| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
Knockout Mouse1
a Departments of Anatomy and
b Urology, University of California, San Francisco, California 94143
c Department of Veterinary Biosciences, University of Illinois, Urbana, Illinois 61802
d MRC Reproductive Biology Unit, Edinburgh, EH3 9ET, United Kingdom
e Departments of Biochemistry and Child Health, University of Missouri, Columbia, Missouri 65211
f Department of Medical Nutrition and Biosciences, Karolinska Institute, NOVUM, S-141 86 Huddinge, Sweden
ABSTRACT
Regulation of progesterone receptor (PR) in uterine stroma (endometrial stroma plus myometrium) by estrogen was investigated in estrogen receptor-
(ER) knockout (ERKO) mice. 17ß-Estradiol (E2) increased PR levels in uterine stroma of ovariectomized ERKO mice, and ICI 182 780 (ICI) inhibited this E2-induced PR expression. Estrogen receptor-ß (ERß) was detected in both uterine epithelium and stroma of wild-type and ERKO mice by immunohistochemistry. In organ cultures of ERKO uterus, both E2 and diethylstilbestrol induced stromal PR, and ICI inhibited this induction. These findings suggest that estrogen induces stromal PR via ERß in ERKO uterus. However, this process is not mediated exclusively by ERß, because in ERß knockout mice, which express ER, PR was up-regulated by E2 in uterine stroma. In both wild-type and ERKO mice, progesterone and mechanical traumatization were essential and sufficient to induce decidual cells, even though E2 and ER were also required for increase in uterine weight. Progesterone receptor was strongly expressed in decidual cells in ERKO mice, and ICI did not inhibit decidualization or PR expression. This study suggests that up-regulation of PR in endometrial stroma is mediated through at least three mechanisms: 1) classical estrogen signaling through ER, 2) estrogen signaling through ERß, and 3) as a result of mechanical stimulation plus progesterone, which induces stromal cells to differentiate into decidual cells. Each of these pathways can function independently of the others.
decidua, estradiol, progesterone receptor, uterus
INTRODUCTION
Progesterone (P4) is a key coordinator of female reproduction, and in female mammals, P4 directly regulates many functions, such as sexual behavior [1], ovulation [2], uterine growth [3], implantation [4], and mammary gland development [5, 6]. Most actions of P4 are mediated via progesterone receptor (PR). Hence, female mice null for PR (PR knockout) are infertile and exhibit impaired secondary sexual development [7]. Progesterone receptor is a member of the steroid-retinoid receptor superfamily and functions as a P4-modulated transcription factor [8]. In the most species, including human and mouse, PR is composed of two major ligand-binding forms (the long B form and a N-terminal truncated A form) encoded by one gene [9]. Progesterone receptor is one of most well-studied estrogen-regulated genes and is widely recognized as a marker for estrogen action [10]. Regulation of the PR gene by estrogen has been extensively studied. Within the PR gene promoter, clusters of estrogen response element (ERE) half-sites are present, which are essential for transactivation of the PR gene by liganded estrogen receptor (ER) [11–13]. Because the PR protein level is thought to be a critical determinant of sensitivity to P4, almost all hormone treatment protocols designed to elicit effects of P4 usually involve previous estrogen priming to induce PR.
Estrogen actions are mediated by ER and/or ERß, the two forms of ER, which have been identified in many species, including rat, human, and mouse [14–16]. The amino acid sequences of ER and ERß in these three species are highly homologous in the DNA-binding domain and moderately homologous in the ligand-binding domain. Both ER and ERß can bind 17ß-estradiol (E2) and transactivate genes that are regulated through ERE [17, 18]. Expression of ER in the mouse uterus has been well documented [19], and ERß mRNA has also been detected in wild-type and ER knockout (ERKO) uteri [20]. Similarly, ERß mRNA and protein have been detected in the rat uterus [21–25], and ERß mRNA has been detected in the human uterus [26]. However, the ERß level in uterus is generally low compared to that of granulosa cells in ovary, where a very high level of ERß is detected [22–24, 27–29]. Moreover, in the uterus of ERKO mice, estrogen did not increase known markers of estrogen action, such as DNA synthesis, or transcription of the lactoferrin and glucose-6-phosphate dehydrogenase genes, despite the presence of ERß [30, 31]. In contrast, the uterus of ERß knockout (ßERKO) mice appears to be normal [27]. These studies suggest that ER is the major receptor mediating E2 action in the mouse uterus, and that ERß may not have unique functions in the rodent uterus. In a previous study, we observed that PR is up-regulated by E2 in the uterine stroma of ERKO mice [32]. In the current study, regulation of PR by estrogen in uterine endometrial stroma and myometrium of ERKO mice was studied in greater depth.
Both P4 and E2 are essential for establishment and maintenance of pregnancy. Whereas P4 is required throughout pregnancy, E2 is essential only during early pregnancy, especially around the time of implantation. In rodents, E2 is essential in addition to P4 to ready the uterus for implantation of the blastocyst [33]. No obvious morphological changes occur in endometrial stroma before implantation, but contact with the blastocyst is an essential stimulus to induce decidualization of endometrial stroma cells [34]. The importance of P4 and PR is well established, but the roles of E2 and ERs in implantation and decidualization are not clear. Decidualization can be induced artificially by a combination of P4-treatment and a mechanical traumatization of the uterine horn in the absence of estrogen [35]. Implantation of transferred blastocysts can occur without priming ovariectomized mice with E2 [36]. Recent studies have shown that decidualization occurs in ERKO mice; thus, classical estrogen actions through ER are not required for decidualization [37, 38]. On the other hand, E2 is required for natural decidualization during normal pregnancy or artificial decidualization induced by mild stimuli such as intrauterine oil injection [39]. The continued presence of E2 reduces the dosage of P4 required to maintain pregnancy after implantation, and E2 deficiency can be overcome by high doses of P4 [40]. These facts suggest that E2 action through ER itself does not generate signals that are essential for implantation and decidualization but, instead, modifies the sensitivity of uterine cells to the effects of P4, perhaps by regulating PR level. For these reasons, we studied the effects of E2 and P4 on PR expression in endometrial stroma of ERKO mice during decidualization.
