Identification of Murine Uterine Genes Regulated in a Ligand-Dependent Manner by the Progesterone Receptor
http://www.100md.com
内分泌学杂志 2005年第8期
Department of Molecular and Cellular Biology (J.-W.J., K.Y.L., I.K., J.P.L., F.J.D.); Microarray Core Facility, Department of Molecular and Human Genetics (L.D.W.); and Breast Center (S.G.H.), Baylor College of Medicine, Houston, Texas 77030
Address all correspondence and requests for reprints to: Dr. Francesco J. DeMayo, Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, Texas 77030. E-mail: fdemayo@bcm.tmc.edu.
Abstract
Progesterone (P4) acting through its cognate receptor, the progesterone receptor (PR), plays an important role in uterine physiology. The PR knockout (PRKO) mouse has demonstrated the importance of the P4-PR axis in the regulation of uterine function. To define the molecular pathways regulated by P4-PR in the mouse uterus, Affymetrix MG U74Av2 oligonucleotide arrays were used to identify alterations in gene expression after acute and chronic P4 treatments. PRKO and wild-type mice were ovariectomized and then treated with vehicle or 1 mg P4 every 12 h. Mice were killed either 4 h after the first injection (acute P4 treatment) or after the fourth injection of P4 (chronic P4 treatment). At the genomic level, the major change in gene expression after acute P4 treatment was an increase in the expression of 55 genes. Conversely, the major change in gene expression after chronic P4 treatment was an overall reduction in the expression of 102 genes. In the analysis, retinoic acid metabolic genes, cytochrome P 450 26a1 (Cyp26a1), alcohol dehydrogenase 5, and aldehyde dehydrogenase 1a1 (Aldh1a1); kallikrein genes, Klk5 and Klk6; and specific transcription factors, GATA-2 and Cited2 [cAMP-corticosterone-binding protein/p300-interacting transactivator with glutamic acid (E) and aspartic acid (D)-rich tail], were validated as regulated by the P4-PR axis. Identification and analysis of these responsive genes will help define the role of PR in regulating uterine biology.
Introduction
THE OVARIAN STEROID hormone progesterone (P4) is an essential regulator of reproductive events associated with all aspects of the establishment and maintenance of pregnancy (1, 2). Most of the physiological affects of P4 are mediated through its receptor, the progesterone receptor (PR). This signaling axis, the P4-PR axis, has been investigated by dissecting the role of PR (3). PR is a transcription factor that belongs to the nuclear receptor superfamily (4, 5, 6). PR is encoded in one gene and exists as one of two isoforms, PR-A and PR-B. These isoforms arise from the alternate translation start sites in the PR gene (7). Genetic ablation of the PR gene in mice (PRKO) leads to pleiotropic reproductive abnormalities, including defects in female reproductive behavior (8), failure to ovulate, failure of the uterus to support embryo implantation, and defects in branching and glandular development in the mammary glands (2, 9). Site-directed mutagenesis of the PR gene in vivo has demonstrated that the PR-A isoform is the major mediator of P4 signaling in the mouse uterus regulating uterine function, whereas the PR-B isoform regulates uterine epithelial cell proliferation (10, 11). Although the physiological processes governed by the P4-PR signaling axis in the uterus have been identified, the molecular pathways governed by PR are not fully understood. Therefore, elucidation of the molecular pathways regulated by PR in the uterus by identification of the target genes whose transcription is regulated by PR is of great importance.
To date, only a few P4-PR-regulated genes have been identified. These include the genes encoding amphiregulin (Areg) (12), histidine decarboxylase (Hdc) (13), Hoxa-10 and -11 (14), calcitonin (15, 16), calbindin-D9K (17), Indian hedgehog (Ihh) (18), hypoxia-inducible factor 1 (HIF1A) (19), and immune-responsive gene 1 (20). These target genes have been identified by testing candidate genes (12), differential library screening (15), and DNA microarray approaches (18, 20). High-density DNA microarray technology has immensely improved the ability to identify PR-regulated genes in the uterus.
Cheon and co-workers (3) used high-density DNA microarray technology to identify PR-responsive genes in the mouse uterus by treating female mice on d 3 of pregnancy with the antiprogestin RU486 and assaying the impact on uterine target genes 24 h later. This approach identified PR-regulated genes by inhibiting PR action at a time when PR levels were elevated in all compartments. In a sense, this approach identified the impact of withdrawal of P4-PR signaling on uterine gene expression and successfully identified 148 genes that were regulated by this axis (3). In this report, high-density DNA microarray technology in combination with the PRKO mouse were used to identify genes affected by acute and chronic stimulation of the P4-PR axis.
In this study, we have taken a pharmacological approach to identify the impact of activation of the P4-PR axis on the mouse uterus. This analysis shows the impact of acute and chronic pharmacological stimulation of the P4-PR axis, allowing additional identification of how these target genes function to regulate uterine biology. The groups of genes that were validated were those involved in retinoic acid (RA) metabolism, the kallikrein (Klk) family of genes, and specific developmentally important transcription factors. The identification of these factors as targets of P4-PR regulation will shed light on the role of P4 in uterine biology.
Materials and Methods
Animals and tissue collection
Mice were maintained at the designated animal care facility at Baylor College of Medicine according to the institutional guidelines for the care and use of laboratory animals. Thirty PRKO and 30 wild-type mice were ovariectomized at 6 wk of age. Two weeks later, ovariectomized mice were injected with either vehicle (sesame oil) or P4 (in sesame oil; Sigma-Aldrich Corp., St. Louis, MO; 1 mg/mouse in 100 μl). The injections were repeated every 12 h, and groups of mice were killed 4 h after the first injection (acute P4 treatment) or 4 h after the fourth injection (chronic P4 treatment). Each treatment consisted of 15 PRKO and 15 wild-type mice. The mice were killed by cervical dislocation after placing them under anesthesia (Avertin, 2,2-tribromoethyl alcohol, Sigma-Aldrich Corp.). Uterine tissues were flash-frozen at the time of dissection and stored at –80 C.
To collect uteri from pseudopregnant mice, wild-type females were mated with vasectomized males after superovulation [5 IU pregnant mare’s serum gonadotropin (VWR Scientific Products, West Chester, PA), followed 48 h later with 5 IU human chorionic gonadotropin (Pregnyl, Organon International, Roseland, NJ)]. The morning of the vaginal plug was designated d 0.5.
RNA isolation and microarray hybridization
Total RNA was extracted from uterine tissues using the RNeasy total RNA isolation kit (Qiagen, Valencia, CA). RNA was pooled from the uteri of five mice per genotype and treatment. All RNA samples were analyzed with a Bioanalyser 2100 (Agilent Technologies, Palo Alto, CA) before microarray hybridization. RT by oligo(deoxythymidine), followed by in vitro transcription and biotin labeling of cRNA, were performed (Enzo Biochem, Farmingdale, NY). The fragmented, labeled cRNA (15 μg) was hybridized to murine genome U74Av2 mouse oligonucleotide arrays (Affymetrix, Santa Clara, CA), which have approximately 12,000 genes represented, of which 6,000 are known and 6,000 are established sequence tags. Each transcript is represented on the chip as a set of 16–20 probe pairs, with each pair containing a perfect match and a mismatch. The microarrays were stained in an Affymetrix Fluidics station with streptavidin-phycoerythrin conjugates, followed by staining with an antistreptavidin antibody and streptavidin-phycoerythrin. The arrays were scanned at 3 μm with a GeneArray scanner (Affymetrix). All experiments were performed in triplicate with independent pools of RNA using 24 Affymetrix MG 74Av2 arrays.
Data analysis
After scanning and low-level quantification using Microarray Suite (Affymetrix), we used DNA-Chip Analyzer dChip (version 1.3) to adjust arrays to a common baseline using invariant set normalization (21) and to estimate expression using the PM-only model described by Li and colleagues (22, 23). All 24 Genechips were normalized to the same baseline (the Genechip with the median average intensity happened to be in the 48-h wild-type P4 treatment group), and all were modeled together. Data quality was reviewed using present call rates from MAS5 (average, 54%; range, 50–58%), ratios of 3' to 5' glyceraldehyde 3-phosphate dehydrogenase probe sets from MAS5 (average, 0.94; range, 0.8 to 1.5), and array outlier rates (average, 0.43%; range, 0.024–2.7%; only two of 24 chips >1%) from dChip. Based on the above parameters, all chips were considered of good quality and were included in subsequent analyses. We selected differentially expressed genes within each time exposure (acute and chronic) using a two-sample comparison according to the following criteria: lower boundary of 90% confidence interval of fold change greater than 1.2 and an absolute value of difference between group means greater than 80. The software uses resampling and SEs of model parameters to partially account for measurement error and unstable estimates of variability (24). Finally, we used the median number of detected genes in 50 permuted samples as an overall estimate of the false discovery rate. After excluding expressed sequence tags with no functional annotation, differentially expressed genes were classified according to Gene Ontology function using the Affymetrix annotation, a literature search in PubMed, and GenMAPP (25). Using these methods, functional categories were assigned for 197 differentially expressed genes.
Quantitative real-time PCR
Expression levels of selected genes found to be regulated by microarray analysis were validated by real-time RT-PCR TaqMan analysis using the ABI PRISM 7700 Sequence Detector System according to the manufacturer’s instructions (Applied Biosystems, Foster City, CA). For cytochrome P 450 26a1 (Cyp26a1); alcohol dehydrogenase 5 (ADH5); aldehyde dehydrogenase 1a1 (Aldh1a1); retinol-binding protein 1 genes Rbp1 and Rbp4; Klk genes Klk5 and Klk6; GATA-2; Cited2 [cAMP-corticosterone-binding protein/p300-interacting transactivator with glutamic acid (E) and aspartic acid (D)-rich tail]; and 18S RNA, prevalidated probes and primers were purchased from Applied Biosystems. RT-PCRs were performed using One-Step RT-PCR Universal Master Mix reagent and TaqMan Gene Expression Assays (Applied Biosystems) according to the manufacturer’s instructions. Standard curves were generated by serial dilution of a preparation of total RNA isolated from whole mouse uterus. All real-time PCR was performed using the three independent RNA sets. mRNA quantities were normalized against 18S RNA using ABI rRNA control reagents.
In situ hybridization
The protocol for in situ hybridization was essentially as described previously by Simmons et al. (26). Uterine tissues were fixed in 10% formalin. After overnight fixation at room temperature, tissues were dehydrated through a series of ethanol and then processed for paraffin embedding. Paraffin sections were mounted onto poly-L-lysine-coated slides (VWR Scientific Products) and then used for in situ hybridization. Riboprobes were labeled with [35S]UTP by in vitro transcription of amplified DNA products containing the T7 polymerase promoter sequence flanking the desired nucleotide primer sequence (Promega Corp., Madison, WI). cRNA probes were generated for Cyp26a1, ADH5, Aldh1a1, Aldh1a2, Klk6, GATA-2, and Cited2 with T7 polymerase. Slides were incubated for 7 min at room temperature in proteinase K (20 μg/ml) in a buffer containing 50 mM Tris and 5 mM EDTA (pH 8.0). Slides were then acetylated with acetic anhydride, dehydrated, and exposed to either denatured antisense or sense probes in hybridization buffer [50% formamide, 10% dextran sulfate, 5x Denhardt’s solution, 300 mM NaCl, 5 mM EDTA (pH 8), 20 mM Tris (pH 8.0), and 0.05 mg/ml yeast tRNA]. Hybridization was performed at 55 C overnight in a humidity chamber containing 5x standard saline citrate and 50% formamide. Hybridized slides were exposed to 20 μg/ml ribonuclease A for 30 min at 37 C. Slides were washed in 50% formamide, 2x standard saline citrate, and 100 mM 2-mercaptoethanol, followed by 2x standard saline citrate at 55 C for 30 min, dehydrated in a graded series of ethanol in 0.3 M ammonium acetate, and exposed to Biomax MR film overnight (Eastman Kodak Co., Rochester, NY). The following morning, slides were dipped in autoradiography emulsion (Amersham Biosciences, Piscataway, NJ) and placed at 4 C in a light-proof box for several days. After development, slides were counterstained with hematoxylin.
Results
Global gene expression profiles of P4-treated uterine samples in PRKO and wild-type mice
P4- and PR-regulated genes in the mouse uterus were identified by comparing the mRNA expression pattern of uteri from ovariectomized wild-type and PRKO mice given an acute or chronic regimen of P4 as previously described. The goal of acute P4 treatment was to identify rapid responsive genes by P4 and PR (direct target genes); the goal of the chronic hormone regimen was to identify secondary, potentially indirect target genes of P4-PR regulation. Of approximately 12,000 genes represented on the Affymetrix murine genome U74Av2 oligonucleotide array (Affymetrix), 48–55% of the genes were recorded as present in the uterine samples using MAS 5.0 (Affymetrix). The data were analyzed as detailed in Materials and Methods. As generally adopted for oligonucleotide microarray profile analysis, we selected differentially expressed genes by a lower boundary of a 90% confidence interval of fold change greater than 1.2 and an absolute value of difference between group means greater than 80. This was applied to select increased and decreased genes.
