to treatment. The fish were exposed to a 14:10 h light:dark photoperiod and fed commercial feed (Antwerp Aquaria, Aartselaar, Belgium) at a ratio of 2% of the maximum body weight (estimated at 1.2 g). Two dosing methods were applied; in a first series of experiments, fish received an aqueous exposure to three concentrations (A1, A2, and A3) of each substance, in parallel with a control (A0), using a semistatic experimental setup (renewal at 48 h). After 24 and 96 h of exposure, liver tissue was dissected from 10 fish per exposure condition. Additionally, in a second series of experiments, fish were ip injected with a single dose of each compound and dissected 24 h postinjection. Control fish received an ip injection of the vehicle control, without any chemical. Concentrations for both delivery methods were based on literature data and are summarized in Table 1.
RNA was prepared using to the Totally RNA Isolation kit (Ambion, Austin, TX) and followed by a DNase treatment (Fermentas, St. Leon-Rot, Germany) and a phenol/chloroform extraction. The concentration was determined by spectrophotometry, and RNA integrity was assessed by denaturing formamide-agarose gel-electrophoresis.
Fluorescent target labeling and microarray hybridizations.
For the aqueous exposure experiments, hybridization series were performed using a loop design, by which pairs of labeled cDNA from the three exposure concentrations of each compound were hybridized in the following way: A0-A1, A2-A1, A2-A3, and A0-A3. Using this strategy, a replicate hybridization is built-in for every exposure condition, creating more reliable data sets. Hybridizations for the ip injection experiments were carried out in duple. The cDNA microarrays were prepared as described in Van der Ven et al. (2005) by mechanical spotting of a collection of 960 gene fragments, previously isolated by SSH, and specifically enriched for hormone-responsive and gender-related genes (Moens et al., in press a,b). A list of the SSH-derived gene fragments showing significant homology to National Center for Biotechnology Information sequences is available as supplementary data. Each cDNA clone was spotted four times on the slides, allowing replicate analyses. In addition, to assist the evaluation of the quality of the array data, a set of artificial control genes (designed from yeast intergenic regions) was spotted in 36 replicates (Lucidea Universal ScoreCard, Amersham Biosciences, Roosendaal, the Netherlands) on each array: 10 calibration controls, 8 ratio controls, and 2 negative controls. These spiked-in–labeled controls, premixed at known concentrations and ratios, provide information on cDNA-labeling efficiency, sensitivity, and intra-array variability of replicates. Negative controls were included to assess nonspecific hybridization and quality of blocking at the prehybridization step. Fluorescent target labeling and hybridization protocols were identical to those reported in Van der Ven et al. (2005). In short, duplicate-labeled cDNA targets were prepared by converting 7 µg DNase-treated total RNA into aminoallyl-dUTP–labeled cDNA using the Superscript II Reverse transcriptase kit (Invitrogen, Paisly, UK). Lucidea reference and test mRNA spikes, corresponding to the calibration and ratio controls spotted on the arrays, were added to the RNA samples from reference and test populations, respectively. RNA was hydrolyzed, and unincorporated nucleotides were removed (QiaQuick PCR purification kit, Qiagen, Crawley, UK). The aminoallyl-labeled cDNA samples were then covalently coupled to Cy3- (for A0 and A2 cDNA populations) or Cy5- (for A1 and A3 cDNA populations) esters (Amersham Biosciences). Reaction mixtures were purified once more, and the labeling efficiency was determined by spectrophotometry. The threshold for optimal dye incorporation was 150 pmol, and a frequency of incorporation of 20–50% was considered appropriate for hybridizations.
Following analysis of incorporation, the fluorescently labeled samples were dried to completion in a vacuum centrifuge and redissolved in hybridization solution. Microarray slides were prehybridized, and the target cDNA solution was denatured and added to the slides. Hybridization took place overnight (1600 h–1800 h) at 42°C. After hybridization, slides were washed and dried with N2. Finally, slides were scanned using the Genepix Personal 4100A scanner (Axon Instruments, Union City, CA).
Data acquisition, preprocessing, and detection of differential expression.
Scanned images were analyzed using Genepix pro 4.1 software (Axon Instruments) for spot identification and quantification of raw fore- and background intensities of the spots.
Raw data were background corrected and normalized using the Variance Stabilization and Normalization method in R (Huber et al., 2002). This method incorporates data normalization, a model for the dependence of the variance of the mean intensity, and a variance-stabilizing data transformation. Subsequently, a linear model was used to compute the contrasts A0-A1, A0-A2, and A0-A3 from the loops A0-A1, A2-A1, A2-A3, and A0-A3, for each of the aqueous exposure experiments.
To determine differential gene expression, an empirical Bayes test (a moderated t test) was run, at a p value of 0.05. Additionally, differentially expressed genes need to have an absolute M value > 1 (twofold change) and an A value > 8 (with M = log2(Cy5/Cy3) and A = log2 (Cy3 x Cy5)1/2). For a gene to be "zero" (unaffected by the exposure), its A should still be > 8, and its absolute M value < 0.3 (fold change 1.23). Using this combination of a cutoff fold change value and a statistical significance test as a criterion for differential gene expression, both the biological relevance and a more statistically sound approach were taken into account.
Clustering analysis and discriminatory gene selection.
Prior to hierarchical cluster analysis, data sets were processed in the following way: for every compound, the expression value corresponding to the most effective treatment condition (i.e., the ip injection experiment or one of the different water exposure conditions) was selected for each individual gene. This means that it is possible that for one gene the most effective expression value was taken from, for example, the low concentration exposure, after 24 h, whereas for another gene the high exposure level at 96 h was more effective. By using this strategy, every single differentially expressed gene resulting from the different exposure conditions is included in the analyses, so that potential concentration- and/or time-related differences in effects can be taken into account.
