| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
Institute of Anatomy,3 University Hospital, 45122 Essen, Germany
NIEHS,4 Research Triangle Park, North Carolina 27709
Institut of Genetics,5 University of Bonn, 53117 Bonn, Germany
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ERß knockout mice confirmed these different pathways, demonstrating that in the endometrium, estrogen-mediated Cx26 gene induction, but not induction during decidualization, is dependent on functional ER. To evaluate potential embryonic signals regulating Cx26 expression, uteri of pseudopregnant animals were incubated with different mediators in an organ-culture model, showing that catechol estrogen and mediators of the inflammatory cascade such as prostaglandin F2 and interleukin-1ß are able to induce Cx26 expression through the ER-independent pathway. Thus, the present study demonstrates that endometrial expression of Cx26 and Cx43 is induced via estrogen and ER during preimplantation but then utilizes an ER-independent signaling pathway during embryo implantation and decidualization.
estradiol receptor, female reproductive tract, gene regulation, implantation, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Uterine receptivity is defined as a restricted time period when the uterus is conducive to blastocyst attachment and implantation [3]. The establishment of this endometrial transition, which supports embryo implantation, is primarily coordinated by ovarian hormones, progesterone and estrogen, that modulate uterine events in a spatiotemporal manner and that prepare the endometrium to respond to blastocyst signals [4, 5]. In addition to the physical interaction of the embryonic tissue with the uterine cells, various substances such as steroid hormones, prostaglandins (PGs), and growth factors have been proposed to play a role in embryonic signaling [6–9]. Although numerous molecules involved in implantation have been identified in rodents [4, 5] and humans [10], microarray analysis of human endometrium from the receptive phase has given insight regarding involvement of certain molecules [11, 12], but the molecular mechanisms, regulation of genes, and signal cascades that govern this process of endometrium-blastocyst interaction remain poorly understood.
In previous studies, we showed that gap junction connexin expression is regulated in a precise spatiotemporal pattern during the receptive phase in rat endometrium [13, 14] as well as during cycling in humans [15]. Gap junction channels mediate direct intercellular communication and allow transfer of small molecules (
1 kDa) between the cytoplasm of neighboring cells, thereby coupling those cells both electrically and metabolically [16, 17]. They are composed of transmembrane proteins (connexins), with 19 members identified in the murine and 20 in the human genome that belong to one multigene family and show a very high sequence identity between different species [18].
We demonstrated a strong correlation between expression of connexins (Cx) 26 and 43 and the implantation reaction in rat endometrium: Expression of both connexins is regulated by ovarian hormones during preimplantation, leading to a suppression during the receptive phase, and is induced by the blastocyst shortly before and during implantation [13, 14, 19]. In the present study, we demonstrate that endometrial expression of Cx26 and Cx43 during preimplantation utilizes an estrogen receptor (ER) -regulated mechanism. The same genes are induced, however, via an ER-independent blastocyst signaling pathway before and during embryo recognition.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adult female Sprague-Dawley rats were housed under defined conditions with a temperature of 22 ± 1°C, an atmospheric humidity of 55% ± 10% (mean±SEM), and a 12L:12D photoperiod. They were fed standard pellet food and provided with water ad libitum. All animal experiments were approved by the institutional animal care committee of the government.
The ER knockout (ERKO) and ERß knockout (ßERKO) mice were obtained from the breeding colony at Taconic Farms (Germantown, NY). Mice were maintained in plastic cages under a 12L:12D photoperiod in a temperature-controlled room (21–22°C) and were fed NIH 31 mouse chow (National Institutes of Health, Bethesda, MD) and fresh water ad libitum. All procedures were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals.
Hormonal Treatment
17ß-Estradiol (E2) and progesterone (both from Sigma, Taufkirchen, Germany) as well as the antiestrogen ICI 182780 (kindly provided by A. Wakeling, Zeneca Pharmaceuticals, Cheshire, U.K.) were dissolved in benzyl benzoate and administered s.c. in sesame oil (100 µl for mice, 200 µl for rats). Corresponding to estrogenic effects in pseudopregnant rats described previously [14, 19], rats were injected with 1 µg of E2 and 0.5 mg of ICI 182780, and mice were injected with 0.1 µg of E2 and 0.1 mg of ICI 182780. Controls were given an equal volume of vehicle only. At least three animals were used for each experimental approach.