The above-described experimental design allowed comparison not only of the effect of acute vs. chronic treatment of P4 in these mice, but also of the effects of vehicle and hormone on both wild-type and PRKO mice. Using this experimental design, the identification of differentially expressed genes can be derived from five physiologically relevant comparisons. The five comparisons are shown graphically in Fig. 1. Comparison 1 consisted of analyzing differential gene expression of wild-type mice treated with vehicle vs. wild-type mice treated with P4. This treatment identified P4-regulated genes. Comparison 2 analyzed wild-type and PRKO mice treated with P4. This comparison identified genes regulated by PR. Comparison 3 analyzed PRKO mice treated with vehicle compared with wild-type mice treated with P4. This comparison was expected to identify genes regulated by both P4 and PR. Comparison 4 identified differentially expressed genes between wild-type and PRKO mice treated with vehicle. This comparison identified genes whose expression was due to the developmental consequences of ablation of the PR. Finally, comparison 5 consisted of PRKO mice treated with either vehicle or P4. This comparison was expected to show P4-regulated genes independent of PR.
FIG. 1. The five physiologically relevant comparisons used to identify significantly P4- and/or PR-regulated uterine genes: comparison 1, vehicle (veh)-treated wild-type (WT) vs. P4-treated wild-type mice; comparison 2, P4-treated PRKO vs. P4-treated wild-type mice; comparison 3, vehicle-treated PRKO vs. P4-treated wild-type mice; comparison 4, vehicle-treated wild-type vs. vehicle-treated PRKO mice; and comparison 5, vehicle-treated PRKO vs. P4-treated PRKO mice. Differentially expressed genes were selected using two-sample comparison according to the lower boundary of a 90% confidence interval of fold change greater than 1.2 and an absolute value of difference between group means greater than 80.
The summary of the number of differentially expressed genes after acute treatment of P4 for the five comparisons is shown in Table 1. Overall, at the 4 h point, the number of increased genes was greater than that of decreased genes. Comparison 1, which identified the effects of P4 on the mouse uterus, showed 90 genes that were differentially expressed. Of the 90 genes, 67 were more highly expressed in the P4-treated, wild-type uteri, and 23 genes were decreased more than 1.2-fold. When examining the effect of PR on the ability of the mouse uteri to respond to an acute P4 treatment, comparison 2, 183 genes were differentially expressed between wild-type and PRKO in P4-treated uteri samples at 4 h. Of the 183 genes, 120 were increased, and 63 genes were decreased. A total of 64 genes were identified by both comparisons; vehicle-treated vs. P4-treated wild-type, and P4-treated PRKO vs. wild-type. Comparison 3, which identifies both P4- and PR-regulated genes, showed the greatest number of differentially expressed genes (n = 235), with 139 genes increased and 96 genes decreased. Figure 2A diagrammatically shows the overlap in the analysis of P4-responsive (comparison 1), PR-responsive (comparison 2), and both P4- and PR-responsive (comparison 3) genes. The high degrees of overlap among these three groups, which are based on different animal groups, demonstrate the robustness of the experimental design. Comparison 4 shows that 45 genes were identified as differentially expressed when comparing the nonhormone-stimulated, ovariectomized, wild-type to the PRKO uterus. This difference could be accounted for by the impact of the lack of P4-PR stimulation throughout development. Finally, comparison 5, which shows the effect of P4 on the PRKO mouse uterus, identified five genes that were differentially expressed. Based on permutation testing, we estimated the false discovery rate to be 14% or less for the first three comparisons, whereas the rate was 70% or higher for comparisons 4 and 5, suggesting that most differentially expressed genes in these two analyses are due to chance. A complete list of the increased and decreased genes that are identified as significant for comparisons 1–3 and comparisons 4 and 5 with acute P4 treatment are presented in supplemental Tables 1 and 2, respectively (published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org).
TABLE 1. Number of significantly increased and decreased uterine genes at 4 h
FIG. 2. Venn diagrams demonstrating the relationship between genes modulated in the wild-type (WT) and PRKO uterus in response to acute (A) or chronic (B) treatment with P4. Red circles indicate genes selected by comparison 1 [vehicle-treated (veh) wild-type vs. P4-treated wild-type]; green, by comparison 2 (P4-treated PRKO vs. P4-treated wild-type); and blue, by comparison 3 (vehicle-treated PRKO vs. P4-treated wild-type). The numbers within the intersections of the circles indicate the common genes by two (or three) comparisons. The number at the bottom right corner of the panel indicates the total number of genes analyzed. C, Clustering analysis of P4- and PR-regulated genes in the murine uterus. One hundred and ninety-seven P4- and PR-regulated genes were clustered in two dimensions according to their gene expression pattern by hierarchical-tree algorithm (dChip version 1.3). The color code for the signal strength in the classification scheme is shown in the box at the bottom, in which induced genes are indicated by red, and repressed genes are indicated by green.
The results of the chronic hormone stimulation of mouse uteri with P4 are summarized in Table 2. The goal of this analysis was to identify late responsive genes by this axis. In the comparison with the acute stimulation of this axis, the majority of genes differentially expressed were decreased. As with the analysis of the acute affects of P4 treatment, this analysis identified genes that were differentially impacted by P4 treatment (comparison 1), PR ablation (comparison 2), and the combination of P4 treatment and PR ablation (comparison 3). As with the acute P4 treatment ablation, significantly more genes were impacted by ablation of the PR than by administration of P4. Figure 2B shows the analysis of the overlap in genes impacted by P4 administration or PR ablation. Again, a significant number of genes were identified as common by all three comparisons. As with the analysis of the acute injection protocol, analysis of PRKO and wild-type mice treated with vehicle demonstrated that similar numbers of genes were shown to be significantly affected. That is, 45 genes were significantly different with both acute and chronic injections of vehicle, respectively, with 15 genes identified at both time points. This comparison showed that the injection protocol did not impact the expression of uterine genes. In the final comparison, in which PRKO mice treated chronically with oil or P4 were analyzed, 44 genes were significantly changed by P4 in the PRKO uteri. This is more than the six genes identified by the acutely stimulated uteri and may reflect either pharmacological stimulation of other steroid hormone receptors, such as the glucocorticoid receptor, or activation of genes through a PR-independent pathway. A complete list of genes that showed significant change for comparisons 1–3 and comparisons 4 and 5 with chronic P4 treatment are presented in supplemental Tables 2 and 3, respectively (published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org).
TABLE 2. Number of significantly increased and decreased uterine genes at 40 h
Grouping of P4- and PR-regulated genes into early, intermediate, and late responsive genes
The overall goal of this investigation was to identify genes that are activated by the P4-PR axis. Therefore, the rest of this manuscript will focus on genes that were identified as in common among comparisons 1, 2, and 3 (Fig. 1). The list of the common genes for acute and chronic P4 treatment is listed in Tables 3 and 4, respectively. This approach identified genes that were significantly impacted by both P4 administration and PR ablation. Combining the gene lists obtained by acute and chronic P4 treatments yielded a total of 197 differentially regulated genes, of which 10 genes were differentially expressed with both acute and chronic P4 treatment, 54 genes were differentially expressed with acute P4 treatment (Fig. 2A), and 133 genes were impacted by chronic P4 treatment (Fig. 2B). All subsequent data analyses and experiments were based on these 197 genes.
TABLE 3. The list of uterine genes significantly increased and decreased by acute P4 treatment
TABLE 3A. Continued
TABLE 4. The list of genes significantly increased and decreased uterine genes by chronic P4 treatment
TABLE 4A. Continued
To identify trends in the changes in gene expression of P4- and PR-stimulated genes, stringent data filtering and clustering also allowed grouping of the P4- and PR-regulated genes into sets of biological relevance. The 197 regulated genes were analyzed by mathematical clustering using a hierarchical tree (Fig. 2C). Gene expression changes were observed after acute P4 treatment. Early responsive genes are presented first, followed by intermediate responsive genes, differentially expressed genes that changed at both time points, then by late responsive genes, whose expression changed only at the late time point. Tables 3 and 4 list the genes, the GenBank accession numbers, and the fold changes as grouped by gene ontology.
This analysis shows that 54 genes were identified as early responsive genes, whereas 130 genes were identified as late responsive genes. Thirteen genes were identified as intermediate responsive genes. The validity of this observation was confirmed by checking already known P4-regulated genes. Ihh (18) and HIF1A (19) belong to the early responsive genes. Amphiregulin (12) and histidine decarboxylase (13), both of which were up-regulated at both time points, were grouped with intermediate responsive genes. Calbindin-D9k (27), calcitonin (15, 16), and immune-responsive gene 1 (20, 28) genes were identified as late responsive genes. The expression profiles of these genes correlated to previously published results.
Functional grouping of P4- and PR-regulated genes
To categorize the P4- and PR-regulated genes into specific functional groups, the significantly increased and decreased genes were functionally categorized by gene ontology terms. The results of sorting the genes that responded to acute and chronic P4 treatments are shown in Fig. 3, A and B, respectively. Transcription factors, transport proteins, signal transduction, cell growth, and enzyme genes were increased by acute treatment with P4 in contrast with the effect of chronic P4 treatment (Fig. 3A). Likewise, more of these regulatory genes were decreased after chronic treatment with P4 (Fig. 3B).
FIG. 3. The functional categorization of increased and decreased genes by P4 and PR in the murine uterus with acute (A) or chronic (B) P4 treatment; genes were annotated and assigned to various functional categories using the Gene Ontology and Affymetrix annotation. , Increased genes; , decreased genes.
P4 regulates spatially and temporally RA metabolism-related genes
RA plays important roles in the maintenance of pregnancy and embryo development (29, 30, 31, 32). Analysis of the genes regulated by the P4-PR axis showed that genes involved in the production, degradation, and transport of RA were regulated by this axis. Enzymes involved in the production of RA, ADH5 and Aldh1a1, were also identified as early responsive genes. Interestingly, although Aldh1a2 was identified as a late responsive gene, i.e. a gene whose expression was increased after chronic P4 treatment, Cyp26a1, an enzyme involved in RA metabolism was identified as an early responsive gene and, in fact, represented the largest fold change in this microarray analysis. Finally, two Rbps were identified as late responsive genes. Rbp1 was identified as a late responsive increased gene, and Rbp4 was identified as a late responsive decreased gene. Although the P4-PR axis altered the expression of genes involved in the production and transport of RA, this axis did not affect the genes involved in retinoic signal transduction. The expression profiles of retinoid X receptor (RXR) and RA receptor (RAR) genes were not changed by P4 treatment or at different time points (data not shown). Real-time PCR was used to validate the microarray results for Cyp26a1, ADH5, and Aldh1a1 genes using three additional sets of RNA. The expressions of Cyp26a1, ADH5, and Aldh1a1 were significantly increased by P4 in wild-type mice (104.09-, 2.13-, and 1.54-fold, respectively), and these inductions were decreased in PRKO mice (2.0%, 46.1%, and 71.1%, respectively). The results were consistent with the microarray data identification of these genes as early responsive genes. Cyp26a1, ADH5, and Aldh1a1 exhibited high expression only in the acute P4-treated, wild-type uterus (Fig. 4A), whereas differential expression of those genes was not observed with chronic P4 treatment (data not shown). Furthermore, confirmation of the increase in Rbp1 (1.92-fold) and the decrease in Rbp4 (41%) after chronic stimulation of the P4-PR axis was demonstrated by real-time PCR (Fig. 4B). These P4 effects of Rbp1 and Rbp4 were significantly decreased in the PRKO.
FIG. 4. A, Real-time RT-PCR analysis of Cyp26a1, ADH5, and Aldh1a1 in vehicle-treated (Veh; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. B, Real-time PCR of Rbp1 and Rbp4 in chronic P4-treated uteri. The results represent the mean ± SE of three independent RNA sets. *, P < 0.05; **, P < 0.01.
Next, the temporal and spatial expressions of these genes were examined by in situ hybridization. Cyp26a1 and ADH5 transcripts were undetectable, and Aldh1a1 was expressed at luminal epithelium and glandular epithelium after vehicle treatment. As shown in Fig. 5, Cyp26a1, ADH5, and Aldh1a1 transcripts exhibited an increase only with acute P4 treatment (Fig. 5A) and not with chronic P4 treatment (Fig. 5B). The expression of Cyp26a1 by P4 was limited to the luminal epithelium. ADH5 and Aldh1a1 transcripts were shown in the luminal epithelium and glandular epithelium at 4 h. Cyp26a1 and ADH5 transcripts were undetectable, and Aldh1a1 transcripts were decreased with chronic P4 treatment (Fig. 5B). In contrast, Aldh1a2 expression was not detected at an early time point, but appeared with chronic P4 treatment in the endometrial stroma.
