| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Knockout Uterus Demonstrate That Implantation but Not Decidualization-Associated Signaling Is Estrogen Dependent
a Laboratory of Reproductive and Developmental Toxicology, NIEHS, NIH, Research Triangle Park, North Carolina 27709
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
knockout (ERKO) mice in the absence of the estrogen dependence normally seen in wild-type (WT) mice for this response. While the ERKO stromal compartment demonstrated the necessary decidual response, embryo implantation is a process initiated in the epithelial layer, a uterine component that lacks estrogen responsiveness in the ERKO. To determine if the ERKO uterus would be competent for implantation, donor embryos were transferred into the uterine lumen of WT and ERKO females that had been ovariectomized and treated with exogenous estradiol and progesterone to mimic early pregnancy. No implantation occurred in the ERKO, while implantation sites containing live embryos were seen in similarly treated WT uteri, indicating that functional estrogen receptor (ER) is required for implantation. Previous observations of estrogen-independent decidualization in the ERKO prompted investigation of the mechanism leading to estrogen independence of this process. The disruption of progesterone receptor (PR), Hoxa10, Cox2, or LIF in transgenic mice results in the loss of decidualization response. Therefore, the expression of these genes was studied in WT and ERKO uteri by comparing expression following vehicle, progesterone alone (P), or estradiol priming followed by progesterone with nidatory estradiol (E+Pe) and by comparing expression following the above hormonal manipulations in addition to luminal infusion of oil used previously as decidualization-initiating stimulus. The whole-uterus level of PR and Hoxa10 mRNAs did not vary; however, the PR protein was induced in the stroma 24 h after oil infusion. Interestingly, in the WT, this induction was most apparent in samples receiving E+Pe, while in the ERKO samples, the induction occurred independent of any hormone priming. Cox2 protein and mRNA increased in both WT and ERKO samples 2 h after oil infusion in all three of the treatment groups. In the WT samples, Cox2 levels remained elevated 24 h after oil infusion only in the E+Pe treatment group; however, the elevated Cox2 was seen in samples taken 24 h after oil infusion in all three ERKO treatment groups. The ERKO uterine tissue appeared to sustain more extensive damage when examined 24 h after oil infusion. Severe trauma, such as crushing of the uterine tissue, has previously been shown to remove the requirement for nidatory estradiol for deciduomas to develop, indicating that the greater susceptibility of ERKO uterine tissue to damage from intraluminal oil infusion is contributing to decidualization in the absence of ER. Leukemia inhibitory factor (LIF) mRNA was also induced following estradiol treatment in the WT, but also following oil infusion in WT samples that were not treated with estradiol. In contrast, estradiol does not induce LIF mRNA in the ERKO, but oil infusion leads to a robust increase in LIF in all ERKO sample groups. LIF binds and activates its membrane receptor, which initiates responses including the phosphorylation and nuclear translocation of Stat3 transcription factor. Thus, Stat3 phosphorylation was studied in WT and ERKO samples and found to be induced following oil infusion in all samples. Together, these and previous observations illustrate that estrogen is essential for epithelial proliferation and embryo implantation and that estrogen is dispensable for stromal decidualization in the ERKO, as the essential genes and signals required for the response are still induced.
cytokines, decidua, estradiol receptor, female reproductive tract, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
knockout (ERKO) mouse, which lacks functional estrogen receptor (ER), is infertile [2]. While ovarian dysfunction is a major component of the ERKO female infertility [3, 4], estrogen insensitivity of the uterus is also a contributing factor [5, 6]. Previous characterization of the ERKO uterine function showed that ERKO stroma could be experimentally induced to decidualize using exogenous progesterone treatment and infusion of oil into the uterine lumen as a stimulus to mimic embryo apposition and initiate decidualization [7, 8]. Those studies indicated that development of deciduoma is independent of ER in this model [7, 8]. This was a striking observation considering studies indicating the requirement for estrogen priming for the initiation and development of a decidual response in the mouse [9–11]. However, the requirement for estrogen priming can be relieved when traumatic stimulus, such as crushing of the tissue, is administered [12].
In contrast with the decidualization competence of the stroma, the luminal epithelium of the ERKO is refractory to estradiol stimulation, lacking both a proliferative response and the ability to induce synthesis of estrogen-regulated genes [5]. Because epithelial estrogen responsiveness is thought to be important for successful embryo implantation, this study examined whether the ERKO uterus was competent for donor embryo implantation.
