FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by RHEN, T.
Right arrow Articles by CIDLOWSKI, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by RHEN, T.
Right arrow Articles by CIDLOWSKI, J. A.
(The FASEB Journal. 2003;17:1849-1870.)
© 2003 FASEB

Dexamethasone blocks the rapid biological effects of 17ß-estradiol in the rat uterus without antagonizing its global genomic actions

TURK RHEN, SHERRY GRISSOM, CYNTHIA AFSHARI and JOHN A. CIDLOWSKI1

Laboratory of Signal Transduction, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina, USA

1Correspondence: Laboratory of Signal Transduction, National Institute of Environmental Health Sciences, National Institutes of Health, 111 T.W. Alexander Dr., Research Triangle Park, NC 27709, USA. E-mail: cidlows1{at}niehs.nih.gov


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Estrogens and glucocorticoids have opposing effects on the female reproductive tract, but the molecular basis for this antagonism is poorly understood. We therefore examined the biological and transcriptional programs induced by estrogens and glucocorticoids in the uterus of immature female rats. Estradiol 17ß (E2) rapidly induced morphological changes reminiscent of an acute inflammatory response, including infiltration of eosinophils, edema in the stroma and myometrium, and a decrease in the height of luminal epithelial cells, whereas dexamethasone (Dex) only altered stromal cell morphology. When coadministered with E2, Dex completely blocked the proinflammatory effects of E2. Surprisingly, examination of E2 and Dex effects on gene expression using cDNA microarrays and real-time PCR revealed that these hormones had similar effects on the expression of many genes and that very few genes displayed antagonistic regulation. Together, these results indicate strong discord between the early biologic and genomic actions of estrogens and glucocorticoids and highlight a complex regulatory role for glucocorticoids and GR in the mammalian uterus.—Rhen, T., Grissom, S., Afshari, C., Cidlowski, J. A. Dexamethasone blocks the rapid biological effects of 17ß-estradiol in the rat uterus without antagonizing its global genomic actions.


Key Words: 17ß-estradiol • dexamethasone • uterus • microarray


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
INTENSE PAIN AND EMOTIONAL TRAUMA, as well as less dramatic forms of stress like caloric restriction, agonistic social encounters, and general anxiety, can increase circulating levels of glucocorticoids (1 , 2) . In turn, glucocorticoids bring about profound changes in the physiology and function of numerous tissues and organs, including the promotion of gluconeogenesis in the liver, degradation of proteins to free amino acids in muscle, and lipolysis (3 4 5) . This class of hormones has well-characterized effects on the brain and behavior (6 , 7) and important immunosuppressive and anti-inflammatory effects (8 9 10) . Also of considerable importance is the fact that stress and glucocorticoids can inhibit reproduction. Although many of the antagonistic effects of glucocorticoids on reproduction occur in the hypothalamus, pituitary, or gonad (11 12 13 14 15 16) , there are direct effects on other tissues like the female reproductive tract. Whereas estrogens normally induce growth and differentiation of the uterus, glucocorticoids antagonize estrogen-induced uterine growth and can even block embryo implantation in the rat (17 18 19 20 21) . Despite the negative impact of glucocorticoids in the uterus and the prevalence of stress-induced reproductive problems, the molecular basis of glucocorticoid antagonism of estrogen-induced uterine growth and differentiation has not been defined. Moreover, it is not clear which cell types in the rat uterus express the glucocorticoid receptor (GR) even though receptors for glucocorticoids have been identified in uterine extracts using biochemical assays (22 , 23) .

Given that the classic mode of action for steroid hormones is to stimulate or inhibit transcription of target genes (24) , we hypothesized that glucocorticoids might widely block estrogen-regulated gene expression just as they obstruct estrogen-induced uterine growth. To test this hypothesis, we examined the early effects of 17ß-estradiol and dexamethasone on global patterns of gene expression in the rat uterus using cDNA microarray experiments. Patterns of gene expression were then confirmed by real-time PCR for a representative set of novel genes (i.e., ESTs) and for genes not previously recognized to be regulated by these hormones or to be expressed in the uterus. We found diverse patterns of gene regulation in response to 17ß-estradiol and dexamethasone, including hormone-specific effects, overlapping effects, additive effects, synergistic effects, as well as antagonism for a few genes.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Animals, hormone treatments, and tissue collection
Animals were treated according to a protocol approved by the Institutional Animal Care and Use Committee at NIEHS. Steroids were dissolved in ethanol to make stock solutions, which were then added to saline vehicle for the following treatments. Immature female rats (Charles River Laboratory; strain CD; 21–23 days old) were injected intraperitoneally with 0.1 mL of sterile 0.9% saline (containing 3.7 µL ethanol), 1 µg of 17ß-estradiol (E2) dissolved in 0.1 mL of 0.9% saline (containing 3.7 µL ethanol), 1 mg dexamethasone (Dex) in 0.1 mL of 0.9% saline (containing 3.7 µL ethanol), or 1 µg E2 and 1 mg Dex in 0.1 mL of 0.9% saline (containing 3.7 µL ethanol). The dose of E2 used here produces serum levels of E2 in the range observed on the evening of proestrus in intact, cycling female rats (25) . Corticosterone, the endogenous rodent glucocorticoid, activates both the glucocorticoid and mineralocorticoid receptors. In contrast, the synthetic glucocorticoid Dex is a highly specific glucocorticoid receptor agonist with essentially no mineralocorticoid activity (26) . The dose of Dex used in the current study is generally in the therapeutic range (26) . However, a dose ~one-sixth (i.e., 0.15 mg/50 g) of that used in our study produces serum concentrations of Dex in the range of 65 ng/mL in rats (27) , which is ~one-tenth of the serum concentration of corticosterone found in rats subjected to restraint stress (28) .

This basic experiment was replicated five times, twice for histology and immunohistochemistry, twice for microarray studies, and a fifth time for a real-time PCR study. Animals were killed by decapitation 4 h post-treatment and their uteri immediately dissected. In the initial studies, uteri were cleaned of fat and connective tissue, weighed, fixed in PBS buffered 4% formalin, and finally rinsed and stored in 70% ethanol for histology and immunohistochemistry. In the first microarray study and in the PCR study, uteri were cleaned, frozen in liquid nitrogen, and stored at -70°C until extraction of total RNA. In the second microarray study, uteri were cleaned, placed in RNAlater and stored at 4°C until total RNA was extracted (Ambion, Austin, TX, USA).

Histology and immunohistochemistry
Uteri were embedded in paraffin and cut at 4 microns. Cross sections were deparaffinized in xylene, 100% ethanol, 95% ethanol, and rehydrated in 1x automation buffer (2x for 5 min in each solution). Some sections were stained with hematoxylin and eosin or hematoxylin and Sirius Red (Direct Red 80) for histology while others were processed for immunohistochemistry. Sirius Red was used to stain eosinophils as described (29) . In brief, 1 g of Sirius Red was dissolved in 90 mL distilled water, 100 mL ethanol, and 2 mL of 1% (w/v) NaOH. Approximately 6 mL of 20% (w/v) NaCl was added until a slight precipitate formed. This solution was left overnight at room temperature and filtered the following day. Deparaffinized slides were stained with hematoxylin for 2 min, rinsed in 1x automation buffer (1x for 3 min), rinsed in 70% ethanol (1x for 3 min), then stained with the Sirius Red solution for 1 h. Slides were dehydrated in ascending alcohols and xylenes and coverslipped using Permount.