Hierarchical cluster analyses were performed based on the (uncentered) Pearson correlation similarity metric (Eisen et al., 1998), combined with a complete linkage-clustering algorithm, using the Acuity 4.0 software (Axon Instruments).
To build a diagnostic gene set, we aimed to find a subset of genes—out of the collection of differentially expressed fragments—that discriminates the compounds in an optimal way. To that purpose, a quaternary value was assigned to each gene: (1) "overexpressed" (when M was strongly positive), (2) "underexpressed" (when M was strongly negative), (3) "not expressed" (M around 0), or (4) "unclear" (M intermediate). The expression of two genes was termed sufficiently different if
- both gene A and gene B have consistent behavior, that is, the genes cannot have been measured overexpressed in one condition (exposure time, concentration) and underexpressed in another and
- one of the following occurs:
- gene A is overexpressed in some conditions and gene B underexpressed in some conditions, or vice versa
- gene A is over- or underexpressed in some conditions, and gene B is not expressed in some conditions, or vice versa.
- gene A is overexpressed in some conditions and gene B underexpressed in some conditions, or vice versa
The distance between two compounds was then calculated as the number of genes whose expression differed sufficiently between the compounds. Using a generational genetic algorithm, subsets of 8–12 genes were selected such that the total number of pairwise small compound distances < 3 was minimal. The stop criterion was 100 generations without improvement. Population size was 20, mutation was applied on a per-individual basis with probability 0.5, crossover was also applied with probability 0.5. Fifty independent runs were replicated. The resulting gene sets each corresponded to a distance matrix summing up all the distances for the different compound combinations.
Hierarchical clustering of the expression data from the discriminatory gene set was performed as described above (using Pearson's uncentered clustering analysis).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Gene Expression Analysis
In order to determine the toxicant-induced gene expression changes, liver RNA was collected 24 and 96 h following aqueous exposure to the compounds or 24 h postinjection. Competitive hybridizations of fluorescently labeled cDNA targets derived from control versus treated livers were used to measure relative abundance of the probes on a custom carp microarray, containing a cDNA collection enriched for hormone-responsive and gender-related genes (Moens et al., in press a,b). For each exposure condition, duplicate hybridizations were performed. We conducted statistical analysis of the microarray data and determined differential expression using an empirical Bayes test (p < 0.05), combined with a cutoff fold change of 2. The quality of individual microarrays was evaluated using the Lucidea ScoreCard references, as described in Moens et al. (in press b). All microarrays included in the analysis fulfilled our prerequisite quality parameters. More details on experimental system validation using the Lucidea control genes (for a representative array) are given as supplementary data. All treatments caused transcriptional changes with respect to their corresponding time-matched controls. It was noted that for a few compounds only a small number of gene expression responses were obtained for some of the treatment conditions, which was probably due to a suboptimal concentration range, exposure time, or exposure route. However, in those cases, data collected from the two alternative dosing methods complemented each other very well, such that for every compound an informative gene expression data set was obtained. In total, the two exposure series resulted in 267 different genes that were significantly altered by at least one of the compounds. A list of expression data of the differentially expressed gene fragments is available as supplementary data. In Table 2, the number of regulated genes for each compound and some examples of relevant gene names and their corresponding biological function are listed. These include the induction of vitellogenin by all the estrogens and MT, the inhibition of several transferrin transcripts by bisphenol A (BPA), 17alpha-ethinylestradiol (EE2), and flutamide (FLUT), and the stimulation of cytochrome c oxidase genes by 4-nonylphenol (NP) and cortisol (hydrocortisone [CORT]).
|
Clustering Analysis and Discriminatory Gene Selection
In the present study we aimed to investigate if our developed cDNA microarray may be used to reveal chemical-specific signature patterns. To that end, we first subjected the complete gene expression data set to a two-dimensional hierarchical clustering (Eisen et al., 1998
). For every compound, the expression value corresponding to the most effective treatment condition was selected for each gene. Both transcripts and compounds were then clustered on the basis of similarity, and the resulting relationships between compounds, based on gene expression profiles, and vice versa, are highlighted in a dendrogram (Fig. 1). A visual inspection of the compound dendrogram indicates that our microarray has the potential to distinguish RNA samples derived from fish exposed to the different chemicals. There are both some striking expression similarities and dissimilarities between the compounds. The estrogenic compounds (17beta-estradiol [E2], NP, BPA, and EE2), for example, are well grouped together, and MT, an androgen which was shown to be metabolically converted into estrogen (Hornung et al., 2004), is also found in the "estrogen subnode." Yet, in spite of some marked expression similarities between these compounds, it is clear that each individual member of the estrogen class has a unique gene expression pattern, distinct from the other compounds.
|
Next to the estrogens, the cortisol modulators (CORT and copper chloride [CuCl2]) are also clustered together, although the correlation is not very strong. Further, dibutyl phthalate (DBP), FLUT, and thyroid hormone (L-triiodothyronine–L-thyroxine [T3-T4]) were found to be quite closely related, and also, tamoxifen (TAM) and propylthiouracil (PTU) share some similarities.
A close-up view of some of the major expression analogies between the different (groups of) compounds is given in Figures 1A–1D.