Several transgenic models with gene disruptions exhibit a phenotype of loss of decidual response, indicating that the expression and activity of those genes are essential to the process. These include the progesterone receptor (PR) [13], Hoxa10 [14], a homeobox protein expressed in developing reproductive tract, the enzyme prostaglandin synthase 2 [15] (also called cyclooxygenase 2 or Cox2), leukemia inhibitory factor (LIF) [16, 17], and interleukin 11 receptor (IL-11 R) [18], which are a cytokine and a cytokine receptor, respectively. To determine whether altered expression of these genes might lead to the observed relief of ER dependence for decidualization, samples obtained from wild-type (WT) and ERKO mice at the time of the initiation of decidualization were compared.
Cytokine signaling is an important component in implantation and initiation of decidualization, as is illustrated by the loss of these functions in the LIF-disrupted and IL-11 receptor-disrupted mouse models [16–18]. One mode of cytokine signaling activity through membrane receptors involves activation of Jak kinases, resulting in phosphorylation of Stat proteins, which then dimerize and translocate to the nucleus where they act as transcription factors [19, 20]. Stat3 is similarly activated in mouse uterine luminal epithelial cells when exposed to LIF [21]. Therefore, Stat3 activation, as a reflection of cytokine signaling, was also studied following intraluminal oil infusion in WT and ERKO uteri.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All studies were carried out according to an approved protocol as dictated in NIH guidelines and the NIEHS Animal Use and Care committee. For most studies WT, and ERKO females were obtained from our breeding colony at Taconic Farms (Germantown, NY). Adult (3- to 5-mo-old) females were ovariectomized and treated to mimic early pregnancy, as described previously [7]. After housing ovariectomized animals for 2 wk to clear endogenous ovarian hormones, the following hormones were administered to three experimental groups. 1) The E+Pe group was injected s.c. daily for 3 days with 100 ng estradiol (Steraloids, Newport, RI) in 100 µl sesame oil (Sigma, St. Louis, MO). On Days 4 and 5, no hormones were administered. Beginning the morning of Day 6, mice were injected daily s.c. with 1 mg of progesterone (Sigma) and 6.7 ng estradiol in 100 µl sesame oil. 2) The vehicle group was injected s.c. daily with 100 µl sesame oil on Days 1–3 and Day 6 until the day of tissue collection. 3) The P group was injected s.c. daily with 100 µl sesame oil on Days 1–3. Beginning the morning of Day 6, mice were injected daily s.c. with 1 mg progesterone in 100 µl sesame oil. Six hours following the injection on Day 8, mice from all three hormone treatment groups were lightly anesthetized and sesame oil was forced into the uterine lumen through the vaginal opening as previously described [7]. Uteri were collected 2 or 24 h after oil infusion or on Day 14 for fully developed deciduomas. Uteri from mice treated in parallel with the three experimental groups with the same hormones that were not infused with oil on Day 8 were collected at the same time as the oil-infused groups. One-half horn was fixed in 10% formalin and processed for immunohistochemistry. The remaining uterine tissue was frozen in liquid nitrogen for RNA or protein isolation.
For LIF induction, ovariectomized WT mice were injected with 100 ng estradiol or ethanol vehicle in 100 µl normal saline i.p., and uteri were collected 2 h later and frozen in liquid nitrogen for later RNA preparation.
For the implantation studies, WT and ERKO females were ovariectomized and treated with E+Pe as described for group 1 above. Six hours following the hormone injection on Day 8, donor embryos (12–20/per uterine horn) were transferred into the uterine lumen as described in [22]. In some animals, only the left horn received transferred embryos. Daily Pe injections were continued, and uteri were collected on Day 15 and inspected visually for implantation sites. Nodules (apparent implantation sites) were dissected and analyzed for the presence of embryonic tissue under a dissecting scope.
RNase Protection Assays
The cDNA for mouse leukemia inhibitory factor (mLIF) was a gift from Colin Stewart (NCI, Frederick, MD). Antisense mLIF riboprobe was prepared by linearizing the pGEM4-mLIF plasmid with BamHI and transcription with T7 RNA polymerase, producing a 600-nt transcript. The cDNA for mouse Hoxa10 was a gift from Richard Maas (Harvard University Medical School, Boston, MA). The Hoxa10(p1B1)/pGEM-3Z plasmid containing the mouse Hoxa10 cDNA was linearized using XhoI, and antisense mHoxa10 RNA was synthesized using SP6 RNA polymerase, resulting in a 320-nt riboprobe. The riboprobes for mouse progesterone receptor and mouse Cox2 were produced as previously described [4]. The mouse cyclophilin cDNA was purchased from Ambion (Austin, TX) and used to generate antisense riboprobe to normalize sample loading.
Frozen uteri were pulverized into powder and RNA was isolated using Trizol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. Riboprobes were labeled using the Ambion Maxiscript kit (Ambion) and 32PCTP (Amersham, Piscataway, NY). RNase protection assays were carried out using the RPAIII kit and protocols (Ambion). Protected RNA fragments were analyzed following separation on 6% TBE-urea gels (Invitrogen). Bands were detected and quantified using a Storm 850 phosporimager and Image Quant software (Amersham).