Pregnancy
Mating was performed overnight with male animals, and the day of vaginal plug (mice) and sperm finding (rats) was designated as 0 days postcoitum (dpc). Delayed implantation was achieved in rats by ovariectomy on 2 dpc and supplying progesterone (4 mg rat–1 day–1) from the day of ovariectomy onward. In rats, pregnancy was confirmed up to 4 dpc and during delayed implantation by flushing the uterus and counting the number of blastocysts.
Pseudopregnancy and Artificial Decidualization
Pseudopregnancy was achieved in ovariectomized rats and mice by treating them with E2 (1 and 0.05 µg/day, respectively) for 2 days, leaving them untreated for 2 days, and then substituting progesterone (5 and 1 mg/day, respectively) for 3 days. In ovariectomized rats, decidualization was obtained under anesthesia by scratching the antimesometrial luminal surface of one uterine horn with a hooked needle by way of a lateral abdominal incision and supplying E2 (0.1 µg/day) and progesterone (5 mg/day) for 24 h. The nonscratched contralateral horn served as a control [14]. In mice, decidualization was obtained by vaginally injecting 100 µl of oil and supplying E2 (0.05 µg/day) and progesterone (1 mg/day) for 24 h [20].
Tissue Collection
Rats were killed by ether and mice by cervical dislocation. Uterine horns were removed, and small pieces were frozen in liquid nitrogen for subsequent histochemical analysis. Rat uteri were opened longitudinally on an ice-cold glass plate, and the endometrium was carefully scraped off. Histological examination of the removed endometrium revealed no contamination with myometrial tissue (data not shown). The rat endometrial samples as well as the whole-mice uteri were frozen in liquid nitrogen and stored at –80°C.
Organ Culture
For in vitro organ-culture experiments, uteri of pseudopregnant rats were opened longitudinally and incubated for 18 h in culture medium (Dulbecco modified Eagle medium [Gibco BRL, Eggenstein, Germany] + 10% heat-inactivated fetal calf serum [Biochrome, Berlin, Germany] + 100 U/ml of penicillin/streptomycin [Boehringer, Mannheim, Germany]) or in culture medium containing E2 (5 ng/ml [21]), ICI 182780 (5 µg/ml [22]), actinomycin D (Sigma; 5 µg/ml [23]), dbcAMP (Sigma; 0.5 mg/ml [24]), 12-O-tetradecanolyphorbol-13-acetate (TPA; Sigma; 20 ng/ml [25]), catechol estrogen (Sigma; 1 µg/ml [26]), interleukin (IL)-1ß (Sigma; 10 ng/ml [27]), or PGF2 (Sigma; 1 µg/ml [28]). After incubation, endometrium was scraped off for RNA isolation, frozen in liquid nitrogen, and stored at –70°C.
Northern Blot Analysis
Total RNA was extracted from endometrial tissue using the RNeasy midi kit (Qiagen, Hilden, Germany). Five micrograms of total RNA (as estimated from optical absorbance measurements at 260 nm) were electrophoresed on a denaturing agarose-formaldehyde gel and blotted onto nylon membranes (Hybond-M; Amersham-Bucher GmbH, Braunschweig, Germany). The cDNA probes were random prime-labeled with a [32P]dCTP and hybridized with the RNA blots overnight at 42°C in a solution containing 55% deionized formamide, 1 M NaCl, 1% SDS, 10% dextran sulfate, and 100 µg/ml of salmon sperm DNA. The following connexin cDNAs were used for hybridization: a 1.1-kilobase (kb) cDNA corresponding to part of the coding region of rat Cx26 gene [29], a 1.4-kb cDNA corresponding to the coding region of rat Cx43 gene [30], and a rat actin-specific cDNA probe [31] or 18S RNA probe. Blots were washed at 60°C in 1x SSC (0.15 M sodium chloride and 0.015 M sodium citrate) and 0.1% SDS for 1 h, in 0.5x SSC and 0.1% SDS for 30 min, and in 0.2x SSC and 0.1% SDS for 30 min. Exposure to Kodak XAR-5 films (Eastman Kodak, Rochester, NY) took place at –70°C with intensifying screens.