For GR immunohistochemistry, endogenous peroxidase was quenched by placing deparaffinized slides in 3% H2O2 for 15 min at room temperature. Slides were then washed 2x for 5 min each in 1x automation buffer. Antigens were unmasked in 1x sodium citrate buffer using a pressure cooker for 5 min. After cooling, slides were washed 2x for 3 min each in distilled water and equilibrated in 1x automation buffer for 5 min. Nonspecific binding was blocked by preincubating slides in 5% normal goat serum for 20 min at room temperature. Endogenous avidin and biotin binding sites were blocked according to the manufacturer’s instructions (Avidin-Biotin Blocking Kit; Vector Laboratories, Burlingame, CA, USA). The blocking solution was aspirated, primary rabbit anti-GR antiserum GR57 was applied (1:7500), and sections were incubated overnight at 4°C; GR57 is a well-characterized polyclonal antiserum raised against two synthetic peptides in the amino terminus of the GR (30) . Normal rabbit serum was used as a negative control. Slides were then washed with 1x automation buffer for 5 min and the secondary (biotinylated goat anti-rabbit) antibody applied for 20 min. The secondary antibody was washed with 1x automation buffer for 5 min. An avidin-biotin peroxidase complex was then applied for 20 min and washed with 1x automation buffer for 5 min. Freshly prepared Chromagen DAB was applied for 6 min in the dark and rinsed in tap water for 3 min. Slides were counterstained with hematoxylin, dried in ascending alcohols, xylene, and coverslipped using Permount. Liver sections were processed in parallel as a positive control for GR.

Immunohistochemistry for ER{alpha} was carried out according to the manufacturer’s instructions for paraffin sections (ABC Vectastain Elite ABC kit; Vector Laboratories) with the following modifications. An avidin-biotin blocking step was included after the initial block with 5% normal goat serum. Sections were incubated with primary rabbit anti-ER{alpha} antibody (1:1000; Upstate Cell Signaling, Lake Placid, NY, USA) for 1 h instead of 30 min at room temperature. Slides were washed twice for 5 min each with 1x automation buffer after each blocking step and after incubation with the primary and secondary antibodies. All other steps were in line with the manufacturer’s instructions.

Protein isolation and Western blots
We collected uteri, liver, skeletal muscle, and hypothalamus from female rats for Western blots. Approximately 25 mg of each tissue was placed in ice-cold RIPA buffer containing protease inhibitors (Complete Mini, Roche Diagnostics, Germany). Tissues were thoroughly homogenized, incubated for 20 min on ice, and spun at 20,000 g for 10 min. Supernatant was collected and total protein measured. Protein extracts (30 µg) were then separated under reducing and denaturing conditions on 8% Tris-glycine gels (Novex, San Diego, CA, USA) and transferred to nitrocellulose membranes. Membranes were washed in TBS-T (Tris-buffered saline with 0.1% Tween-20) and blocked in TBS-T containing 5% nonfat milk for 1 h at room temperature. Blots were incubated with anti-GR antiserum GR57 (1:2000) overnight at 4°C or with anti-ER{alpha} antibody (1:1000; Upstate Cell Signaling) for 1 h at room temperature. After washing in TBS-T (3x for 10 min each), blots were probed with peroxidase-conjugated goat anti-rabbit secondary antibody (1:10,000) for 1 h at room temperature. After blots were washed in TBS-T (3x for10 min each), bands were visualized using ECL (Amersham Pharmacia Biotech Arlington Heights, IL, USA).

RNA isolation
Total RNA was isolated from pooled uterine tissue using the Qiagen RNeasy kit (Qiagen, Valencia, CA, USA). Uteri were pooled from 60 vehicle-treated females in the first microarray study. Twenty uteri were pooled for each hormone-treated group. Pooled, frozen uteri from each group were processed in batches totaling 200–250 mg. Each batch was added to ice-cold buffer, immediately homogenized, and processed according to the standard Qiagen midiprep protocol. Total RNA from each batch was pooled to generate four RNA samples: vehicle-treated, E2-treated, Dex-treated, and E2 and Dex-treated.

Uteri were pooled from 40 vehicle-treated females in the second microarray study. Thirteen uteri were pooled for each hormone-treated group. Uteri from each group were removed from the RNAlater solution, patted dry, and processed in batches totaling 200–250 mg. Each batch was added to ice-cold buffer and immediately homogenized. In contrast to the first study, the homogenate from the second study was processed according to the Qiagen protocol for heart, muscle, and skin tissues, which is designed to remove contractile proteins, connective tissues, and collagen. This procedure differs from the standard protocol in the addition of a proteinase K digestion step after tissue homogenization. As in the first study, total RNA from each batch was pooled to generate four RNA samples: vehicle-treated, E2-treated, Dex-treated, and E2 and Dex-treated. The total RNA from both experiments had 260 nm/280 nm absorbance ratios of 1.6–1.8 in pure water and displayed discrete 18S and 28S rRNA bands on formaldehyde-agarose gels.

cDNA microarray hybridization and analysis
The rat cDNA chip was developed at NIEHS. A list of all genes on the chip is located at http://www.dir.niehs.nih.gov/microarray/chips.htm. The cDNA chips were prepared as described previously (31) , except that ~6800 genes were arrayed on the current version of the chip. cDNA clones consisted of 500–2000 base pairs of the 3' end of sequence verified clones (Research Genetics, Huntsville, AL, USA). Antisense cDNA was amplified from purified plasmid DNA by PCR and an aliquot was separated and checked for quality on agarose gels. The rest of the PCR products were purified by ethanol precipitation, suspended in ArrayIt buffer (Telechem, San Jose, CA, USA), and spotted onto glass slides using a robotic DNA arrayer (Beecher Instruments, Bethesda, MD, USA).

The conditions for labeling first-strand cDNAs from RNA samples and hybridizing those cDNAs to the NIEHS rat chip have been described in detail (31 , 32) . Purified total RNA from the vehicle and hormone-treated groups was reverse transcribed to incorporate fluorescently labeled Cy3 or Cy5 deoxy-UTP into first-strand cDNA. Labeled cDNAs from each of the three hormone treatment groups (i.e., Cy3) were mixed individually with labeled cDNA from the vehicle group (i.e., Cy5) and hybridized to three separate cDNA chips. The labeling and hybridization procedures were repeated four times for the three contrasts between vehicle and hormone-treated groups, flipping the fluor used to label the control and the treated groups. Thus, there were 12 hybridizations for each microarray study. This design provided an a priori contrast between baseline levels of gene expression vs. hormone-regulated levels of gene expression, i.e., vehicle-E2, vehicle-Dex, and vehicle-E2&Dex. Chips were scanned for intensity of Cy3 and Cy5 using the GenePix 4000 Microarray Scanner (Axon Instruments, Foster City, CA, USA) with excitation of the two fluors at 532 and 635 nm wavelengths, respectively.