Statistical Analysis
Data sets were first tested for homoscedasticity of variance using the Levene test. In cases where data were found to lack homoscedasticity of variance (P < 0.05), values were log transformed prior to further statistical analysis. In all cases, data sets were analyzed by one-way ANOVA (http://www.physics.csbsju.edu/stats/anova.html) followed by individual post hoc comparisons (GraphPad.com). Statistical significance was assigned at P < 0.05.
Immunohistochemistry
Uterine cross sections were embedded in paraffin and mounted on Superfrost/Plus slides (Daigger and Co., Wheeling, IL). Slides were deparaffinized in xylene, processed through 100, 95, and 70% ethanol, and decloaked in citrate buffer (Biocare Medical, Walnut Creek, CA) for 3 min in a Biocare Medical Decloaking Chamber. Endogenous peroxidase was inactivated with 3% H2O2, then slides were rinsed in Automation Buffer (Biocare). Samples were washed with Tris-buffered saline between all steps. Samples were first blocked for 1 h with MOM blocking reagent (Vector Labs, Burlingame, CA), then incubated with anti-PR antibody (Immunotech, Marseilles, France) diluted 1:100 in diluent reagent (Vector) for 30 min. Negative control slides were incubated with diluent. Slides were then incubated with biotinylated anti-mouse IgG (Vector) diluted 1:250 in diluent reagent for 10 min, followed by ABC reagent (Vector) for 5 min, and rinsed in water. The peroxidase activity was visualized with Nova RED (Vector). Slides were counterstained briefly in hematoxylin (Sigma).
Western Blot Analysis
Proteins were isolated from frozen uterine tissue by homogenization with a Polytron homogenizer (Brinkmann, Westbury, NY) in homogenization/immunoprecipitation buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% Triton X-100, 0.25% sodium deoxycholate) supplemented with protease and phosphatase inhibitors (1 mM Na3VO4, 1 mM NaF, and 0.05 mM Na2MoO4, 20 µg/ml aprotinin, 20 µg/ml leupeptin, 4 µg/ml A-PMSF). Protein concentration was assayed using a BCA assay (Pierce, Rockford, IL) and 20 µg of protein was boiled in SDS-sample buffer (Invitrogen) with added DTT and loaded on an 8% Tris-glycine gel for Stat 3 analysis or a 10% Tris-glycine gel for Cox2 analysis (Invitrogen). Proteins were separated and transferred to nitrocellulose (Invitrogen) using XcelII apparatus and the manufacturer's protocol (Invitrogen). Efficient transfer was verified using Ponceu Red stain (Sigma). Membranes were blocked for 1 h with 10 ml blocking buffer (5% nonfat dry milk; Kroger, Cincinnati, OH) in TBS-T (10 mM Tris, 137 mM NaCl, pH 7.6, with 0.1% Tween-20; Sigma). Anti-Cox2 was obtained from Cayman Chemicals (Ann Arbor, MI) and was diluted 1:1000 in 10 ml blocking buffer. Anti-Stat3 and anti-phospho Stat3 (pTry705) antibodies were purchased from Cell Signaling (Beverly, MA) and were diluted 1:1000 in blocking buffer. Phospho Stat3 blots were stripped according to the protocol provided with the ECL reagent (Amersham) and reprobed with the Stat 3 antibody. All blots were processed as follows: Antibodies were incubated with the blots overnight at 4°C. The membrane was washed in TBS-T (three times for 10 min), then incubated with horseradish peroxidase-linked anti-rabbit Ig (Cell Signaling) diluted 1:2000 in blocking buffer for 1 h at room temperature. The blot was washed as before, and bands were visualized using ECL reagents according to the manufacturer's protocol (Amersham).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Earlier studies demonstrated that the ERKO uterus could undergo decidualization, indicating that this pregnancy-related uterine function does not require functional ER [7]. To determine whether embryos would be able to implant in the ERKO uterus, WT and ERKO females were ovariectomized and treated with exogenous estradiol and progesterone to mimic early pregnancy. This was necessary to achieve a pregnancy-like hormonal environment in which to examine embryo implantation because the ERKO is anovulatory and, additionally, LH, testosterone, and estradiol levels are chronically elevated due to dysregulation in the hypothalamic-pituitary-gonadal axis [6]. Embryos collected from donor mice were transferred into the uterine lumen of ERKO and similarly treated wild-type (WT) mice. Nodules that appeared to be implantation sites were apparent in WT uteri (Fig. 1), whereas significantly fewer sites were seen in ERKO uteri (Table 1). Nodules were formed in several ERKO uteri: 26 ERKO horns were analyzed, with 7 containing 1–3 nodules (average of 2), whereas 16 of the 21 WT horns inspected had nodules (1–10 per horn, average of 4). Additionally, WT nodules showed evidence of embryonic tissue, while the ERKO nodules consisted of decidualized uterine tissue with no evidence of embryonic tissue (Table 1). Therefore, the ERKO uterine tissue is able to decidualize in response to embryo transfer but with low efficiency compared with WT. Additionally, the lack of transferred embryo implantation in the ERKOs and successful implantations in the WT mice indicate that functional ER is required for implantation to occur.