MATERIALS AND METHODS
Animals and Treatments
All animals were maintained in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals, and all procedures described here were approved by the University of California-San Francisco and University of Illinois animal care and usage committees. Mice were maintained under controlled temperature and lighting conditions during the experiment and were given food (Purina Laboratory Chow; Ralston Purina Company, St. Louis, MO) and water ad libitum. The ERKO [30] and ßERKO mice [27], with a C57BL6J/129Svj mixed genetic background, were produced as described previously. Adult female athymic nude mice were purchased from Charles River (Wilmington, MA).
To test the effect of E2 on PR expression, all animals were ovariectomized at least 2 wk before hormone treatment. Small pieces of uterus were collected with ovary at the time of ovariectomy. These uterine tissue samples were used as uteri from intact mice. Ovaries were used as positive control for ERß staining. Daily dosages of 125 ng of E2 (Sigma, St. Louis, MO), with or without 50 µg of ICI 182 780 (ICI; Zeneca, Cheshire, UK), were injected i.p. in 0.1 ml of peanut oil (Sigma). Ovariectomized adult (60–80 days old) female wild-type, ERKO, ßERKO, and host athymic mice bearing grafts of ERKO uterine tissues received oil, E2, or E2 plus ICI for 3 days. Twenty-four hours after the last hormone injection, animals were sacrificed to harvest tissue samples. To study the effect of E2 on uteri of ERKO and wild-type mice under identical endocrine conditions, portions of uteri from mature adult ERKO and wild-type mice were surgically isolated and grafted under the contralateral renal capsules of female athymic nude mice. Approximately 1 mo after grafting, all hosts were ovariectomized and then received hormone treatments described above.
For artificial decidualization, hormone treatment was given as described previously [41]. The ERKO and wild-type mice were ovariectomized at least 2 wk before the hormone treatments. All hormones were injected i.p. in 0.1 ml of peanut oil (Sigma) to give the following groups: 1) oil-treated group, oil on Days 1–3 and 6–12; 2) P4-treated group, oil on Days 1–3 and 500 µg/day of P4 on Days 6–12; 3) E2+P4-treated group, 100 ng/day of E2 on Days 1–3 and 500 µg/day of P4 plus 10 ng/day of E2 on Days 6–12; and 4) ICI+P4-treated group, 50 µg/day of ICI on Days 1–3 and 500 µg/day of P4 plus 50 µg/day of ICI on Days 6–12. On Day 9, the body cavity was opened and the right uterine horn scratched from the inside with a needle tip at 6 h after the last hormone treatment (traumatization). On Day 13, animals were killed, and the wet weight and length of the uterine horns were measured.
Preparation of Total RNA
The procedure for RNA preparation has been described previously [32, 42]. Adult female ERKO mice were ovariectomized. Two weeks later, mice were given three daily treatments with oil or E2 as described above. All mice were killed 24 h after the last injection, and uteri were removed. Epithelium and stroma were isolated as described previously [43] and then flash-frozen in liquid nitrogen. Uteri from three mice were used for each treatment group. The experiment was repeated three times. Total RNA was prepared from frozen uterine stromal tissue using the RNeasy Mini Kit (Qiagen, Chatsworth, CA). Purity and concentration of the RNA were determined by ultraviolet (UV; 260/280 nm) absorbance using a spectrophotometer.
Northern Blotting
For all Northern blots used, equal amounts of RNA from control (oil-treated) and E2-treated mice were electrophoresed on a 1.5% v:v agarose formaldehyde gel. The gel was blotted to nylon membranes, and the RNA was fixed to the membrane by UV cross-linking. Murine PR cDNA [44] was a gift from Dr. G. Shyamala (Lawrence Berkeley Laboratory, Berkeley, CA). Human 28S rRNA [45] was used for a loading control. The PR cDNA probe was labeled with 32P-deoxycytidine triphosphate using the multiprime DNA-labeling system (Amersham, Arlington Heights, IL) and hybridized at 68°C. The hybridized membrane was washed and exposed to Kodak X-omat x-ray film (Kodak, Rochester, NY) with intensifying screens. After hybridization with the PR cDNA probe, membranes were stripped of probe and then reprobed with the 28S rRNA cDNA probe as described previously [46].
Immunohistochemistry
Anti-human PR rabbit polyclonal immunoglobulin (Ig) G (Dako, Carpenteria, CA) and anti-human-ER mouse monoclonal IgG 1D5 (Dako) were used at 1:100 and 1:50 dilutions (v:v), respectively. Alkaline phosphatase-conjugated, anti-smooth muscle actin mouse monoclonal IgG (Sigma) was used at a 1:1000 dilution (v:v). Anti-human-ERß chicken polyclonal antibody (ERß 503 IgY) [47] was used at a 1:600 dilution (v:v). Anti-human-ERß sheep polyclonal antibody was generated against a peptide directed against the hinge (D) domain of human ERß (CAGKAKRSGGHAPRVREL; Affinity Reagents, Exeter, UK) as described previously [48]. Briefly, the peptide was conjugated via cysteine residue to keyhole limpet hemocyanin and injected into a sheep; immunization and recovery of antisera were all carried out by Diagnostic Scotland Carluke (Lanarkshire, UK). Antisera were fractionated by sodium acetate precipitation, and the supernatant was dialyzed against PBS (pH 7.4) and applied to affinity peptide columns. The specificity of this antibody has been studied by Western blotting [48]. This antibody was used for immunohistochemistry at a 1:1000 dilution (v:v).