In a second phase, we investigated whether a small set of informative genes could be identified, whose expression pattern is sufficient to discriminate between compounds. Therefore, distances (i.e., the number of genes whose expression differs sufficiently) between the compounds were calculated and subjected to a genetic algorithm, which enabled the selection of subsets of genes that separated the compounds in an optimal way. Figure 2 shows the distance matrix and transcriptional profile of the gene set with the highest discriminating power between the 14 compounds. The distance matrix, summarizing all pairwise distances between the 14 compounds, as calculated in material and methods, demonstrates that all chemicals can be distinguished from any other chemical by a combination of at least three genes. Smallest distances were found between T3-T4 and DBP and between TAM and PTU. When compounds were hierarchically clustered based on the expression levels of this "discriminatory gene set" (Fig. 2B), the estrogens were clearly clustered together again. Furthermore, CORT and CuCl2 were also grouped, and a node was formed by DBP, T3-T4, and FLUT. Finally, PTU and TAM were also clustered together.
|
It is noted that all the described compound associations based on the expression pattern of this small subset of genes, correspond to those observed after clustering the whole array data set (Fig. 1). This indicates that our discriminatory gene set comprises the most informative genes of the array, making it a valuable tool for describing key gene expression effects resulting from chemical exposure.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Over the last few decades, increasing concern has been raised regarding the discharge, presence, and potential adverse effects of EDCs in the environment. To date, most available literature on endocrine disruption in aquatic organisms focuses on estrogenic responses, while chemical-induced interference with other hormonal signaling pathways received relatively little attention so far. Advances in functional genomics have stimulated the generation of a subdiscipline of (eco)toxicology: (eco)toxicogenomics. These technologies offer great potential to evaluate a variety of endocrine-disruptive effects at the basis of all physiological responses, the molecular level, and their application in aquatic toxicology is now starting to burgeon. In our laboratory, we have developed a custom cDNA microarray for the detection and evaluation of endocrine disruption in common carp (Moens et al., in press a,b). This array, enriched for endocrine relevant carp genes, was now used to determine gene expression profiles in livers of fish exposed to 14 model EDCs, representing six major toxic classes (Table 1). The main goal of our study was to investigate if the developed custom array could discriminate the molecular effects of the different EDCs.
In order to evaluate the usefulness of our custom array in detecting and discriminating EDCs, we selected a relevant set of reference chemicals encompassing a wide range of endocrine-signaling pathways. Our selection of reference compounds was based on recommendations made by OECD and Environmental Protection Agency in their screening programs for in vitro and in vivo endocrine disruptor testing (EDMVS, 2003). In addition, 11KT and cortisol were also included in our set of test chemicals, considering their major importance in androgenic and glucocorticoid pathways in teleost fish, respectively. The postulated modes of action of the other chemicals (Table 1) were adopted from said EDC-screening programs. As for DBP, several authors have described developmental and/or reproductive effects of this compound in a variety of species (Higuchi et al., 2003; Jensen et al., 2001; Kim et al., 2004; Lee et al., 2005), but reports on the mechanism by which these effects occur are quite conflicting. Some authors proposed that DBP has a weak estrogenic effect (Sharpe et al., 1995), based upon in vitro data (Harris et al., 1997; Jobling et al., 1995). However, this effect could not be confirmed in vivo, neither in mammalian (Zacharewski et al., 1998) nor in aquatic (Van den Belt et al., 2003) organisms. Others have ascribed antiandrogenic properties to this compound (Gray et al., 1999; Mylchreest et al., 1998; 1999), and although it is clear that these effects are not caused by interference with the androgen receptor (AR) (Mylchreest et al., 1998; Foster, 2001), the exact mechanistic basis for the endocrine-disrupting properties of DBP yet remains to be determined.
For all compounds, aqueous exposures to three different concentrations (low, intermediate, and high levels), and during two different time points (24 and 96 h), were performed. Concentration levels were selected from literature or from our own personal experience (see Table 1). As for the estrogenic compounds, concentrations were chosen to induce similar levels of the estrogenic biomarker vitellogenin, based on dose-response curves in juvenile rainbow trout and zebra fish (Van den Belt et al., 2003, 2004). Incorporation of a low, intermediate, and high concentration for each chemical exposure, integrates the dose-dependent differences in effects. Also, by evaluating two different time points, both short-term and longer term effects were taken into account. In addition to the aqueous exposures, a series of ip injection experiments were conducted, in which fish received a single dose of each compound, during 24 h. By using two alternative dosing methods, we tried to cope with possible differences in the uptake route of the various compounds. It is evident that the application of more than one delivery method—which is a common exposure strategy in in vivo mammalian toxicology studies (Bulera et al., 2001; Steiner et al., 2004)—results in more comprehensive data sets.
In Table 2, a number of relevant genes, including both some generic and compound-specific effects, are listed. Some examples of marked effects will be discussed below. The upregulation of vitellogenin by the estrogenic compounds (E2, NP, BPA, and EE2) was expected considering its involvement in the estrogen-regulated process of oogenesis in fish (Chen, 1983). The observation that MT (an androgen) also potently induced the expression of this gene is consistent with the estrogenic effects of MT observed in fathead minnow (Ankley et al., 2001; Parrot and Wood, 2002; Zerulla et al., 2002). NP potently enhanced the transcription of cytochrome c oxidase subunits I, II, and III, as did cortisol. The activation of the cytochrome c oxidase transcripts by NP may reflect a nonestrogenic effect of this compound, as none of the other estrogens induced the expression of these genes. Noteworthily, TAM was found to exert some antiestrogenic effects: for example, it repressed the expression level of calmodulin, a calcium-signaling protein, which was shown to be under estrogenic control in current and previous work (Moens et al., in press a), and seems to be involved in the regulation of the transcriptional activity of the estrogen receptor (Li et al., 2005). Furthermore, TAM induced the expression of 14-kDa apolipoprotein and apolipoprotein A-I, which were found to be downregulated by estrogen exposure in both current and earlier studies (Moens et al., in press a). Thyroid hormones (T3 and T4) are known to play a critical role in controlling the metabolic balance in all vertebrates (Oppenheimer et al., 1987). Their most obvious and well-documented action is an increase in basal energy expenditure, including lipid, protein, and carbohydrate metabolism, as well as respiration. Here we observed a distinctive effect of the T3-T4 mixture on protein turnover: both transcripts involved in protein synthesis (such as several genes for ribosomal proteins, the translation initiation factor EF-1-alpha) and protein catabolism (such as carboxypeptidase A, elastase 2, chymotrypsinogen A) were upregulated by T3-T4 exposure. Furthermore, thyroid hormone was also found to have an effect on cellular respiration, by increasing the transcription of cytochrome c oxidase subunit I (but not subunits II and III). Interestingly, PTU, a widely used thyroid inhibitor, was often found to have the opposite effect of T3-T4 on these energy metabolism–related genes.
