Differential Gene Expression in Response to Methoxychlor and Estradiol through ER{alpha}, ERß, and AR in Reproductive Tissues of Female Mice

Katrina M. Waters*, Stephen Safe{dagger} and Kevin W. Gaido*,1

* Department of Endocrine, Reproductive and Developmental Toxicology, CIIT Centers for Health Research, 6 Davis Drive, Research Triangle Park, North Carolina 27709; and {dagger} Department of Veterinary Physiology and Pharmacology, Texas A&M University, College Station, Texas 77843

Received March 30, 2001; accepted June 11, 2001


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The reproductive and developmental effects of 17ß-estradiol (E2) and methoxychlor (MXC) observed in treated rodents appear to be linked to some unique but also overlapping patterns of gene expression. The MXC metabolite 2,2-bis(p-hydroxyphenyl)-1,1,1-trichloroethane (HPTE) was previously shown to have selective agonist activity through estrogen receptor {alpha} (ER{alpha}) and antagonist activity through ERß and androgen receptor (AR). To discover gene families regulated by HPTE and E2, and to characterize similarities and differences in patterns of gene expression induced by these selective ER ligands, we analyzed tissues from mice treated for 3 days with a combined treatment of E2 and HPTE (E2 + HPTE), or the antiandrogen flutamide (FLU). RNA from uteri and ovaries was analyzed with cDNA microarrays and real-time RT-PCR. Results indicate that HPTE and E2 acted similarly to regulate most gene families in the uterus, which expresses predominantly ER{alpha}. However, in both the uterus and the ovary, there were a few genes that displayed differential patterns of gene regulation by E2 or HPTE treatment, presumably through ERß, AR, or other unidentified pathways. In the uterus, progesterone receptor, ER{alpha}, AR, insulin-like growth factor 1, insulin-like growth factor binding protein 5, and clusterin mRNAs were significantly reduced with both E2 or HPTE treatments, whereas cathepsin B was induced. Conversely, in the ovary, induction of cathepsin B by E2 was reversed after cotreatment with HPTE, and ERß expression was induced similarly by HPTE and FLU but not by E2. In addition, E2 uniquely regulated glutathione peroxidase 3, glutathione S-transferase, and cytochrome P450 17{alpha}-hydroxylase, with no effect of HPTE or FLU treatments. This analysis demonstrated several gene families that appear to be regulated in a ligand-specific pattern, which may explain the unique but overlapping reproductive tissue pathologies following exposure to E2 and MXC.

Key Words: methoxychlor; HPTE; estrogen; estrogen receptor {alpha} (ER{alpha}); estrogen receptor ß (ERß); androgen receptor (AR); microarray; gene expression.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Methoxychlor (MXC) is an organochlorine pesticide developed to replace 1,1,1-trichloro-2,2-bis(p-chlorophenyl)ethane (DDT). MXC is estrogenic in vivo, despite the fact that its binding affinity for estrogen receptor {alpha} (ER{alpha}) is 10,000 times less than 17ß-estradiol (E2) (Kuiper et al., 1997Go). Numerous studies show that MXC mimics E2 action in the female rodent reproductive tract, causing a uterotropic response in ovariectomized (ovx) rats and adverse developmental and reproductive effects in gestational or chronic exposure studies in rodents. These include embryonic toxicity, precocious puberty, decreased fertility, and ovarian atrophy (Alm et al., 1996Go; Chapin et al., 1997Go; Cummings and Gray, 1989Go; Gray et al., 1989Go; Hall et al., 1997Go). Molecular studies with ovx rodents have also shown that MXC and E2 regulate the activity of many of the same uterine proteins, including epidermal growth factor receptor (Metcalf et al., 1996Go), uterine peroxidase (Cummings and Metcalf, 1994Go), and ER (Eroschenko et al., 1996Go). MXC also appears to antagonize E2 action under some conditions. Reproductive tract weight changes and induction of uterine albumin levels by E2 is blocked with co-administration of MXC (Eroschenko et al., 2000Go). Intact mice given low doses of MXC, but not high doses, showed a delay in the normal reproductive aging process (which is accelerated by E2 administration) by maintaining large ovaries filled with corpora lutea at 6 months of age (Eroschenko et al., 1995Go). MXC restored ovx-reduced pituitary prolactin concentration, but reduced it in intact rats (Gray et al., 1988Go). Similarly, MXC-induced increases in uterine weight in immature mice that could be blocked by raloxifene (a tissue-specific ER antagonist), while those induced by E2 could not (Al-Jamal and Dubin, 2000Go). These studies suggest that MXC displays a unique pattern of ER agonist and antagonist properties in vivo. MXC is also reported to have pharmacological activities apparently unrelated to ER binding. More recently, it was demonstrated that MXC-induced lactoferrin and glucose-6-phosphate dehydrogenase expression in ER{alpha} knockout mice could not be blocked by administration of the antiestrogen ICI 182780 (Ghosh et al., 1999Go). These studies clearly indicate a mechanism for the estrogenic effects of MXC that is different from that of E2.

MXC is metabolized in vivo into mono- and bisphenol demethylated derivatives; however, the bisphenol metabolite, 2,2-bis(p-hydroxyphenyl)-1,1,1-trichloroethane (HPTE), is believed to be responsible for the in vivo estrogenic action of MXC (Bulger et al., 1978Go; Kupfer and Bulger, 1987Go). HPTE is uterotropic in a 3-day bioassay (Shelby et al., 1996Go) and stimulates proliferation of uterine leiomyoma cells in vitro (Hodges et al., 2000Go). We have shown that HPTE binds directly to the ER and AR in vitro (Gaido et al.1999Go); however, the transcriptional effect of this binding is receptor-specific. HPTE is an ER{alpha} agonist and ERß and androgen receptor (AR) antagonist (Gaido et al., 2000Go). The physiological effects of MXC treatment in vivo may be due to actions mediated by HPTE binding to ER{alpha}, ERß, or AR. To date, only genes that are regulated similarly by E2 and HPTE, such as gonadotropin-releasing hormone (Roy et al., 1999Go) and uterine insulin-like growth factor 1 (Klotz et al., 2000Go), have been identified.