Tissues were fixed with 4% paraformaldehyde for 3 h on ice, processed into paraffin, and then sectioned at 6 µm. Slides were heated in 10 mM citrate buffer solution (pH 6.0) by microwave or pressure cooker. Immunohistochemical detection of ER [32], ERß [22], and PR [49] by the horseradish peroxidase-avidin biotin complex method has been described previously. Immunoreactivity for ER, ERß, and PR was visualized utilizing 3,3'-diaminobenzidine tetrahydrochloride (DAB; Sigma). The immunoreactivity for smooth muscle actin was developed by the VectorRed alkaline phosphatase staining kit (Vector Laboratories, Burlingame, CA). Specificity of ERß immunohistochemistry in the mouse uterus was checked using anti-human-ERß sheep IgG preabsorbed with immobilized peptide corresponding to the hinge (D) domain of human ERß. To test the specificity, ovary and uterus of wild-type and ERKO mice were stained with ERß 503 IgY antibody preabsorbed with human recombinant ERß (PanVera Corporation, Madison, WI).
Image Analysis
To determine the percentage of PR-positive endometrial stromal cells, images of PR immunohistochemistry were captured with a DC330 camera (Dage-MTI, Michigan City, IN) interfaced with a PowerBase 200 computer (Power Computing, Round Rock, TX) and analyzed with Scion Image 1.62a (Scion, Frederick, MD) software. All images were captured in CMYK (cyan, magenta, yellow, and black) mode under the exact same lighting conditions. In the combination of cyan and magenta channels (C+M channel), nuclei of PR-positive cells (stained brown from DAB) and negative cells (stained blue from hematoxylin) were equally detected. In the combination of yellow and black channels (Y+K channel), only nuclei of PR-positive cells were detected. Nuclei of all endometrial stromal cells were manually selected by adjusting the threshold for optical density (OD) in the C+M channel. Total selected area (i.e., total nuclear area) and area with an OD higher than the average OD of the negative control in the Y+K channel (i.e., PR-positive nuclear area) were measured. The percentage of PR-positive endometrial stromal cells was calculated as (PR-positive nuclear area)/(total nuclear area) x 100. In each group, the nuclear area equivalent to approximately 3000 nuclei in 6 to 10 pieces of uteri from three to five ERKO mice was analyzed. To determine the average intensity of PR staining, OD in the Y+K channel was measured in the total nuclear area of untraumatized uterine horns or in the manually selected nuclear area for decidual cells. The average OD of nonimmune IgG control was used as the baseline (OD = 0). In each group, an area equivalent to at least approximately 1000 nuclei in four to six pieces of uteri from two to four ERKO mice was analyzed. Statistical analysis was performed using the statistical analysis package StatView (Abacus Concepts, Berkeley, CA). To compare OD between hormone treatments, factorial ANOVA was used for overall analysis, and Fisher's protected least-significant-difference (PLSD) test was used for follow-up analysis to determine the difference between groups (P < 0.05).
Organ Culture
Mature ERKO mice (7 to 8 wk old) were ovariectomized, and 2 to 3 wk after ovariectomy, uteri were dissected and cut into small pieces (10–20 pieces/uterine horn) in a 1:1 mixture of Dulbecco's modified Eagle's medium and Ham's F-12 medium (DME/F-12 medium; Gibco, Gaithersburg, NY). Uterine tissue pieces were embedded in an 1:4 mixture of growth factor-reduced Matrigel (Becton Dickinson Labware, Bedford, MA) and rat-tail collagen gel and then cultured in DME/F-12 medium containing 10% fetal calf serum overnight. The medium was then changed to DME/F-12 medium with transferrin (5 µg/ml; Sigma) and insulin (10 µg/ml; Sigma). After 2 days of cultivation, the medium was changed to DME/F-12 containing 10-6 M ICI and/or 10-8 M E2 or diethylstilbestrol (DES; Sigma). After four additional days of cultivation, uterine cultures were fixed as described above. To quantitate PR level in endometrial stromal cells, the morphometric analysis was performed on slides immunostained for PR as described above. A one-tailed statistical test was used to verify if estrogens (E2 and DES) could increase PR-positive cells in endometrial stroma in organ culture. Use of a one-tailed statistical test is appropriate when a directional, a priori hypothesis is specified [50].
RESULTS
Expression of PR Induced by E2 in Uterine Stroma of ERKO Mice In Vivo
Expression of PR was assessed by immunohistochemistry in the uterus of ERKO mice. In uteri of these mice, PR staining was strong in the luminal epithelium and the subepithelial endometrial stroma (Fig. 1a) and was weak or undetectable in the peripheral endometrial stroma and myometrium (Fig. 1b). In ERKO mice treated for 3 days with oil 14 days after ovariectomy (OVX), the PR level was dramatically reduced in the endometrial stroma (Fig. 1c), but epithelial PR staining remained intense. Whereas PR expression in uterine epithelium did not change with E2 treatment in ERKO mice, three injections of 125 ng of E2 induced strong PR expression in the endometrial stroma (Fig. 1, c and e). Similarly, E2 induced PR in the myometrium of ERKO mice (Fig. 1, d and f). Up-regulation of PR by E2 in uterine endometrial stroma and myometrium was completely inhibited by coadministration of ICI (Fig. 1, g and h).