FIG. 5. Localization pattern of RA metabolism genes by in situ hybridization in the acute (A) and chronic (B) P4-treated wild-type mouse uterus. C, Localization pattern of ADH5 and Cyp26a1 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
The spatial and temporal expressions of Aldh1a1, Aldh1a2, and Cyp26a1 in the mouse uterus during pregnancy have been reported (29). Aldh1a1 glandular expression has been reported to be sharply induced on d 2.5, whereas Aldh1a2 stromal expression increased more steadily until the implantation phase (29). Cyp26a1 epithelial expression is strongly induced between d 3.5 and 4.5. In this report, we performed in situ hybridization on pseudopregnant uterine samples to determine the expression patterns of ADH5 and Cyp26a1. The results of this analysis are shown in Fig. 6. ADH5 transcripts were almost undetectable on d 0.5 of pseudopregnancy. However, they were induced by d 2.5 throughout the endometrial stroma. Expression levels increased on d 4.5 and decreased on d 6.5. Cyp26a1 transcripts were not detected on d 0.5 or 2.5. Expression appeared in the uterine luminal epithelium on d 4.5, and strong expression persisted throughout the uterine epithelium until d 6.5 (Fig. 5C). ADH5, Aldh1a, and Aldh1a2 are RA synthesis enzymes, and Cyp26a1 is a RA-catabolizing enzyme. Interestingly, these RA-metabolizing enzymes exhibited distinct spatial and temporal profiles of expression in pseudopregnant uteri. These in situ hybridization results demonstrate that the RA synthesis and degradation enzymes are always spatially and temporally expressed in a discrete nonoverlapping manner in the pseudopregnant uterus.
FIG. 6. The expression pattern of Klk genes in the murine wild-type (WT) and PRKO mouse uterus. A, Real-time PCR of Klk genes in vehicle-treated (Veh.; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. The results represent the mean ± SE of three independent RNA sets. *, P < 0.05. B, Localization pattern of Klk6 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
Kallikrein genes were increased early responsive genes by P4
In the functional analysis of P4- and PR-regulated genes, the enzyme group of regulated genes showed that a large number of kallikreins genes were shown to be regulated by this axis. Klk genes are a highly conserved gene family of serine proteases (33, 34, 35). Microarray analysis showed that Klk5 and Klk6 were identified as early responsive genes, i.e. significantly increased by acute P4 treatment, but no difference was detected with chronic P4 treatment. Although, Klk13, Klk16, and Klk26 were shown to be early responsive genes, the expression of these genes was not significant, as judged by analysis of the microarray data. The microarray analysis also indicated that the expression level of Klk genes was very low in the uterus. The microarray data for the Klk genes were confirmed by real-time PCR. Real-time PCR could only validate the differential expression of Klk5 and Klk6 by the P4-PR axis, as shown in Fig. 6A. In situ hybridization was conducted to determine the regulation and location of the Klk genes. Only Klk6 could be detected by in situ hybridization in pseudopregnant uteri (Fig. 6B). Attempts to detect the expression of the other Klk genes in the uterus were unsuccessful due to the low levels of the mRNAs for these genes. Klk6 transcripts were not detected by d 2.5. Expression appeared in the uterine endometrial stroma by d 4.5. Expression in luminal epithelium and glandular epithelium was seen on d 6.5.
Transcriptional regulators, GATA2 and Cited2, were rapidly induced by P4
Several transcription factors were identified as being impacted by the P4-PR axis using microarray analysis. HIF1 (19) and Hoxa-10 (14) genes are known to be increased by P4. These transcription factor genes were highly expressed by P4 in our microarray data. This microarray analysis also identified the developmentally important transcription factors, GATA-2 (GATA-binding protein 2) and Cited2, as being regulated by the P4-PR axis. GATA-2 is a transcription factor and plays a critical role in hemopoiesis (36, 37). Regulation of these genes by the P4-PR axis was confirmed by real-time RT-PCR and in situ hybridization assays. Real-time PCR validated that GATA-2 expression was elevated in P4-treated wild-type uteri compared with P4-treated PRKO (1.54-fold) and vehicle-treated wild-type (1.78-fold) and PRKO (1.84-fold) mouse uteri, as shown in Fig. 7A. In situ hybridization analysis of mouse uteri showed that GATA-2 transcripts were rapidly induced in the luminal epithelium and glandular epithelium by P4 (Fig. 7B). Having determined the spatial expression of GATA-2 in the uterus after pharmacological administration of P4, the spatial and temporal expressions of GATA-2 were determined during pseudopregnancy. GATA-2 transcripts were first detected in the luminal epithelium and glandular epithelium of pseudopregnant uterine samples on d 2.5. The expression in luminal epithelium was decreased, and the signal in glandular epithelium disappeared on d 4.5. By d 6.5, the transcripts were undetectable (Fig. 7C). Thus, as observed with Ihh (18), GATA-2, a P4-regulated gene, was physiologically expressed before the window of receptivity in the mouse uterus.
FIG. 7. The expression pattern of GATA-2 in the murine uterus. A, Real-time PCR of GATA-2 in vehicle-treated (Veh.; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. The results represent the mean ± SE of three independent RNA sets. **, P < 0.01. B, Localization pattern of GATA-2 transcripts by in situ hybridization in vehicle- and acute P4-treated wild-type mouse uterus. C, Localization pattern of GATA-2 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
The expression level of Cited2 also was confirmed by real-time PCR and in situ hybridization assays (Fig. 8, A and B). Cited2 transcripts were highly expressed in the luminal epithelium, glandular epithelium, and myometrium by P4 at 4 h. In the analysis of the spatial and temporal patterns of expression of Cited2 transcripts during pseudopregnancy, Cited2 transcripts were seen at low levels in the endometrial stroma on d 0.5–4.5 and increased in the luminal epithelium, glandular epithelium, and myometrium on d 6.5 (Fig. 8C). Therefore, pharmacological doses of P4 induced Cited2 expression in an appropriate compartment of the uterus.
FIG. 8. The expression pattern of Cited2 in the murine uterus. A, Real-time PCR of Cited2 in vehicle-treated (Veh.; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. The results represent the mean ± SE of three independent RNA sets. **, P < 0.01. B, Localization pattern of Cited2 transcripts by in situ hybridization in vehicle- and acute P4-treated uterus at 4 h. C, Localization pattern of Cited2 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
Discussion
In this study, we have identified P4- and PR-regulated uterine genes using the PRKO mouse in combination with high-density DNA microarray analysis. Acute and chronic administrations of P4 identified early responsive and late responsive genes. Although this microarray analysis identified genes regulated by P4, PR, or both components of the P4-PR axis, the majority of these genes were identified as regulated by both P4 and PR. The genes that were separately affected by either P4 treatment or PR ablation may be the result of several phenomena: 1) developmental differences between the PRKO and the wild-type uterus, 2) ligand-independent activation of the PR (38), 3) P4 having effects via a membrane receptor (39), or 4) pharmacological activation of other steroid/nuclear receptors, such as the glucocorticoid receptor. Hereafter, this report focuses on the identification of P4- and PR-regulated target genes. These target genes were identified by selecting only genes whose expression was significantly altered in comparisons 1, 2, and 3, as described in Fig. 1.
In the analysis of the impact of either acute or chronic stimulation of the P4-PR axis on the regulation of global uterine gene expression, it is clear that acute stimulation of this axis results in the overall activation of uterine gene expression. The majority of genes activated by acute P4 treatment are regulatory genes. This observation agrees with the premise that PR is an activator of gene transcription (40). The other observation that can be made for the global changes in gene expression is that with chronic treatment of P4, the major result was repression of gene expression. This could be partially explained by the fact that chronic administration of P4 will repress the expression of PR and estrogen receptor (41). This would explain why some of the early responsive genes lost activation with chronic P4 treatment. However, it does not fully explain why the majority of genes were repressed at this time point compared with the ovariectomized or PRKO uteri treated with vehicle. The down-regulation of these genes is most likely due to downstream effects initiated by activation of the P4-PR axis. Chronic treatment of the uterus may result in differentiation of uterine tissue to a tissue in which these regulatory genes are expressed at lower levels than in the unstimulated uterus. The target genes identified in this report may serve as a resource for a bioinformatics analysis in which the promoter regions of these genes could reveal the common pathways that result in repression of these genes by chronic P4 treatment.
The choice of conducting the microarray analysis on uteri isolated 4 h after a single dose of P4 was for the purpose of identifying direct targets of P4 activity. Previous attempts at identifying the impact of P4 on uterine gene expression consisted of analyzing the mice 6 h after hormone administration (12). However, in analyzing the temporal response of the uterus to this regimen of P4, it was demonstrated that changes in gene expression can be observed at 4 h or earlier (18). Therefore, the time point of 4 h was chosen to identify genes that are potentially direct targets of P4.
We have validated the microarray results for nine individual genes by real-time PCR using three additional sets of RNA. As a whole, the assessments of fold change by array analysis are markedly less that those determined by real-time PCR. The underestimation of the fold change by microarray analysis compared with real-time PCR analysis highlights the limitations of using microarray analysis as a means of quantifying changes in gene expression. Also, because the uterus consists of several different compartments, in situ hybridization analysis was conducted to localize the uterine compartment in which the change in gene expression was observed. Cyp26a1, Aldh1a1, and GATA-2 were only expressed at the luminal epithelium. ADH5 and Aldh1a2 were expressed at the stroma. These P4- and PR-regulated genes were regulated only at a specific compartment that can reduce the fold change in the entire tissue analysis. Thus, the fold changes seen on the array analysis are probably an underestimation of the true differences in gene expression.
One of the major metabolic pathways identified by this array is that of genes involved in RA metabolism. RA, the natural metabolite of vitamin A, has been shown to play a key role in the regulation of reproduction. In the female reproductive tract, RA has been shown to regulate differentiation of the reproductive tract epithelium (42). RA has also been show to mimic the ability of P4 to regulate the expression of matrix metalloproteases in endometriotic tissue (43). These previous observations indicate that RA signaling is important for the function of the female reproductive tract and that RA signaling may in part mediate P4 signaling in the uterus. The results of this microarray analysis has demonstrated that both RA synthesis (ADH5 and Aldh1a1) and RA catabolic enzyme (Cyp26a1) were rapidly induced by the P4-PR axis, whereas levels of RAR and RXR were not affected. In addition, we showed that Rbp1 and Rbp4 were P4 regulated. These results are similar to observations in clinical samples, where the intracellular levels of RA and the expression of cytoplasmic Rbps fluctuate, and the nuclear RAR and RXR receptors remain at similar levels throughout the menstrual cycle (44, 45).
The regulation of enzymes involved in both synthesis and breakdown of RA by the P4-PR axis seems paradoxical. However, one must keep in mind the conditions under which the microarrays were performed. The microarrays were conducted by treating ovariectomized mice with pharmacological doses of P4. Therefore, these mice were given P4 at a time when the levels of PR were highest in the endometrial epithelium. Not surprisingly, the majority of genes detected by this array were found to be expressed in the uterine epithelium. However, when analyzing the expression patterns of these RA metabolic enzymes, it was shown that their expressions were compartmentally and temporarily regulated. During pregnancy, ADH5 transcripts were induced by d 2.5 throughout the endometrial stroma. Expression levels increased on d 4.5 and decreased on d 6.5. Aldh1a1 expression was sharply induced in the endometrial glandular compartment on gestational d 2.5, whereas Aldh1a2 stromal expression increased more steadily until the implantation phase (29). Cyp26a1 transcripts appeared in the uterine epithelium and glandular epithelium on d 4.5, and strong expression persisted throughout the uterine epithelium until d 6.5. The in situ hybridization results in the pseudopregnant uterus demonstrate that the RA synthesis and degradation enzymes are always spatially and temporally expressed in a discrete, nonoverlapping manner in early pregnancy. Moreover, the balance between synthesis and degradation of RA is very important in early pregnancy to allow RA signaling to act on the endometrium to prepare for implantation without having deleterious affects on differentiation of the embryo.
One group of genes that has been shown to be up-regulated by the P4-PR axis is the Klk genes. Klks are a gene family associated with regulation of the inflammatory process. These genes regulate a cascade of events resulting in vasodilatation, an increase in vascular permeability, and edema (46). Because implantation of the embryo in the mouse uterus requires vasodilatation and changes in vascular permeability, it is not surprising that the P4-PR axis is a primary regulator of uterine Klk gene expression. We have identified Klk5 and Klk6, which are regulated by the P4-PR axis. The expression levels of the Klk genes regulated in the uterus were low, and although validation could be accomplished by real-time PCR, only Klk6 transcripts could be detected by in situ hybridization in the mouse uterus. In the pseudopregnant uterus, Klk6 appeared in the uterine endometrial stroma by pregnancy d 4.5 and then in the luminal epithelium and glandular epithelium on d 6.5. This expression pattern of the Klks is similar to that in the rat uterus, where Klk expression is increased during the implantation period (47, 48).
The two transcription regulatory genes validated in this report were GATA-2 and Cited2. GATA-2 is a member of the GATA transcription factor family. Ablation of GATA-2 results in an embryonic lethal phenotype due to defects in hemopoiesis (36). Interestingly, GATA-2 activity in regulating hemopoiesis is linked to the action of RAR (49). This microarray analysis has shown that the P4-PR axis also regulates the expression of enzymes involved in the production of RA. Therefore, modulation of GATA-2 and regulation of levels of the ligand for RAR signaling may be a means by which the P4-PR axis regulates the activity of the RAR pathway in the uterus during pregnancy. Aside from interacting with RAR, GATA family members have been shown to interact with chromatin-remodeling proteins. In the rabbit, GATA-4 is regulated by P4 and interacts with RUSH (proteins with RING finger motifs that bind to the uteroglobin promoter), a member of the Smarca3 family of SWI/SNF-related chromatin-remodeling proteins, in the mediation of prolactin-augmented, P4-induced gene expression (50). Therefore, GATA-2 may play a pivotal role in regulating uterine function.