Signals detected by autoradiography were quantified by densitometry. Densitometric values for connexin expression were calculated relative to the ß-actin level of the corresponding lane for possible differences in RNA loading.
Statistical Analysis
Analysis of variance was used for statistical testing, and when necessary, the Scheffé test was used as a post-hoc test. For Figure 5, the Mann-Whitney U-test was performed. The level of significance was set at P < 0.05.
|
Immunohistochemistry
Immunohistochemical staining was performed on cryostat sections (thickness, 4–6 µm) as described previously [32] using affinity-purified rabbit antibodies (1 µg/ml) to Cx26 from mouse liver gap junctions [33] and to the C-terminal 22 amino acids of rat Cx43 [34], respectively, after rat heart (Cx43) and liver (Cx26) were tested for positive controls. For negative controls, preimmune serum was used instead of the primary antibody showing no staining. Photographs were taken with an Axiophot microscope (Zeiss, Jena, Germany) equipped for epifluorescence. The intensity of immunostaining was semiquantitatively divided into strong (+++), moderate (++), low (+), or negative (–).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Induction of the gap junction Cx26 in the epithelial cells and Cx43 in the developing decidual cells of rat endometrium is dependent on the blastocyst during implantation [13], but it is highly regulated by ovarian steroid hormones during preimplantation [14, 19]. To further analyze the signaling pathways involved in connexin gene regulation, estrogen was given during early pregnancy as well as during pseudopregnancy either alone or in combination with the antiestrogen ICI 182780. Basal expression of Cx26 transcripts was suppressed in the rat endometrium on Day 3 of pseudopregnancy as well as during the receptive phase on Day 3 of pregnancy before the blastocyst reaches the uterus [14]. This reduction of Cx26 mRNA expression could be overcome by treatment with E2 (Fig. 1). Transcription of the Cx26 gene was significantly enhanced in the endometrium of pseudopregnant rats, revealing a 13-fold increase in mRNA expression compared to controls within 24 h after a single injection of 1 µg of E2 and a 22-fold increase in the endometrium of pregnant rats on Day 3 of pregnancy when E2 was administered during the receptive phase at 1 and 2 dpc (Fig. 1). In both experimental groups, this E2 effect on Cx26 gene transcription could be significantly inhibited by simultaneous application of the pure antiestrogen ICI 182780 (Fig. 1). The ICI 182780-treated rats showed no significant difference in Cx26 expression compared to controls. Basal Cx43 gene transcription also was suppressed during pseudopregnancy, but basal transcription could be detected during the receptive phase of pregnancy at 3 dpc. In both experimental groups, no significant regulation of Cx43 expression could be observed by treatment with E2 and/or ICI 182780. Thus, these two gap junction genes responded differently where Cx43 was less sensitive to E2-mediated regulation as well as to antiestrogen treatment compared to Cx26, pointing to additional regulatory pathways that may not be dependent on E2.
|
The observed changes in RNA levels of these transcripts were confirmed by protein analyses. The Cx43 protein was seen at the cell membrane of stromal cells on Day 3 of pseudopregnancy (Fig. 2a) and in the receptive phase of pregnancy at 3 dpc (Fig. 2d), respectively. No change in Cx43 protein level was apparent in the endometrial tissue 24 h after treatment with estrogen alone or in combination with antiestrogen (data not shown). Staining for Cx26 protein was observed at the membranes of the uterine epithelial cells of pseudopregnant rats 24 h after estrogen treatment (Fig. 2b) and on Day 3 of pregnancy when rats had been treated with estrogen at 1 and 2 dpc (Fig. 2e). In both experiments, the protein was not present when the antiestrogen was used (Fig. 2, c and f).