|
|
Expression of Decidualization-Associated Transcripts
Previous observations of estrogen-independent decidualization in the ERKO prompted investigation of the mechanism leading to the relief of estrogen dependence of this process. The disruption of PR, Hoxa10, Cox2, or LIF in transgenic mice results in loss of the decidualization response [13–18]. Therefore, the expression of these genes was studied in WT and ERKO uteri as affected by the following parameters: 1) Comparing expression following vehicle, progesterone alone (P), or estradiol priming followed by progesterone with nidatory estradiol (E+Pe) and 2) comparing expression following the above hormonal manipulations in addition to luminal injection of oil as used previously as decidualization-initiating stimulus. As described previously [7] and in the Materials and Methods section, 6 h after s.c. injection of vehicle, P, or E+Pe on Day 8, sesame oil is infused into the uterine lumen through the vaginal opening. This oil infusion, with proper E+Pe priming (WT) or P priming (ERKO), leads to eventual development of deciduoma [7]. For these studies, however, samples were analyzed 2 or 24 h after the oil infusion to allow analysis of early signaling events that initiate the processes required for development of deciduomas.
When PR (Fig. 2A) and Hoxa10 (Fig. 2B) expression were measured by RNase protection assay (RPA), they were both readily detected in WT and ERKO uteri following E+Pe treatment. In addition, expression was not altered when measured 24 h after intraluminal oil infusion, indicating this stimulus does not cause changes in expression of PR and Hoxa10. Samples from mice treated with vehicle or P alone also had comparable expression of PR and Hoxa10. Samples isolated from decidualized tissues of WT or ERKO mice showed elevated expression of PR and Hoxa10. Overall, it appears that the regulation of the expression of these genes was not altered in the ERKO.
|
Progesterone Receptor Immunohistochemistry
Although no variation in the expression of PR mRNA in whole uterine samples was apparent (Fig. 2), the protein expression level or localization might vary. Therefore, PR expression was studied using immunohistochemical staining of uterine cross-sections from WT and ERKO mice treated as described above. In the WT samples, P or E+Pe treatment shifted PR localization from the epithelial compartment to the stromal compartment (Fig. 3, A–C). In addition, 24 h after intraluminal oil infusion, stromal cells in E+Pe-treated and, to a lesser extent, in P-treated samples showed an accumulation of fibrillar to vacuolated material and cytoplasmic enlargement consistent with early decidual changes (Fig. 3, E and F). Nuclear PR staining was more intense in the WT stroma after E+Pe treatment and oil infusion in comparison with the P treatment after oil. This indicates that, although the overall PR mRNA expression detected in the RPA (Fig. 2A) was not altered by estradiol (E) priming or oil infusion, the types of cells in which PR was expressed was affected. Although the stromal cells in the P-treated sample following oil infusion showed early changes of decidualization, E priming was required for the full development of deciduomas [7].
|
In the ERKO samples, P or E+Pe treatment also shifted PR localization from the epithelial to the stromal compartment. Oil infusion in the ERKO resulted in increased intensity of nuclear PR staining (Fig. 3, J–L) even in vehicle-treated samples. Although PR mRNA expression was comparable in most samples (Fig. 2), the localization of PR protein (Fig. 3) showed increased expression in the stroma following intraluminal oil infusion that is hormone-independent in the ERKO but estrogen-dependent in the WT. In addition, the oil infusion appeared to result in damage to the ERKO tissue, with disrupted stromal cell contacts as well as damage to the luminal epithelial cell layer (Fig. 3).