In this study, we utilized the unique properties of MXC as a model system to identify differential gene expression profiles that may relate to its receptor-specific activity. HPTE was used instead of MXC to eliminate mixtures of compounds and metabolites. The dose of E2 was selected to induce a slightly stronger uterotropic response to that seen with a uterotropic dose of HPTE, based upon a previous study (Shelby et al., 1996Go). Sexually immature female mice were treated with either E2 (50 µg/kg) or HPTE (500 mg/kg) for 3 days, uterine and ovarian tissues were obtained, and extracted mRNA from these tissues was analyzed by microarray and real-time PCR for changes in gene expression. A combined dose of E2 and HPTE (E2 + HPTE) was administered to examine HPTE antagonism of ERß-mediated gene regulation by E2, and a flutamide (FLU) treatment group (45 mg/kg) was included to identify genes regulated through AR antagonism. The unique and overlapping patterns of gene expression produced by E2 or HPTE treatment were linked to the reproductive and developmental effects of E2 and MXC observed in treated rodents.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals.
HPTE (2971–36–0) was synthesized and purified by preparative thin-layer chromatography as described previously (Gaido et al., 1999Go). E2 (50–28–2) and FLU (13311–84–7) were obtained from Sigma Chemical Co. (St. Louis, MO). All chemicals were >= 97% pure. Prior to dosing, homogenous preparations of each chemical were prepared in the appropriate amount of corn oil (Mazola, Best Foods) for an approximate dosing volume of 100 µl/10 g mouse.

Animals.
This study was approved by the Institutional Animal Care and Use Committee at CIIT, and the CIIT animal-care unit is accredited by the American Association for Accreditation of Laboratory Animal Care, with humidity- and temperature-controlled rooms, maintained on a 12-h light-dark cycle. All procedures followed the criteria of the National Research Council for the care and use of laboratory animals in research. C57BL6/J mice were obtained from Taconic (Germantown, NY) at postnatal day 21 (weaning). Upon arrival, animals were assigned to treatment groups by randomization of body weights. Rodent diet NIH-07 and reverse-osmosis water was provided ad libitum, and animals were acclimatized for at least 48 h prior to dosing.

Study design.
Sexually immature female mice received daily subcutaneous injections of either vehicle (corn oil), E2 (50 µg/kg), HPTE (500 mg/kg), E2 + HPTE, or FLU (45 mg/kg) for 3 days. Three animals were included in each treatment group. On the morning of day 4, animals were euthanized by cervical dislocation and target tissues were harvested. Body weights were measured, as well as wet weights for uterus and ovaries. Tissues were submerged in the RNA preservative reagent RNA later (Ambion, Austin, TX), kept at 4°C overnight, and then stored at –80°C until processed for RNA isolation.

cDNA array analysis.
Total RNA from mouse tissues was isolated using the Totally-RNA kit and subsequently DNase-treated with RNase-free DNase, both reagents from Ambion (Austin, TX). Reverse-transcription (RT) reactions were performed with 5 µg of total RNA, (P-32)-dATP, and Superscript-II MMLV reverse transcriptase (Life Technologies, Gaithersburg, MD) for 30 min at 50°C. Following purification, probes were added to each Clontech (Palo Alto, CA) Atlas Mouse cDNA expression array (588 genes) or stress/tox array membrane (140 genes), with hybridization and washing performed according to manufacturer's instructions. Membranes were exposed to film for 1–3 days to acquire optimum images for analysis with Clontech Atlas Image software (Palo Alto, CA) and pairwise comparisons of log2 transformed raw data. Array analysis was performed on one animal per treatment group for both the expression membrane and the stress/tox membrane.

Real-time RT-PCR.
Changes in gene expression were confirmed and quantified using real time RT-PCR on the ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA). Total RNA (2.5 µg) from each animal tissue was aliquoted in quadruplicate for RT using the TaqMan Reverse Transcription Reagents (Applied Biosystems) and SuperScript-II reverse transcriptase (Life Technologies), with one aliquot designated to receive no enzyme, according to manufacturer's instructions. PCR was performed using SYBR Green PCR Core Reagents (Applied Biosystems) according to manufacturer's instructions, with one modification: reaction volume was adjusted to 25 µl. Primer sets were selected by the Primer Express software package, production of a single PCR product was confirmed using gel electrophoresis, and primer efficiency was determined according to manufacturer's recommended protocol (Applied Biosystems). Sequences of gene primer sets are given in Table 1Go. RT-PCR analysis was performed on all 3 animals in each treatment group, in triplicate. Quality of RT reactions was confirmed by comparison of triplicate RT versus no enzyme control for each RNA sample, using the GAPDH primer set according to manufacturer's protocols (Applied Biosystems).


View this table:
[in this window]
[in a new window]
 
TABLE 1 Sequences of Primer Sets Used for Real-Time RT-PCR Analysis of Gene Expression on ABI 7700 Sequence Detection System
 
Statistics.
Each data point is an average of 3 animals per group, with each analysis performed in triplicate. Error bars represent the standard error (SE), with all values represented as fold change compared to the control treatment group average of 1.0. Significance was determined by one-tailed, non-paired Student`s t-test with p < 0.05.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
HPTE causes a uterotropic response in female mice.
Female weanling mice were subjected to a 3-day uterotropic assay that included treatments with E2 or HPTE, as described previously (Shelby et al., 1996Go), with the addition of a treatment group receiving E2 + HPTE and a group treated with FLU. Uterine wet weights increased 3.8- and 2.6-fold with E2 and HPTE treatments, respectively (Fig. 1Go). Combined E2 + HPTE treatment resulted in an apparent additive effect; uterine weight was increased 8.9-fold. This additive uterotropic response with combined E2 and HPTE treatment is different from a study described previously, where ovariectomized adult mice were given much lower doses of E2 and MXC and no additive uterotropic response was observed (Eroschenko et al., 2000Go). However, it has been demonstrated that the uterine cell population responding to E2 treatment in the immature mouse is different from that in the adult mouse uterus (Quarmby and Korach, 1984Go), which could explain the differences between our study and previous studies with E2 using adult ovariectomized animals. The large SE in the E2 value is due to a single animal in the treatment group whose uterus had the same uterotropic morphology, i.e., dilated and fluid-filled, but it was shorter in length and therefore weighed the same as the controls. There was relatively little change in ovarian wet weight, although the combined E2 + HPTE treatment group had slightly smaller ovaries than the E2 treatment group. There was no significant change in body weights across the treatment groups (data not shown). FLU had no effect on uterine or ovarian wet weight.