|
ER-Independent Regulation of Cx26 and Cx43 Gene Expression in Rat Endometrium
Pregnancy During early pregnancy, expression of connexin transcripts is suppressed during preimplantation, and expression of Cx26 is strongly induced and that of Cx43 is increased in the rat endometrium in the presence of the blastocyst from 4 dpc onward [14]. A significant increase in Cx26 as well as Cx43 transcripts can be seen at 5 dpc compared to 3 dpc (Fig. 3). Application of the antiestrogen ICI 182780 at 2–4 dpc did not inhibit induction of Cx26 and Cx43 at 5 dpc (Fig. 3). Thus, this blastocyst-mediated connexin induction during early pregnancy apparently acts via a signaling pathway independent of the ER.
|
Delayed implantation To test whether Cx26 induction at the time of implantation in the uterine epithelium is mediated by presence of the blastocyst itself or is caused by the uterine implantation reaction, we used the delayed implantation approach. Rats with experimentally generated delayed implantation were substituted exclusively with progesterone. In the presence of blastocysts but the absence of the implantation reaction, a significant induction of Cx26 transcripts took place in the endometrium (Fig. 3). However, in contrast to normal pregnancy, in which the Cx26 protein is locally restricted to the uterine epithelium of the implantation chamber (Fig. 4a), delayed implantation leads to a Cx26 induction throughout the luminal epithelium on Day 5 (Fig. 4b). The Cx26 proteins were not detected in the stromal compartment, reflecting the absence of the decidual response from the implanting blastocyst. This also most likely is the reason for the lack of induction of Cx43 in delayed implantation at 5 dpc (Fig. 3). As in a normal pregnancy, the blastocyst-mediated Cx26 gene induction could not be inhibited by simultaneous application of ICI 182780 (Fig. 3). However, transcript levels of Cx43 were significantly increased in the endometrium after antiestrogen treatment (Fig. 3).
|
Decidualization The decidual implantation reaction can also be induced artificially by a traumatic stimulus in pseudopregnant rats, leading to an induction of the gap junction genes Cx26 and Cx43 in the deciduoma [14]. As in the case of connexin gene induction by the blastocyst, expression of Cx26 and Cx43 is significantly increased 24 h after induction of decidualization and cannot be abolished by application of the antiestrogen ICI 182780 on either the mRNA level (Fig. 3) or the protein level (Fig. 4, c and d), illustrating again the ER-independent nature of this response.
Regulation of Connexin Expression in Wild-Type, ERKO, and ßERKO Mice
In the rodent, ER is the main ER in the endometrium, whereas ERß is only very weakly expressed in the glandular epithelial cells and in the stroma [35–37]. To confirm the role of an ER-dependent as well as an ER-independent pathway and to discriminate between the actions of ER and ERß in connexin gene regulation in the endometrium, experiments were performed in pseudopregnant ERKO (ER–/–) and ßERKO (ERß–/–) mice using the corresponding experimental design as described for the rat.
Connexin gene expression during early pregnancy in wild-type mice Before evaluation of connexin regulation in the mutant mouse model, we compared their regulation in wild-type mice to that previously described for the rat. Similar to our observations in the rat, the expression of Cx26 in the mouse uterus was clearly suppressed during the receptive phase up to Day 3.5 of pregnancy and was induced with the beginning of the implantation reaction at 4.5 dpc (Fig. 5). However, unlike in the rat, expression of Cx26 in the mouse is limited to the uterine epithelium (Fig. 4e) and is not expressed in the primary decidua cells with ongoing implantation, as already shown by Pauken and Lo [38]. Thus, only a short peak of Cx26 transcript expression could be observed with the beginning of implantation. This Cx26 expression vanished from the epithelium of the implantation chamber and showed no transcription by 6.5 dpc (Fig. 5). Transcripts of Cx43 showed a slight decrease during the receptive phase, with a significantly reduced expression shortly before implantation on 3.5 dpc and a significant increase in transcription during implantation from 4.5 dpc onward (Fig. 5). Expression of this gap junction protein in the developing decidual tissue of the mice (Fig. 4f) is comparable to the Cx43 expression pattern during early pregnancy in the rat.