|
The PR mRNA level was analyzed in isolated uterine epithelium and stroma (i.e., endometrial stroma plus myometrium) to confirm results in the immunohistochemical study. A representative Northern blot is demonstrated in Figure 2. Northern analysis showed induction of PR mRNA by E2 in stroma of ovariectomized ERKO mice (Fig. 2, UtS), whereas in uterine epithelium, the PR mRNA level remained high in oil- and E2-treated, ovariectomized ERKO mice (Fig. 2, UtE). The PR mRNA level in uterine stroma was extremely low compared to that of the epithelium in both oil- and E2-treated groups. This suggests that the increase in the level of stromal PR mRNA induced by estrogen can be masked by constitutive expression of epithelial PR mRNA in an analysis of whole-uterine extract.
|
The PR immunohistochemistry was quantitated via image analysis by measuring the percentage of PR-positive endometrial stromal cells (Fig. 3). Differences in the percentage of PR-positive stromal cells between all four groups (intact, OVX+oil, OVX+E2, and OVX+E2+ICI) were all significant (P < 0.05). Values ranged from highest to lowest in the following order: OVX+E2 > intact > OVX+oil > OVX+E2+ICI (Fig. 3A). The percentage of PR-positive cells in endometrial stroma decreased significantly 2 wk after ovariectomy, and E2-treatment increased the PR-positive population (Fig. 3A). However, ICI inhibited the E2-induced up-regulation of PR in endometrial stroma. The percentage of PR-positive endometrial stromal cells was significantly lower in the ICI+E2-treated group than in the oil-treated group. This suggests residual estrogenic activity (perhaps of adrenal origin) in the uterus even 2 wk after ovariectomy.
|
To eliminate the possible effect of abnormal systemic hormone levels in ERKO mice, E2 effects on PR expression were studied in uteri of ERKO and wild-type mice grafted contralaterally under the renal capsules of female athymic nude mice. For this purpose, ERKO and wild-type uterine samples were grown for 1 mo in intact female athymic nude mice, which were then ovariectomized. Two weeks after ovariectomy, the hosts were treated with oil or E2 as described. Induction of uterine stromal PR by E2 in uterine grafts was identical to that observed in ERKO and wild-type animals treated directly (data not shown). The percentage of PR-positive cells in endometrial stroma of ERKO uterine grafts was significantly higher in E2-treated hosts than in oil-treated hosts (Fig. 3B).
Expression of PR Induced by E2 in Uterine Stroma of ERKO Mice In Vitro
To test whether E2 induces PR via direct action on uterus or via other systemic endocrine effects, uteri from ovariectomized ERKO mice were placed in organ culture. To assess PR expression in myometrium, explants were double-stained for PR (brown) and smooth muscle actin (red).
In organ culture without hormonal supplementation, PR was down-regulated in uterine epithelium, which had been strongly positive for PR in the absence of estrogen in vivo. The endometrial stroma, which had been very weakly positive in vivo (Fig. 1c), became positive for PR without estrogen treatment (Fig. 4a). Both E2and DES increased the PR-positive cell population in the endometrial stroma significantly (P < 0.05; Figs. 4c and 5). In the absence of estrogen, ERKO myometrium was mostly negative for PR (Fig. 4b). Both DES and E2 induced PR expression in myometrial cells (Fig. 4d). Thus, estrogens induce PR in endometrial stroma and myometrium of ERKO mice through direct action on uterine cells.
|
To assess whether estrogens induce PR in uterine stroma via ER, effects of ICI on PR expression were studied in ERKO uterine organ cultures. We found that ICI inhibited estrogen-induced PR expression in myometrium and endometrial stroma (Figs. 4, e and f, and 5). These results suggest that estrogen induces PR in endometrial stroma and myometrium of ERKO mice through ERß or another unidentified pathway whose function is blocked by ICI.
Localization of ERß in Mouse Uterus
Localization of ERß in wild-type and ERKO uterus was studied by immunohistochemistry. Mouse monoclonal anti-ER IgG 1D5 stained uterine tissues from wild-type (Fig. 6a) but not ERKO (Fig. 6b) mice. In contrast, anti-human-ERß sheep polyclonal antibody stained uterine tissues of both wild-type (Fig. 6c) and ERKO mice (Fig. 6d). In the uterus of both wild-type and ERKO mice, localization of ERß was very similar to that of ER in wild-type mice and, thus, was detected in the nuclei of endometrial stroma, myometrium, luminal, and glandular epithelia. In both oil- and E2-treated ERKO mice, ERß protein was detected in endometrial stroma and myometrium, in which PR was also detected in E2-treated ERKO mice (Figs. 1e and 6d). Ovaries were used as a positive control, because granulosa cells in the ovary express a high level of ERß [22–24, 27, 28]. In both wild-type (data not shown) and ERKO (Fig. 6h) mice, nuclei of granulosa, theca, and interstitial cells were stained with this anti-human-ERß sheep polyclonal antibody (Fig. 6h). In the negative controls with the peptide-absorbed primary antibody, ERß signal was completely absent in the ovary (data not shown) and uterus (Fig. 6g). Localization of ERß protein was confirmed with another anti-ERß chicken antibody, ERß 503 IgY [47]. In both wild-type (Fig. 6e) and ERKO (Fig. 6f) uteri, the immunostaining pattern obtained with ERß 503 IgY was essentially identical to that obtained with the anti-human-ERß sheep polyclonal antibody. Preabsorption with human recombinant ERß completely abolished the positive nuclear signal from uterus and ovary of both wild-type and ERKO mice (data not shown).