The other transcriptional regulator validated in this microarray analysis is Cited2. Cited2 is a member of the Cited transcription coactivator family that binds the p300/corticosterone-binding protein coactivator. Ablation of Cited2 results in cardiac, neural crest, and adrenal developmental defects (51). Cited2 is induced by various inflammatory stimuli and has been shown to regulate matrix metalloproteinase gene expression (52). Cited2 has been shown to be a coregulator of peroxisome proliferator-activated receptors and and has been shown to be a regulator of peroxisome proliferator-activated receptor induction of angiogenic factors in vitro (53). The induction of matrix metalloproteinases and angiogenic factors is critical in remodeling of the uterus for establishment of embryonic implantation. The induction of Cited2 by the P4-PR axis and the expression of Cited2 in the uterus during pregnancy indicate that Cited2 may play a role as a mediator of P4 action in remodeling of the uterus in preparation for pregnancy.
In summary, this study has demonstrated that acute stimulation of the P4-PR axis stimulates the expression of many regulatory genes, whereas chronic stimulation of this axis is inhibitory. Analysis of the genes regulated by acute stimulation of the P4-PR axis will identify genes and pathways that are directly regulated by P4 and may help determine what pathways are important for preparation of the uterus for the support and maintenance of embryonic implantation.
Acknowledgments
We thank Jinghua Li, Bryan Ngo, Jie Li, and Janet DeMayo, M.S., for technical assistance, and John Ellsworth for help with manuscript preparation.
References
Clarke CL, Sutherland RL 1990 Progestin regulation of cellular proliferation. Endocr Rev 11:266–301
Lydon JP, DeMayo FJ, Funk CR, Mani SK, Hughes AR, Montgomery Jr CA, Shyamala G, Conneely OM, O’Malley BW 1995 Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 9:2266–2278
Cheon YP, Li Q, Xu X, DeMayo FJ, Bagchi IC, Bagchi MK 2002 A genomic approach to identify novel progesterone receptor regulated pathways in the uterus during implantation. Mol Endocrinol 16:2853–2871
Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895
Tsai MJ, O’Malley BW 1994 Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu Rev Biochem 63:451–486
O’Malley BW, Conneely OM 1992 Orphan receptors: in search of a unifying hypothesis for activation. Mol Endocrinol 6:1359–1361
Conneely OM, Kettelberger DM, Tsai MJ, Schrader WT, O’Malley BW 1989 The chicken progesterone receptor A and B isoforms are products of an alternate translation initiation event. J Biol Chem 264:14062–14064
Stone DM, Hynes M, Armanini M, Swanson TA, Gu Q, Johnson RL, Scott MP, Pennica D, Goddard A, Phillips H, Noll M, Hooper JE, de Sauvage F, Rosenthal A 1996 The tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 384:129–134
Lydon JP, DeMayo FJ, Conneely OM, O’Malley BW 1996 Reproductive phenotpes of the progesterone receptor null mutant mouse. J Steroid Biochem Mol Biol 56:67–77
Mulac-Jericevic B, Lydon JP, DeMayo FJ, Conneely OM 2003 Defective mammary gland morphogenesis in mice lacking the progesterone receptor B isoform. Proc Natl Acad Sci USA 100:9744–9749
Mulac-Jericevic B, Mullinax RA, DeMayo FJ, Lydon JP, Conneely OM 2000 Subgroup of reproductive functions of progesterone mediated by progesterone receptor-B isoform. Science 289:1751–1754
Das SK, Chakraborty I, Paria BC, Wang XN, Plowman G, Dey SK 1995 Amphiregulin is an implantation-specific and progesterone-regulated gene in the mouse uterus. Mol Endocrinol 9:691–705
Paria BC, Das N, Das SK, Zhao X, Dileepan KN, Dey SK 1998 Histidine decarboxylase gene in the mouse uterus is regulated by progesterone and correlates with uterine differentiation for blastocyst implantation. Endocrinology 139:3958–3966
Lim H, Ma L, Ma WG, Maas RL, Dey SK 1999 Hoxa-10 regulates uterine stromal cell responsiveness to progesterone during implantation and decidualization in the mouse. Mol Endocrinol 13:1005–1017
Zhu LJ, Cullinan-Bove K, Polihronis M, Bagchi MK, Bagchi IC 1998 Calcitonin is a progesterone-regulated marker that forecasts the receptive state of endometrium during implantation. Endocrinology 139:3923–3934
Kumar S, Zhu LJ, Polihronis M, Cameron ST, Baird DT, Schatz F, Dua A, Ying YK, Bagchi MK, Bagchi IC 1998 Progesterone induces calcitonin gene expression in human endometrium within the putative window of implantation. J Clin Endocrinol Metab 83:4443–4450
Nie GY, Li Y, Wang J, Minoura H, Findlay JK, Salamonsen LA 2000 Complex regulation of calcium-binding protein D9k (calbindin-D(9k)) in the mouse uterus during early pregnancy and at the site of embryo implantation. Biol Reprod 62:27–36
Takamoto N, Zhao B, Tsai SY, DeMayo FJ 2002 Identification of Indian hedgehog as a progesterone-responsive gene in the murine uterus. Mol Endocrinol 16:2338–2348
Daikoku T, Matsumoto H, Gupta RA, Das SK, Gassmann M, DuBois RN, Dey SK 2003 Expression of hypoxia-inducible factors in the peri-implantation mouse uterus is regulated in a cell-specific and ovarian steroid hormone-dependent manner. Evidence for differential function of HIFs during early pregnancy. J Biol Chem 278:7683–7691
Cheon YP, Xu X, Bagchi MK, Bagchi IC 2003 Immune-responsive gene 1 (Irg1) is a novel target of progesterone receptor and plays a critical role during implantation in the mouse. Endocrinology 144:5623–5630
Schadt EE, Li C, Ellis B, Wong WH 2001 Feature extraction and normalization algorithms for high-density oligonucleotide gene expression array data. J Cell Biochem 37(Suppl):120–125
Li C, Wong WH 2001 Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci USA 98:31–36
Li C, Hung Wong W 2001 Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol 2:RESEARCH0032
Li C, Wong WH 2003 DNA-chip analyzer (dChip). In: Parmigiani G, Garrett ES, Irizarry R, Zeger SL, eds. The analysis of gene expression data: methods and software. New York: Springer; 120–141
Dahlquist KD, Salomonis N, Vranizan K, Lawlor SC, Conklin BR 2002 GenMAPP, a new tool for viewing and analyzing microarray data on biological pathways. Nat Genet 31:19–20
Simmons DM, Arriza JL, Swanson LW 1989 A complete protocol for in situ hybridization of messenger RNAs in brain and other tissues with radiolabeled single-stranded RNA probes. J Histotechnol 12:169–181
Tatsumi K, Higuchi T, Fujiwara H, Nakayama T, Itoh K, Mori T, Fujii S, Fujita J 1999 Expression of calcium binding protein D-9k messenger RNA in the mouse uterine endometrium during implantation. Mol Hum Reprod 5:153–161
Chen B, Zhang D, Pollard JW 2003 Progesterone regulation of the mammalian orthologue of methylcitrate dehydratase (immune response gene 1) in the uterine epithelium during implantation through the PKC pathway. Mol Endocrinol 17:2340–2354
Vermot J, Fraulob V, Dolle P, Niederreither K 2000 Expression of enzymes synthesizing (aldehyde dehydrogenase 1 and reinaldehyde dehydrogenase 2) and metabolizing (Cyp26) retinoic acid in the mouse female reproductive system. Endocrinology 141:3638–3645
van der Leede BM, van den Brink CE, Pijnappel WW, Sonneveld E, van der Saag PT, van der Burg B 1997 Autoinduction of retinoic acid metabolism to polar derivatives with decreased biological activity in retinoic acid-sensitive, but not in retinoic acid-resistant human breast cancer cells. J Biol Chem 272:17921–17928
Morriss-Kay GM, Ward SJ 1999 Retinoids and mammalian development. Int Rev Cytol 188:73–131
Sapin V, Ward SJ, Bronner S, Chambon P, Dolle P 1997 Differential expression of transcripts encoding retinoid binding proteins and retinoic acid receptors during placentation of the mouse. Dev Dyn 208:199–210
Yousef GM, Diamandis EP 2003 An overview of the kallikrein gene families in humans and other species: emerging candidate tumour markers small star, filled. Clin Biochem 36:443–452
Yousef GM, Diamandis EP 2002 Expanded human tissue kallikrein family–a novel panel of cancer biomarkers. Tumour Biol 23:185–192
Yousef GM, Diamandis EP 2002 Human tissue kallikreins: a new enzymatic cascade pathway? Biol Chem 383:1045–1057
Li Y, Wang M 2000 The role of transcription factor GATA-2 in early hematopoiesis. Zhongguo Shi Yan Xue Ye Xue Za Zhi 8:66–70
Tsai FY, Keller G, Kuo FC, Weiss M, Chen J, Rosenblatt M, Alt FW, Orkin SH 1994 An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371:221–226
Power RF, Mani SK, Codina J, Conneely OM, O’Malley BW 1991 Dopaminergic and ligand-independent activation of steroid hormone receptors. Science 254:1636–1639
Zhu Y, Rice CD, Pang Y, Pace M, Thomas P 2003 Cloning, expression, and characterization of a membrane progestin receptor and evidence it is an intermediary in meiotic maturation of fish oocytes. Proc Natl Acad Sci USA 100:2231–2236
Li X, Lonard DM, O’Malley B W 2004 A contemporary understanding of progesterone receptor function. Mech Ageing Dev 125:669–678
Tibbetts TA, Mendoza-Meneses M, O’Malley BW, Conneely OM 1998 Mutual and intercompartmental regulation of estrogen receptor and progesterone receptor expression in the mouse uterus. Biol Reprod 59:1143–1152
Ponnamperuma RM, Kirchhof SM, Trifiletti L, De Luca LM 1999 Ovariectomy increases squamous metaplasia of the uterine horns and survival of SENCAR mice fed a vitamin A-deficient diet. Am J Clin Nutr 70:502–508
Bruner-Tran KL, Eisenberg E, Yeaman GR, Anderson TA, McBean J, Osteen KG 2002 Steroid and cytokine regulation of matrix metalloproteinase expression in endometriosis and the establishment of experimental endometriosis in nude mice. J Clin Endocrinol Metab 87:4782–4791
Kumarendran MK, Loughney AD, Prentice A, Thomas EJ, Redfern CP 1996 Nuclear retinoid receptor expression in normal human endometrium throughout the menstrual cycle. Mol Hum Reprod 2:123–129
Loughney AD, Kumarendran MK, Thomas EJ, Redfern CP 1995 Variation in the expression of cellular retinoid binding proteins in human endometrium throughout the menstrual cycle. Hum Reprod 10:1297–1304
Bhoola KD, Figueroa CD, Worthy K 1992 Bioregulation of kinins: kallikreins, kininogens, and kininases. Pharmacol Rev 44:1–80[Free Full Text]
Corthorn J, Valdes G 1994 Variations in uterine kallikrein during cycle and early pregnancy in the rat. Biol Reprod 50:1261–1264
Valdes G, Corthorn J, Scicli AG, Gaete V, Soto J, Ortiz ME, Foradori A, Saed GM 1993 Uterine kallikrein in the early pregnant rat. Biol Reprod 49:802–808 !-- null -->
Tsuzuki S, Kitajima K, Nakano T, Glasow A, Zelent A, Enver T 2004 Cross talk between retinoic acid signaling and transcription factor GATA-2. Mol Cell Biol 24:6824–6836
Hewetson A, Hendrix EC, Mansharamani M, Lee VH, Chilton BS 2002 Identification of the RUSH consensus-binding site by cyclic amplification and selection of targets: demonstration that RUSH mediates the ability of prolactin to augment progesterone-dependent gene expression. Mol Endocrinol 16:2101–2112
Bamforth SD, Braganca J, Eloranta JJ, Murdock JN, Marques FI, Kranc KR, Farza H, Henderson DJ, Hurst HC, Bhattacharya A 2001 Cardiac malformations, adrenal agenesis, neural crest defects and exencephaly in mice lacking Cited2, a new Tfap2 co-activator. Nat Genet 29:469–474
Yokota H, Goldring MB, Sun HB 2003 CITED2-mediated regulation of MMP-1 and MMP-13 in human chondrocytes under flow shear. J Biol Chem 278:47275–47280
Tien ES, Davis JW, Vanden Heuvel JP 2004 Identification of the CREB-binding protein/p300-interacting protein CITED2 as a peroxisome proliferator-activated receptor coregulator. J Biol Chem 279:24053–24063(Jae-Wook Jeong, Kevin Y. )
Address all correspondence and requests for reprints to: Dr. Francesco J. DeMayo, Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, Texas 77030. E-mail: fdemayo@bcm.tmc.edu.