Figure 1 gives a general overview of the compound-associated transcriptional profile across the whole microarray. Hierarchical cluster analysis of the complete gene expression data set showed a close association in transcriptional responses between the different estrogenic compounds. MT, which was initially selected as a prototypical androgen, was also found to cluster with the estrogenic compounds. As stated before, this kind of effect is in correspondence with the observation made by other authors that MT produces both androgenic and estrogenic effects in fathead minnows (Ankley et al., 2001; Parrot and Wood, 2002; Zerulla et al., 2002). Most likely, the mechanistic basis for the estrogenic effect of this compound is the aromatization of MT to 17-alpha-methylestradiol (Hornung et al., 2004).
Next to its estrogenic properties, MT is expected to exert some androgenic effects as well. However, we do not find a lot of expression analogies between this compound and 11KT, a typical androgen. Furthermore, the antiandrogenic compounds vinclozolin (VINC) and FLUT were not clustered together either; FLUT seems to be more homologous to another presumed antiandrogen, DBP, than to VINC. This is quite surprising, since VINC and FLUT are known to have a quite similar mechanism of action; unlike DBP, they both exert their antiandrogenic effects through an inhibition of AR-binding and AR-dependent transcriptional activity (Kemppainen et al., 1992; Wong et al., 1995). Our expression data thus seem to indicate that FLUT and VINC do have different modes of action, apart from their known interference with the AR. Yet, considering the fact that neither the AR agonists (11KT and MT) nor the AR antagonists (FLUT and VINC) reveal much gene expression analogies on our array, it is possible that the specific (anti)androgen-responsive target genes shared by these compounds are not completely represented on our array. Moreover, androgens and antiandrogens may have effects on different organs within the hypothalamus-pituitary-gonadal axis, rather than on the liver.
Interestingly, it is observed that the cortisol modulators are clustered together. Furthermore, T3-T4, FLUT, and DBP are quite closely related, and also TAM and PTU share some homologies. Figures 1A–1D show a close-up view of the major expression homologies between (groups of) compounds. For example, it seems that the association between PTU and TAM is largely caused by a parallel inhibition of several mitochondrial clones, among others (Fig. 1A). These genes at the same time play an important role in linking cortisol and CuCl2, although in the opposite direction. Moreover, NP and BPA also cause an induction of said gene transcripts, indicating that these compounds and CuCl2 and cortisol, might share some common mechanism of action.
Figure 1B reveals both some striking expression similarities and dissimilarities between the compounds; very close gene expression homologies are observed between some of the estrogens (NP and EE2), cortisol, thyroid hormone, FLUT, and DBP. BPA and VINC, on the other hand, seem to have the opposite effect. Since most of the affected genes in this group are unknown clones, it does not make much sense to try to unravel the observed transcript profiles or to distil mechanisms of action. Nevertheless, we want to point out that said gene expression effects provide an excellent illustration of the potential of microarray experiments to reveal both common (or overlapping) mechanisms between compounds and individual effects; although BPA was found to stimulate the expression of several genes in a similar way as did cortisol and NP (see Fig. 1A), it is clear that these compounds also have unique, compound-specific effects.
In Figure 1C, the analogies between the estrogen agonists are represented. Evidently, the different vitellogenin transcripts displayed the most striking expression homologies. Yet, also glutathione-S-transferase, calmodulin, and a clone similar to ribosomal protein L41 were found to correlate well with this compound class.
Next to these common estrogenic effects, EE2 and BPA exhibit some other striking expression parallelisms (Fig. 1D). These analogies are mainly represented by the inhibition of several complement and transferrin transcripts. As both these genes were shown to have an estrogen response element in their 5'-end region (Mikawa et al., 1996; Vik et al., 1991), their regulation by EE2 and BPA is not unexpected. However, the other estrogenic compounds did not have these effects. Furthermore, it is noted that the expression of the transferrin genes is also inhibited by the antiandrogens FLUT and, to a less extent, VINC, whereas DBP did not affect the expression of these transcripts.
Again, these results indicate that although individual members of each class are assumed to act upon the same hormonal pathways, they may also stimulate other unique pathways that may be secondary to this effect or that may represent a slightly different mechanism of action. This kind of "individuality" of compounds belonging to the same (toxic) class has repeatedly been described in literature. For example, in a study by Naciff et al. (2002), where the expression pattern of three estrogenic compounds (EE2, BPA, and genistein) was compared, it was found that each of these chemicals induced changes in the expression of a set of unique transcripts—in addition to a common pool of genes whose expression was similarly regulated by these three compounds—suggesting that not all compounds with estrogenic activity act alike. Other researchers reported that in the liver of female mouse, NP induced markedly more genes than E2, which indicated that NP could activate a specific set of genes that are distinct from the estrogen-responsive genes (Watanabe et al., 2004), an observation that was also made in the present study. However, it has to be kept in mind that differences in gene expression responses induced by different chemicals could not only be a sign of distinct modes of action of these compounds but also might be attributed to intrinsic dissimilarities in potency, bioavailability, pharmacokinetics, and pharmacodynamics of the compounds. Although the integration of multiple concentrations, time points, and dosing routes in our study is believed to address this issue to some extent, a more extensive survey covering a broad range of concentrations and time points will be needed to thoroughly investigate this matter.