Induction of Cox2 by Oil Infusion
Cox2 mRNA expression has been previously shown to increase in response to oil infusion of the uterine lumen with two peaks of induction [15, 23]. An early robust increase occurs within 2 h of oil infusion, while a second peak of lower magnitude occurs 24 h after oil infusion. To determine whether Cox2 expression was altered in the ERKO, mRNA and protein levels were studied following vehicle, P alone, or E+Pe treatment as well as without oil infusion or 2 or 24 h after oil infusion and compared in WT and ERKO samples. RPA analysis of RNA samples collected from P only or E+Pe-treated mice shows that Cox2 expression could not be detected prior to oil infusion (Fig. 4A). Oil infusion lead to robust induction of Cox2 mRNA in WT and ERKO samples after 2 h, independent of E priming. Western blot analysis of WT and ERKO samples showed E independent induction of Cox2 protein 2 h after oil infusion as well (Fig. 4B). A second, less robust induction occurred 24 h after oil infusion. The Cox2 mRNA level induced 24 h after oil infusion was not significantly elevated over the control (Fig. 4A), although a protected band was visible. The lack of significant elevation is due to the difficulty quantifying low-level RNA fragments. Lack of E priming, however, seemed to diminish this induction 24 h after oil infusion at the mRNA level in the WT sample (Fig. 4A). Interestingly, E priming of the ERKO was not required for the Cox2 increase 24 h after oil infusion (Fig. 4A). Further analysis, including Western blot, confirmed the requirement for E priming for Cox2 mRNA or protein induction 24 h after oil infusion in the WT (Fig. 5). Although bands were visible on the RPA analysis, they were again at a low level and, when quantified (not shown), did not reveal significant elevation above noninfused. Western blot analysis clearly confirmed the Cox2 mRNA induction at the protein level. In the ERKO, however, the Cox2 mRNA and protein induction 24 h after oil infusion was completely hormone independent, as vehicle-treated samples also showed induction (Fig. 5).
|
|
Induction of LIF mRNA
The cytokine LIF is induced in the mouse uterus around the time of implantation; in addition, LIF can be induced in response to estradiol treatment [24, 25]. LIF is required for decidualization and implantation, as illustrated by the lack of these responses in the LIF-deficient transgenic mouse [16, 17]. LIF expression was studied in WT and ERKO uterine samples in order to determine whether it was altered in the ERKO. LIF could be detected in 5-µg RNA samples from ERKO uteri; however, because the level of expression was lower in WT samples, 15 µg of RNA from these samples had to be used. LIF mRNA is increased within 1–2 h following treatment with E in the WT but not the ERKO uterus (Fig. 6, lanes 13–16). LIF is also induced 1–2 h after intraluminal oil infusion, and this induction occurs in WT samples that were not treated with E (Fig. 6, lanes 1–4). Interestingly, WT samples that were treated with E+Pe expressed LIF mRNA at a level comparable with that seen following oil infusion in the vehicle or P-only-treated WT samples (Fig. 6, lane 5). Oil infusion did not lead to further increase in LIF in the WT E+Pe group (Fig. 6, lane 6), indicating the E-stimulated expression of LIF was not further increased. In addition, this oil infusion-stimulated induction is more robust in the ERKO samples and is not dependent on hormone priming (Fig. 6, lanes 7–12).
|
Oil Infusion Results in Stat3 Phosphorylation
The oil infusion stimulus used for induction of deciduomas in these studies is reminiscent of a cardiomyocyte model in which stretching of the cells initiates membrane receptor mediated signaling, including cytokine signaling [26]. One target of cytokine receptor activation is Stat3, which is phosphorylated and translocated to the nucleus, where it acts as a transcription factor. Indeed, LIF signaling through its receptor results in Stat3 activation in the uterine luminal epithelium at the time of implantation [21]. Therefore, as an indicator of activation of signaling that might occur following oil infusion, Stat3 activation was studied in samples from animals treated with vehicle, P alone, or E+Pe. Initial studies in WT animals indicated Stat3 was phosphorylated beginning 1 h after oil infusion (not shown). When measured 2 h after oil infusion, phosphorylated Stat3 was increased in WT and ERKO samples independent of administration of hormones (Fig. 7), although some phospho-Stat3 was present in samples that did not have oil infusion. Reprobing the same blots with anti-Stat3 antibody, which recognizes both unactivated Stat3 and phospho-Stat3, indicated that the treatments did not alter the amount of Stat3 protein (Fig. 7, bottom panels); however, the Stat3 level is lower in all the ERKO samples. These results indicate that the Stat3 is present and becomes phosphorylated in both the WT and ERKO uterine tissues in response to oil infusion.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ERKO. First, the ability of the ERKO uterus to implant donor embryos was evaluated. Second, the expression of components that might account for the previously observed lack of estrogen dependence for development of deciduoma in the ERKO was studied.