Abstract
Progesterone (P4) acting through its cognate receptor, the progesterone receptor (PR), plays an important role in uterine physiology. The PR knockout (PRKO) mouse has demonstrated the importance of the P4-PR axis in the regulation of uterine function. To define the molecular pathways regulated by P4-PR in the mouse uterus, Affymetrix MG U74Av2 oligonucleotide arrays were used to identify alterations in gene expression after acute and chronic P4 treatments. PRKO and wild-type mice were ovariectomized and then treated with vehicle or 1 mg P4 every 12 h. Mice were killed either 4 h after the first injection (acute P4 treatment) or after the fourth injection of P4 (chronic P4 treatment). At the genomic level, the major change in gene expression after acute P4 treatment was an increase in the expression of 55 genes. Conversely, the major change in gene expression after chronic P4 treatment was an overall reduction in the expression of 102 genes. In the analysis, retinoic acid metabolic genes, cytochrome P 450 26a1 (Cyp26a1), alcohol dehydrogenase 5, and aldehyde dehydrogenase 1a1 (Aldh1a1); kallikrein genes, Klk5 and Klk6; and specific transcription factors, GATA-2 and Cited2 [cAMP-corticosterone-binding protein/p300-interacting transactivator with glutamic acid (E) and aspartic acid (D)-rich tail], were validated as regulated by the P4-PR axis. Identification and analysis of these responsive genes will help define the role of PR in regulating uterine biology.
Introduction
THE OVARIAN STEROID hormone progesterone (P4) is an essential regulator of reproductive events associated with all aspects of the establishment and maintenance of pregnancy (1, 2). Most of the physiological affects of P4 are mediated through its receptor, the progesterone receptor (PR). This signaling axis, the P4-PR axis, has been investigated by dissecting the role of PR (3). PR is a transcription factor that belongs to the nuclear receptor superfamily (4, 5, 6). PR is encoded in one gene and exists as one of two isoforms, PR-A and PR-B. These isoforms arise from the alternate translation start sites in the PR gene (7). Genetic ablation of the PR gene in mice (PRKO) leads to pleiotropic reproductive abnormalities, including defects in female reproductive behavior (8), failure to ovulate, failure of the uterus to support embryo implantation, and defects in branching and glandular development in the mammary glands (2, 9). Site-directed mutagenesis of the PR gene in vivo has demonstrated that the PR-A isoform is the major mediator of P4 signaling in the mouse uterus regulating uterine function, whereas the PR-B isoform regulates uterine epithelial cell proliferation (10, 11). Although the physiological processes governed by the P4-PR signaling axis in the uterus have been identified, the molecular pathways governed by PR are not fully understood. Therefore, elucidation of the molecular pathways regulated by PR in the uterus by identification of the target genes whose transcription is regulated by PR is of great importance.
To date, only a few P4-PR-regulated genes have been identified. These include the genes encoding amphiregulin (Areg) (12), histidine decarboxylase (Hdc) (13), Hoxa-10 and -11 (14), calcitonin (15, 16), calbindin-D9K (17), Indian hedgehog (Ihh) (18), hypoxia-inducible factor 1 (HIF1A) (19), and immune-responsive gene 1 (20). These target genes have been identified by testing candidate genes (12), differential library screening (15), and DNA microarray approaches (18, 20). High-density DNA microarray technology has immensely improved the ability to identify PR-regulated genes in the uterus.
Cheon and co-workers (3) used high-density DNA microarray technology to identify PR-responsive genes in the mouse uterus by treating female mice on d 3 of pregnancy with the antiprogestin RU486 and assaying the impact on uterine target genes 24 h later. This approach identified PR-regulated genes by inhibiting PR action at a time when PR levels were elevated in all compartments. In a sense, this approach identified the impact of withdrawal of P4-PR signaling on uterine gene expression and successfully identified 148 genes that were regulated by this axis (3). In this report, high-density DNA microarray technology in combination with the PRKO mouse were used to identify genes affected by acute and chronic stimulation of the P4-PR axis.
In this study, we have taken a pharmacological approach to identify the impact of activation of the P4-PR axis on the mouse uterus. This analysis shows the impact of acute and chronic pharmacological stimulation of the P4-PR axis, allowing additional identification of how these target genes function to regulate uterine biology. The groups of genes that were validated were those involved in retinoic acid (RA) metabolism, the kallikrein (Klk) family of genes, and specific developmentally important transcription factors. The identification of these factors as targets of P4-PR regulation will shed light on the role of P4 in uterine biology.
Materials and Methods
Animals and tissue collection
Mice were maintained at the designated animal care facility at Baylor College of Medicine according to the institutional guidelines for the care and use of laboratory animals. Thirty PRKO and 30 wild-type mice were ovariectomized at 6 wk of age. Two weeks later, ovariectomized mice were injected with either vehicle (sesame oil) or P4 (in sesame oil; Sigma-Aldrich Corp., St. Louis, MO; 1 mg/mouse in 100 μl). The injections were repeated every 12 h, and groups of mice were killed 4 h after the first injection (acute P4 treatment) or 4 h after the fourth injection (chronic P4 treatment). Each treatment consisted of 15 PRKO and 15 wild-type mice. The mice were killed by cervical dislocation after placing them under anesthesia (Avertin, 2,2-tribromoethyl alcohol, Sigma-Aldrich Corp.). Uterine tissues were flash-frozen at the time of dissection and stored at –80 C.
To collect uteri from pseudopregnant mice, wild-type females were mated with vasectomized males after superovulation [5 IU pregnant mare’s serum gonadotropin (VWR Scientific Products, West Chester, PA), followed 48 h later with 5 IU human chorionic gonadotropin (Pregnyl, Organon International, Roseland, NJ)]. The morning of the vaginal plug was designated d 0.5.
RNA isolation and microarray hybridization
Total RNA was extracted from uterine tissues using the RNeasy total RNA isolation kit (Qiagen, Valencia, CA). RNA was pooled from the uteri of five mice per genotype and treatment. All RNA samples were analyzed with a Bioanalyser 2100 (Agilent Technologies, Palo Alto, CA) before microarray hybridization. RT by oligo(deoxythymidine), followed by in vitro transcription and biotin labeling of cRNA, were performed (Enzo Biochem, Farmingdale, NY). The fragmented, labeled cRNA (15 μg) was hybridized to murine genome U74Av2 mouse oligonucleotide arrays (Affymetrix, Santa Clara, CA), which have approximately 12,000 genes represented, of which 6,000 are known and 6,000 are established sequence tags. Each transcript is represented on the chip as a set of 16–20 probe pairs, with each pair containing a perfect match and a mismatch. The microarrays were stained in an Affymetrix Fluidics station with streptavidin-phycoerythrin conjugates, followed by staining with an antistreptavidin antibody and streptavidin-phycoerythrin. The arrays were scanned at 3 μm with a GeneArray scanner (Affymetrix). All experiments were performed in triplicate with independent pools of RNA using 24 Affymetrix MG 74Av2 arrays.
Data analysis
After scanning and low-level quantification using Microarray Suite (Affymetrix), we used DNA-Chip Analyzer dChip (version 1.3) to adjust arrays to a common baseline using invariant set normalization (21) and to estimate expression using the PM-only model described by Li and colleagues (22, 23). All 24 Genechips were normalized to the same baseline (the Genechip with the median average intensity happened to be in the 48-h wild-type P4 treatment group), and all were modeled together. Data quality was reviewed using present call rates from MAS5 (average, 54%; range, 50–58%), ratios of 3' to 5' glyceraldehyde 3-phosphate dehydrogenase probe sets from MAS5 (average, 0.94; range, 0.8 to 1.5), and array outlier rates (average, 0.43%; range, 0.024–2.7%; only two of 24 chips >1%) from dChip. Based on the above parameters, all chips were considered of good quality and were included in subsequent analyses. We selected differentially expressed genes within each time exposure (acute and chronic) using a two-sample comparison according to the following criteria: lower boundary of 90% confidence interval of fold change greater than 1.2 and an absolute value of difference between group means greater than 80. The software uses resampling and SEs of model parameters to partially account for measurement error and unstable estimates of variability (24). Finally, we used the median number of detected genes in 50 permuted samples as an overall estimate of the false discovery rate. After excluding expressed sequence tags with no functional annotation, differentially expressed genes were classified according to Gene Ontology function using the Affymetrix annotation, a literature search in PubMed, and GenMAPP (25). Using these methods, functional categories were assigned for 197 differentially expressed genes.
Quantitative real-time PCR
Expression levels of selected genes found to be regulated by microarray analysis were validated by real-time RT-PCR TaqMan analysis using the ABI PRISM 7700 Sequence Detector System according to the manufacturer’s instructions (Applied Biosystems, Foster City, CA). For cytochrome P 450 26a1 (Cyp26a1); alcohol dehydrogenase 5 (ADH5); aldehyde dehydrogenase 1a1 (Aldh1a1); retinol-binding protein 1 genes Rbp1 and Rbp4; Klk genes Klk5 and Klk6; GATA-2; Cited2 [cAMP-corticosterone-binding protein/p300-interacting transactivator with glutamic acid (E) and aspartic acid (D)-rich tail]; and 18S RNA, prevalidated probes and primers were purchased from Applied Biosystems. RT-PCRs were performed using One-Step RT-PCR Universal Master Mix reagent and TaqMan Gene Expression Assays (Applied Biosystems) according to the manufacturer’s instructions. Standard curves were generated by serial dilution of a preparation of total RNA isolated from whole mouse uterus. All real-time PCR was performed using the three independent RNA sets. mRNA quantities were normalized against 18S RNA using ABI rRNA control reagents.
In situ hybridization
The protocol for in situ hybridization was essentially as described previously by Simmons et al. (26). Uterine tissues were fixed in 10% formalin. After overnight fixation at room temperature, tissues were dehydrated through a series of ethanol and then processed for paraffin embedding. Paraffin sections were mounted onto poly-L-lysine-coated slides (VWR Scientific Products) and then used for in situ hybridization. Riboprobes were labeled with [35S]UTP by in vitro transcription of amplified DNA products containing the T7 polymerase promoter sequence flanking the desired nucleotide primer sequence (Promega Corp., Madison, WI). cRNA probes were generated for Cyp26a1, ADH5, Aldh1a1, Aldh1a2, Klk6, GATA-2, and Cited2 with T7 polymerase. Slides were incubated for 7 min at room temperature in proteinase K (20 μg/ml) in a buffer containing 50 mM Tris and 5 mM EDTA (pH 8.0). Slides were then acetylated with acetic anhydride, dehydrated, and exposed to either denatured antisense or sense probes in hybridization buffer [50% formamide, 10% dextran sulfate, 5x Denhardt’s solution, 300 mM NaCl, 5 mM EDTA (pH 8), 20 mM Tris (pH 8.0), and 0.05 mg/ml yeast tRNA]. Hybridization was performed at 55 C overnight in a humidity chamber containing 5x standard saline citrate and 50% formamide. Hybridized slides were exposed to 20 μg/ml ribonuclease A for 30 min at 37 C. Slides were washed in 50% formamide, 2x standard saline citrate, and 100 mM 2-mercaptoethanol, followed by 2x standard saline citrate at 55 C for 30 min, dehydrated in a graded series of ethanol in 0.3 M ammonium acetate, and exposed to Biomax MR film overnight (Eastman Kodak Co., Rochester, NY). The following morning, slides were dipped in autoradiography emulsion (Amersham Biosciences, Piscataway, NJ) and placed at 4 C in a light-proof box for several days. After development, slides were counterstained with hematoxylin.
Results
Global gene expression profiles of P4-treated uterine samples in PRKO and wild-type mice
P4- and PR-regulated genes in the mouse uterus were identified by comparing the mRNA expression pattern of uteri from ovariectomized wild-type and PRKO mice given an acute or chronic regimen of P4 as previously described. The goal of acute P4 treatment was to identify rapid responsive genes by P4 and PR (direct target genes); the goal of the chronic hormone regimen was to identify secondary, potentially indirect target genes of P4-PR regulation. Of approximately 12,000 genes represented on the Affymetrix murine genome U74Av2 oligonucleotide array (Affymetrix), 48–55% of the genes were recorded as present in the uterine samples using MAS 5.0 (Affymetrix). The data were analyzed as detailed in Materials and Methods. As generally adopted for oligonucleotide microarray profile analysis, we selected differentially expressed genes by a lower boundary of a 90% confidence interval of fold change greater than 1.2 and an absolute value of difference between group means greater than 80. This was applied to select increased and decreased genes.