The different examples mentioned above demonstrate that, on the basis of our gene collection, a clear-cut distinction can be made between the different EDCs, even within one class. In a next step, we wanted to investigate if a small subset of genes could be identified that also allows to discriminate between the compounds. Using a generational genetic algorithm, we aimed to select a subset of genes that separated the compounds in an optimal way. It is noted that, during this gene selection process, we focused on maximizing distances between "all" the compounds (also within classes), without taking grouping of compounds into account.
The gene set with the highest discriminating power between the 14 compounds consisted of 12 genes, including some mitochondrial clones, a vitellogenin transcript, calmodulin, and two transferrin variants. Interestingly, these are the same genes that were in the foregoing discussion already found to play an important role in both the expression differences and similarities between the compounds.
In Figure 2A, pairwise distances between the compounds, based on expression data of these 12 diagnostic genes, are summarized. Here, the "distance" between two compounds is defined as "the number of genes whose expression differed sufficiently (i.e., induced vs. reduced or altered vs. not altered) between those compounds." It is noted that all the compounds can be separated by a combination of at least three genes. In fact, only two pairs of compounds were found with a mutual distance of no more than three genes: DBP and T3-T4 and PTU and TAM. On the other hand, 62 of the 91 possible compound combinations ( 68%) were separated by six or more genes.
When compounds were hierarchically clustered based on the expression values of these 12 discriminatory genes (Pearson's uncentered clustering, Eisen et al., 1998), we see that the resulting dendrogram (Fig. 2B) very closely resembles the clustering result of the whole array data set. This implies that our discriminatory gene set really succeeds in grasping the essence of the full gene collection, thereby opening the door to a wide range of perspectives for future applications.
Our study is one of the first reports describing microarray-based expression profiling of a set of compounds interfering with a diversity of hormonal signaling pathways in fish. Using a customized cDNA microarray for C. carpio, we have generated gene expression signatures for 14 reference EDCs, representing various mechanisms of action. We were able to show that each individual compound produces its own unique expression pattern on the array. Furthermore, a small discriminatory gene set, consisting of 12 informative genes, could be identified that was also able to discriminate between the different compounds. Although it was noted that some of the compounds preferentially clustered with compounds outside their presumed mode of action, we also showed some marked expression similarities between groups of analogous compounds. For actual classification purposes, however, a much larger database comprising numerous compound-associated expression profiles will be needed. Also, it is expected that an extension of the current gene collection with additional (anti)androgen-responsive gene fragments could further add to the value of this array. Overall, our findings suggest that the developed cDNA array has great promise as a screening tool for new or existing chemicals with an unknown mode of action to assess their endocrine-disruptive potential and to identify distinct classes of EDCs.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
A list of SSH-derived gene fragments showing significant homology to sequences in the National Center for Biotechnology Information databases, a short description of microarray quality analysis using the Lucidea Universal ScoreCard Controls and boxplots of the calibration M, A, log2(R), and log2(G) values for a representative array, and expression values of the differentially expressed genes in carp exposed to the different compounds are available online at http://toxsci.oxfordjournals.org/.
|
ACKNOWLEDGMENTS |
|---|
This work was financially supported by the Promotion of Innovation by Science and Technology in Flanders (IWT) (grant no 1274) and partially funded by the Fund for Scientific Research Flanders (FWO-Flanders) (G.0358.02).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Ankley, G. T., Jensen, K. M., Kahl, M. D., Korte, J. J., and Makynen, E. A. (2001). Description and evaluation of a short-term reproductive test with the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 20, 1276–1290.[CrossRef][ISI][Medline]
Arukwe, A., Celius, T., Walther, B. T., and Goksoyr, A. (2000). Effects of xenoestrogen treatment on zona radiata protein and vitellogenin expression in Atlantic salmon (Salmo salar). Aquat. Toxicol. 49, 159–170.[CrossRef][ISI][Medline]
Bartosiewicz, M. J., Jenkins, D., Penn, S., Emery, J., and Buckpitt, A. (2001). Unique gene expression patterns in liver and kidney associated with exposure to chemical toxicants. J. Pharmacol. Exp. Ther. 297, 895–905.
Bowman, C. J., Kroll, K. J., Gross, T. G., and Denslow, N. D. (2002). Estradiol-induced gene expression in largemouth bass (Micropterus salmoides). Mol. Cell. Endocrinol. 196, 67–77.[CrossRef][ISI][Medline]
Bulera, S. J., Eddy, S. M., Ferguson, E., Jatkoe, T. A., Reindel, J. F., Bleavins, M. R., and De La Iglesia, F. A. (2001). RNA expression in the early characterization of hepatotoxicants in Wistar rats by high-density DNA microarrays. Hepatology 33, 1239–1258.[CrossRef][ISI][Medline]
Chen, T. T. (1983). Identification and characterization of estrogen-responsive gene products in the liver of rainbow trout. Can. J. Biochem. Cell Biol. 61, 802–810.[ISI][Medline]
Cravedi, J. P., and Perdu-Durand, E. (2002). The phthalate diesters DEHP and DBP do not induce lauric acid hydroxylase activity in rainbow trout. Mar. Environ. Res. 54, 787–791.[CrossRef][ISI][Medline]
De, S., Ray, A. K., and Medda, A. K. (1989). Effects of L-thyroxine and L-triiodothyronine on protein and nucleic acid contents of liver of 6N-2-propylthiouracil treated hypothyroid singi fish, Heteropneustes fossilis bloch. Horm. Metab. Res. 21, 416–420.[ISI][Medline]
De Boeck, G., De Smet, H., and Blust, R. (1995). The effect of sublethal levels of copper on oxygen consumption and ammonia excretion in common carp, Cyprinus carpio. Aquat. Toxicol. 32, 127–141.