After scanning, raw pixel intensity was analyzed using the ArraySuite v2.0 image processing software (Scanalytics, Fairfax, VA, USA). This software locates spots on the array, measures local background for each spot, and subtracts background florescence using an algorithm developed by Chen et al. (33) . We then plotted the intensity of the Cy3 and Cy5 signals for all genes in each hybridization and determined 95% confidence limits for each regression. Genes were considered up- or down-regulated if they fell outside the 95% confidence limits for 3/4 or 4/4 hybridizations. Ratios (i.e., log2 of the intensity ratios) were then calculated between the signal intensity in the treated groups vs. the signal intensity in the control group for each regulated gene (or spot) on each chip. Thereafter, we calculated mean expression values using the four hybridizations in each microarray study. All subsequent analyses were conducted using the two independent estimates of hormone-regulated expression values (i.e., the means for each microarray study). In a second round of screening, we calculated 95% confidence intervals for expression values in response to E2, Dex, and E2&Dex for each gene using the mean for each microarray study as a replicate (i.e., 6 observations per gene). The two-tailed t value from the Student’s t distribution with 5 degrees of freedom and a pooled estimate for the standard error of the mean were used to calculate the 95% confidence intervals. Genes whose confidence limits overlapped 0 for E2-, Dex-, and E2&Dex-treated groups were excluded from further analysis. Consequently, we only report genes that are significantly regulated using a conservative, two-tiered screen of expression levels.

Real-time quantitative PCR
Total RNA was isolated from individual uteri of eight animals per treatment group as in the first microarray and stored at -70°C. Real-time PCR was performed using the TaqMan Gold RT-PCR kit and primers and probes designed using Primer Express software (Table 1 ; Applied Biosystems). Total RNA (0.4 µg) was added to a 50 µL reaction mixture containing 1x TaqMan buffer A, 300 µM dATP, 300 µM dCTP, 300 µM dGTP, 600 µM dUTP, 5.5 mM MgCl2, 200 nM forward primer, 200 nM reverse primer, 100 nM probe, 0.05 U/µL of Taq polymerase, 0.25 U/µL MultiScribe Reverse Transcriptase, and 0.4 U/µL Rnase inhibitor. Reactions were run on an ABI Prism 7700 sequence detection system. Reverse transcription was carried out for 30 min at 50°C, the sample heated to 95°C for 10 min, and 40 cycles of two-step PCR performed (15 s at 95°C and 1 min at 60°C). As a negative control, 18S rRNA was measured for each sample.


View this table:
[in this window]
[in a new window]
  		Table 1. Primer and probe sets for real-time PCR analysis of indicated genesa

Pooled total RNA from uteri of vehicle-treated females was used as a control and to generate a standard curve for each gene. This RNA was added to a tube containing no reverse transcriptase to control for DNA contamination, which could interfere with measurements of mRNA. Total RNA from each of the 32 samples was run without reverse transcriptase in one assay using primers and probes for alcohol dehydrogenase. No signal was detected after 40 PCR cycles in the absence of reverse transcriptase, indicating that all samples were free of DNA. In addition, no signal was detected when reverse transcriptase was added but RNA template was not, indicating that there was no contamination from exogenous RNA or DNA. Control RNA from uteri of vehicle-treated females was diluted and added to reaction mixtures in the following quantities to generate a standard curve for each gene: 1 µg, 0.5 µg, 0.25 µg, 0.125 µg, 0.0625 µg, and 0.03125 µg. The standard curve for all genes displayed an increase of one threshold cycle for each halving of template concentration. Threshold cycle or CT was recorded for each of the 32 samples. The CT value for each sample was then subtracted from the mean CT value for the vehicle-treated group for each gene. This procedure results in a mean of zero for the vehicle-treated group and allows direct comparison of the TaqMan data to the microarray data: i.e., expression values are on the same scale because a change of 1 CT value equals a twofold change in mRNA levels just as a change of 1 on the log2 scale for the microarray indicates a twofold difference in relative fluorescent intensity. All statistics were done using JMP 3.2 software (SAS Institute, Cary, NC, USA).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Injection of E2 induced a 1.5-fold increase in wet uterine mass 4 h after treatment (Fig. 1 A). In parallel with the change in uterine mass, E2 induced changes like an acute inflammatory response: i.e., vasodilation, edema in the stroma and myometrium, and accumulation of fluid in the lumen (Fig. 1B ). Coadministration of 1 mg Dex blocked these E2-induced changes. Examination of the stroma at a higher magnification shows accumulation of interstitial fluid in the E2-treated group and that this effect was blocked by Dex (Fig. 1C ). Histological changes in the luminal epithelium from columnar to cuboidal in response to E2 were similarly blocked by Dex. Finally, stromal cells were smaller and more tightly packed in Dex-treated uteri than in vehicle-treated uteri. Although E2 did not block these Dex effects on the stroma, E2 did appear to block the Dex-induced decrease in nucleus size. Sirius Red was used to specifically stain eosinophils as an additional marker of uterine inflammation. Treatment with E2 resulted in a fourfold increase in the number of eosinophils found per cross section of uterus, while coadministration of Dex completely antagonized E2-induced recruitment of eosinophils (Fig. 1D ).



View larger version (40K):
[in this window]
[in a new window]
  		Figure 1. A) Effects of 1 µg of E2, 1 mg Dex, or 1 µg E2 and 1 mg Dex on wet uterine mass of immature female rats 4 h post-treatment compared with vehicle-treated controls. Mass is expressed as the mean ± 1 standard error. B) Cross sections of representative uteri from vehicle-treated, E2-treated, Dex-treated, and E2&Dex-treated females. C) Higher magnification of cross sections of uterine stroma and luminal epithelium from vehicle-treated, E2-treated, Dex-treated, and E2&Dex-treated females. D) Number of eosinophils in 4 micron cross sections of uteri from immature female rats treated with vehicle, 1 µg of E2, 1 mg Dex, or 1 µg E2 and 1 mg Dex for 4 h. Number of eosinophils per cross section are expressed as the mean ± 1 standard error (n=20 cross sections/group {5 cross sections/animal and 4 animals/group}).

Despite the clear effects of Dex when given alone or in combination with E2, it is not known which uterine cell types express the GR. The pattern of GR expression in the uterus was therefore analyzed by Western blot and immunohistochemistry. Levels of GR in the uterus were comparable to or even higher than those found in the liver when analyzed by Western blot (Fig. 2 A). Immunostaining for GR in liver (a classical glucocorticoid-responsive tissue) and uterus was also similar. The GR was detected in all uterine cell types, including luminal epithelium, glandular epithelium, stroma, and myometrium (Fig. 2B ). Staining for GR in the liver and uterus was primarily nuclear because animals were not adrenalectomized.



View larger version (107K):
[in this window]
[in a new window]
  		Figure 2. A) Western blot for glucocorticoid receptor (GR) using protein extracts from immature female rat brain, liver, muscle, and uterus. B) Tissue sections of liver and uterus processed in parallel to localize GR protein by immunohistochemistry. Areas stained light and dark brown are immunoreactive using a GR-specific primary antibody.