View larger version (26K):
[in this window]
[in a new window]
 
FIG. 1. Organ weights of uteri and ovaries. Wet weights of uterus and ovaries (mg) were normalized to animal body weight (g). Data represent means ± SE of n = 3 animals per group. Asterisk indicates p < 0.05 (compared to control); diamond indicates p < 0.05 (compared to E2).

 
cDNA microarray analysis demonstrates a large number of changes in gene expression with treatment.
To characterize the similarities and differences in the mechanism of action of MXC and E2 in the female reproductive tract through changes in gene expression, uterus and ovary RNA samples from each treatment group were subjected to microarray analysis. One animal from each treatment group was used for both the cDNA expression array (588 genes) and the stress/tox array (140 genes). Data from E2, HPTE, and FLU treatments were compared to untreated controls, and E2 and E2 + HPTE treatment groups were compared. Pairwise comparisons of log2 transformed raw data (588 data points) for uterine samples hybridized to cDNA expression arrays are shown in Figure 2Go. Linear regression analysis was performed, and r2 values demonstrated similarity between expression levels of most genes on the arrays. When using the xy scatterplot and regression analysis, differences in hybridization intensity are more easily interpreted without the need for normalization to housekeeping genes that may change in response to treatment (Smid-Koopman et al., 2000Go). Similar results were obtained from analysis of the ovary samples on the cDNA expression and stress/tox data (data not shown). About 50 genes were identified in each tissue that showed at least a 2-fold change in gene expression on either the cDNA expression array or the stress/tox array. Based upon a potential relationship to the observed reproductive tissue effects, a select number of these genes were chosen for confirmation and quantitation using real-time RT-PCR analysis. Genes selected with a minimum 2-fold change in more than one treatment group included steroid hormone receptors, growth factors, cytokines, apoptosis-associated proteins, proteins involved in xenobiotic metabolism, protein turnover and steroid metabolism, oncogenes, transcription factors, and DNA binding proteins.



View larger version (25K):
[in this window]
[in a new window]
 
FIG. 2. Pairwise comparisons of microarray data from uterus samples. Data from uterus cDNA expression microarray blots were log2 transformed and are shown in an xy-scatterplot. Linear regression analysis was performed, with equation and r2 values given.

 
Steroid receptor mRNAs are downregulated by estrogen and HPTE in the uterus and upregulated in the ovary.
Using real-time RT-PCR analysis of RNA samples from all animals in each treatment group, the change in gene expression in response to treatment was quantified. Starting with hormone receptors, uterine progesterone receptor (PR), ER{alpha}, and AR mRNAs were all significantly downregulated after treatment with either E2 or HPTE (Fig. 3AGo). E2 reduced PR expression by 61%, ER{alpha} by 73%, and AR by 79%, and HPTE reduced PR expression by 71%, ER{alpha} by 64%, and AR by 65%. E2 + HPTE treatment showed the same pattern of regulation of PR, ER{alpha}, and AR as observed for E2 or HPTE alone. These are most likely ER{alpha}-mediated responses, since HPTE is an ER{alpha} agonist (Gaido et al., 1999Go) and ER{alpha} is the predominant estrogen receptor in the mouse uterus (Couse et al., 1997Go).



View larger version (39K):
[in this window]
[in a new window]
 
FIG. 3. Gene expression changes of steroid hormone receptors progesterone receptor (PR), estrogen receptor alpha (ER{alpha}), ERß, and androgen receptor (AR). Total RNA from uteri (A) and ovaries (B) was subjected to real-time RT-PCR analysis. Data represent means ± SE of n = 3 animals per group. Asterisk indicates p < 0.05 (compared to control); diamond indicates p < 0.05 (compared to E2).

 
In contrast, E2 upregulated PR by 80% in the ovary (Fig. 3BGo). HPTE did not cause a similar increase in PR expression in the ovary, but it upregulated ERß by 90%. E2 + HPTE treatment resulted in a significantly higher induction of PR than E2 alone in the ovary, to almost 200% above control, and up-regulation of ERß (104% above control) was similar to that observed for HPTE alone. FLU also increased ERß expression by 82% in the ovary. The regulation of ERß by HTPE is most likely an AR-mediated response, since it is mimicked by FLU treatment. ER{alpha} expression was not significantly altered by any treatment in the ovary, and FLU only slightly downregulated AR in the uterus. ERß expression was too low to be quantified in the uterus.

Insulin-like growth factor 1A and insulin-like growth factor-binding protein 5 genes are downregulated in the uterus but not in the ovary.
In the uterus, insulin-like growth factor 1A (IGF1A) expression was reduced 65 and 64% by E2 or HPTE, respectively (Fig. 4AGo). Likewise, insulin-like growth factor-binding protein 5 expression was downregulated 81 and 51% by E2 or HPTE, respectively. E2 + HPTE treatment showed the same pattern for regulation of IGF1A and IGFBP5 as the compounds alone. Again, these are most likely ER{alpha}-mediated responses. In the ovary, E2 decreased IGFBP5 expression by 38% (p = 0.08), and HPTE had a slight effect in reversing the repression of IGFBP5 by E2 when given at the same time (Fig. 4BGo). For these genes, FLU had no effect on expression in either the uterus or the ovary.



View larger version (38K):
[in this window]
[in a new window]
 
FIG. 4. Gene expression changes of insulin-like growth factor 1 (IGF1A) and insulin-like growth factor binding protein 5 (IGFBP5). Total RNA from uteri (A) and ovaries (B) was subjected to real-time RT-PCR analysis. Data represent means ± SE of n = 3 animals per group. Asterisk indicates p < 0.05 (compared to control); diamond indicates p < 0.05 (compared to E2).