Denslow, N. D., Bowman, C. J., Ferguson, R. J., Lee, H. S., Hemmer, M. J., and Folmar, L. (2001). Induction of gene expression in sheepshead minnows (Cyprinodon variegatus) treated with 17beta-estradiol, diethylstilbestrol, or ethinylestradiol: The use of mRNA fingerprints as an indicator of gene regulation. Gen. Comp. Endocrinol. 121, 250–260.[CrossRef][ISI][Medline]
Endocrine Disruptor Methods Validation Subcommittee (EDMVS). (2003). Reference chemicals used in endocrine disruptor screening program tier 1 prevalidation studies. Available at: www.epa.gov/scipoly/oscpendo/pubs/edmvs/refchemsanddisclaimer121003.pdf. Accessed July 24, 2006.
Eisen, M. B., Spellman, P. T., Brown, P. O., and Botstein, D. (1998). Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA 95, 14863–14868.
Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC). (1998). Final report. 63 Fed. Reg. 71541–71568. Available at: http://www.epa.gov/scipoly/oscpendo/edspoverview/finalrpt.htm.
Environmental Protection Agency (EPA) (2003). Comparative evaluation of fathead minnow assays for detecting endocrine-disrupting chemicals. Draft Final Report, July 2003. EPA Contract Number 68-W-01-023. Work Assignments 2-18 and 3-8. Available at: http://www.epa.gov/scipoly/oscpendo/docs/edmvs/section_1_2_wa_2_18_frontintromethods.pdf. Accessed July 24, 2006.
Foster, P. M. D., Mylchreest, E., Gaido, K. W., and Sar, M. (2001). Effects of phthalate esters on the developing reproductive tract of male rats. Hum. Reprod. Update 7, 231–235.
Gray, L. E., Jr, Wolf, C., Lambright, C., Mann, P., Price, M., Cooper, R. L., and Ostby, J. (1999). Administration of potentially antiandrogenic pesticides (procymidone, linuron, iprodione, chlozolinate, p,p'-DDE, and ketoconazole) and toxic substances (dibutyl- and diethylhexyl phthalate, PCB 169, and ethane dimethane sulphonate) during sexual differentiation produces diverse profiles of reproductive malformations in the male rat. Toxicol. Ind. Health 15, 94–118.
Hamadeh, H. K., Bushel, P. R., Jayadev, S., DiSorbo, O., Bennett, L., Li, L., Tennant, R., Stoll, R., Barrett, J. C., Paules, R. S., et al. (2002a). Prediction of compound signature using high density gene expression profiling. Toxicol. Sci. 67, 232–240.
Hamadeh, H. K., Bushel, P. R., Jayadev, S., Martin, K., DiSorbo, O., Sieber, S., Bennett, L., Tennant, R., Stoll, R., Barrett, J. C., et al. (2002b). Gene expression analysis reveals chemical-specific profiles. Toxicol. Sci. 67, 219–231.
Harris, C. A., Henttu, P., Parker, M. G., and Sumpter, J. P. (1997). The estrogenic potency of phthalate esters in vitro. Environ. Health Perspect. 105, 802–811.[ISI][Medline]
Higuchi, T. T., Palmer, J. S., Gray, L. E., Jr, and Veeramachaneni, D. N. (2003). Effects of dibutyl phthalate in male rabbits following in utero, adolescent, or postpubertal exposure. Toxicol. Sci. 72, 301–313.
Hornung, M. W., Jensen, K. M., Korte, J. J., Michael, D., Durhan, E. J., Denny, J. S., Henry, T. R., and Ankley, G. T. (2004). Mechanistic basis for estrogenic effects in fathead minnow (Pimephales promelas) following exposure to the androgen 17
-methyltestosterone: Conversion of 17-methyltestosterone to 17-methylestradiol. Aquat. Toxicol. 66, 15–23.[CrossRef][ISI][Medline]
Hoyt, P. R., Doktycz, M. J., Beattie, K. L., and Greeley, M. S., Jr. (2003). DNA microarrays detect 4-nonylphenol-induced alterations in gene expression during zebrafish early development. Ecotoxicology 12, 469–474.[CrossRef][ISI][Medline]
Huber, W., Von Heydebreck, A., Sultmann, H., Poustka, A., and Vingron, M. (2002). Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics 18, S96–S104.[Abstract]
Jensen, J., van Langevelde, J., Pritzl, G., and Krogh, P. H. (2001). Effects of di(2-ethylhexyl) phthalate and dibutyl phthalate on the collembolan Folsomia fimetaria. Environ. Toxicol. Chem. 20, 1085–1091.[CrossRef][ISI][Medline]
Jobling, S., Reynolds, T., White, R., Parker, M. G., and Sumpter, J. P. (1995). A variety of environmental persistent chemicals, including some phthalate plasticizers are weakly estrogenic. Environ. Health Perspect. 103, 582–587.[ISI][Medline]
Kajimura, S., Hirano, T., Visitacion, N., Moriyama, S., Aida, K., and Grau, E. G. (2003). Dual mode of cortisol action on GH/IGF-I/IGF binding proteins in the tilapia, Oreochromis mossambicus. J. Endocrinol. 178, 91–99.[Abstract]
Kemppainen, J. A., Lane, M. V., Sar, M., and Wilson, E. M. (1992). Androgen receptor phosphorylation, turnover, nuclear transport, and transcriptional activation. Specificity for steroids and antihormones. J. Biol. Chem. 267, 968–974.