The above-described experimental design allowed comparison not only of the effect of acute vs. chronic treatment of P4 in these mice, but also of the effects of vehicle and hormone on both wild-type and PRKO mice. Using this experimental design, the identification of differentially expressed genes can be derived from five physiologically relevant comparisons. The five comparisons are shown graphically in Fig. 1. Comparison 1 consisted of analyzing differential gene expression of wild-type mice treated with vehicle vs. wild-type mice treated with P4. This treatment identified P4-regulated genes. Comparison 2 analyzed wild-type and PRKO mice treated with P4. This comparison identified genes regulated by PR. Comparison 3 analyzed PRKO mice treated with vehicle compared with wild-type mice treated with P4. This comparison was expected to identify genes regulated by both P4 and PR. Comparison 4 identified differentially expressed genes between wild-type and PRKO mice treated with vehicle. This comparison identified genes whose expression was due to the developmental consequences of ablation of the PR. Finally, comparison 5 consisted of PRKO mice treated with either vehicle or P4. This comparison was expected to show P4-regulated genes independent of PR.
FIG. 1. The five physiologically relevant comparisons used to identify significantly P4- and/or PR-regulated uterine genes: comparison 1, vehicle (veh)-treated wild-type (WT) vs. P4-treated wild-type mice; comparison 2, P4-treated PRKO vs. P4-treated wild-type mice; comparison 3, vehicle-treated PRKO vs. P4-treated wild-type mice; comparison 4, vehicle-treated wild-type vs. vehicle-treated PRKO mice; and comparison 5, vehicle-treated PRKO vs. P4-treated PRKO mice. Differentially expressed genes were selected using two-sample comparison according to the lower boundary of a 90% confidence interval of fold change greater than 1.2 and an absolute value of difference between group means greater than 80.
The summary of the number of differentially expressed genes after acute treatment of P4 for the five comparisons is shown in Table 1. Overall, at the 4 h point, the number of increased genes was greater than that of decreased genes. Comparison 1, which identified the effects of P4 on the mouse uterus, showed 90 genes that were differentially expressed. Of the 90 genes, 67 were more highly expressed in the P4-treated, wild-type uteri, and 23 genes were decreased more than 1.2-fold. When examining the effect of PR on the ability of the mouse uteri to respond to an acute P4 treatment, comparison 2, 183 genes were differentially expressed between wild-type and PRKO in P4-treated uteri samples at 4 h. Of the 183 genes, 120 were increased, and 63 genes were decreased. A total of 64 genes were identified by both comparisons; vehicle-treated vs. P4-treated wild-type, and P4-treated PRKO vs. wild-type. Comparison 3, which identifies both P4- and PR-regulated genes, showed the greatest number of differentially expressed genes (n = 235), with 139 genes increased and 96 genes decreased. Figure 2A diagrammatically shows the overlap in the analysis of P4-responsive (comparison 1), PR-responsive (comparison 2), and both P4- and PR-responsive (comparison 3) genes. The high degrees of overlap among these three groups, which are based on different animal groups, demonstrate the robustness of the experimental design. Comparison 4 shows that 45 genes were identified as differentially expressed when comparing the nonhormone-stimulated, ovariectomized, wild-type to the PRKO uterus. This difference could be accounted for by the impact of the lack of P4-PR stimulation throughout development. Finally, comparison 5, which shows the effect of P4 on the PRKO mouse uterus, identified five genes that were differentially expressed. Based on permutation testing, we estimated the false discovery rate to be 14% or less for the first three comparisons, whereas the rate was 70% or higher for comparisons 4 and 5, suggesting that most differentially expressed genes in these two analyses are due to chance. A complete list of the increased and decreased genes that are identified as significant for comparisons 1–3 and comparisons 4 and 5 with acute P4 treatment are presented in supplemental Tables 1 and 2, respectively (published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org).
TABLE 1. Number of significantly increased and decreased uterine genes at 4 h
FIG. 2. Venn diagrams demonstrating the relationship between genes modulated in the wild-type (WT) and PRKO uterus in response to acute (A) or chronic (B) treatment with P4. Red circles indicate genes selected by comparison 1 [vehicle-treated (veh) wild-type vs. P4-treated wild-type]; green, by comparison 2 (P4-treated PRKO vs. P4-treated wild-type); and blue, by comparison 3 (vehicle-treated PRKO vs. P4-treated wild-type). The numbers within the intersections of the circles indicate the common genes by two (or three) comparisons. The number at the bottom right corner of the panel indicates the total number of genes analyzed. C, Clustering analysis of P4- and PR-regulated genes in the murine uterus. One hundred and ninety-seven P4- and PR-regulated genes were clustered in two dimensions according to their gene expression pattern by hierarchical-tree algorithm (dChip version 1.3). The color code for the signal strength in the classification scheme is shown in the box at the bottom, in which induced genes are indicated by red, and repressed genes are indicated by green.
The results of the chronic hormone stimulation of mouse uteri with P4 are summarized in Table 2. The goal of this analysis was to identify late responsive genes by this axis. In the comparison with the acute stimulation of this axis, the majority of genes differentially expressed were decreased. As with the analysis of the acute affects of P4 treatment, this analysis identified genes that were differentially impacted by P4 treatment (comparison 1), PR ablation (comparison 2), and the combination of P4 treatment and PR ablation (comparison 3). As with the acute P4 treatment ablation, significantly more genes were impacted by ablation of the PR than by administration of P4. Figure 2B shows the analysis of the overlap in genes impacted by P4 administration or PR ablation. Again, a significant number of genes were identified as common by all three comparisons. As with the analysis of the acute injection protocol, analysis of PRKO and wild-type mice treated with vehicle demonstrated that similar numbers of genes were shown to be significantly affected. That is, 45 genes were significantly different with both acute and chronic injections of vehicle, respectively, with 15 genes identified at both time points. This comparison showed that the injection protocol did not impact the expression of uterine genes. In the final comparison, in which PRKO mice treated chronically with oil or P4 were analyzed, 44 genes were significantly changed by P4 in the PRKO uteri. This is more than the six genes identified by the acutely stimulated uteri and may reflect either pharmacological stimulation of other steroid hormone receptors, such as the glucocorticoid receptor, or activation of genes through a PR-independent pathway. A complete list of genes that showed significant change for comparisons 1–3 and comparisons 4 and 5 with chronic P4 treatment are presented in supplemental Tables 2 and 3, respectively (published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org).
TABLE 2. Number of significantly increased and decreased uterine genes at 40 h
Grouping of P4- and PR-regulated genes into early, intermediate, and late responsive genes
The overall goal of this investigation was to identify genes that are activated by the P4-PR axis. Therefore, the rest of this manuscript will focus on genes that were identified as in common among comparisons 1, 2, and 3 (Fig. 1). The list of the common genes for acute and chronic P4 treatment is listed in Tables 3 and 4, respectively. This approach identified genes that were significantly impacted by both P4 administration and PR ablation. Combining the gene lists obtained by acute and chronic P4 treatments yielded a total of 197 differentially regulated genes, of which 10 genes were differentially expressed with both acute and chronic P4 treatment, 54 genes were differentially expressed with acute P4 treatment (Fig. 2A), and 133 genes were impacted by chronic P4 treatment (Fig. 2B). All subsequent data analyses and experiments were based on these 197 genes.
TABLE 3. The list of uterine genes significantly increased and decreased by acute P4 treatment
TABLE 3A. Continued
TABLE 4. The list of genes significantly increased and decreased uterine genes by chronic P4 treatment
TABLE 4A. Continued
To identify trends in the changes in gene expression of P4- and PR-stimulated genes, stringent data filtering and clustering also allowed grouping of the P4- and PR-regulated genes into sets of biological relevance. The 197 regulated genes were analyzed by mathematical clustering using a hierarchical tree (Fig. 2C). Gene expression changes were observed after acute P4 treatment. Early responsive genes are presented first, followed by intermediate responsive genes, differentially expressed genes that changed at both time points, then by late responsive genes, whose expression changed only at the late time point. Tables 3 and 4 list the genes, the GenBank accession numbers, and the fold changes as grouped by gene ontology.
This analysis shows that 54 genes were identified as early responsive genes, whereas 130 genes were identified as late responsive genes. Thirteen genes were identified as intermediate responsive genes. The validity of this observation was confirmed by checking already known P4-regulated genes. Ihh (18) and HIF1A (19) belong to the early responsive genes. Amphiregulin (12) and histidine decarboxylase (13), both of which were up-regulated at both time points, were grouped with intermediate responsive genes. Calbindin-D9k (27), calcitonin (15, 16), and immune-responsive gene 1 (20, 28) genes were identified as late responsive genes. The expression profiles of these genes correlated to previously published results.
Functional grouping of P4- and PR-regulated genes
To categorize the P4- and PR-regulated genes into specific functional groups, the significantly increased and decreased genes were functionally categorized by gene ontology terms. The results of sorting the genes that responded to acute and chronic P4 treatments are shown in Fig. 3, A and B, respectively. Transcription factors, transport proteins, signal transduction, cell growth, and enzyme genes were increased by acute treatment with P4 in contrast with the effect of chronic P4 treatment (Fig. 3A). Likewise, more of these regulatory genes were decreased after chronic treatment with P4 (Fig. 3B).
FIG. 3. The functional categorization of increased and decreased genes by P4 and PR in the murine uterus with acute (A) or chronic (B) P4 treatment; genes were annotated and assigned to various functional categories using the Gene Ontology and Affymetrix annotation. , Increased genes; , decreased genes.
P4 regulates spatially and temporally RA metabolism-related genes
RA plays important roles in the maintenance of pregnancy and embryo development (29, 30, 31, 32). Analysis of the genes regulated by the P4-PR axis showed that genes involved in the production, degradation, and transport of RA were regulated by this axis. Enzymes involved in the production of RA, ADH5 and Aldh1a1, were also identified as early responsive genes. Interestingly, although Aldh1a2 was identified as a late responsive gene, i.e. a gene whose expression was increased after chronic P4 treatment, Cyp26a1, an enzyme involved in RA metabolism was identified as an early responsive gene and, in fact, represented the largest fold change in this microarray analysis. Finally, two Rbps were identified as late responsive genes. Rbp1 was identified as a late responsive increased gene, and Rbp4 was identified as a late responsive decreased gene. Although the P4-PR axis altered the expression of genes involved in the production and transport of RA, this axis did not affect the genes involved in retinoic signal transduction. The expression profiles of retinoid X receptor (RXR) and RA receptor (RAR) genes were not changed by P4 treatment or at different time points (data not shown). Real-time PCR was used to validate the microarray results for Cyp26a1, ADH5, and Aldh1a1 genes using three additional sets of RNA. The expressions of Cyp26a1, ADH5, and Aldh1a1 were significantly increased by P4 in wild-type mice (104.09-, 2.13-, and 1.54-fold, respectively), and these inductions were decreased in PRKO mice (2.0%, 46.1%, and 71.1%, respectively). The results were consistent with the microarray data identification of these genes as early responsive genes. Cyp26a1, ADH5, and Aldh1a1 exhibited high expression only in the acute P4-treated, wild-type uterus (Fig. 4A), whereas differential expression of those genes was not observed with chronic P4 treatment (data not shown). Furthermore, confirmation of the increase in Rbp1 (1.92-fold) and the decrease in Rbp4 (41%) after chronic stimulation of the P4-PR axis was demonstrated by real-time PCR (Fig. 4B). These P4 effects of Rbp1 and Rbp4 were significantly decreased in the PRKO.
FIG. 4. A, Real-time RT-PCR analysis of Cyp26a1, ADH5, and Aldh1a1 in vehicle-treated (Veh; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. B, Real-time PCR of Rbp1 and Rbp4 in chronic P4-treated uteri. The results represent the mean ± SE of three independent RNA sets. *, P < 0.05; **, P < 0.01.
Next, the temporal and spatial expressions of these genes were examined by in situ hybridization. Cyp26a1 and ADH5 transcripts were undetectable, and Aldh1a1 was expressed at luminal epithelium and glandular epithelium after vehicle treatment. As shown in Fig. 5, Cyp26a1, ADH5, and Aldh1a1 transcripts exhibited an increase only with acute P4 treatment (Fig. 5A) and not with chronic P4 treatment (Fig. 5B). The expression of Cyp26a1 by P4 was limited to the luminal epithelium. ADH5 and Aldh1a1 transcripts were shown in the luminal epithelium and glandular epithelium at 4 h. Cyp26a1 and ADH5 transcripts were undetectable, and Aldh1a1 transcripts were decreased with chronic P4 treatment (Fig. 5B). In contrast, Aldh1a2 expression was not detected at an early time point, but appeared with chronic P4 treatment in the endometrial stroma.
FIG. 5. Localization pattern of RA metabolism genes by in situ hybridization in the acute (A) and chronic (B) P4-treated wild-type mouse uterus. C, Localization pattern of ADH5 and Cyp26a1 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
The spatial and temporal expressions of Aldh1a1, Aldh1a2, and Cyp26a1 in the mouse uterus during pregnancy have been reported (29). Aldh1a1 glandular expression has been reported to be sharply induced on d 2.5, whereas Aldh1a2 stromal expression increased more steadily until the implantation phase (29). Cyp26a1 epithelial expression is strongly induced between d 3.5 and 4.5. In this report, we performed in situ hybridization on pseudopregnant uterine samples to determine the expression patterns of ADH5 and Cyp26a1. The results of this analysis are shown in Fig. 6. ADH5 transcripts were almost undetectable on d 0.5 of pseudopregnancy. However, they were induced by d 2.5 throughout the endometrial stroma. Expression levels increased on d 4.5 and decreased on d 6.5. Cyp26a1 transcripts were not detected on d 0.5 or 2.5. Expression appeared in the uterine luminal epithelium on d 4.5, and strong expression persisted throughout the uterine epithelium until d 6.5 (Fig. 5C). ADH5, Aldh1a, and Aldh1a2 are RA synthesis enzymes, and Cyp26a1 is a RA-catabolizing enzyme. Interestingly, these RA-metabolizing enzymes exhibited distinct spatial and temporal profiles of expression in pseudopregnant uteri. These in situ hybridization results demonstrate that the RA synthesis and degradation enzymes are always spatially and temporally expressed in a discrete nonoverlapping manner in the pseudopregnant uterus.