Immunohistochemistry and Western blots were used to determine whether hormone treatments would alter expression of GR in the uterus. After 4 h, there were no detectable differences in the pattern of GR expression in uteri from vehicle-treated, E2-treated, Dex-treated, or E2&Dex-treated females (Fig. 3 A). Levels of GR expression were very similar among uteri from vehicle and hormone-treated females when Western blots were analyzed by semiquantitative densitometry (Fig. 3B, C ).



View larger version (58K):
[in this window]
[in a new window]
  		Figure 3. A) Immunohistochemistry for glucocorticoid receptor (GR) in uteri from vehicle-treated females and females treated with 1 µg of E2, 1 mg of Dex, or 1 µg E2 and 1 mg Dex. Areas stained brown are immunoreactive using a GR-specific primary antibody. B) Western blot for GR using protein extracts from uteri of vehicle-treated, E2-treated, Dex-treated, and E2&Dex-treated females. C) Densitometric analysis of immunoblots for GR expression in the uteri of vehicle-treated, E2-treated, Dex-treated, and E2&Dex-treated females. Optical density was measured in arbitrary units and is expressed as the mean ± 1 standard error (n=4 animals/ group).

Regulation of uterine ER{alpha} expression by E2 and Dex was also examined by immunohistochemistry and Western blots. After 4 h there were clear differences in ER{alpha} expression among uteri from vehicle-treated, E2-treated, Dex-treated, or E2&Dex-treated females (Fig. 4 A). Specifically, E2 treatment appeared to cause a decrease in ER{alpha} expression in luminal and glandular epithelium. The same effect was observed even when Dex was coadministered with E2. Administration of Dex alone had no noticeable effect on ER{alpha} expression. In accord with these findings, treatment with E2 significantly decreased expression of ER{alpha} in the uterus as determined by semiquantitative densitometry of Western blots (Fig. 4B, C ). Coadministration of Dex did not block E2-induced down-regulation of ER{alpha} and had no effect on ER{alpha} expression on its own.



View larger version (54K):
[in this window]
[in a new window]
  		Figure 4. A) Immunohistochemistry for estrogen receptor (ER) in uteri from vehicle-treated females and females treated with 1 µg of E2, 1 mg of Dex, or 1 µg E2 and 1 mg Dex. Areas stained brown are immunoreactive using an ER-specific primary antibody. B) Western blot for ER using protein extracts from uteri of vehicle-treated, E2-treated, Dex-treated, and E2&Dex-treated females. C) Densitometric analysis of immunoblots for ER expression in the uteri of vehicle-treated, E2-treated, Dex-treated, and E2&Dex-treated females. Optical density was measured in arbitrary units and is expressed as the mean ± 1 standard error (n=4 animals/ group).

Given that Dex blocks many of the early E2-induced morphological and histological changes, effects of E2 and Dex on global patterns of gene expression in the rat uterus were analyzed using cDNA microarray experiments. Approximately 4946 of the 6817 genes on the chip had expression levels above background. In all, 472 of these genes were regulated by E2 and/or Dex, which represents 7% of the 6817 genes on the chip. Approximately 429 genes were regulated by E2 and 249 genes by Dex. We observed a close correspondence between results from the microarray experiments. The correlation coefficients for expression values (i.e., log2 of the intensity ratios) between experiments 1 and 2 were 0.91 for the E2-treated groups, 0.79 for the Dex-treated groups, and 0.95 for the groups treated with E2&Dex. The relatively low correlation for Dex-treated groups reflects the fact that all 472 genes were used in calculating the correlation coefficient, not only the 249 Dex-regulated genes. In support of this interpretation, the correlation coefficient increased to 0.86 for the Dex-treated groups when 11 outliers were excluded from the analysis. Genes already known to be regulated by E2 and/or Dex were also detected in the current study (Table 2 ). Moreover, a number of novel hormone-regulated genes were confirmed in the real-time PCR study described below. These findings indicate that the microarray results are highly repeatable and strongly suggest that the genes identified here are truly regulated by E2 and/or Dex.


View this table:
[in this window]
[in a new window]
  		Table 2. Genes significantly regulated by 17ß-estradiol (E2) and/or dexamethasone (Dex) in the immature rat uterus as determined by microarray analysisa

Given the large number of genes regulated in this study, we used principal components analysis (PCA) to examine overall patterns of gene induction and repression in response to E2, Dex, and E2&Dex. The eigen vector for the first principal component (PC1), which explained 80.6% of the variation in gene expression, indicates that E2, Dex, and the combined treatment with E2&Dex have similar effects on the expression of many genes (Table 3 ). For example, many genes induced by E2 were also induced by Dex (see upper right quadrant of Fig. 5 A). Conversely, many genes repressed by E2 were also repressed by Dex (see lower left quadrant in Fig. 5A ). The eigen vector for PC2, which explained 12.2% of the variation in gene expression, reflects hormone-specific effects of E2 treatment vs. Dex treatment; there were genes induced by E2 and repressed by Dex, and vice versa (see upper left and lower right quadrants in Fig. 5A ). The eigen vectors for three of the remaining principal components (PC3, PC4, and PC6), which explained just 3.6%, 1.7%, and 0.6% of the variation in gene expression, generally capture differences between the biological replicates for Dex-, E2-, and E2&Dex-treated groups, respectively. The eigen vector for PC5, which explained 1.3% of the variation in gene expression, generally represents the difference between the effect of E2 and Dex when administered alone vs. their effect when administered together. Although Dex completely blocked the E2-induced increase in uterine mass and the histological changes observed at 4 h, the PCA clearly shows that Dex had a disproportionately small effect on E2-regulated gene expression. The same conclusions were drawn from cluster analysis; the E2&Dex-treated groups cluster closely with the E2-treated groups, which are somewhat removed from the Dex-treated groups (Fig. 5B ). Relative branch lengths in the dendrogram are proportional to the relative strength of correlations among biological replicates and among hormone treatments.


View this table:
[in this window]
[in a new window]
  		Table 3. Principal components analysis of expression values (i.e., log2 of the intensity ratios) for 472 genes regulated by 17ß-estradiol (E2) and/or dexamethasone (Dex)a



View larger version (45K):
[in this window]
[in a new window]
  		Figure 5. A) Scatter plot of average gene expression values (i.e., log2 of the intensity ratios) for 17ß-estradiol (E2) and dexamethasone (Dex) regulated genes from the microarray studies. Red points in the upper right quadrant represent genes induced by E2 and by Dex. Green points in the lower left quadrant represent genes repressed by E2 and by Dex. Blue points in the upper left quadrant represent genes repressed by E2 but induced by Dex. Yellow points in the lower right quadrant represent genes induced by E2, but repressed by Dex. B) Cluster analysis of expression values for genes regulated by E2 and/or Dex. The similarity between biological replicates and among hormone treatments is proportional to the branch lengths in the dendrogram shown on the right. A black background indicates that no difference was detected between the hormone-treated and vehicle control samples. For genes induced by a given hormone treatment, the level of induction is correlated with the intensity of the red color. For genes repressed by a given hormone treatment, the level of repression is correlated with the intensity of the green color. Note the resemblance between gene expression in response to E2 alone and E2 in the presence of Dex.