 
Estrogen upregulates glutathione peroxidase 3 mRNA in both the uterus and the ovary and downregulates clusterin and glutathione S-transferase mRNAs in the uterus, but not the ovary.
Clusterin (Clus), also known as testosterone-repressed prostate message 2 (TRPM-2), was downregulated 43 and 73% by E2 or HPTE, respectively, in the uterus (Fig. 5AGo). Glutathione S-transferase (GST) was also downregulated 58 and 24% (p = 0.08) by E2 or HPTE, respectively. E2 + HPTE followed a similar pattern of down-regulation for Clus (68%) and GST (37%) as observed for E2 or HPTE alone. These could be ER{alpha}-mediated responses. Conversely, glutathione peroxidase 3 (GPX3) was induced 231% above control with E2 treatment, but not with HPTE treatment. The E2 + HPTE combined effect was only 134% above control, significantly lower than the E2-induced change. This could be an example of an ERß-mediated response, even in the uterus, where ERß has been localized to the epithelium and stroma in immature mice (Weihua et al.2000Go).



View larger version (34K):
[in this window]
[in a new window]
 
FIG. 5. Gene expression changes of the apoptosis-related protein clusterin (Clus), and xenobiotic metabolism proteins glutathione peroxidase 3 (GPX3) and glutathione S-transferase (GST). Total RNA from uteri (A) and ovaries (B) was subjected to real-time RT-PCR analysis. Data represent means ± SE of n = 3 animals per group. Asterisk indicates p < 0.05 (compared to control); diamond indicates p < 0.05 (compared to E2).

 
In the ovary, E2 induced GPX3 108% above control (Fig. 5BGo), again with no effect from HPTE alone. The E2 + HPTE combined treatment induced GPX3 59% above control, lower than the E2 alone treatment but not significantly different from the E2 value. Clus and GST expression patterns were not downregulated in the ovary after treatment with E2 or HPTE. Once again, FLU treatment had no effect on expression of these genes in the uterus or ovary.

GPX3 mRNA was upregulated by E2 in both the uterus and the ovary, but not affected by HPTE (Fig. 5Go). As in the uterus, cotreatment with E2 + HPTE decreased induction compared with E2 treatment alone, although it was not statistically significant in the ovary.

Cathepsin B gene is differentially regulated and steroid metabolism is significantly downregulated through the Cyp17 gene in the ovary.
Cathepsin B (CTSB) was upregulated 75 and 49% by E2 or HPTE, respectively, and upregulated 54% with combined E2 + HPTE treatment in the uterus (Fig. 6AGo). This could also be an ER{alpha}-mediated response. In the ovary, CTSB is also upregulated by E2 (45% above control); however, HPTE does not upregulate CTSB (Fig. 6BGo). In fact, E2 + HPTE combined treatment results in significantly lower expression of CTSB than observed for E2 alone. Interestingly, FLU treatment downregulates CTSB by 27% in the uterus and 35% in the ovary. This could be an ERß-mediated response, since HPTE is an ERß antagonist (Gaido et al.1999Go) and ERß is highly expressed in the ovary, specifically in the granulosa cells (Sar and Welsch, 1999Go). However, it could also be an antiandrogenic response, since FLU treatment downregulated CTSB in the ovary, and AR co-localizes with ERß in granulosa cells (Pelletier et al.2000Go).



View larger version (32K):
[in this window]
[in a new window]
 
FIG. 6. Gene expression changes of protein turnover and steroid metabolism genes cathepsin B (CTSB) and cytochrome P-450 17{alpha}-hydroxylase (Cyp17). Total RNA from uteri (A) and ovaries (B) was subjected to real-time RT-PCR analysis. Cyp17 mRNA was too low to be quantified in the uterus. Data represent means ± SE of n = 3 animals per group. Asterisk indicates p < 0.05 (compared to control); diamond indicates p < 0.05 (compared to E2).

 
Cytochrome P450 17{alpha}-hydroxylase (Cyp17) expression could be quantified only in the ovary, and it demonstrated significant regulation with E2 treatment. Cyp17 was decreased 87 and 92% in animals treated with E2 and with E2 + HPTE combined, respectively. HPTE alone caused a 90% reduction in 1 animal, but the other 2 mice in the treatment group did not respond (data not shown). FLU induced Cyp17 expression; however, this response was not statistically significant because one animal in the treatment group did not respond, even though 220 and 360% induction of Cyp17 was observed in the other 2 mice (data not shown).

Differential regulation of cFMS and STAT1 mRNAs in the uterus and JunD mRNA in the ovary.
Two genes showing up-regulation after treatment with both E2 and HPTE on the microarray were c-FMS proto-oncogene and signal transducer and activator of transcription 1 (STAT1) in the uterus. RT-PCR analysis of these gene changes showed no significant change with E2 (Fig. 7AGo), although HPTE induced cFMS more than 100% (p = 0.08). Two animals in this treatment group responded well (146 and 258% above control), while one did not respond (data not shown). FLU treatment downregulated STAT1 expression by 76%, a significant change that was not predicted by the array analysis (data not shown).



View larger version (34K):
[in this window]
[in a new window]
 
FIG. 7. Gene expression changes of cFMS protooncogene and signal transducer and activator of transcription 1 (STAT1) in uterus and JunD in ovary. Total RNA from uteri (A) and ovaries (B) was subjected to real-time RT-PCR analysis. Data represent means ± SE of n = 3 animals per group. Asterisk indicates p < 0.05 (compared to control); diamond indicates p < 0.05 (compared to E2).