|
Expression of PR Inducted by E2 in Uterine Stroma of ßERKO Mice In Vivo
Because ERß appears to mediate E2 up-regulation of uterine stromal PR, E2 regulation of uterine stromal PR was compared in wild-type and ßERKO mice, which both express ER (Fig. 6, a and d). In the uterus of both wild-type and ßERKO mice, intensity of PR signal and number of PR-positive cells in the endometrial stroma and myometrium were low in ovariectomized oil-treated mice (Fig. 6, b and e) but high in E2-treated mice (Fig. 6, c and f). Thus, ERß is not essential for up-regulation of uterine stromal PR by E2. The PR level in uterine epithelium is high in the absence of E2 and down-regulated by E2 via stromal ER [32]. Thus, down-regulation of epithelial PR by E2 did not occur in ERKO uterus (Fig. 1). In contrast to ERKO mice [32], uterine epithelial PR was down-regulated by E2 in wild-type and ßERKO mice (Fig. 6, b, c, e, and f). Overall, the effect of ovariectomy and E2-treatment was identical in the uteri of ßERKO mice and wild-type mice. Thus, our observations confirm the idea that ER is the major mediator of estrogen action in the mouse female reproductive tract.
Decidualization in Wild-Type and ERKO Mice
Because PR expression in uterine stroma is induced by E2 in ERKO mice, we determined whether E2 modifies the action of P4 to induce decidualization via an ER-independent pathway. Mature female wild-type and ERKO mice were ovariectomized and, 2 wk later, received hormone treatment.
Uterine weight
In wild-type mice, wet weight of the untraumatized uterine horn was significantly higher in the E2+P4-treated group than in the oil- and P4-treated groups. In contrast, in ERKO mice, no significant difference was found in the wet weight of untraumatized uterine horn among all groups (oil, P4, E2+P4, and ICI+P4) (Fig. 7, ERKO). Thus, E2 increases the wet weight of untraumatized uterine horns via an ER-dependent pathway. In both wild-type and ERKO mice, P4 treatment plus traumatization increased the wet weight of the uterine horn significantly compared to oil treatment plus traumatization or to P4 treatment alone. In wild-type mice, the traumatized uterine horn was significantly heavier in the E2+P4-treated group than in P4- and oil-treated groups (Fig. 7, WT). In contrast, E2 did not synergize P4 action to increase the wet weight of traumatized uterine horn in ERKO mice (Fig. 7, ERKO). In ERKO mice, no significant difference was found in the wet weights of traumatized uterine horns between the ICI+P4-treated group and the P4- or E2+P4-treated groups (Fig. 7, ERKO). Because pretreatment plus coadministration of ICI with P4 did not inhibit the increase in wet weight of traumatized uterine horns, actions of residual estrogen in the body are not involved in the increased wet weight of traumatized uterine horns in ovariectomized ERKO mice. In comparing wild-type and ERKO mice, the weight of traumatized uterine horn in P4-treated wild-type mice was statistically identical to that of P4-, E2+P4-, and ICI+P4-treated ERKO mice. These data suggest that during decidualization, P4 and traumatization are essential to increase the wet weight of uterine horns, but that E2 signal via ER also has a synergistic effect on the wet weight of uterine horns. In conclusion, ER-independent estrogen action in ERKO uterus does not modify the wet weight of uterine horns, even though estrogen induces PR in endometrial stroma of ERKO mice.
|
Decidual cell differentiation and PR expression
In both ERKO and wild-type mice, decidual cells were never detected in untraumatized uterine horns of all groups (Fig. 8, a–e). Likewise, decidual cells were not observed in traumatized uterine horns of the oil-treated group (Fig. 8g). In contrast, decidual cells were observed in all other groups receiving P4-treatment and traumatization (Fig. 8). Thus, both P4-treatment and traumatization were essential and sufficient for differentiation of uterine stroma to decidual cells in both wild-type and ERKO mice (Fig. 8).
|
The histology of artificially induced decidual cells in ERKO mice (Fig. 8, h–j) was comparable to that of wild-type mice (Fig. 8f). In the decidual cells, PR was strongly expressed irrespective of the hormone treatment (P4, E2+P4, or ICI+P4) (Fig. 8, h–j). In ERKO mice, ICI at a dosage of 50 µg/mouse per day did not inhibit decidualization or induction of high levels of PR expression in decidual cells (Fig. 8, g and h). These data suggest that the high level of PR resulted from decidual differentiation requiring P4 treatment and traumatization but not E2 and ERs. Although decidual differentiation was never detected, the P4-treatment (Fig. 8, c–e) or traumatization (Fig. 8g) alone slightly increased PR in endometrial stroma (Fig. 9). However, the level of PR induced by P4 without traumatization or by traumatization alone was minimal and, thus, not comparable to that induced by E2+P4 without traumatization or decidualization (P4+traumatization) (Fig. 9). The PR level in endometrial stroma appeared to increase slightly in untraumatized uterine horns of oil-treated mice (Fig. 8b) compared to those in oil-treated ovariectomized mice (Fig. 1a). The difference in treatments of these two groups is the presence or absence of mechanical stimulation to the other uterine horn. This suggests that mechanical stimulation increases PR in endometrial stroma, at least partially, via a systemic effect.