Regulation of connexin expression in ERKO and ß ERKO mice
Ovariectomized wild-type mice showed an eightfold and ßERKO mice a sixfold enhancement of Cx26 gene expression 24 h after a single s.c. injection of E2 (1 µg/mouse) (Fig. 6). No induction of Cx26 transcripts could be seen in ERKO mice using the same protocol (Fig. 6). However, Cx26 transcripts were induced in the uteri of pseudopregnant wild-type and ERKO mice 24 h after a traumatic stimulus, leading to a decidualization reaction (Fig. 6). This signal was transient; it disappeared 3 days after oil infusion (data not shown).
|
These results confirm that E2-mediated induction of Cx26 in the endometrium is dependent on ER but that induction by artificial decidualization acts via an ER-independent pathway.
Mediators Possibly Involved in ER-Independent Cx26 Gene Regulation in Rat Endometrium
To identify possible mediators involved in the ER-independent Cx26 induction in rat endometrium, we established an organ-culture system. To prove the value of this model for investigating endometrial Cx26 regulation, the initial set of experiments evaluated ER-dependent induction of Cx26. Expression of the Cx26 gene was significantly enhanced after 18 h of culture in medium containing E2. This induction was suppressed by simultaneously adding either the pure antiestrogen ICI 182780 or the transcription-inhibitor actinomycin D (Fig. 7). Results from the organ-culture model confirmed that the endometrium response to estrogen is similar to those from the in vivo experiments. To test other activators, uteri were incubated in medium containing dbcAMP, TPA, catechol estrogen, IL-1ß, or PGF2. Although neither dbcAMP nor TPA had any significant effect on Cx26 gene expression, levels of the Cx26 transcripts were significantly enhanced by catechol estrogen, IL-1ß, and PGF2, even in the presence of ICI 182780 (Fig. 7). These observations suggest an ER-independent signaling pathway that is activated by these compounds, resulting in Cx26 gene regulation. In this in vitro model, none of the treatments had a significant effect on Cx43 gene expression in the endometrial tissue (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-dependent signaling pathway during preimplantation; the second, an ER-independent pathway, occurs during embryo implantation or artificial decidualization. As shown previously, maternal progesterone suppresses expression of Cx26 and Cx43 in rat endometrium during the receptive phase of pregnancy as well as during pseudopregnancy, however, only if progesterone serum levels reach concentrations similar to those in pregnancy [19]. In the present study, we showed that shifting the hormonal level in favor of estrogen during the receptive phase leads to a nonphysiological expression of Cx26 in the uterine epithelium, confirming earlier observations that suppression on Cx26 gap junction channels requires a balance between the relative levels of progesterone and estrogen as established during early pregnancy [19]. Antiestrogen treatment revealed that this estrogen-mediated induction of connexin transcripts during pseudopregnancy as well as during the receptive phase of pregnancy involves an ER-dependent signaling pathway for the Cx26 gene. However, the amount of Cx43 transcripts was not significantly reduced by ICI 182780, pointing to an additional signaling pathway involved in the regulation of this gene in the stromal compartment. In contrast to connexin gene regulation during preimplantation, induction of connexin transcripts during embryo implantation is mediated via an ER-independent signal cascade. In pregnant rats, application of the pure antiestrogen had no effect on epithelial Cx26 transcription induced by the blastocysts. Stromal expression of Cx43 after antiestrogen treatment was weaker compared to untreated animals. This effect, however, was more related to loss or diminishment of the decidual reaction in the endometrial stroma caused by the lack of implantation, because E2 is necessary for initiation of the implantation reaction [20, 39].