Kim, H. S., Kim, T. S., Shin, J. H., Moon, H. J., Kang, I. H., Kim, I. Y., Oh, J. Y., and Han, S. Y. (2004). Neonatal exposure to di(n-butyl) phthalate (DBP) alters male reproductive-tract development. J. Toxicol. Environ. Health A 67, 2045–2060.[CrossRef][ISI][Medline]
Kulczykowska, E., Sokolowska, E., Takvam, B., Stefansson, S., and Ebbesson, L. (2004). Influence of exogenous thyroxine on plasma melatonin in juvenile Atlantic salmon (Salmo salar). Comp. Biochem. Physiol. B Biochem. Mol. Biol. 137, 43–47.[CrossRef][Medline]
Larkin, P., Folmar, L. C., Hemmer, M. J., Poston, A. J., and Denslow, N. D. (2003). Expression profiling of estrogenic compounds using a sheepshead minnow cDNA macroarray. EHP Toxicogenom. 111, 29–36.
Larkin, P., Sabo-Attwood, T., Kelso, J., and Denslow, N. D. (2002). Gene expression analysis of largemouth bass exposed to estradiol, nonylphenol, and p,p'-DDE. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 133, 543–557.[CrossRef][Medline]
Lee, S. K., Owens, G. A., and Veeramachaneni, D. N. (2005). Exposure to low concentrations of di-n-butyl phthalate during embryogenesis reduces survivability and impairs development of Xenopus laevis frogs. J. Toxicol. Environ. Health A 68, 763–772.[ISI][Medline]
Li, L., Li, Z., and Sacks, D. B. (2005). The transcriptional activity of estrogen receptor-alpha is dependent on Ca2+/calmodulin. J. Biol. Chem. 280, 13097–13104.
MacLatchy, D. L., and Eales, J. G. (1988). Short-term treatment with testosterone increases plasma 3,5,3'-triiodo-L-thyronine and hepatic L-thyroxine 5'-monodeiodinase levels in arctic charr, Salvelinus alpinus. Gen. Comp. Endocrinol. 71, 10–16.[CrossRef][ISI][Medline]
Makynen, E. A., Kahl, M. D., Jensen, K. M., Tietge, J. E., Wells, K. L., Van Der Kraak, G., and Ankley, G. T. (2000). Effects of the mammalian antiandrogen vinclozolin on development and reproduction of the fathead minnow (Pimephales promelas). Aquat. Toxicol. 48, 461–475.[CrossRef][ISI][Medline]
Matta, S. L., Vilela, D. A., Godinho, H. P., and Franca, L. R. (2002). The goitrogen 6-n-propyl-2-thiouracil (PTU) given during testis development increases Sertoli and germ cell numbers per cyst in fish: the tilapia (Oreochromis niloticus) model. Endocrinology 143, 970–978.
Mikawa, N., Hirono, I., and Aoki, T. (1996). Structure of medaka transferrin gene and its 5'-flanking region. Mol. Mar. Biol. Biotechnol. 5, 225–229.[ISI][Medline]
Moens, L. N., Soetaert, A., van der Ven, K., Del-Favero, J., and De Coen, W. M. (2006). Use of Suppression Subtractive Hybridization PCR for the development of cDNA arrays for the detection of endocrine disruption in carp (Cyprinus carpio). Comp. Biochem. Physiol. (in press).
Moens, L. N., van der Ven, K., Van Remortel, P., Del-Favero, J., and De Coen W. M. (2006). Gender related differences in gene expression in common carp (Cyprinus carpio): A microarray study. Gen. Comp. Endocrinol. (in press).
Mylchreest, E., Cattley, R. C., and Foster, P. M. (1998). Male reproductive tract malformations in rats following gestational and lactational exposure to di(n-butyl) phthalate: An antiandrogenic mechanism? Toxicol. Sci. 43, 47–60.
Mylchreest, E., Sar, M., Cattley, R. C., and Foster, P. M. (1999). Disruption of androgen-regulated male reproductive development by di(n-butyl) phthalate during late gestation in rats is different from flutamide. Toxicol. Appl. Pharmacol. 156, 81–95.[CrossRef][ISI][Medline]
Naciff, J. M., Jump, M. L., Torontali, S. M., Carr, G. J., Tiesman, J. P., Overmann, G. J., and Daston, G. P. (2002). Gene expression profile induced by 17alpha-ethynyl estradiol, bisphenol A, and genistein in the developing female reproductive system of the rat. Toxicol. Sci. 68, 184–199.
Oppenheimer, J. H., Schwartz, H. L., Mariash, C. N., Kinlaw, W. B., Wong, N. C., and Freake, H. C. (1987). Advances in our understanding of thyroid hormone action at the cellular level. Endocr. Rev. 8, 288–308.[ISI][Medline]
Parhar, I. S., Soga, T., and Sakuma, Y. (2000). Thyroid hormone and estrogen regulate brain region-specific messenger ribonucleic acids encoding three gonadotropin-releasing hormone genes in sexually immature male fish, Oreochromis niloticus. Endocrinology 141, 1618–1626.
Parrott, J. L., and Wood, C. S. (2002). Fathead minnow lifecycle test for detection of endocrine-disrupting substances in effluent. Water Qual. Res. J. Can. 37, 651–667.
Pawlowski, S., Sauer, A., Shears, J. A., Tyler, C. R., and Braunbeck, T. (2004). Androgenic and estrogenic effects of the synthetic androgen 17alpha-methyltestosterone on sexual development and reproductive performance in the fathead minnow (Pimephales promelas) determined using the gonadal recrudescence assay. Aquat. Toxicol. 68, 277–291.[CrossRef][ISI][Medline]
Pedersen, S. N., Christiansen, L. B., Pedersen, K. L., Korsgaard, B., and Bjerregaard, P. (1999). In vivo estrogenic activity of branched and linear alkylphenols in rainbow trout (Oncorhynchus mykiss). Sci. Total Environ. 233, 89–96.[CrossRef][Medline]
Sharpe, R. M., Fisher, J. S., Millar, M. M., Jobling, S., and Sumpter, J. P. (1995). Gestational and lactational exposure of rats to xenoestrogens results in reduced testicular size and sperm production. Environ. Health Perspect. 103, 1136–1143.[ISI][Medline]
Steiner, G., Suter, L., Boess, F., Gasser, R., de Vera, M. C., Albertini, S., and Ruepp, S. (2004). Discriminating different classes of toxicants by transcript profiling. Environ. Health Perspect. 112, 1236–1248.[ISI][Medline]
Terasaka, S., Aita, Y., Inoue, A., Hayashi, S., Nishigaki, M., Aoyagi, K., Sasaki, H., Wada-Kiyama, Y., Sakuma, Y., Akaba, S., et al. (2004). Using a customized DNA microarray for expression profiling of the estrogen-responsive genes to evaluate estrogen activity among natural estrogens and industrial chemicals. Environ. Health Perspect. 112, 773–781.