|
DISCUSSION
In this study, we have shown that estrogen can induce PR expression in endometrial stroma and myometrium through ER-independent pathways. Recently, Curtis et al. [38] reported that PR is not up-regulated by E2 in ERKO uterus, even though the R5020-binding level in whole-uterine homogenates decreased after ovariectomy. In their study, PR levels were assessed in whole uteri. The uterus is composed of three major tissue types (i.e., epithelium, endometrial stroma, and myometrium), and the total PR level in whole uterus is the average of the PR levels in all tissue types. Thus, measurement of mRNA or proteins in whole uterus is affected by the individual PR level in each cell type and the relative abundance of each cell type in the uterus. In this regard, McCormick and Glasser [51] have demonstrated that in whole-uterine homogenates, changes in estrogen-binding activity in uterine epithelium is masked by changes in myometrium because the myometrium is the most abundant tissue, even though epithelium contains the highest concentration of estrogen-binding activity on a per-cell basis among the three tissue components of the uterus. Because PR is differentially regulated in the three tissue compartments by E2, either directly via ER in the responding cell type or indirectly via paracrine mechanisms [32], changes in protein, mRNA, or hormone-binding levels must be studied in each isolated tissue type to gain biologically meaningful data. Analysis of whole-uterine homogenates can be misleading in regard to gene regulation at the tissue and cellular levels.
A null mutation in a key element of hormone action, such as ER and PR, often alters the entire systemic endocrine profile [52, 53]. For this reason, hormonal levels in intact ERKO and wild-type mice differ substantially, and thus, ERKO and wild-type uteri of intact females are exposed to vastly different hormone levels. Grafting two different groups of hormone target organs to contralateral renal capsules of athymic hosts eliminates such hormonal differences so that mutant and wild-type organs can be studied under identical hormonal conditions. In this study, regulation of PR by E2 in ERKO uterine grafts was identical as in the experiments using animals directly.
When a hormonal effect is observed in an organ, the effect may be mediated directly through receptors in the organ or indirectly through receptors in other target organs that may affect systemic hormones. Organ culture was utilized to assess whether estrogen action on the uterine stroma of ERKO mice is direct or indirect. Although in organ culture PR expression was not exactly identical to that in vivo, up-regulation of stromal/myometrial PR in uterine explants grown in vitro confirmed that the signaling pathways within the uterus of ERKO mice can mediate the effect of estrogen to up-regulate PR in the uterine stroma/myometrium. The reason why uterine cells express PR differentially in vivo and in vitro is not clear.
The discrepancy between PR expression and regulation in vivo versus in vitro, though not fully understood, is a potentially serious problem for the field. As demonstrated recently by several groups, administration of E2 to ovariectomized mice dramatically down-regulates PR in mouse uterine epithelium, and the effect is mediated via stromal ER [32]. In contrast, when mouse uterine epithelial cells are cultured by themselves in vitro, E2 up-regulates PR [54]. Likewise, the PR profile in the mouse uterine organ cultures is slightly different from that in vivo as reported herein. Clearly, it is the in vivo observations to which all other artificial conditions must be compared. A potential explanation that may reconcile our in vitro and in vivo observations may involve the insulin pathway. Insulin-like growth factor-I can up-regulate PR in endometrial stromal cells in vitro [55]. The medium used for the organ culture contained a relatively high level of insulin (10 µg/ml), which may activate expression of PR in endometrial stroma. Also, extracellular matrix contains undefined substances, and a molecule in the extracellular matrix may induce PR in endometrial stroma. Uterine epithelial PR is regulated by a paracrine factor produced by uterine stroma [32, 42]. Insulin or molecules in extracellular matrix may modify epithelial PR expression directly on epithelium or indirectly via activation of stromal paracrine signals. This study has shown that mechanical stimulation can alter expression patterns of PR in endometrial stroma. For the organ-culture experiment, uterine horns were cut into small pieces by a sharp blade, and the mechanical force used during the preparation of uterine tissue may modify expression of PR in uterine organ culture.
The 5'-region of the PR gene contains clusters of ERE half-sites that are essential for transactivation of the PR gene by ligand-occupied ER [11–13]. However, the mouse PR gene is not a typical ERE-regulated gene, and its 5'-region contains many other putative binding sites for other regulatory transcription factors [56]. To our knowledge, it has not been reported if E2 can activate PR gene transcription via ERß, or if ICI can block the ERß-mediated transcription of the PR gene. Whether ERß can mediate up-regulation of PR in endometrial stroma is not clear, but ERß protein was detected in exactly the same ERKO endometrial stromal cells induced by E2 to up-regulate their PR. Thus, our findings suggest that up-regulation of PR by E2 is mediated by ERß in the ERKO uterus. However, even if E2 can induce PR in endometrial stroma of ERKO mice via ERß, PR induction in endometrial stroma is not regulated solely by ERß, because PR was also up-regulated by E2 in endometrial stroma of ßERKO mice, which express ER. These data suggest that E2 can up-regulate uterine stromal PR via both ER and ERß, and that the both ER and ERß pathways can work independently.
Recently, Weihua et al. [57] reported that down-regulation of uterine epithelial PR induced by E2 requires ERß in immature mice. Our data show that ERß is not required for E2-induced down-regulation of uterine epithelial PR in ovariectomized adult female mice. In uteri of immature mice, Weihua et al. demonstrated that levels of ER and ERß mRNA were equivalent. In contrast, in adult mice, the level of uterine ER mRNA vastly exceeds the level of ERß mRNA [20]. Given the substantial difference in ERß and ER mRNA levels and the difference in the mode of PR regulation in these two studies, the role of ERß in the uterus may differ in immature versus adult mice. Another major difference between these two studies is the involvement of the ovary. In the study of Weihua et al., intact immature mice were treated with E2, whereas in our study, adult mice were ovariectomized 2 wk before hormone treatment to minimize or eliminate the effects of endogenous estrogen and progesterone. The ßERKO mice have defects in ovarian differentiation [27]; thus, the profile of systemic hormone levels in ßERKO mice likely differs from that of ovariectomized adult wild-type mice. In this regard, it is worth noting that progesterone inhibits estrogen-induced down-regulation of uterine epithelial PR [42]. Thus, levels of P4 in the immature ßERKO mice could affect the level of PR. In the present study, ovariectomy controlled levels of ovarian sex hormones, whereas in the immature mouse study, these hormones were not controlled. For years, the immature uterine model has been used interchangeably with the ovariectomized adult uterine model. This discussion shows, however, that these models may be significantly different.