As shown in the present study using the delayed implantation model, Cx26 transcript induction in the uterine epithelium is mediated by the presence of the blastocyst even in the absence of the implantation reaction. In this experimental model, blastocysts remain closely apposed to the uterine luminal epithelium without initiating the attachment reaction. This condition can be maintained by continued progesterone treatment and is terminated by an injection of estrogen with blastocyst activation and implantation [40, 41]. It has been stated that during delayed implantation in mice, the uterus becomes nonresponsive to the presence of a blastocyst when exposed to progesterone alone [5, 42]. In the present study, we demonstrate that even in the absence of E2 and implantation, the uterine epithelium is responsive to the blastocyst. In contrast to the local induction of Cx26, which is restricted to the luminal epithelium of the implantation chamber at 4 dpc of normal pregnancy, expression of Cx26 in delayed implantation is induced with less intensity but throughout the uterine epithelium. The mere presence of the blastocyst results in the ability to induce Cx26 gene expression. Thus, this model is well suited for the investigation of very early blastocyst signals occurring before the onset of implantation. Induction of Cx43 was not observed in the stromal compartment in this model related to the lack of implantation and decidualization reaction in the absence of E2 [5, 42].
Induction of Cx26 and Cx43 in the rat endometrium can also be achieved by artificial decidualization in a manner similar to the ER-independent connexin induction by the blastocyst. Whereas E2 leads to an induction of Cx26 transcripts only in the uterine epithelium, not in the stromal cells, a traumatic stimulus such scratching effects a strong induction of Cx26 in addition to a clear enhancement of Cx43 in the developing rat deciduomata. Thus, induction of Cx26 in the stromal cells, in contrast to induction in epithelial cells, seems to be regulated solely via the ER-independent signaling pathway. Traumatic stimuli relieve estrogen dependence for decidualization in mice [43], which has been recently confirmed using ERKO mice [20, 44, 45]. As described previously by Curtis Hewitt et al. [20], the stromal cell response is primarily dependent on progesterone, whereas the luminal epithelium, which controls the initiation of implantation, is highly sensitive to estrogen. Those authors conclude that estrogen is essential for epithelial proliferative and implantation responses but is dispensable for signaling leading to the decidual response.
The experiments in ERKO mice confirmed the results gained from ICI 182780-treated rats and revealed that E2-mediated induction of Cx26 and Cx43 is mediated via ER, but not ERß, in the endometrium. This goes along with the findings that ER has been shown previously to be the most important form of ER in the uterus, whereas ERß is strongly expressed in the ovary and the mammary gland but shows only weak expression in the uterus of mice [35] and in the glandular epithelium and stromal cells of rats [36, 37]. The mere presence of the ERß obviously is not sufficient for estrogen-mediated connexin induction in the endometrium. The ER-dependent induction of Cx26 by E2 is regulated on the transcriptional level, because addition of the transcription-inhibitor actinomycin D prevents Cx26 gene induction in the organ-culture model.
In the mouse, induction of connexin genes by an artificial stimulation of the endometrium is independent from ER. Unlike in the rat, Cx26 in the mouse is not expressed in the primary decidual zone; here, Cx43 is the only connexin of the decidua, as described earlier by Pauken and Lo [38]. As a consequence, Cx26 expression peaks shortly before implantation exclusively because of expression in the luminal epithelium in pregnant wild-type mice as well as during artificial decidualization of pseudopregnant wild-type and ERKO mice, but this expression is lost with progression of the decidualization process. The reasons for this difference in the pattern of connexin expression between the rat and the mouse are unknown.
An apparent enhancement of E2-independent connexin regulation was observed after application of the antiestrogen ICI 182780 during delayed implantation or artificially induced decidualization. This is in accordance with previous results of Curtis Hewitt et al. [20], who observed an increased sensitivity of stromal cells during deciduoma development in ERKO mice, apparently because of an increased stromal progesterone receptor (PR) expression following oil infusion compared to wild-type mice. Thus, it is possible that ER may limit and, thus, regulate the rate of decidua formation during early pregnancy.