For pregnancy to be successful, the uterus must be competent for embryo implantation, which requires the establishment of a functional implantation site and stromal decidualization. Earlier studies have indicated that decidualization can be induced in the absence of ER [7, 8]. However, this is primarily a progesterone-dependent stromal cell response. The initiation of implantation occurs at the luminal epithelium, a uterine component that is highly sensitive to estrogen stimulus. Interestingly, tissue recombination studies using separated stromal and epithelial uterine components from ERKO and WT mice have indicated that the epithelial proliferative response as well as the attenuation of PR expression of the uterine luminal epithelium following estradiol stimulation requires stromal but not epithelial ER [27, 28]. Such studies illustrate that the uterine epithelial and stromal cells are separate yet interacting components, indicating stromal competence might compensate for epithelial dysfunction in some processes. Implantation in the mouse is highly sensitive to estrogen, as illustrated by the delayed implantation model in which the implantation window can be delayed by withdrawal of estrogen while maintaining progesterone [29]. Subsequent estrogen replacement results in implantation in this model, which is initiated by embryo attachment to and invasion of the epithelial cell layer. Therefore, one might expect implantation, unlike decidualization, would be more dependent on estrogen responsiveness and epithelial competence. Indeed, no evidence of implanted embryos was found in the ERKO following donor embryo transfer. Because the uterine tissue was analyzed 7 days after embryo transfer, if early implantation events did occur or if embryos implanted but were then reabsorbed in the ERKO, they might not have been detected in our analysis. However, the few nodules formed in the ERKOs lacked embryonic tissue, whereas WT sites either contained live embryos developed to Day 10 or contained dead embryos that were arrested at Day 7 or later. Thus, tissue from implanted embryos that ceased developing 2–3 days prior to analysis was still detectable.
The second component of this study was determining whether the hormone and/or intraluminal oil infusion responsiveness of several genes involved in the decidual response was altered in a manner that might account for the relief of estrogen dependence. The whole-tissue mRNA levels of Hoxa10 and PR were not altered in WT samples compared with ERKO samples, nor did hormonal treatments vary the expression. Thus, although PR and Hoxa10 play important roles in decidualization, it appears that their expression is not regulated by the components manipulated for initiation of decidualization. In fact, they are expressed even under conditions that do not lead to deciduoma formation (ovariectomy and vehicle treatment, lack of oil infusion), indicating it is not induction of expression that is important for their function in decidualization response but their interaction in other responses that they mediate.
Although the overall uterine PR mRNA expression did not vary (Fig. 2), the localization of the PR protein was altered both by hormonal treatment as well as oil infusion. As previously observed, PR is primarily localized in the epithelial cells in ovariectomized WT and ERKO (Fig. 3, [28, 30]). P treatment in both the WT and ERKO shifted PR expression from the epithelial cells to the stroma. In addition, oil infusion led to further increases in stromal PR, with the greatest increase in the WT seen with E+Pe priming. In the ERKO, however, even vehicle-treated samples showed a significant increase in PR following oil infusion (Fig. 3). Interestingly, previous research indicated that estrogen could increase stromal PR expression in the ERKO [28]. This response was not studied here, yet our observations indicate that stromal PR is also induced readily 24 h after oil infusion independent of estrogen. The lack of a requirement for any E priming for this response in the ERKO rules out a role for ERß for the increase in stromal PR staining observed following oil infusion. P priming is required for development of the deciduoma; thus, the increased stromal PR expression without the requirement for E priming in the ERKO might increase the sensitivity of the ERKO stromal cells during deciduoma development. In contrast, although there is stromal PR expression in the WT, the E-dependent increase in the WT following oil infusion may be required for full progesterone responsiveness and development of deciduoma. Indeed, despite some indication of cytoplasmic enlargement consistent with early decidual changes in the WT progesterone-treated sample 24 h after intraluminal oil injection (Fig. 3E), this treatment does not lead to a full deciduoma development [7, 30].
Cox2 induction is essential to decidualization, and in the case of the WT, it is apparent that, although the initial Cox2 mRNA and protein induction 2 h after oil infusion is independent of hormone priming, the increase observed 22 h later is only apparent following E+Pe priming. This 24-h post-oil-infusion Cox2 increase remains independent of hormone in the ERKO. It appears that this later-maintained Cox2 induction in combination with progesterone stimulation might allow deciduoma to develop. Interestingly, Cox2 remains elevated in both WT and ERKO fully developed deciduomas (not shown).