Estrogen can induce PR expression in the forebrain of ERKO mice [58]. In both forebrain and uterine stroma of ERKO mice, PR is induced by E2 through an ER-independent mechanism. On the other hand, PR expression in decidual cells is independent of both estrogens and ERs. Because a high level of PR expression always coincided with the decidual cell phenotype, PR expression appeared to be determined by the differentiation status (i.e., decidualization) of endometrial stroma. Uterine and vaginal epithelia are other examples in which PR expression patterns are determined through differentiation. Both the uterus and the upper part of the vagina develop from paramesonephric duct (i.e., Müllerian duct). In rat and mouse uterine epithelial cells, estrogen and ER are not required for a high level of PR expression [32]. In contrast, in vaginal epithelial cells, a high level of PR is expressed only in the presence of ER and estrogen [32]. In rodent uterine epithelial cells, in which PR expression is independent of estrogen and ER, transcription factors regulating the PR gene likely are totally different from transcription factors in rodent vaginal epithelial cells, in which PR expression is dependent on estrogen and ER. Likewise, the profile of transcription factors regulating PR expression in endometrial stromal cells must be vastly different before and after the decidualization.
Our finding that E2 and ER action are not essential for endometrial stromal cells to differentiate into decidual cells agrees with those of the classic studies in which decidualization was induced artificially by P4 and traumatization without E2 [35]. This concept has been confirmed in a study on ERKO mice by Paria et al. [37]. However, differentiation of decidual cells is only part of the implantation reaction. Requirement for ER actions must be assessed in terms of the various aspects of decidualization. Recently, Curtis et al. [38] concluded that E2 is essential for decidualization in wild-type, but not in ERKO, mice based on increases in uterine weight. An increase in uterine weight is only one aspect of decidualization, however, and is not necessarily linked directly to decidual differentiation of endometrial stromal cells. In both ERKO and wild-type mice, decidual differentiation can be induced by P4 and mechanical stimulation (E2 is not required) [37]. The weight of the uterus is mainly determined by the water content, and the water content results from many complex physiological reactions (e.g., congestion, inflammation, etc.) involving many different cell types (e.g., endometrial cells, endothelial cells, smooth muscle cells, and lymphocytes) and factors (e.g., steroids, lymphokines, and growth factors). Because so many factors can be involved in regulation of uterine weight, how ER-mediated E2 action synergizes with P4 and traumatization to increase uterine weight is not clear.
Although the level was low, P4 or mechanical stimulation alone could induce a detectable level of PR in endometrial stroma. This may explain why endometrial cells, in which PR levels are extremely low in the absence of E2, can respond to P4 without E2 priming.
Action of estrogenic compounds on lactoferrin expression independent of ER has been reported [59]. Chlordecone and 4-hydroxyestradiol-E2 up-regulate lactoferrin in ERKO uterus [59]. However, the mechanism of lactoferrin induction appears to differ from up-regulation of PR, because in ERKO mice, E2 has no effect on lactoferrin whereas PR is up-regulated in endometrial stroma by E2. Also, ICI does not block the induction of lactoferrin by chlordecone and 4-hydroxyestradiol-E2, whereas ICI inhibited E2-induced up-regulation of PR in uterine stroma of ERKO mice. Identification of the molecules that mediate estrogen action in the uterus of ERKO mice is unclear, and it may involve both direct estrogen action and paracrine circuitry.
|
FOOTNOTES
First decision: 22 March 2000.
1 Supported by National Institutes of Health grants AG-13784 and DK47517 to G.R.C., AG-15500 to P.S.C., R01-ES08272 to D.B.L., AG-16870 to R.D., and U.S. Army grant DAMD17-97-1-7171 to D.B.L. 
2 Correspondence: Gerald R. Cunha, P.O. Box 0452, Department of Anatomy, University of California-San Francisco, San Francisco, CA 94143. FAX: 415 502 2270; gcunha{at}itsa.ucsf.edu
Accepted: August 25, 2000.
Received: February 18, 2000.