Thomas, R. S., Rank, D. R., Penn, S. G., Zastrow, G. M., Hayes, K. R., Pande, K., Glover, E., Silander, T., Craven, M. W., Reddy, J. K., et al. (2001). Identification of toxicologically predictive gene sets using cDNA microarrays. Mol. Pharmacol. 60, 1189–1194.
Timmermans, L. P., Chmilevsky, D. A., Komen, H., and Schipper, H. (1997). Precocious onset of spermatogenesis in juvenile carp (Cyprinus carpio L., Teleostei) following treatment with low doses of L-thyroxine. Eur. J. Morphol. 35, 344–353.[CrossRef][ISI][Medline]
Van den Belt, K., Berckmans, P., Vangenechten, C., Verheyen, R., and Witters, H. (2004). Comparative study on the in vitro/in vivo estrogenic potencies of 17beta-estradiol, estrone, 17alpha-ethynylestradiol and nonylphenol. Aquat. Toxicol. 66, 183–195.[CrossRef][ISI][Medline]
Van den Belt, K., Verheyen, R., and Witters, H. (2003). Comparison of vitellogenin responses in zebrafish and rainbow trout following exposure to environmental estrogens. Ecotoxicol. Environ. Saf. 56, 271–281.[CrossRef][ISI][Medline]
Van der Ven, K., De Wit, M., Keil, D., Moens, L., Van Leemput, K., Naudts, B., and De Coen, W. (2005). Development and application of a brain-specific cDNA microarray for effect evaluation of neuro-active pharmaceuticals in zebrafish (Danio rerio). Comp. Biochem. Physiol. B Biochem. Mol. Biol. 141, 408–417.[CrossRef][Medline]
Verslycke, T., Vandenbergh, G. F., Versonnen, B., Arijs, K., and Janssen, C. R. (2002). Induction of vitellogenesis in 17alpha-ethinylestradiol-exposed rainbow trout (Oncorhynchus mykiss): A method comparison. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 132, 483–492.[CrossRef][ISI][Medline]
Versonnen, B. J., and Janssen, C. R. (2004). Xenoestrogenic effects of ethinylestradiol in zebrafish (Danio rerio). Environ. Toxicol. 19, 198–206.[CrossRef][ISI][Medline]
Vik, D. P., Amiguet, P., Moffat, G. J., Fey, M., Amiguet-Barras, F., Wetsel, R. A., and Tack, B. P. (1991). Structural features of the human C3 gene: intron/exon organization, transcriptional start site, and promoter region sequence. Biochemistry 30, 1080–1085.[CrossRef][Medline]
Waring, J. F., Ciurlionis, R., Jolly, R. A., Heindel, M., and Ulrich, R. G. (2001a). Microarray analysis of hepatotoxins in vitro reveals a correlation between gene expression profiles and mechanisms of toxicity. Toxicol. Lett. 120, 359–368.[CrossRef][ISI][Medline]
Waring, J. F., Jolly, R. A., Ciurlionis, R., Lum, P. Y., Praestgaard, J. T., Morfitt, D. C., Buratto, B., Roberts, C., Schadt, E., and Ulrich, R. G. (2001b). Clustering of hepatotoxins based on mechanism of toxicity using gene expression profiles. Toxicol. Appl. Pharmacol. 175, 28–42.[CrossRef][ISI][Medline]
Watanabe, H., Suzuki, A., Goto, M., Lubahn, D. B., Handa, H., and Iguchi, T. (2004). Tissue-specific estrogenic and non-estrogenic effects of a xenoestrogen, nonylphenol. J. Mol. Endocrinol. 33, 243–252.[Abstract]
Watanuki, H., Yamaguchi, T., and Sakai, M. (2002). Suppression in function of phagocytic cells in common carp Cyprinus carpio L. injected with estradiol, progesterone or 11-ketotestosterone. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 132, 407–413.[CrossRef][ISI][Medline]
Williams, T. D., Gensberg, K., Minchin, S. D., and Chipman, J. K. (2003). A DNA expression array to detect toxic stress response in European flounder (Platichthys flesus). Aquat. Toxicol. 65, 141–157.[CrossRef][ISI][Medline]
Wong, C., Kelce, W. R., Sar, M., and Wilson, E. M. (1995). Androgen receptor antagonist versus agonist activities of the fungicide vinclozolin relative to hydroxyflutamide. J. Biol. Chem. 270, 19998–20003.
Zacharewski, T. R., Meek, M. D., Clemons, J. H., Wu, Z. F., Fielden, M. R., and Matthews, J. B. (1998). Examination of the in vitro and in vivo estrogenic activities of eight commercial phthalate esters. Toxicol. Sci. 46, 282–293.
Zerulla, M., Länge, R., Steger-Hartmann, T., Panter, G., Hutchinson, T., and Dietrich, D. R. (2002). Morphological sex reversal upon short-term exposure to endocrine modulators in juvenile fathead minnow (Pimephales promelas). Toxicol. Lett. 131, 51–63.[CrossRef][ISI][Medline]