When we divided Fig. 6 A into nine bins with expression values greater than 0.4, less than -0.4, or between 0.4 and -0.4 on the log2 scale for each hormone, we observed a significant association between the effects of E2 and Dex (Likelihood Ratio {chi}2=102.7, P<0.0001; see Table 4 ). To be exact, there were more genes induced by both E2 and by Dex than expected by chance. There were also more genes repressed by both hormones than expected by chance. Conversely, there were fewer genes induced by E2 and repressed by Dex than expected, as well as fewer genes induced by Dex and repressed by E2 than expected. There were close to the expected number of genes induced or repressed by E2 (and weakly regulated by Dex). Finally, there were close to the expected number of genes induced or repressed by Dex (and weakly regulated by E2). While these results support the general conclusions drawn from the PCA and cluster analyses, they should be interpreted with more caution because expression values demarcating the bins were chosen arbitrarily and expression values actually exhibit continuous variation in response to both hormones.



View larger version (61K):
[in this window]
[in a new window]
  		Figure 6. Diverse effects of 1 µg of 17ß-estradiol (E2), 1 mg dexamethasone (Dex), or 1 µg E2 and 1 mg Dex on gene expression levels (i.e., log2 of the intensity ratios) in the immature rat uterus as determined by microarray analysis. A) Genes induced by E2 and partially or completely antagonized by Dex: D3 (Iodothyronine deiodinase, type III); Rdc1 (chemokine orphan receptor 1); AA859368 (EST); AA956007 (EST); AA819269 (EST); AA818744 (EST); PSTI-II (pancreatic secretory trypsin inhibitor, type II). B) Genes induced by Dex and partially or completely antagonized by E2: AA955251 (EST); AA900944 (EST); AD1 (alcohol dehydrogenase, class 1 {alpha} polypeptide); MGla (matrix Gla protein); Amph II (amphiphysin II); AA899930 (EST). C) Representative genes induced to a similar degree by E2 and Dex when administered alone or together: AA900046 (EST); Rpl27 (ribosomal protein L27); AA924400 (EST); AA955888 (EST); AA926143 (EST); AA818945 (EST). D) Representative genes induced in an additive manner by E2 and Dex: Gal (galanin); Mt2 (metallothionein II); Mt1a (metallothionein Ia); Zfp36 (zinc finger protein 36); AA899629 (EST); AA955619 (EST); AA955826 (EST). E) Synergistic gene induction by combined treatment with E2 and Dex: AI137780 (EST); Ykt6 (prenylated SNARE protein); AA901294 (EST); Rgs2 (regulator of G-protein signaling protein 2).


View this table:
[in this window]
[in a new window]
  		Table 4. Tabulation of genes induced, not (or weakly) regulated, and repressed by 17ß-estradiol (E2) administered alone or dexamethasone (Dex) administered alonea

Altogether, Dex antagonized only seven E2-induced genes (Fig. 6A , Table 2 ). Three of these genes were completely blocked by Dex. Expressed sequence tag (EST) AA819269 displays some sequence similarity to the retinal short-chain dehydrogenase/reductase family (i.e., accession # XM 129697). EST AA818744 displays high sequence similarity to mouse lymphocyte antigen 68. Pancreatic secretory trypsin inhibitor type II has been shown to display a pattern of regulation like acute-phase proteins. Four E2-induced genes were partially blocked by coadministration of Dex, including iodothyronine deiodinase type III, chemokine orphan receptor 1, and ESTs AA859368 and AA956007. The converse pattern was also observed in which E2 antagonized 11 genes induced by Dex (Fig. 6B , Table 2 ). Alcohol dehydrogenase class 1, matrix Gla protein, collagen {alpha} 1 (Xv), follistatin, mitochondrial 3-hydroxy-3-methylglutaryl CoA synthase, and amphysyin 2 were the only genes in this cluster with known functions. The other five genes induced by Dex and antagonized by E2 were ESTs (AA900830, AA925553, AA900944, AA955251, and AA899930) that displayed little similarity to any known genes.

As indicated in Fig. 5 , E2 and Dex regulated many genes in a similar manner. Representative genes illustrate the overlapping effects of E2 and Dex: these hormones have essentially the same effect when administered alone or together (Fig. 6C ). Additive effects between E2 and Dex were also observed for many genes (Fig. 6D ). All genes exhibiting overlapping and additive effects are listed in Table 2 under the heading "Genes induced by both E2 and Dex." Genes repressed by both E2 and Dex are also listed in Table 2 , as are those specifically regulated by each hormone. We detected potential synergistic interactions between E2 and Dex for seven genes (Fig. 6E , Table 2 ), including prenylated SNARE protein (Ykt6), regulator of G-protein signaling protein 2, preproenkephalin, protein PM5 precursor, AI137780, EST AA923876, and AA901294. EST AI137780 shows high homology to vigilin, a protein involved in stabilization of vitellogenin mRNA.

Although groups of genes with similar patterns of expression in response to E2 and Dex may be controlled by similar mechanisms, these clusters do not necessarily give any indication of gene function. This was especially true for the current study because there was no clear correlation between patterns of hormone responsiveness, putative function of known genes, and the morphological and histological changes induced by E2 and Dex at 4 h. We therefore grouped genes into functional categories using the GeneSpring program and PubMed searches (Table 2) . Although more than half of the regulated genes were ESTs of unknown function, a large number of genes regulated by E2 and/or Dex code for proteins that may play a role in long-term tissue remodeling. Such proteins include cytoskeletal proteins (19 genes) and extracellular matrix proteins (13 genes) that control cell shape, cell adhesion, and cell–cell interactions. Most cytoskeletal genes were induced or repressed by E2 alone and not influenced by Dex. The same was true of extracellular matrix proteins, with the exception of collagen {alpha}1 (Xv) and matrix Gla protein, which were induced by Dex and antagonized by E2. In general, nucleic acid binding proteins (16 genes) and transcription factors (4 genes) were induced by E2 alone and not regulated or antagonized by Dex. However, translation initiation factor eIF3 (p110 subunit), inhibitor of DNA binding 3, and GATA binding protein 3 were repressed by both E2 and Dex. Other functional categories of genes with significant representation include various enzymes (19 genes), GTP proteins (14 genes), signal transduction proteins (14 genes), proteins involved in immunity/inflammation (12 genes), and various transporters (11 genes).