In both the rat and the mouse, we could show that induction of connexins by the blastocyst during early pregnancy as well as during artificially induced decidualization is mediated by a signal cascade independent of the ER. Molecular signals guiding interactions between the blastocyst and uterus to initiate the process of implantation remain ill-defined. However, a number of potential blastocyst-derived molecules, such as leukemia inhibitory factor [20], have been either described [46] or shown to be involved in artificial decidualization. Recently, growth factors, cytokines, and their receptors as well as various other molecules have been implicated in the implantation process [4, 5, 47, 48]. Using an organ-culture model, ER-independent regulation of Cx26 gene expression can be induced by IL-1ß, catechol estrogen, and PGF2, which are factors involved in the implantation process [49, 50]. Current data suggest that IL-1ß can activate multiple signaling pathways regulating insulin-like growth factor binding protein-1 gene expression and decidualization in vitro and may be involved in the events leading to decidualization in baboons [50]. During human embryonic implantation, the interactions between embryo and endometrium are mediated through the embryonic IL-1 and IL-1ß [51], and blockade of the endometrial IL-1 receptor prevents implantation in the mouse by interfering with embryonic attachment [52]. Furthermore, the embryo is able to synthesize PGs [53]. In addition, recent discoveries suggest important and critical roles for PGs in embryonic and uterine functions during implantation [54, 55], because the cyclooxygenase Cox-2 was shown to be restricted to the implantation site [55, 56] and Cox-2-deficient mice have defective implantation and decidualization [5, 55]. An elevated expression of Cox-2 in artificial decidualization after oil infusion was shown previously in wild-type [57] and ERKO mice [20].
Like the in vivo situation, Cx26 transcription in this in vitro model is regulated via an ER-dependent as well as an ER-independent pathway. Although an estrogen-responsive element could not be found in the Cx26 promoter region in mice [58] or in rats [59], GC-rich regions were identified in the Cx26 promoter region of rodents and humans [58, 59]; these regions are putative binding sites for the transcription factor SP1 [60] and, thus, may be involved in estrogen-mediated gene regulation [61, 62]. On the other hand, a binding site for the transcription factor NFkB could be found in both species [58, 59], which may be involved in ER-independent Cx26 induction via ILs secreted by the blastocyst resulting in inflammatory responses.
Expression of the Cx26 gene is transactivated by two different pathways in the rodent endometrium. On the one hand, it is regulated by ovarian steroid hormones via steroid hormone receptors during preimplantation; on the other hand, it is induced by an as-yet-unknown signal of the blastocyst, even in the absence of the implantation reaction, involving a signaling pathway independent of the ER. Future analysis of these regulatory mechanisms will be helpful in determining different signal cascades involved in embryonal signaling and uterine responses necessary for successful implantation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Ruth Grümmer, Institut für Anatomie, Universitätsklinikum Essen, Hufelandstr. 55, D-45122 Essen, Germany. FAX: 49 201 7235916; ruth.gruemmer{at}uni-essen.de
Received: 6 October 2003.
First decision: 1 November 2003.
Accepted: 4 March 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
knockout uterus demonstrate that implantation but not decidualization-associated signaling is estrogen dependent. Biol Reprod 2002 67:1268-1277 represses androgen sensitivity in the LNCaP prostate cancer cell line. J Urol 2000 164:800-805[CrossRef][Medline] (ER) and estrogen receptor-ß (ERß) messenger ribonucleic acid in the wild-type and ER-knockout mouse. Endocrinology 1997 138:4613-4621) and beta (ER-ß) mRNA in the rat pituitary, gonad, and reproductive tract. Steroids 1998 63:498-504[CrossRef][Medline]-deficient mice. Endocrinology 1999 140:2704-2710This article has been cited by other articles:
|
|
 |
|
  M. J. Laws, R. N. Taylor, N. Sidell, F. J. DeMayo, J. P. Lydon, D. E. Gutstein, M. K. Bagchi, and I. C. Bagchi Gap junction communication between uterine stromal cells plays a critical role in pregnancy-associated neovascularization and embryo survival Development, August 1, 2008; 135(15): 2659 - 2668. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