REFERENCES
and ERß at AP1 sites. Science 1997; 277:1508–1510.. Mol Endocrinol 1997; 11:1486–1496. (ER) and estrogen receptor-ß (ERß) messenger ribonucleic acid in the wild-type and ER-knockout mouse. Endocrinology 1997; 138:4613–4621. and ß. Endocrinology 1997; 138:863–870. and ß in rat ovary and uterus. J Mol Endocrinol 1999; 22:37–44.[Abstract] and ß in the rat uterus. Biol Reprod 1999; 61:955–964.-deficient mice. Endocrinology 1999; 140:2704–2710. knockout mouse uterus. Proc Natl Acad Sci U S A 1999; 96:3646–3651. gene in female mice: characterization of ovarian responses and phenotype in the adult. Endocrinology 1999; 140:2733–2744. gene-disrupted mice. J Neurosci 1998; 18:9556–9563. deficient mice reveal a distinct estrogen signaling pathway. Proc Natl Acad Sci U S A 1997; 94:12786–12791.This article has been cited by other articles:
|
|
 |
|
  O. Bukulmez, D. B. Hardy, B. R. Carr, R. A. Word, and C. R. Mendelson Inflammatory Status Influences Aromatase and Steroid Receptor Expression in Endometriosis Endocrinology, March 1, 2008; 149(3): 1190 - 1204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. A. Neves, K. Stovall, J. Joyner, G. Valdes, P. E. Gallagher, C. M. Ferrario, D. C. Merrill, and K. B. Brosnihan ACE2 and ANG-(1-7) in the rat uterus during early and late gestation Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2008; 294(1): R151 - R161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A Wood, J. E Fata, K. L M Watson, and R. Khokha Circulating hormones and estrous stage predict cellular and stromal remodeling in murine uterus Reproduction, May 1, 2007; 133(5): 1035 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Jackson, A. Brudney, J. M. Hastings, P. A. Mavrogianis, J. J. Kim, and A. T. Fazleabas The Altered Distribution of the Steroid Hormone Receptors and the Chaperone Immunophilin FKBP52 in a Baboon Model of Endometriosis Is Associated With Progesterone Resistance During the Window of Uterine Receptivity Reproductive Sciences, February 1, 2007; 14(2): 137 - 150. [Abstract] [PDF]
|
|
|
|
 |
|
 O. Wada-Hiraike, H. Hiraike, H. Okinaga, O. Imamov, R. P. A. Barros, A. Morani, Y. Omoto, M. Warner, and J.-A. Gustafsson Role of estrogen receptor beta in uterine stroma and epithelium: Insights from estrogen receptor beta-/- mice PNAS, November 28, 2006; 103(48): 18350 - 18355. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. O'Brien, T. J. Peterson, M. H. Tong, E.-J. Lee, L. E. Pfaff, S. C. Hewitt, K. S. Korach, J. Weiss, and J. L. Jameson Estrogen-induced Proliferation of Uterine Epithelial Cells Is Independent of Estrogen Receptor {alpha} Binding to Classical Estrogen Response Elements J. Biol. Chem., September 8, 2006; 281(36): 26683 - 26692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Peluso Multiplicity of Progesterone's Actions and Receptors in the Mammalian Ovary Biol Reprod, July 1, 2006; 75(1): 2 - 8. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. S. Huggins, J. Y.Y. Wong, S. E. Hankinson, and I. De Vivo GATA5 Activation of the Progesterone Receptor Gene Promoter in Breast Cancer Cells Is Influenced by the +331G/A Polymorphism Cancer Res., February 1, 2006; 66(3): 1384 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M A J Herve, G Meduri, F G Petit, T S Domet, G Lazennec, S Mourah, and M Perrot-Applanat Regulation of the vascular endothelial growth factor (VEGF) receptor Flk-1/KDR by estradiol through VEGF in uterus J. Endocrinol., January 1, 2006; 188(1): 91 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Vallejo, C. Ballare, J. Lino Baranao, M. Beato, and P. Saragueta Progestin Activation of Nongenomic Pathways via Cross Talk of Progesterone Receptor with Estrogen Receptor {beta} Induces Proliferation of Endometrial Stromal Cells Mol. Endocrinol., December 1, 2005; 19(12): 3023 - 3037. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Shimizu, T. Maruyama, K. Tamaki, H. Uchida, H. Asada, and Y. Yoshimura Impairment of Decidualization in SRC-Deficient Mice Biol Reprod, December 1, 2005; 73(6): 1219 - 1227. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Durrer, K. Maerkel, M. Schlumpf, and W. Lichtensteiger Estrogen Target Gene Regulation and Coactivator Expression in Rat Uterus after Developmental Exposure to the Ultraviolet Filter 4-Methylbenzylidene Camphor Endocrinology, May 1, 2005; 146(5): 2130 - 2139. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Zenclussen, K. Gerlof, M. L. Zenclussen, A. Sollwedel, A. Z. Bertoja, T. Ritter, K. Kotsch, J. Leber, and H.-D. Volk Abnormal T-Cell Reactivity against Paternal Antigens in Spontaneous Abortion: Adoptive Transfer of Pregnancy-Induced CD4+CD25+ T Regulatory Cells Prevents Fetal Rejection in a Murine Abortion Model Am. J. Pathol., March 1, 2005; 166(3): 811 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Rider, E. Thomson, and C. Seifert Transit of Rat Uterine Stromal Cells through G1 Phase of the Cell Cycle Requires Temporal and Cell-Specific Hormone-Dependent Changes on Cell Cycle Regulators Endocrinology, December 1, 2003; 144(12): 5450 - 5458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Frasor, D. H. Barnett, J. M. Danes, R. Hess, A. F. Parlow, and B. S. Katzenellenbogen Response-Specific and Ligand Dose-Dependent Modulation of Estrogen Receptor (ER) {alpha} Activity by ER{beta} in the Uterus Endocrinology, July 1, 2003; 144(7): 3159 - 3166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Tamura, T. Hara, M. Yoshie, S. Irie, A. Sobel, and H. Kogo Enhanced Expression of Uterine Stathmin during the Process of Implantation and Decidualization in Rats Endocrinology, April 1, 2003; 144(4): 1464 - 1473. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Parmar, P. Young, J. T. Emerman, R. M. Neve, S. Dairkee, and G. R. Cunha A Novel Method for Growing Human Breast Epithelium in Vivo Using Mouse and Human Mammary Fibroblasts Endocrinology, December 1, 2002; 143(12): 4886 - 4896. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Curtis Hewitt, E. H. Goulding, E.M. Eddy, and K. S. Korach Studies Using the Estrogen Receptor {alpha} Knockout Uterus Demonstrate That Implantation but Not Decidualization-Associated Signaling Is Estrogen Dependent Biol Reprod, October 1, 2002; 67(4): 1268 - 1277. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