General patterns of hormonal regulation were confirmed for most genes analyzed by real-time PCR. EST AA819269 and chemokine orphan receptor 1 were both induced by E2 and antagonized by Dex (Fig. 7 A). Conversely, alcohol dehydrogenase class 1 and matrix Gla protein were induced by Dex and repressed by E2 (Fig. 7B ). Dex and E2 had an additive effect on the induction of EST AA858882 (Fig. 7C ). Inhibitor-of-DNA binding 3 was repressed by Dex and E2, and more so by the combined treatment with both hormones (Fig. 7C ). When analyzed by real-time PCR, E2 and Dex had overlapping effects on expression of Prenylated SNARE protein (Ykt6) (Fig. 7D ) rather than the apparent synergy observed in the microarray study. Although the microarray also suggested that EST AA901294 was regulated synergistically by E2 and Dex, there was no detectable effect of these hormones on expression of AA901294 with real-time PCR (Fig. 7D ). Hormone treatments had no effect on levels of the negative control 18S rRNA in the immature rat uterus (Fig. 7E ).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 7. Effects of 1 µg of 17ß-estradiol (E2), 1 mg dexamethasone (Dex), or 1 µg E2 and 1 mg Dex on relative gene expression levels in the immature rat uterus as determined by real-time PCR analysis. Gene expression values are graphed as the threshold cycle of the vehicle control minus the threshold cycle of the treated group as described in Materials and Methods. A) Genes induced by E2 and completely or partially antagonized by Dex: AA819269 (EST); Rdc1 (chemokine orphan receptor 1). B) Genes induced by Dex and partially antagonized by E2: AD1 (alcohol dehydrogenase, class 1 {alpha} polypeptide); MGla (matrix Gla protein). C) Representative genes induced or repressed in an additive manner by E2 and Dex: TIS11 (Zinc finger protein tristetraproline 11); Inhibitor of DNA binding-3 (ID3). D) Expression of Ykt6 (prenylated SNARE protein) and AA901294 (EST). E) Expression of the negative control 18S rRNA.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Glucocorticoids interfere with estrogen action in the uterus, implantation of the blastocyst, as well as growth of the developing fetus. Nevertheless, relatively little is known about the molecular basis for glucocorticoid effects in the female reproductive tract. Hence, we used microarray analysis to examine E2- and Dex-induced changes in gene expression with the idea that such changes might play a role in the morphological and histological reorganization described above. There were two particularly surprising results from these studies given that members of the steroid/thyroid hormone receptor family are primarily thought to act by inducing or suppressing gene transcription. First, although Dex completely blocked the inflammatory like changes induced by E2 in the uterus, very few genes displayed antagonistic regulation by E2 and Dex (i.e., just seven, or 1.5%) of the regulated genes were induced by E2 and blocked by Dex 4 h post-treatment. Conversely, only two (or 0.4%) genes were repressed by E2 and antagonized by Dex. These results are in stark contrast to our initial hypothesis that Dex might widely block E2-regulated gene expression.

A second surprising finding was that E2 and Dex had similar effects on the expression of many genes. Common E2 and Dex effects as estimated by principal components analysis explain ~80% of the variation in gene expression levels. Cluster analysis also illustrated the broad correspondence between E2 and Dex effects on global patterns of gene expression. A similar conclusion was drawn using arbitrary gene expression thresholds to categorize genes as induced or repressed (i.e., 102. or 21.6%) of the regulated genes were induced by E2 and by Dex. Conversely, 23 (or 4.9%) of the regulated genes were repressed by E2 and by Dex. Genes induced (or repressed) by both hormones cannot logically play a role in the antagonistic biological effects of E2 and Dex.

An important caveat regarding these findings is that cell-specific patterns of gene expression could hinder the identification of antagonistically regulated genes. This constraint, however, is applicable to any microarray study using animal tissues composed of multiple cell types. For example, E2 has opposing effects on expression of the progesterone receptor (PR) in epithelial and stromal cells of the mouse uterus, decreasing PR in the former but increasing PR in the latter (34) . These authors suggested that differential regulation of PR could influence the regulation of progesterone-responsive genes in a cell-specific manner. It was therefore important to characterize E2 and Dex regulation of ER and GR in the rat uterus. We found that E2 and Dex had no detectable effect on the pattern or level of GR expression 4 h after hormone treatment. In contrast, administration of E2 alone resulted in significant down-regulation of ER{alpha} after 4 h. This result is in agreement with previous work showing full suppression of uterine ER occurs by 6 h after E2 treatment (35) . Coadministration of Dex did not alter E2-induced down-regulation of ER{alpha}. Treatment with Dex alone had no effect on the pattern or level of ER expression. These results indicate that Dex antagonism of E2-induced uterine inflammation at 4 h is not mediated by Dex-induced changes in expression of ER{alpha}. It will be important to study long-term effects of estrogens and glucocorticoids on ER and GR expression in the uterus, as 48 h treatment with E2 has been shown to down-regulate GR and glucocorticoid induced gene expression in MCF-7 cells (36) .

While the few antagonistically regulated genes identified here may play a critical role in mediating the pro- and anti-inflammatory effects of E2 and Dex, we think it unlikely. Pancreatic secretory trypsin inhibitor type II (PSTI-II) is associated with acute pancreatitis and displays a pattern of expression like acute-phase proteins (37) but we are unaware of any studies indicating that this gene is itself an inflammatory mediator. A gene with similarity to the short-chain dehydrogenase/reductase family (EST AA819269) was also induced by E2 and completely antagonized by Dex. The short-chain dehydrogenase/reductase family is involved in retinoid metabolism (38) , suggesting that this putative enzyme could be involved in mediating estrogen and glucocorticoid effects on retinoic acid signaling in the uterus. A homologue of mouse lymphocyte antigen 68 (EST AA818744) was also "induced" by E2 and "antagonized" by Dex. Regulation of this gene, however, is probably not direct but more likely a result of the migration of white blood cells into the E2-treated uterus. It is known that neutrophils and macrophages, in addition to the eosinophils examined here, infiltrate the uterus upon estrogenic stimulation and that progesterone, like Dex, antagonizes this effect (39) . Chemokine orphan receptor 1 is also likely to be expressed by infiltrating white blood cells rather than resident uterine cells. Three other genes were induced by E2 and partially antagonized by Dex. Thyroid hormones are known to play an important part in uterine physiology and type 3 iodothyronine deiodinase has been shown to be highly expressed in the pregnant rat uterus (40) . Two other antagonistically regulated ESTs, AA859368 and AA956007, display no similarity to known proteins and have no putative function. Consequently, none of these seven genes are prime candidates for mediating the suite of proinflammatory effects induced by E2 and antagonized by Dex.

A more trivial possibility is that the critical E2-regulated and Dex-antagonized genes were not on the NIEHS chip. However, the general conclusion that glucocorticoids do not widely block E2-regulated gene expression, and vice versa, is sound because the NIEHS chip contains a large and random sample of rat genes (i.e., 1/5th of the ~30,000 predicted protein-coding genes; refs 41 , 42 ). Another prospect is that antagonistic interactions occur between genes that are specifically regulated by each hormone; Dex could theoretically induce a gene that antagonizes the action of a gene induced by E2. However, we do not favor this notion and are currently testing the hypothesis that the early effects of E2 are mediated by nongenomic mechanisms and that Dex antagonizes these mechanisms. Nongenomic effects of steroid hormones, although neglected for a long time, are receiving renewed attention (43) . Moreover, estrogens and glucocorticoids are known to induce rapid changes in the vasculature, including vasodilation, vasoconstriction, and changes in vascular permeability (44 45 46) . Other studies have shown rapid changes (i.e., within 2 min) in microtubule and microvilli structure in lumenal epithelial cells of the uterus in E2-treated rats (47 , 48) .