 
In the ovary, JunD expression was reduced by 46 and 31% by HPTE and E2 + HPTE, treatment, respectively, but not by E2 alone. This may be an antiandrogenic response, as indicated by the 44% down-regulation of ovarian JunD in FLU-treated animals.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The objective of this study was to identify novel genes regulated by MXC through ER{alpha}, ERß, and AR. We demonstrated that HPTE regulated several of the same genes as E2; however, some genes showed differential regulation in both the uterus and the ovary. Table 2Go contains a summary of the data from the RT-PCR analysis, along with the ligand specificity (estrogen-like, antiestrogen-like, or antiandrogen-like) suggested by the gene regulation pattern exhibited by HPTE.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Summary of Gene Expression Data from Real-Time RT-PCR Analysis
 
This study confirmed several gene responses demonstrated previously, as well as identifying several new ones. Beginning with steroid receptors, PR and ER{alpha} have been shown previously to be downregulated by E2 treatment in uterine epithelium, but upregulated in stroma (Kurita et al., 2000Go; Nephew et al., 2000Go; Weihua et al., 2000Go). The epithelial response dominates our analysis, and this is probably due to increased uterine epithelial heights following treatment with either E2 or MXC as previously reported (Eroschenko 1991Go; Swartz et al., 1994Go; Walters et al., 1993Go). The AR has been localized to uterine epithelial and stromal cells (Pelletier et al., 2000Go) and is downregulated by ethinylestradiol and other environmental estrogens in the uterus (Diel et al., 2000Go). Ovarian PR mRNA was induced by both E2 and E2 + HPTE treatments (Fig. 3BGo). This is most likely an indirect effect, mediated through follicle-stimulating hormone (FSH) and cAMP, since E2 does not directly regulate PR in granulosa cells (Clemens et al., 1998Go). Increased ovarian ERß mRNA levels after treatment with HPTE (Fig. 3BGo) may be an AR-mediated effect, since FLU treatment caused the same response. Biologically, this could be due to a change in cell type profile. MXC treatment causes decreased fertility in rodents, corresponding with a decrease in corpora lutea and an increase in granulosa cells in the interstitial tissue (Gray et al., 1989Go). This increase in granulosa cell number would result in an overall increase in ERß expression in the tissue. Although testosterone, dihydroxytestosterone and FLU have been shown to regulate progesterone secretion in the ovary (Chandrasekhar and Armstrong, 1991Go; Mahesh et al., 1987Go), this is the first indication of antiandrogenic regulation of gene expression in the ovary, for ERß and JunD, by HPTE or FLU.

The decrease in IGF1A and IGFBP5 mRNA expression with E2 or HPTE treatment in the uterus was unexpected (Fig. 4AGo). Estrogen increases IGF1 in the adult ovx rodent uterus (Carlsson and Billig, 1991Go; Ghahary and Murphy 1989Go; Kapur et al., 1992Go; Murphy and Friesen 1988Go; Murphy et al., 1987Go) through what has been determined to be an ER{alpha}-dependent mechanism (Kahlert et al., 2000Go; Klotz et al., 2000Go). Adolescent rodents have high levels of circulating growth hormone in preparation for puberty, resulting in high IGF1 in the uterus (Murphy et al., 1987Go) compared to the low levels of IGF1 in adult ovariectomized rodents (Carlsson and Billig, 1991Go). This difference in IGF expression between immature and adult mice may account for the difference seen with this study compared with previously published reports using mature mice. The decreased IGF1 mRNA levels after treatment with E2 may be due to decreased ER{alpha} in the uterus. This effect was also seen in the ovary, where a slight decrease in ER{alpha} mRNA levels with E2 treatment corresponded with a slight decrease in IGF1A and IGFBP5 gene expression (though not significant; Fig. 4BGo). In the ovary, IGF1 amplifies FSH action in granulosa cells, thereby determining the ovulatory fate of follicles (Adashi et al., 1997Go). In developmental studies with E2 and MXC, reduced fertility in adult animals may result from repressed IGF1 mRNA and FSH levels in immature animals, thereby preventing ovulation and causing the buildup of large atretic follicles (Chapin et al., 1997Go).

Clus mRNA is associated with tissue remodeling (loss) in the uterus (Brown et al., 1995Go) and is repressed in rat endometrial tissue by treatment with E2 (Wunsche et al., 1998Go). Our data confirm this result, which is not unexpected given the uterotropic response observed in our animals after 3 days of treatment with E2 and HPTE (Fig. 1Go). This finding is also observed with MXC treatment (Eroschenko, 1991Go; Swartz et al., 1994Go; Walters et al., 1993Go). Induction of GPX activity (but not mRNA) in the uterus and ovary by E2 has previously been reported (Ohwada et al., 1996Go). The time course of estrogen-induced uterotropic response positively correlates with increased GPX and glutathione reductase activities in rats (Suojanen et al., 1980Go), and GPX activity in the human ovary is significantly decreased in premenopausal women in comparison to menopausal women, when circulating estradiol levels are markedly reduced (Okatani et al., 1993Go). It is thought that lipid peroxides may be produced during cell growth and DNA synthesis that is induced by E2 treatment (Ohwada et al., 1996Go). Since HPTE treatment does not cause this increase in GPX3, it is either a unique effect of E2, or it may be an HPTE-inhibited response, given the slight attenuation of the E2 response in both the uterus and the ovary with HPTE cotreatment. Conversely, GST activity is inhibited by quinoid metabolites of estrogen (Chang et al., 1998Go), and GST levels are low in uterine tumors induced by E2 and testosterone (Hudson et al., 1998Go). In the ovary, GST activity is induced with E2 treatment (Singh and Pandey, 1996Go). Our data confirm these findings, and the lack of regulation by HPTE treatment indicates that this effect is unique to E2.

CTSB secretion is induced by E2 in rat uterine tissue (Pietras and Szego, 1975Go), but regulation in ovarian cells has not previously been reported. Although the effects observed in this study are small, CTSB is localized to developing follicles in the ovary, only a small fraction of our tissue sample, and its expression increases in granulosa cells during follicle maturation (Oksjoki et al., 2001Go). Given the localization of ERß to ovarian granulosa cells, the reversal of the E2 regulation of CTSB with HPTE cotreatment may be an ERß-mediated effect. Since E2 regulates its own synthesis through a negative feedback mechanism acting on the hypothalamic-pituitary axis, inhibition of Cyp17 by E2 in the ovary is likely an LH-mediated response. The regulation of Cyp17 by HPTE is more complex because of the competing ER{alpha} agonist activity and the antiandrogenic activity, in which Cyp17 is upregulated by FLU (Fig. 6BGo). The antiandrogenic effect may occur in the granulosa cells, which express AR and FSH receptors as well as secreting inhibin. Inhibin participates in a paracrine mechanism that locally amplifies androgen synthesis (Hillier et al., 1994Go), thereby up-regulating Cyp17.