FIG. 6. The expression pattern of Klk genes in the murine wild-type (WT) and PRKO mouse uterus. A, Real-time PCR of Klk genes in vehicle-treated (Veh.; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. The results represent the mean ± SE of three independent RNA sets. *, P < 0.05. B, Localization pattern of Klk6 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
Kallikrein genes were increased early responsive genes by P4
In the functional analysis of P4- and PR-regulated genes, the enzyme group of regulated genes showed that a large number of kallikreins genes were shown to be regulated by this axis. Klk genes are a highly conserved gene family of serine proteases (33, 34, 35). Microarray analysis showed that Klk5 and Klk6 were identified as early responsive genes, i.e. significantly increased by acute P4 treatment, but no difference was detected with chronic P4 treatment. Although, Klk13, Klk16, and Klk26 were shown to be early responsive genes, the expression of these genes was not significant, as judged by analysis of the microarray data. The microarray analysis also indicated that the expression level of Klk genes was very low in the uterus. The microarray data for the Klk genes were confirmed by real-time PCR. Real-time PCR could only validate the differential expression of Klk5 and Klk6 by the P4-PR axis, as shown in Fig. 6A. In situ hybridization was conducted to determine the regulation and location of the Klk genes. Only Klk6 could be detected by in situ hybridization in pseudopregnant uteri (Fig. 6B). Attempts to detect the expression of the other Klk genes in the uterus were unsuccessful due to the low levels of the mRNAs for these genes. Klk6 transcripts were not detected by d 2.5. Expression appeared in the uterine endometrial stroma by d 4.5. Expression in luminal epithelium and glandular epithelium was seen on d 6.5.
Transcriptional regulators, GATA2 and Cited2, were rapidly induced by P4
Several transcription factors were identified as being impacted by the P4-PR axis using microarray analysis. HIF1 (19) and Hoxa-10 (14) genes are known to be increased by P4. These transcription factor genes were highly expressed by P4 in our microarray data. This microarray analysis also identified the developmentally important transcription factors, GATA-2 (GATA-binding protein 2) and Cited2, as being regulated by the P4-PR axis. GATA-2 is a transcription factor and plays a critical role in hemopoiesis (36, 37). Regulation of these genes by the P4-PR axis was confirmed by real-time RT-PCR and in situ hybridization assays. Real-time PCR validated that GATA-2 expression was elevated in P4-treated wild-type uteri compared with P4-treated PRKO (1.54-fold) and vehicle-treated wild-type (1.78-fold) and PRKO (1.84-fold) mouse uteri, as shown in Fig. 7A. In situ hybridization analysis of mouse uteri showed that GATA-2 transcripts were rapidly induced in the luminal epithelium and glandular epithelium by P4 (Fig. 7B). Having determined the spatial expression of GATA-2 in the uterus after pharmacological administration of P4, the spatial and temporal expressions of GATA-2 were determined during pseudopregnancy. GATA-2 transcripts were first detected in the luminal epithelium and glandular epithelium of pseudopregnant uterine samples on d 2.5. The expression in luminal epithelium was decreased, and the signal in glandular epithelium disappeared on d 4.5. By d 6.5, the transcripts were undetectable (Fig. 7C). Thus, as observed with Ihh (18), GATA-2, a P4-regulated gene, was physiologically expressed before the window of receptivity in the mouse uterus.
FIG. 7. The expression pattern of GATA-2 in the murine uterus. A, Real-time PCR of GATA-2 in vehicle-treated (Veh.; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. The results represent the mean ± SE of three independent RNA sets. **, P < 0.01. B, Localization pattern of GATA-2 transcripts by in situ hybridization in vehicle- and acute P4-treated wild-type mouse uterus. C, Localization pattern of GATA-2 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
The expression level of Cited2 also was confirmed by real-time PCR and in situ hybridization assays (Fig. 8, A and B). Cited2 transcripts were highly expressed in the luminal epithelium, glandular epithelium, and myometrium by P4 at 4 h. In the analysis of the spatial and temporal patterns of expression of Cited2 transcripts during pseudopregnancy, Cited2 transcripts were seen at low levels in the endometrial stroma on d 0.5–4.5 and increased in the luminal epithelium, glandular epithelium, and myometrium on d 6.5 (Fig. 8C). Therefore, pharmacological doses of P4 induced Cited2 expression in an appropriate compartment of the uterus.
FIG. 8. The expression pattern of Cited2 in the murine uterus. A, Real-time PCR of Cited2 in vehicle-treated (Veh.; –) and acute P4-treated (+) wild-type (WT) and PRKO mouse uteri. The results represent the mean ± SE of three independent RNA sets. **, P < 0.01. B, Localization pattern of Cited2 transcripts by in situ hybridization in vehicle- and acute P4-treated uterus at 4 h. C, Localization pattern of Cited2 transcripts by in situ hybridization during pseudopregnancy. Representative panels for bright-field and dark-field photomicrographs are depicted. Nuclei are lightly counterstained with hematoxylin. Original magnification of all slides, x20. Bar, 200 μm. ST, Stroma cells; GE, glandular epithelium; LE, luminal epithelium; Myo, myometrium.
Discussion
In this study, we have identified P4- and PR-regulated uterine genes using the PRKO mouse in combination with high-density DNA microarray analysis. Acute and chronic administrations of P4 identified early responsive and late responsive genes. Although this microarray analysis identified genes regulated by P4, PR, or both components of the P4-PR axis, the majority of these genes were identified as regulated by both P4 and PR. The genes that were separately affected by either P4 treatment or PR ablation may be the result of several phenomena: 1) developmental differences between the PRKO and the wild-type uterus, 2) ligand-independent activation of the PR (38), 3) P4 having effects via a membrane receptor (39), or 4) pharmacological activation of other steroid/nuclear receptors, such as the glucocorticoid receptor. Hereafter, this report focuses on the identification of P4- and PR-regulated target genes. These target genes were identified by selecting only genes whose expression was significantly altered in comparisons 1, 2, and 3, as described in Fig. 1.
In the analysis of the impact of either acute or chronic stimulation of the P4-PR axis on the regulation of global uterine gene expression, it is clear that acute stimulation of this axis results in the overall activation of uterine gene expression. The majority of genes activated by acute P4 treatment are regulatory genes. This observation agrees with the premise that PR is an activator of gene transcription (40). The other observation that can be made for the global changes in gene expression is that with chronic treatment of P4, the major result was repression of gene expression. This could be partially explained by the fact that chronic administration of P4 will repress the expression of PR and estrogen receptor (41). This would explain why some of the early responsive genes lost activation with chronic P4 treatment. However, it does not fully explain why the majority of genes were repressed at this time point compared with the ovariectomized or PRKO uteri treated with vehicle. The down-regulation of these genes is most likely due to downstream effects initiated by activation of the P4-PR axis. Chronic treatment of the uterus may result in differentiation of uterine tissue to a tissue in which these regulatory genes are expressed at lower levels than in the unstimulated uterus. The target genes identified in this report may serve as a resource for a bioinformatics analysis in which the promoter regions of these genes could reveal the common pathways that result in repression of these genes by chronic P4 treatment.
The choice of conducting the microarray analysis on uteri isolated 4 h after a single dose of P4 was for the purpose of identifying direct targets of P4 activity. Previous attempts at identifying the impact of P4 on uterine gene expression consisted of analyzing the mice 6 h after hormone administration (12). However, in analyzing the temporal response of the uterus to this regimen of P4, it was demonstrated that changes in gene expression can be observed at 4 h or earlier (18). Therefore, the time point of 4 h was chosen to identify genes that are potentially direct targets of P4.
We have validated the microarray results for nine individual genes by real-time PCR using three additional sets of RNA. As a whole, the assessments of fold change by array analysis are markedly less that those determined by real-time PCR. The underestimation of the fold change by microarray analysis compared with real-time PCR analysis highlights the limitations of using microarray analysis as a means of quantifying changes in gene expression. Also, because the uterus consists of several different compartments, in situ hybridization analysis was conducted to localize the uterine compartment in which the change in gene expression was observed. Cyp26a1, Aldh1a1, and GATA-2 were only expressed at the luminal epithelium. ADH5 and Aldh1a2 were expressed at the stroma. These P4- and PR-regulated genes were regulated only at a specific compartment that can reduce the fold change in the entire tissue analysis. Thus, the fold changes seen on the array analysis are probably an underestimation of the true differences in gene expression.
One of the major metabolic pathways identified by this array is that of genes involved in RA metabolism. RA, the natural metabolite of vitamin A, has been shown to play a key role in the regulation of reproduction. In the female reproductive tract, RA has been shown to regulate differentiation of the reproductive tract epithelium (42). RA has also been show to mimic the ability of P4 to regulate the expression of matrix metalloproteases in endometriotic tissue (43). These previous observations indicate that RA signaling is important for the function of the female reproductive tract and that RA signaling may in part mediate P4 signaling in the uterus. The results of this microarray analysis has demonstrated that both RA synthesis (ADH5 and Aldh1a1) and RA catabolic enzyme (Cyp26a1) were rapidly induced by the P4-PR axis, whereas levels of RAR and RXR were not affected. In addition, we showed that Rbp1 and Rbp4 were P4 regulated. These results are similar to observations in clinical samples, where the intracellular levels of RA and the expression of cytoplasmic Rbps fluctuate, and the nuclear RAR and RXR receptors remain at similar levels throughout the menstrual cycle (44, 45).
The regulation of enzymes involved in both synthesis and breakdown of RA by the P4-PR axis seems paradoxical. However, one must keep in mind the conditions under which the microarrays were performed. The microarrays were conducted by treating ovariectomized mice with pharmacological doses of P4. Therefore, these mice were given P4 at a time when the levels of PR were highest in the endometrial epithelium. Not surprisingly, the majority of genes detected by this array were found to be expressed in the uterine epithelium. However, when analyzing the expression patterns of these RA metabolic enzymes, it was shown that their expressions were compartmentally and temporarily regulated. During pregnancy, ADH5 transcripts were induced by d 2.5 throughout the endometrial stroma. Expression levels increased on d 4.5 and decreased on d 6.5. Aldh1a1 expression was sharply induced in the endometrial glandular compartment on gestational d 2.5, whereas Aldh1a2 stromal expression increased more steadily until the implantation phase (29). Cyp26a1 transcripts appeared in the uterine epithelium and glandular epithelium on d 4.5, and strong expression persisted throughout the uterine epithelium until d 6.5. The in situ hybridization results in the pseudopregnant uterus demonstrate that the RA synthesis and degradation enzymes are always spatially and temporally expressed in a discrete, nonoverlapping manner in early pregnancy. Moreover, the balance between synthesis and degradation of RA is very important in early pregnancy to allow RA signaling to act on the endometrium to prepare for implantation without having deleterious affects on differentiation of the embryo.
One group of genes that has been shown to be up-regulated by the P4-PR axis is the Klk genes. Klks are a gene family associated with regulation of the inflammatory process. These genes regulate a cascade of events resulting in vasodilatation, an increase in vascular permeability, and edema (46). Because implantation of the embryo in the mouse uterus requires vasodilatation and changes in vascular permeability, it is not surprising that the P4-PR axis is a primary regulator of uterine Klk gene expression. We have identified Klk5 and Klk6, which are regulated by the P4-PR axis. The expression levels of the Klk genes regulated in the uterus were low, and although validation could be accomplished by real-time PCR, only Klk6 transcripts could be detected by in situ hybridization in the mouse uterus. In the pseudopregnant uterus, Klk6 appeared in the uterine endometrial stroma by pregnancy d 4.5 and then in the luminal epithelium and glandular epithelium on d 6.5. This expression pattern of the Klks is similar to that in the rat uterus, where Klk expression is increased during the implantation period (47, 48).
The two transcription regulatory genes validated in this report were GATA-2 and Cited2. GATA-2 is a member of the GATA transcription factor family. Ablation of GATA-2 results in an embryonic lethal phenotype due to defects in hemopoiesis (36). Interestingly, GATA-2 activity in regulating hemopoiesis is linked to the action of RAR (49). This microarray analysis has shown that the P4-PR axis also regulates the expression of enzymes involved in the production of RA. Therefore, modulation of GATA-2 and regulation of levels of the ligand for RAR signaling may be a means by which the P4-PR axis regulates the activity of the RAR pathway in the uterus during pregnancy. Aside from interacting with RAR, GATA family members have been shown to interact with chromatin-remodeling proteins. In the rabbit, GATA-4 is regulated by P4 and interacts with RUSH (proteins with RING finger motifs that bind to the uteroglobin promoter), a member of the Smarca3 family of SWI/SNF-related chromatin-remodeling proteins, in the mediation of prolactin-augmented, P4-induced gene expression (50). Therefore, GATA-2 may play a pivotal role in regulating uterine function.