The final component of this study was examining LIF expression and Stat3 phosphorylation as indicators of cytokine signaling, a pathway known to be important for initiating implantation and decidualization. Disruption of LIF or IL-11 receptor leads to loss of a decidual response [17, 18, 31]. The initiating stimulus for artificial induction of deciduoma development is reminiscent of a model developed to study the effects of stretching on cardiomyocytes [26]. The responses induced in cardiomyocytes following stretching included induction of cytokines, resulting in activation of cytokine receptors and associated Jak-Stat pathways. In the uterus, LIF mRNA was induced by 1–2 h of E treatment in the WT or following oil infusion in non-E-treated WT and all ERKO-treatment-group mice. In addition, IL-11 was similarly induced (not presented). The magnitude of induction of LIF following oil infusion was greater in the ERKO, suggesting loss of estrogen regulation is compensated for by oil infusion-induced expression. Therefore, although ERKO mice lack estradiol induction of LIF, increased LIF induction in response to oil infusion might supply LIF for the initiation of development of the deciduoma. It is clear that LIF is a necessary component for decidual development, as LIF-deficient mice can undergo decidualization when LIF is injected at the time of oil infusion [24]. Additionally, LIF injection can replace the nidatory estrogen normally required for embryo implantation in WT mice [24]. LIF receptors have been localized to the uterine luminal epithelium [21], yet this uterine compartment is not responsive to estradiol and is severely damaged in the ERKO following oil infusion (Fig. 3), appearing to be disrupted or completely missing in some sections. Thus, it will be important in future studies to localize the LIF receptors in the ERKO uterus to determine whether they are still present and thus involved in decidual response.
LIF and other cytokines signal through heterodimeric membrane receptors that activate Jak-Stat pathways. Phosphorylation of Stat proteins results in their dimerization and translocation to the nucleus, where they function as transcription factors through serum response elements in target genes [19, 20]. LIF treatment of cultured uterine luminal epithelial layers results in Stat3 activation [21]. Because we showed LIF mRNA induction following oil infusion and because this might result in cytokine receptor signaling, Stat3 phosphorylation was studied following oil infusion. Stat3 was phosphorylated following oil infusion in both WT and ERKO uteri independent of hormone priming. Although this is not conclusive evidence for LIF-dependent Stat3 phosphorylation, it is suggestive of this mode of cytokine signaling. The observation that oil infusion did not lead to further LIF induction in the WT that was treated with E+Pe but did result in increased Stat3 phosphorylation suggests that there is another component involved in Stat3 activation. Interestingly, in vitro studies have indicated that Stat3 can interact with ER and that this interaction regulates Stat3 transcriptional activity in an estrogen-dependent manner [32, 33]. Therefore, it will be important to determine whether LIF is directly responsible for Stat3 phosphorylation and also whether ER ablation alters Stat3 transcriptional activity in the ERKO uterus.
The appearance of the ERKO uterine tissue histology 24 h after oil infusion is much different than that of the WT uterine tissue (Fig. 3), with much damage to the luminal epithelial and stromal layers. The stromal cells are condensed, the nuclei are hyperchromatic, and there is a decrease in cell-to-cell contact. This increased damage as a result of oil infusion may also account in part for the relief of estrogen dependence for deciduoma development in the ERKO. It is known that severe trauma, such as crushing the uterine tissue, relieves estrogen dependence for decidualization in WT mice [12]. Thus, due to the hypoplasticity of the ERKO uterine tissue [2], the oil infusion might lead to greater damage than the same infusion in a WT, which might allow the decidual development to proceed despite the lack of ER or estradiol. One initiator of decidualization is the penetration and apoptosis of the epithelial layer by the implanting embryo [1]. Because the epithelial layer is so severely damaged in the ERKO by oil infusion, this may serve as a strong stimulus to initiate stromal transformation. The damage in the ERKO has been observed to various degrees in all sections from many uteri examined in the course of this study. However, because the sections were cross-sections and not representative of the entire length of the uterus, a more complete analysis would be required to determine the degree of damage in the whole uterus. Interestingly, damage to the epithelial cells is present in the WT samples as well 2 h after oil infusion (unpublished data), indicating that the difference may not be the degree of damage but the rate of progression of the subsequent response in the tissue. The increased damage in the ERKO may initiate a prolonged inflammatory response, leading to the maintained elevation of Cox2 seen 24 h after oil infusion (Figs. 4 and 5). Finally, as the ERKO stromal cells are expressing elevated PR as a result of the oil injection, their sensitivity to progesterone, together with elevated Cox2, could lead to development of deciduoma. These combined effects are not all directly related to ER ablation but are also a consequence of the immaturity of the uterine tissue as a result of estrogen insensitivity. These observations combined with previously published studies illustrate that 1) estrogen is essential for epithelial proliferative and implantation responses and 2) estrogen is dispensable for signaling that leads to stromal decidual response. Future studies will further explore the signaling mechanisms leading to decidual transformation of the stroma.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received: 13 December 2001.
First decision: 4 January 2002.