In this study, microarray and RT-PCR analysis have shown that several gene families appear to be regulated by E2 or HPTE in some unique, but also overlapping, patterns. These gene changes are mediated either through the steroid receptors present in the given tissue, through the steroid receptors in another tissue leading to a down-stream effect in the tissue examined, or through an alternate undetermined mechanism. Without coordinate immunohistochemistry analyses that co-localizes the proteins, it is impossible to be absolutely certain of the particular mechanism of action. Microarray analyses are useful for this type of study, but they are not without limitations. Qualitative information derived from microarrays should not be confused with the quantitative data generated from standard molecular techniques, such as real-time RT-PCR, that measure relative gene expression levels. The modest changes in mRNA expression levels observed for several genes were far below what was predicted by microarray analysis. Two examples are cFMS and STAT1, which exhibited 3–4-fold changes in uterine expression in mice treated with E2 or HPTE and analyzed by microarray (data not shown); however, these changes were not confirmed by RT-PCR analysis (Fig. 7AGo). Biological variability in the expression of these two genes in the uterus is probably the cause of this discrepancy. Our array analysis was performed on only one animal from each treatment group, whereas the quantitative RT-PCR was performed on all 3 animals per treatment group. Had the array analysis been performed on all 3 animals and then averaged treatment groups compared to each other, the resulting gene ratios would likely have been more reliable for quantitation.

This is the first study to explore ligand-specific patterns of gene regulation through ER{alpha}, ERß, and AR as determined by microarray analysis and confirmed with quantitative RT-PCR. Many xenoestrogens, such as MXC and HPTE, exhibit some agonist, partial agonist, or antagonist properties through ER binding that result in tissue-specific regulation of gene expression. Results from this study highlight how the expression of specific receptors in a given tissue or cell type, and as yet undetermined interaction of the receptors and their corresponding cofactor proteins, can affect the biological consequence of exposure during developmental, prepubertal, or adult life stage. Gene profiling is a useful tool to explore differences between xenobiotic chemicals and the hormones that they mimic, and it can also provide valuable information and insight into the biological mechanisms that result in toxicological pathologies.


    ACKNOWLEDGMENTS
 
The authors thank Kim Lehman and Drs. Valerie Shultz, Norman Barlow, Barry McIntyre, Katie Turner, David Dorman, and Paul Foster for technical assistance and helpful discussions. This work was supported by NIH grants ES09106 and ES04917 (awarded to SS) and by the American Chemistry Council.


    NOTES
 
1 To whom correspondence should be addressed. E-mail: gaido{at}ciit.org. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Adashi, E. Y., Resnick, C. E., Payne, D. W., Rosenfeld, R. G., Matsumoto, T., Hunter, M. K., Gargosky, S. E., Zhou, J., and Bondy, C. A. (1997). The mouse intraovarian insulin-like growth factor I system: Departures from the rat paradigm. Endocrinology 138, 3881–3890.[Abstract/Free Full Text]

Al-Jamal, J. H., and Dubin, N. H. (2000). The effect of raloxifene on the uterine weight response in immature mice exposed to 17ß-estradiol, 1,1,1-trichloro-2, 2-bis(p-chlorophenyl)ethane, and methoxychlor. Am. J. Obstet. Gynecol. 182, 1099–1102.[ISI][Medline]

Alm, H., Tiemann, U., and Torner, H. (1996). Influence of organochlorine pesticides on development of mouse embryos in vitro. Reprod. Toxicol. 10, 321–326.[ISI][Medline]

Brown, T. L., Moulton, B. C., Baker, V. V., Mira, J., and Harmony, J. A. (1995). Expression of apolipoprotein J in the uterus is associated with tissue remodeling. Biol. Reprod. 52, 1038–1049.[Abstract]

Bulger, W. H., Muccitelli, R. M., and Kupfer, D. (1978). Studies on the in vivo and in vitro estrogenic activities of methoxychlor and its metabolites. Role of hepatic mono-oxygenase in methoxychlor activation. Biochem. Pharmacol. 27, 2417–2423.[ISI][Medline]

Carlsson, B., and Billig, H. (1991). Insulin-like growth factor-I gene expression during development and estrous cycle in the rat uterus. Mol. Cell. Endocrinol. 77, 175–180.[ISI][Medline]

Chandrasekhar, Y., and Armstrong, D. T. (1991). Effects of the antiandrogen hydroxyflutamide on progesterone secretion by pre-ovulatory rat follicles in vivo and in vitro. Can. J. Physiol. Pharmacol. 69, 1288–1293.[ISI][Medline]

Chang, M., Zhang, F., Shen, L., Pauss, N., Alam, I., van Breemen, R. B., Blond, S. Y., and Bolton, J. L. (1998). Inhibition of glutathione S-transferase activity by the quinoid metabolites of equine estrogens. Chem. Res. Toxicol. 11, 758–765.[ISI][Medline]

Chapin, R. E., Harris, M. W., Davis, B. J., Ward, S. M., Wilson, R. E., Mauney, M. A., Lockhart, A. C., Smialowicz, R. J., Moser, V. C., Burka, L. T., and Collins, B. J. (1997). The effects of perinatal/juvenile methoxychlor exposure on adult rat nervous, immune, and reproductive system functions. Fundam. Appl. Toxicol. 40, 138–157.[ISI][Medline]

Clemens, J. W., Robker, R. L., Kraus, W. L., Katzenellenbogen, B. S., and Richards, J. S. (1998). Hormone induction of progesterone receptor (PR) messenger ribonucleic acid and activation of PR promoter regions in ovarian granulosa cells: Evidence for a role of cyclic adenosine 3',5'-monophosphate but not estradiol. Mol. Endocrinol. 12, 1201–1214.[Abstract/Free Full Text]