The hypothesis that nongenomic mechanisms mediate the pro- and anti-inflammatory effects of E2 and Dex at 4 h does not diminish the potential importance of the novel genes that we have identified as regulated at the transcriptional level. In fact, many genes identified here have previously been shown to be regulated by E2 and/or Dex and some are known to play a key role in various aspects of uterine physiology and pregnancy. Basagin, for instance, is required for embryo implantation and is regulated by estrogens during the peri-implantation period (49 , 50) . It is therefore interesting that basagin was induced by E2 in our study but was not antagonized by Dex. This result suggests that glucocorticoid obstruction of implantation is not directly mediated by inhibition of basagin expression. Other key steroid-regulated genes in the uterus include galanin, metallothionein, insulin-like growth factor 1, creatine kinase (brain isoform), ornithine decarboxylase, integrins ({alpha}6 and ß1), pregnancy specific glycoprotein, guanylate cyclase, and insulin-like growth factor binding proteins 3 and 5. It is probable that many of the novel E2 and/or Dex regulated genes also play a crucial role in the female reproductive tract.

For example, conversion of retinol to retinal is the rate-limiting step in the production of retinoic acid (51 , 52) , which is a well-established paracrine signal and key regulator of epithelial cell differentiation in the female reproductive tract (53 , 54) . The alcohol dehydrogenase family is involved in retinoid metabolism and is generally thought to play a pivotal role in local retinoic acid synthesis and signaling, with alcohol dehydrogenase class 1 (AD1) catalyzing the initial conversion of retinol to retinal (52) . Whereas E2 significantly repressed AD1 mRNA to nearly half control levels, Dex induced AD1 mRNA >threefold (data from real-time PCR). Consequently, the net effect of both hormones administered together was a twofold induction of AD1 mRNA. These observations suggest that E2, when administered alone, might decrease the level of retinoic acid in the uterus by repressing AD1. In contrast, Dex when administered alone might increase retinoic acid levels by inducing AD1. When administered together, we hypothesize that Dex would override the inhibitory effects of E2 on retinoic acid synthesis in the uterus. E2 and Dex also have antagonistic effects on expression of the putative short-chain dehydrogenase/reductase gene (EST AA819269), further suggesting that estrogen and glucocorticoid signaling pathways may converge to regulate retinoic acid production in the uterus.

Another significant finding is that two distinct classes of steroid hormones—namely, glucocorticoids and estrogens—have diverse effects on gene expression, including hormone-specific effects, overlapping effects, independent (i.e., additive) effects, as well as antagonism. Hormone-specific effects are relatively easy to understand because receptors for each hormone bind with high affinity to distinct consensus sequences in DNA called estrogen-responsive elements (AGGTCAnnnTGACCT) or glucocorticoid-responsive elements (GGRACAnnnTGTTCT). The finding that E2 and Dex have overlapping and additive effects on the expression of many genes was unexpected and deserves further exploration from both regulatory and functional perspectives. Additive effects could rationally be mediated by separate hormone-responsive elements in the same promoter. Overlapping effects might theoretically be mediated by degenerate hormone-responsive elements recognized by either steroid hormone receptor or by interactions with other transcription factors like AP-1. Analyses of genes with dissimilar patterns of regulation will also help elucidate the mechanistic basis for antagonistic interactions between estrogens and glucocorticoids. Ultimately, it will be important to determine how the same two hormones can have such diverse effects on gene transcription in the same tissue or cell type.

General patterns of gene expression were confirmed by real-time PCR for six genes, including four that were antagonistically regulated by E2 and Dex, one that was additively induced by both hormones, and one that was repressed in an additive manner by both hormones. Synergistic effects were observed in the microarray study, but were not confirmed for the two genes analyzed by real-time PCR. One of these genes, EST AA901294, was not regulated in the confirmatory study, indicating a false positive signal was obtained for the combined hormone treatment in the microarray study. Prenylated SNARE protein was induced by all hormone treatments in the real-time PCR study, but only by E2 and the combined treatment in the microarray study, suggesting a false negative signal was obtained for Dex treatment in the microarray study. Yet despite the discrepancy for one hormone treatment for each of these two genes, the overall correspondence between the microarray and real-time PCR studies was high. In fact, there is correspondence for 22 of 24 comparisons (i.e., three hormone treatments for each of the eight genes analyzed). Given that we only used two biological replicates in the microarray study, an apparent error rate of ~4% (i.e., one false positive signal) is quite low and close to the nominal type I error rate of 5%. We also detected one false negative signal, suggesting that the statistical power of our study is relatively high (i.e., there was a high probability of detecting genes that are truly regulated by E2 and/or Dex).

Further studies are required to determine whether the other genes are truly regulated in a synergistic manner by estrogens and glucocorticoids. Notwithstanding, estrogens are known to regulate vigilin, also called high density lipoprotein binding protein: this gene is induced by estrogens in the uterus and pancreas and appears to mediate the estrogen-dependent stabilization of vitellogenin mRNA (55 56 57) . Estrogens and glucocorticoids are known to regulate preproenkephalin mRNA (58 , 59) . There is also some evidence that estrogens might induce Regulator of G-coupled protein signaling 2 (Rgs2) in the ovary (60) and that a related protein interacts with ER{alpha} and represses its ability to trans-activate gene expression (61) . Moreover, a recent microarray study demonstrated that Rgs2 is induced by Dex in a breast cancer cell line (62) . Our finding that a number of other GTP binding proteins were regulated by E2 and/or Dex suggests that steroid hormones might significantly modulate G-protein signaling in the uterus.

In summary, rigorous statistical analysis of microarray data using principal components, cluster, and categorical analyses revealed a general lack of correspondence between the early biologic and genomic actions of estrogens and glucocorticoids in the rat uterus. Whereas Dex completely blocked the early inflammatory-like effects of E2, we detected only a small number of genes for which antagonistic regulation by glucocorticoids and estrogens was apparent. Furthermore, many genes were regulated in a similar manner by E2 and Dex in vivo, which was also unexpected. These results highlight a significant, but highly complex, regulatory role for glucocorticoids and GR in the mammalian uterus and hint at novel nongenomic interactions between estrogens and glucocorticoids.


View this table:
[in this window]
[in a new window]
  		Table 2A. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2B. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2C. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2D. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2E. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2F. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2G. (continued)


View this table:
[in this window]
[in a new window]
  		Table 2H. (continued)


   ACKNOWLEDGMENTS
 
We would like to acknowledge technical support by L. Bennett, N. Clayton, J. Collins, D. Ducharme, and J. Tucker. We also thank S. Hewitt, M. Birnbaumer, and two anonymous reviewers for comments that improved the manuscript.