The other transcriptional regulator validated in this microarray analysis is Cited2. Cited2 is a member of the Cited transcription coactivator family that binds the p300/corticosterone-binding protein coactivator. Ablation of Cited2 results in cardiac, neural crest, and adrenal developmental defects (51). Cited2 is induced by various inflammatory stimuli and has been shown to regulate matrix metalloproteinase gene expression (52). Cited2 has been shown to be a coregulator of peroxisome proliferator-activated receptors and and has been shown to be a regulator of peroxisome proliferator-activated receptor induction of angiogenic factors in vitro (53). The induction of matrix metalloproteinases and angiogenic factors is critical in remodeling of the uterus for establishment of embryonic implantation. The induction of Cited2 by the P4-PR axis and the expression of Cited2 in the uterus during pregnancy indicate that Cited2 may play a role as a mediator of P4 action in remodeling of the uterus in preparation for pregnancy.
In summary, this study has demonstrated that acute stimulation of the P4-PR axis stimulates the expression of many regulatory genes, whereas chronic stimulation of this axis is inhibitory. Analysis of the genes regulated by acute stimulation of the P4-PR axis will identify genes and pathways that are directly regulated by P4 and may help determine what pathways are important for preparation of the uterus for the support and maintenance of embryonic implantation.
Acknowledgments
We thank Jinghua Li, Bryan Ngo, Jie Li, and Janet DeMayo, M.S., for technical assistance, and John Ellsworth for help with manuscript preparation.
References
Clarke CL, Sutherland RL 1990 Progestin regulation of cellular proliferation. Endocr Rev 11:266–301
Lydon JP, DeMayo FJ, Funk CR, Mani SK, Hughes AR, Montgomery Jr CA, Shyamala G, Conneely OM, O’Malley BW 1995 Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 9:2266–2278
Cheon YP, Li Q, Xu X, DeMayo FJ, Bagchi IC, Bagchi MK 2002 A genomic approach to identify novel progesterone receptor regulated pathways in the uterus during implantation. Mol Endocrinol 16:2853–2871
Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895
Tsai MJ, O’Malley BW 1994 Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu Rev Biochem 63:451–486
O’Malley BW, Conneely OM 1992 Orphan receptors: in search of a unifying hypothesis for activation. Mol Endocrinol 6:1359–1361
Conneely OM, Kettelberger DM, Tsai MJ, Schrader WT, O’Malley BW 1989 The chicken progesterone receptor A and B isoforms are products of an alternate translation initiation event. J Biol Chem 264:14062–14064
Stone DM, Hynes M, Armanini M, Swanson TA, Gu Q, Johnson RL, Scott MP, Pennica D, Goddard A, Phillips H, Noll M, Hooper JE, de Sauvage F, Rosenthal A 1996 The tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 384:129–134
Lydon JP, DeMayo FJ, Conneely OM, O’Malley BW 1996 Reproductive phenotpes of the progesterone receptor null mutant mouse. J Steroid Biochem Mol Biol 56:67–77
Mulac-Jericevic B, Lydon JP, DeMayo FJ, Conneely OM 2003 Defective mammary gland morphogenesis in mice lacking the progesterone receptor B isoform. Proc Natl Acad Sci USA 100:9744–9749
Mulac-Jericevic B, Mullinax RA, DeMayo FJ, Lydon JP, Conneely OM 2000 Subgroup of reproductive functions of progesterone mediated by progesterone receptor-B isoform. Science 289:1751–1754
Das SK, Chakraborty I, Paria BC, Wang XN, Plowman G, Dey SK 1995 Amphiregulin is an implantation-specific and progesterone-regulated gene in the mouse uterus. Mol Endocrinol 9:691–705
Paria BC, Das N, Das SK, Zhao X, Dileepan KN, Dey SK 1998 Histidine decarboxylase gene in the mouse uterus is regulated by progesterone and correlates with uterine differentiation for blastocyst implantation. Endocrinology 139:3958–3966
Lim H, Ma L, Ma WG, Maas RL, Dey SK 1999 Hoxa-10 regulates uterine stromal cell responsiveness to progesterone during implantation and decidualization in the mouse. Mol Endocrinol 13:1005–1017
Zhu LJ, Cullinan-Bove K, Polihronis M, Bagchi MK, Bagchi IC 1998 Calcitonin is a progesterone-regulated marker that forecasts the receptive state of endometrium during implantation. Endocrinology 139:3923–3934
Kumar S, Zhu LJ, Polihronis M, Cameron ST, Baird DT, Schatz F, Dua A, Ying YK, Bagchi MK, Bagchi IC 1998 Progesterone induces calcitonin gene expression in human endometrium within the putative window of implantation. J Clin Endocrinol Metab 83:4443–4450
Nie GY, Li Y, Wang J, Minoura H, Findlay JK, Salamonsen LA 2000 Complex regulation of calcium-binding protein D9k (calbindin-D(9k)) in the mouse uterus during early pregnancy and at the site of embryo implantation. Biol Reprod 62:27–36
Takamoto N, Zhao B, Tsai SY, DeMayo FJ 2002 Identification of Indian hedgehog as a progesterone-responsive gene in the murine uterus. Mol Endocrinol 16:2338–2348
Daikoku T, Matsumoto H, Gupta RA, Das SK, Gassmann M, DuBois RN, Dey SK 2003 Expression of hypoxia-inducible factors in the peri-implantation mouse uterus is regulated in a cell-specific and ovarian steroid hormone-dependent manner. Evidence for differential function of HIFs during early pregnancy. J Biol Chem 278:7683–7691
Cheon YP, Xu X, Bagchi MK, Bagchi IC 2003 Immune-responsive gene 1 (Irg1) is a novel target of progesterone receptor and plays a critical role during implantation in the mouse. Endocrinology 144:5623–5630
Schadt EE, Li C, Ellis B, Wong WH 2001 Feature extraction and normalization algorithms for high-density oligonucleotide gene expression array data. J Cell Biochem 37(Suppl):120–125
Li C, Wong WH 2001 Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci USA 98:31–36
Li C, Hung Wong W 2001 Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol 2:RESEARCH0032
Li C, Wong WH 2003 DNA-chip analyzer (dChip). In: Parmigiani G, Garrett ES, Irizarry R, Zeger SL, eds. The analysis of gene expression data: methods and software. New York: Springer; 120–141
Dahlquist KD, Salomonis N, Vranizan K, Lawlor SC, Conklin BR 2002 GenMAPP, a new tool for viewing and analyzing microarray data on biological pathways. Nat Genet 31:19–20
Simmons DM, Arriza JL, Swanson LW 1989 A complete protocol for in situ hybridization of messenger RNAs in brain and other tissues with radiolabeled single-stranded RNA probes. J Histotechnol 12:169–181
Tatsumi K, Higuchi T, Fujiwara H, Nakayama T, Itoh K, Mori T, Fujii S, Fujita J 1999 Expression of calcium binding protein D-9k messenger RNA in the mouse uterine endometrium during implantation. Mol Hum Reprod 5:153–161
Chen B, Zhang D, Pollard JW 2003 Progesterone regulation of the mammalian orthologue of methylcitrate dehydratase (immune response gene 1) in the uterine epithelium during implantation through the PKC pathway. Mol Endocrinol 17:2340–2354
Vermot J, Fraulob V, Dolle P, Niederreither K 2000 Expression of enzymes synthesizing (aldehyde dehydrogenase 1 and reinaldehyde dehydrogenase 2) and metabolizing (Cyp26) retinoic acid in the mouse female reproductive system. Endocrinology 141:3638–3645
van der Leede BM, van den Brink CE, Pijnappel WW, Sonneveld E, van der Saag PT, van der Burg B 1997 Autoinduction of retinoic acid metabolism to polar derivatives with decreased biological activity in retinoic acid-sensitive, but not in retinoic acid-resistant human breast cancer cells. J Biol Chem 272:17921–17928
Morriss-Kay GM, Ward SJ 1999 Retinoids and mammalian development. Int Rev Cytol 188:73–131
Sapin V, Ward SJ, Bronner S, Chambon P, Dolle P 1997 Differential expression of transcripts encoding retinoid binding proteins and retinoic acid receptors during placentation of the mouse. Dev Dyn 208:199–210
Yousef GM, Diamandis EP 2003 An overview of the kallikrein gene families in humans and other species: emerging candidate tumour markers small star, filled. Clin Biochem 36:443–452
Yousef GM, Diamandis EP 2002 Expanded human tissue kallikrein family–a novel panel of cancer biomarkers. Tumour Biol 23:185–192
Yousef GM, Diamandis EP 2002 Human tissue kallikreins: a new enzymatic cascade pathway? Biol Chem 383:1045–1057
Li Y, Wang M 2000 The role of transcription factor GATA-2 in early hematopoiesis. Zhongguo Shi Yan Xue Ye Xue Za Zhi 8:66–70
Tsai FY, Keller G, Kuo FC, Weiss M, Chen J, Rosenblatt M, Alt FW, Orkin SH 1994 An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371:221–226
Power RF, Mani SK, Codina J, Conneely OM, O’Malley BW 1991 Dopaminergic and ligand-independent activation of steroid hormone receptors. Science 254:1636–1639
Zhu Y, Rice CD, Pang Y, Pace M, Thomas P 2003 Cloning, expression, and characterization of a membrane progestin receptor and evidence it is an intermediary in meiotic maturation of fish oocytes. Proc Natl Acad Sci USA 100:2231–2236
Li X, Lonard DM, O’Malley B W 2004 A contemporary understanding of progesterone receptor function. Mech Ageing Dev 125:669–678
Tibbetts TA, Mendoza-Meneses M, O’Malley BW, Conneely OM 1998 Mutual and intercompartmental regulation of estrogen receptor and progesterone receptor expression in the mouse uterus. Biol Reprod 59:1143–1152
Ponnamperuma RM, Kirchhof SM, Trifiletti L, De Luca LM 1999 Ovariectomy increases squamous metaplasia of the uterine horns and survival of SENCAR mice fed a vitamin A-deficient diet. Am J Clin Nutr 70:502–508
Bruner-Tran KL, Eisenberg E, Yeaman GR, Anderson TA, McBean J, Osteen KG 2002 Steroid and cytokine regulation of matrix metalloproteinase expression in endometriosis and the establishment of experimental endometriosis in nude mice. J Clin Endocrinol Metab 87:4782–4791
Kumarendran MK, Loughney AD, Prentice A, Thomas EJ, Redfern CP 1996 Nuclear retinoid receptor expression in normal human endometrium throughout the menstrual cycle. Mol Hum Reprod 2:123–129
Loughney AD, Kumarendran MK, Thomas EJ, Redfern CP 1995 Variation in the expression of cellular retinoid binding proteins in human endometrium throughout the menstrual cycle. Hum Reprod 10:1297–1304
Bhoola KD, Figueroa CD, Worthy K 1992 Bioregulation of kinins: kallikreins, kininogens, and kininases. Pharmacol Rev 44:1–80[Free Full Text]
Corthorn J, Valdes G 1994 Variations in uterine kallikrein during cycle and early pregnancy in the rat. Biol Reprod 50:1261–1264
Valdes G, Corthorn J, Scicli AG, Gaete V, Soto J, Ortiz ME, Foradori A, Saed GM 1993 Uterine kallikrein in the early pregnant rat. Biol Reprod 49:802–808 !-- null -->
Tsuzuki S, Kitajima K, Nakano T, Glasow A, Zelent A, Enver T 2004 Cross talk between retinoic acid signaling and transcription factor GATA-2. Mol Cell Biol 24:6824–6836
Hewetson A, Hendrix EC, Mansharamani M, Lee VH, Chilton BS 2002 Identification of the RUSH consensus-binding site by cyclic amplification and selection of targets: demonstration that RUSH mediates the ability of prolactin to augment progesterone-dependent gene expression. Mol Endocrinol 16:2101–2112
Bamforth SD, Braganca J, Eloranta JJ, Murdock JN, Marques FI, Kranc KR, Farza H, Henderson DJ, Hurst HC, Bhattacharya A 2001 Cardiac malformations, adrenal agenesis, neural crest defects and exencephaly in mice lacking Cited2, a new Tfap2 co-activator. Nat Genet 29:469–474
Yokota H, Goldring MB, Sun HB 2003 CITED2-mediated regulation of MMP-1 and MMP-13 in human chondrocytes under flow shear. J Biol Chem 278:47275–47280
Tien ES, Davis JW, Vanden Heuvel JP 2004 Identification of the CREB-binding protein/p300-interacting protein CITED2 as a peroxisome proliferator-activated receptor coregulator. J Biol Chem 279:24053–24063(Jae-Wook Jeong, Kevin Y. )