Couse, J. F., Lindzey, J., Grandien, K., Gustafsson, J. A., and Korach, K. S. (1997). Tissue distribution and quantitative analysis of estrogen receptor-{alpha} (ER{alpha}) and estrogen receptor-ß (ERß) messenger ribonucleic acid in the wild-type and ER{alpha}-knockout mouse. Endocrinology 138, 4613–4621.[Abstract/Free Full Text]

Cummings, A. M., and Gray, L. E., Jr. (1989). Antifertility effect of methoxychlor in female rats: Dose- and time-dependent blockade of pregnancy. Toxicol. Appl. Pharmacol. 97, 454–462.[ISI][Medline]

Cummings, A. M., and Metcalf, J. L. (1994). Mechanisms of the stimulation of rat uterine peroxidase activity by methoxychlor. Reprod. Toxicol. 8, 477–486.[ISI][Medline]

Diel, P., Schulz, T., Smolnikar, K., Strunck, E., Vollmer, G., and Michna, H. (2000). Ability of xeno- and phytoestrogens to modulate expression of estrogen-sensitive genes in rat uterus: Estrogenicity profiles and uterotropic activity. J. Steroid Biochem. Mol. Biol. 73, 1–10.[ISI][Medline]

Eroschenko, V. P. (1991). Ultrastructure of vagina and uterus in young mice after methoxychlor exposure. Reprod. Toxicol. 5, 427–435.[ISI][Medline]

Eroschenko, V. P., Abuel-Atta, A. A., and Grober, M. S. (1995). Neonatal exposures to technical methoxychlor alters ovaries in adult mice. Reprod. Toxicol. 9, 379–387.[ISI][Medline]

Eroschenko, V. P., Johnson, T. J., and Rourke, A. W. (2000). Estradiol and pesticide methoxychlor do not exhibit additivity or synergism in the reproductive tract of adult ovariectomized mice. J. Toxicol. Environ. Health A 60, 407–421.[Medline]

Eroschenko, V. P., Rourke, A. W., and Sims, W. F. (1996). Estradiol or methoxychlor stimulates estrogen receptor (ER) expression in uteri. Reprod. Toxicol. 10, 265–271.[ISI][Medline]

Gaido, K. W., Leonard, L. S., Maness, S. C., Hall, J. M., McDonnell, D. P., Saville, B., and Safe, S. (1999). Differential interaction of the methoxychlor metabolite 2,2-bis-(p-hydroxyphenyl)-1,1,1-trichloroethane with estrogen receptors alpha and beta. Endocrinology 140, 5746–5753.[Abstract/Free Full Text]

Gaido, K. W., Maness, S. C., McDonnell, D. P., Dehal, S. S., Kupfer, D., and Safe, S. (2000). Interaction of methoxychlor and related compounds with estrogen receptors alpha and beta, and androgen receptor: Structure-activity studies. Mol. Pharmacol. 58, 852–858.[Abstract/Free Full Text]

Ghahary, A., and Murphy, L. J. (1989). Uterine insulin-like growth factor-I receptors: Regulation by estrogen and variation throughout the estrous cycle. Endocrinology 125, 597–604.[Abstract]

Ghosh, D., Taylor, J. A., Green, J. A., and Lubahn, D. B. (1999). Methoxychlor stimulates estrogen-responsive messenger ribonucleic acids in mouse uterus through a non-estrogen receptor (non-ER) alpha and non-ER beta mechanism. Endocrinology 140, 3526–3533.[Abstract/Free Full Text]

Gray, L. E., Jr., Ostby, J., Ferrell, J., Rehnberg, G., Linder, R., Cooper, R., Goldman, J., Slott, V., and Laskey, J. (1989). A dose-response analysis of methoxychlor-induced alterations of reproductive development and function in the rat. Fundam. Appl. Toxicol. 12, 92–108.[ISI][Medline]

Gray, L. E., Jr., Ostby, J. S., Ferrell, J. M., Sigmon, E. R., and Goldman, J. M. (1988). Methoxychlor induces estrogen-like alterations of behavior and the reproductive tract in the female rat and hamster: Effects on sex behavior, running wheel activity, and uterine morphology. Toxicol. Appl. Pharmacol. 96, 525–540.[ISI][Medline]

Hall, D. L., Payne, L. A., Putnam, J. M., and Huet-Hudson, Y. M. (1997). Effect of methoxychlor on implantation and embryo development in the mouse. Reprod. Toxicol. 11, 703–708.[ISI][Medline]

Hillier, S. G., Whitelaw, P. F., and Smyth, C. D. (1994). Follicular oestrogen synthesis: The `two-cell, two-gonadotrophin` model revisited. Mol. Cell. Endocrinol. 100, 51–54.[ISI][Medline]

Hodges, L. C., Bergerson, J. S., Hunter, D. S., and Walker, C. L. (2000). Estrogenic effects of organochlorine pesticides on uterine leiomyoma cells in vitro. Toxicol. Sci. 54, 355–364.[Abstract/Free Full Text]

Hudson, C. E., Kelly, M. M., Schwartz, D. A., Schofield, D. A., DeHaven, J. E., Schulte, B. A., and Norris, J. S. (1998). Glutathione S-transferase in hormonal carcinogenesis. Chem. Biol. Interact. 111–112, 343–350.

Kahlert, S., Nuedling, S., van Eickels, M., Vetter, H., Meyer, R., and Grohe, C. (2000). Estrogen receptor alpha rapidly activates the IGF-1 receptor pathway. J. Biol. Chem. 275, 18447–18453.[Abstract/Free Full Text]

Kapur, S., Tamada, H., Dey, S. K., and Andrews, G. K. (1992). Expression of insulin-like growth factor-I (IGF-I) and its receptor in the peri-implantation mouse uterus and cell-specific regulation of IGF-I gene expression by estradiol and progesterone. Biol. Reprod. 46, 208–219.[Abstract]

Klotz, D. M., Hewitt, S. C., Korach, K. S., and Diaugustine, R. P. (2000). Activation of a uterine insulin-like growth factor I signaling pathway by clinical and environmental estrogens: Requirement of estrogen receptor-alpha. Endocrinology 141, 3430–3439.[Abstract/Free Full Text]