Accepted: 27 May 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Kondoh, T. Okamoto, T. Higuchi, K. Tatsumi, T. Baba, S. K. Murphy, K. Takakura, I. Konishi, and S. Fujii Stress affects uterine receptivity through an ovarian-independent pathway Hum. Reprod., April 1, 2009; 24(4): 945 - 953. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. N. Jefferson, E. Padilla-Banks, E. H. Goulding, S.-P. C. Lao, R. R. Newbold, and C. J. Williams Neonatal Exposure to Genistein Disrupts Ability of Female Mouse Reproductive Tract to Support Preimplantation Embryo Development and Implantation Biol Reprod, March 1, 2009; 80(3): 425 - 431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-W. Xu, N. Yasui, K. Ikeda, W.-J. Pan, J. Watanabe, M. Shiotani, A. Yanaihara, T. Miki, and Y. Yamori Isoflavones regulate secretion of leukemia inhibitory factor and transforming growth factor {beta} and expression of glycodelin in human endometrial epithelial cells J. Endocrinol., February 1, 2008; 196(2): 425 - 433. [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]
|
|
|
|
 |
|
 P.G. Groothuis, H.H.N.M. Dassen, A. Romano, and C. Punyadeera Estrogen and the endometrium: lessons learned from gene expression profiling in rodents and human Hum. Reprod. Update, July 1, 2007; 13(4): 405 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. K. Marriott, K. R. McGann-Gramling, B. Hauss-Wegrzyniak, L. C. Sheldahl, R. A. Shapiro, D. M. Dorsa, and G. L. Wenk Brain Infusion of Lipopolysaccharide Increases Uterine Growth as a Function of Estrogen Replacement Regimen: Suppression of Uterine Estrogen Receptor-{alpha} by Constant, But Not Pulsed, Estrogen Replacement Endocrinology, January 1, 2007; 148(1): 232 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-H. Ma, S.-J. Hu, H. Ni, Y.-C. Zhao, Z. Tian, J.-L. Liu, G. Ren, X.-H. Liang, H. Yu, P. Wan, et al. Serial Analysis of Gene Expression in Mouse Uterus at the Implantation Site J. Biol. Chem., April 7, 2006; 281(14): 9351 - 9360. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Baba, J. Mimura, N. Nakamura, N. Harada, M. Yamamoto, K.-i. Morohashi, and Y. Fujii-Kuriyama Intrinsic Function of the Aryl Hydrocarbon (Dioxin) Receptor as a Key Factor in Female Reproduction Mol. Cell. Biol., November 15, 2005; 25(22): 10040 - 10051. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J Kimber Leukaemia inhibitory factor in implantation and uterine biology Reproduction, August 1, 2005; 130(2): 131 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Couse, M. M. Yates, B. J. Deroo, and K. S. Korach Estrogen Receptor-{beta} Is Critical to Granulosa Cell Differentiation and the Ovulatory Response to Gonadotropins Endocrinology, August 1, 2005; 146(8): 3247 - 3262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Ramos, B. R. Rueda, P. C. Leavis, and R. R. Gonzalez Leptin Serves as an Upstream Activator of an Obligatory Signaling Cascade in the Embryo-Implantation Process Endocrinology, February 1, 2005; 146(2): 694 - 701. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Y Lee and F. J DeMayo Animal models of implantation Reproduction, December 1, 2004; 128(6): 679 - 695. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-B. Teng, H.-L. Diao, H. Ma, J. Cong, H. Yu, X.-H. Ma, L.-B. Xu, and Z.-M. Yang Signal transducer and activator of transcription 3 (Stat3) expression and activation in rat uterus during early pregnancy Reproduction, August 1, 2004; 128(2): 197 - 205. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. M. Simmen, R. R. Eason, J. R. McQuown, A. L. Linz, T.-J. Kang, L. Chatman Jr., S. R. Till, Y. Fujii-Kuriyama, F. A. Simmen, and S. P. Oh Subfertility, Uterine Hypoplasia, and Partial Progesterone Resistance in Mice Lacking the Kruppel-like Factor 9/Basic Transcription Element-binding Protein-1 (Bteb1) Gene J. Biol. Chem., July 9, 2004; 279(28): 29286 - 29294. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Grummer, S.W. Hewitt, O. Traub, K.S. Korach, and E. Winterhager Different Regulatory Pathways of Endometrial Connexin Expression: Preimplantation Hormonal-Mediated Pathway Versus Embryo Implantation-Initiated Pathway Biol Reprod, July 1, 2004; 71(1): 273 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Dey, H. Lim, S. K. Das, J. Reese, B. C. Paria, T. Daikoku, and H. Wang Molecular Cues to Implantation Endocr. Rev., June 1, 2004; 25(3): 341 - 373. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Hewitt, B. J. Deroo, K. Hansen, J. Collins, S. Grissom, C. A. Afshari, and K. S. Korach Estrogen Receptor-Dependent Genomic Responses in the Uterus Mirror the Biphasic Physiological Response to Estrogen Mol. Endocrinol., October 1, 2003; 17(10): 2070 - 2083. [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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