Received for publication December 19, 2002. Accepted for publication June 10, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Delbende, C., Delarue, C., Lefebvre, H., Bunel, D. T., Szafarczyk, A., Mocaer, E., Kamoun, A., Jegou, S., Vaundry, H. (1992) Glucocorticoids, transmitters and stress. Br. J. Psychiatry Suppl. 15,24-35
  2. Miller, D. B., O'Callaghan, J. P. (2002) Neuroendocrine aspects of the response to stress. Metabolism 51,5-10[CrossRef]
  3. Dallman, M. F., Strack, A. M., Akana, S. F., Bradbury, M. J., Hanson, E. S., Scribner, K. A., Smith, M. (1993) Feast and famine: critical role of glucocorticoids with insulin in daily energy flow. Front. Neuroendocrinol. 14,303-347[CrossRef][Medline]
  4. Mitch, W. E. (2000) Mechanisms accelerating muscle atrophy in catabolic diseases. Trans. Am. Clin. Climatol. Assoc. 111,258-269[Medline]
  5. Leal-Carro, A., Soto, A., Martinez, M. A., Dieguez, C., Casanueva, F. F. (2001) Influence of cortisol status on leptin secretion. Pituitary 4,111-116[CrossRef][Medline]
  6. Sapolsky, R. M. (1993) Potential behavioral modification of glucocorticoid damage to the hippocampus. Behav. Brain Res. 57,175-182[CrossRef][Medline]
  7. Welberg, L. A., Seckl, J. R. (2001) Prenatal stress, glucocorticoids and the programming of the brain. J Neuroendocrinol. 13,113-128[CrossRef][Medline]
  8. Cidlowski, J. A., King, K. L., Evans-Storms, R. B., Montague, J. W., Bortner, C. D., Hughes, F. M., Jr (1996) The biochemistry and molecular biology of glucocorticoid-induced apoptosis in the immune system. Recent Prog. Horm. Res. 51,457-490
  9. McKay, L. I., Cidlowski, J. A. (1999) Molecular control of immune/inflammatory responses: interactions between nuclear factor-kappa B and steroid receptor-signaling pathways. Endocr. Rev. 20,435-459[Abstract/Free Full Text]
  10. Ashwell, J. D., Lu, F. W., Vacchio, M. S. (2000) glucocorticoids in T cell development and function. Annu. Rev. Immunol. 18,309-345[CrossRef][Medline]
  11. Chrousos, G. P., Torpy, D. J., Gold, P. W. (1998) Interactions between the hypothalamic-pituitary-adrenal axis and the female reproductive system: clinical implications. Ann. Intern. Med. 129,229-240[Abstract/Free Full Text]
  12. Chandran, U. R., DeFranco, D. B. (1999) Regulation of gonadotropin-releasing hormone gene transcription. Behav. Brain Res. 105,29-36[CrossRef][Medline]
  13. Tilbrook, A. J., Turner, A. I., Clark, I. J. (2000) Effects of stress on reproduction in non-rodent mammals: the role of glucocorticoids and sex differences. Rev. Reprod. 5,105-113[Abstract]
  14. Sapolsky, R. M., Romero, L. M., Munck, A. U. (2000) How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr. Rev. 21,55-89[Abstract/Free Full Text]
  15. Huang, T. J., Shirley Li, P. (2001) Dexamethasone inhibits luteinizing hormone- induced synthesis of steroidogenic acute regulatory protein in cultured rat preovulatory follicles. Biol. Reprod. 64,163-170[Abstract/Free Full Text]
  16. Viau, V. (2002) Functional cross-talk between the hypothalamic-pituitary-gonadal and -adrenal axes. J. Neuroendocrinol. 14,506-513[CrossRef][Medline]
  17. Velardo, J. T., Hisaw, F. L., Bever, A. T. (1956) Inhibitory action of desoxycorticosterone acetate, cortisone acetate, and testosterone on uterine growth induced by estradiol-17ß. Endocrinology 59,165-169
  18. Bitman, J., Cecil, H. C. (1967) Differential inhibition by cortisol of estrogen- stimulated uterine responses. Endocrinology 80,423-429[Medline]
  19. Johnson, D. C., Dey, S. K. (1980) Role of histamine in implantation: dexamethasone inhibits estradiol-induced implantation in the rat. Biol. Reprod. 22,1136-1141[Abstract]
  20. Markaverich, B. M., Upchurch, S., Clark, J. H. (1981) Progesterone and dexamethasone antagonism of uterine growth: a role for a second nuclear binding site for estradiol in estrogen action. J. Steroid Biochem. 14,125-132[CrossRef][Medline]
  21. Gunin, A. G., Mashin, I. N., Zakharov, D. A. (2001) Proliferation, mitosis orientation and morphogenetic changes in the uterus following chronic treatment with both estrogen and glucocorticoid hormones. J. Endocrinol. 169,23-31[Abstract]
  22. Panko, W. B., Clark, J. H., Walters, M. R. (1981) Glucocorticoid receptors: documentation in the rat uterus. J. Recept. Res. 2,29-45[Medline]
  23. Izawa, M., Satoh, Y., Iwasaki, K., Ishii, S. (1984) Glucocorticoid receptor in the rat uterus. Endocrinology (Jpn.) 31,491-500
  24. Aranda, A., Pascual, A. (2001) Nuclear hormone receptors and gene expression. Physiol. Rev. 81,1269-1304[Abstract/Free Full Text]
  25. Gibbs, R. B. (1996) Fluctuations in relative levels of choline acetyltransferase mRNA in different regions of the rat basal forebrain across the estrous cycle: effects of estrogen and progesterone. J. Neurosci. 16,1049-1055[Abstract/Free Full Text]
  26. McKay, L. I., Cidlowski, J. A. (2000) Corticosteroids. Bast, R. C. Kufe, D. W. Pollock, R. E. Weichselbaum, R. R. Holland, J. F. Frei, E. eds. Cancer Medicine ,730-742 B.C. Decker, Inc. London.
  27. Mizoshige, K., Yuzurihara, M., Ishige, A., Sasaki, H., Chui, D. H., Tabira, T. (2001) Chronic stress differentially regulates glucocorticoid negative feedback response in rats. Psychoneuroendocrinology 26,443-459[CrossRef][Medline]
  28. Viau, V., Meaney, M. J. (1991) Variations in the hypothalamic-pituitary-adrenal response to stress during the estrous cycle in the rat. Endocrinology 129,2503-2511[Abstract]
  29. Reibiger, I., Spanel-Borowski, K. (2000) Differences in localization of eosinophils and mast cells in the bovine ovary. J. Reprod. Fertil. 118,243-249[Abstract]
  30. McGimsey, W. C., Cidlowski, J. A., Stumpf, W. E., Sar, M. (1991) Immunocytochemical localization of the glucocorticoid receptor in rat brain, pituitary, liver, and thymus with two new polyclonal antipeptide antibodies. Endocrinology 129,3064-3072[Abstract]
  31. 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 (2002) Gene expression analysis reveals chemical-specific profiles. Toxicol. Sci. 67,219-231[Abstract/Free Full Text]
  32. Lobenhofer, E. K., Bennett, L., Cable, P. L., Li, L., Bushel, P. R., Afshari, C. A. (2002) Regulation of DNA replication fork genes by 17beta-estradiol. Mol. Endocrinol. 16,1215-1229[Abstract/Free Full Text]
  33. Chen, Y., Dougherty, E., Bittner, M. (1997) Ratio-based decisions and the quantitative analysis of cDNA microarray images. J. Biomed. Opt. 2,364-374[CrossRef]
  34. Tibbets, T. A., Mendoza-Meneses, M., O’Malley, B. W., Conneely, O. M. (1998) Mutual and intercompartmental regulation of estrogen receptor and progesterone receptor expression in the mouse uterus. Biol. Reprod. 59,1143-1152[Abstract/Free Full Text]
  35. Medlock, K. L., Forrester, T. M., Sheehan, D. M. (1991) Short-term effects of physiological and pharmacological doses of estradiol on estrogen receptor and uterine growth. J. Recept. Res. 11,743-756[Medline]
  36. Kinyamu,