Kuiper, G. G., Carlsson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S., and Gustafsson, J. A. (1997). Comparison of the ligand-binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 138, 863–870.[Abstract/Free Full Text]

Kupfer, D., and Bulger, W. H. (1987). Metabolic activation of pesticides with proestrogenic activity. Fed. Proc. 46, 1864–1869.[ISI][Medline]

Kurita, T., Lee, K. J., Cooke, P. S., Taylor, J. A., Lubahn, D. B., and Cunha, G. R. (2000). Paracrine regulation of epithelial progesterone receptor by estradiol in the mouse female reproductive tract. Biol. Reprod. 62, 821–830.[Abstract/Free Full Text]

Mahesh, V. B., Mills, T. M., Bagnell, C. A., and Conway, B. A. (1987). Animal models for study of polycystic ovaries and ovarian atresia. Adv. Exp. Med. Biol. 219, 237–257.[Medline]

Metcalf, J. L., Laws, S. C., and Cummings, A. M. (1996). Methoxychlor mimics the action of 17 beta-estradiol on induction of uterine epidermal growth factor receptors in immature female rats. Reprod. Toxicol. 10, 393–399.[ISI][Medline]

Murphy, L. J., and Friesen, H. G. (1988). Differential effects of estrogen and growth hormone on uterine and hepatic insulin-like growth factor I gene expression in the ovariectomized hypophysectomized rat. Endocrinology 122, 325–332.[Abstract]

Murphy, L. J., Murphy, L. C., and Friesen, H. G. (1987). Estrogen induces insulin-like growth factor-I expression in the rat uterus. Mol. Endocrinol. 1, 445–450.[Abstract]

Nephew, K. P., Long, X., Osborne, E., Burke, K. A., Ahluwalia, A., and Bigsby, R. M. (2000). Effect of estradiol on estrogen receptor expression in rat uterine cell types. Biol. Reprod. 62, 168–177.[Abstract/Free Full Text]

Ohwada, M., Suzuki, M., Sato, I., Tsukamoto, H., and Watanabe, K. (1996). Glutathione peroxidase activity in endometrium: Effects of sex hormones and cancer. Gynecol. Oncol. 60, 277–282.[ISI][Medline]

Okatani, Y., Morioka, N., Wakatsuki, A., Nakano, Y., and Sagara, Y. (1993). Role of the free radical-scavenger system in aromatase activity of the human ovary. Horm. Res. 39, 22–27.[ISI][Medline]

Oksjoki, S., Soderstrom, M., Vuorio, E., and Anttila, L. (2001). Differential expression patterns of cathepsins B, H, K, L, and S in the mouse ovary. Mol. Hum. Reprod. 7, 27–34.[Abstract/Free Full Text]

Pelletier, G., Labrie, C., and Labrie, F. (2000). Localization of oestrogen receptor alpha, oestrogen receptor beta, and androgen receptors in the rat reproductive organs. J. Endocrinol. 165, 359–370.[Abstract/Free Full Text]

Pietras, R. J., and Szego, C. M. (1975). Surface modifications evoked by estradiol and diethylstilbestrol in isolated endometrial cells: Evidence from lectin probes and extracellular release of lysosomal protease. Endocrinology 97, 1445–1454.[Abstract]

Quarmby, V. E., and Korach, K. S. (1984). The influence of 17ß-estradiol on patterns of cell division in the uterus. Endocrinology 114, 694–702.[Abstract]

Roy, D., Angelini, N. L., and Belsham, D. D. (1999). Estrogen directly represses gonadotropin-releasing hormone (GnRH) gene expression in estrogen receptor-alpha (ERalpha)- and ERbeta-expressing GT1–7 GnRH neurons. Endocrinology 140, 5045–5053.[Abstract/Free Full Text]

Sar, M., and Welsch, F. (1999). Differential expression of estrogen receptor-beta and estrogen receptor- alpha in the rat ovary. Endocrinology 140, 963–971.[Abstract/Free Full Text]

Shelby, M. D., Newbold, R. R., Tully, D. B., Chae, K., and Davis, V. L. (1996). Assessing environmental chemicals for estrogenicity using a combination of in vitro and in vivo assays. Environ. Health Perspect. 104, 1296–1300.[ISI][Medline]

Singh, D., and Pandey, R. S. (1996). Glutathione S-transferase in rat ovary: Its changes during estrous cycle and increase in its activity by estradiol-17ß. Indian J. Exp. Biol. 34, 1158–1160.[Medline]

Smid-Koopman, E., Blok, L. J., Chadha-Ajwani, S., Helmerhorst, T. J., Brinkmann, A. O., and Huikeshoven, F. J. (2000). Gene expression profiles of human endometrial cancer samples using a cDNA-expression array technique: Assessment of an analysis method. Br. J. Cancer 83, 246–251.[ISI][Medline]

Suojanen, J. N., Gay, R. J., and Hilf, R. (1980). Influence of estrogen on glutathione levels and glutathione-metabolizing enzymes in uteri and R3230AC mammary tumors of rats. Biochim. Biophys. Acta 630, 485–496.[ISI][Medline]

Swartz, W. J., Wink, C. S., and Johnson, W. D. (1994). Response of adult murine uterine epithelium to 50% methoxychlor. Reprod. Toxicol. 8, 81–87.[ISI][Medline]

Walters, L. M., Rourke, A. W., and Eroschenko, V. P. (1993). Purified methoxychlor stimulates the reproductive tract in immature female mice. Reprod. Toxicol. 7, 599–606.[ISI][Medline]

Weihua, Z., Saji, S., Makinen, S., Cheng, G., Jensen, E. V., Warner, M., and Gustafsson, J. A. (2000). Estrogen receptor (ER) beta, a modulator of ER{alpha} in the uterus. Proc. Natl. Acad. Sci. U.S.A. 97, 5936–5941.[Abstract/Free Full Text]

Wunsche, W., Tenniswood, M. P., Schneider, M. R., and Vollmer, G. (1998). Estrogenic regulation of clusterin mRNA in normal and malignant endometrial tissue. Int. J. Cancer 76, 684–688.[ISI][Medline]