Gap Junction Connexin Genes cx26 and cx43 Are Differentially ...

0013-7227/99/$03.00/0 Endocrinology Copyright © 1999 by The Endocrine Society

Vol. 140, No. 6 Printed in U.S.A.

Gap Junction Connexin Genes cx26 and cx43 Are Differentially Regulated by Ovarian Steroid Hormones in Rat Endometrium ¨ MMER, OTTO TRAUB, RUTH GRU

AND

ELKE WINTERHAGER

Institute of Anatomy, University Hospital (R.G., E.W.), 45122 Essen; and Institut of Genetics, University of Bonn (O.T.), 53117 Bonn, Germany ABSTRACT In rat endometrium, expression of gap junction connexin-26 (cx26) in the epithelium and cx43 in the uterine stroma is suppressed by progesterone before implantation. For further study of connexin gene regulation we analyzed expression of cx26, cx43, and cx32 in the endometrium of ovariectomized rats treated with different ratios of 17b-estradiol (E2) and progesterone (P). A hormonal ratio of E2 to P that mimics conditions during pregnancy (0.1 mg E2 and 4 mg P) suppressed expression of cx26 and cx43. By changing the ratio to higher E2 levels (1 mg E2), cx26, in contrast to cx43, was not suppressed even by application of a high P concentration (10 mg). Timecourse experiments supplying E2 alone led to an early gene response

I

MPLANTATION of a mammalian blastocyst into the endometrium involves a complex series of precisely synchronized physiological and cell biological events that prepare the developing blastocyst and the endometrium for interacting with one another. Disruption of this synchrony in the differentiation process of both the blastocyst and the uterine epithelium leads to failure in the implantation process. In humans, 20% of spontaneous abortions during pregnancy are estimated to occur before pregnancy is detected clinically (1). In farm animals, this spontaneous abortion rate around implantation is increased to 80% (2). Implantation of the embryo at the time of uterine nidation capability depends upon interactions between factors that are under the control of progesterone (P) and 17b-estradiol (E2) (3, 4). The onset and timing of the transient uterine receptivity, which is described as the implantation window, is subject to a shifting ratio of P to E2 in uterine tissue (4 – 6). However, the exact cell biological mechanisms, regulation of genes, and signal cascades that control this transformation into the receptive state are complex and still poorly defined. The expression pattern of the adhesion molecule b3-integrin has been suggested as a marker for uterine receptivity (7), and the presence of leukemia inhibitory factor in the receptive endometrium has been shown to be necessary for embryo implantation (8). In previous studies we could show that gap junction connexin expression is regulated in a precise spatial and temporal pattern during the receptive phase in rat endometrium (9, 10) and during cycling in humans (11).

Received June 10, 1998. Address all correspondence and requests for reprints to: Dr. Ruth Gru¨mmer, Institut fu¨r Anatomie, Universita¨tsklinikum Essen, D-45122 Essen, Germany. E-mail: [email protected].

of cx26 within 3 h, whereas induction of cx43 transcripts was not detected until 14 h after E2 treatment. Simultaneous application of the antiestrogen ICI 182780 abolished E2-mediated induction of both connexins. No hormonal regulation of cx32 could be detected. As already shown for cx43 gene induction in the myometrium, E2-mediated induction of cx26 expression in the endometrium also required newly synthesized transcription factors. It can be concluded that only a hormonal ratio resembling conditions during pregnancy is able to suppress the expression of both cx26 and cx43 and that cx26 gene expression is induced earlier by E2 and is likely to be more sensitive to a shift in the E2 to P ratio than cx43. (Endocrinology 140: 2509 – 2516, 1999)

Gap junction channels, responsible for direct intercellular communication, connect the cytoplasms of neighboring cells and allow transfer of small molecules up to 1 kDa, such as intercellular signaling molecules and ions from one cell to another, thereby coupling the cells both electrically and metabolically (12, 13). One channel is formed by two hemichannels (connexons), each composed of six transmembrane proteins (connexins) radially arranged around a hydrophilic pore. More than 13 different connexins (cx) that all belong to a multigene family and show a very high sequence identity between different species are known in the murine genome (13, 14). Some members of the connexin family are broadly expressed in many tissues, whereas others show a highly restricted pattern of distribution. The knowledge of the physiological functions of the different channels is increasing, as connexin-deficient mice have been established, e.g. for cx43 (15), cx26 (16), cx32 (17), cx37 (18), and cx46 (19). In most tissues connexin expression is in a steady state, but some organs show a tissue-specific regulation of connexin expression, e.g. hormonal target organs such as ovary and uterus (10, 20 –22). From several studies it is well known that connexin gene expression in the uterus can be regulated by ovarian steroid hormones (22, 23). Most of those studies, however, focus on cx43 expression in the myometrium during late pregnancy and around the time of labor. Gap junction proteins expressed in rat endometrium in a defined spatial and temporal pattern during pregnancy have been identified as cx26 and cx43 (9, 24). In addition, cx32 was demonstrated in the uterine epithelium of immature rats as well as in late pregnancy (25). We previously reported that the expression of cx26 and cx43 is hormonally regulated by

2509

2510

DIFFERENT HORMONAL RESPONSE OF cx26 AND cx43

Endo • 1999 Vol 140 • No 6

Materials and Methods Animal care Adult female Sprague-Dawley rats were housed under defined conditions with a temperature of 22 6 1 C, an atmospheric humidity of 55 6 10%, and a 12-h dark, 12-h light cycle. They were fed standard pellet food and provided with water ad libitum. All animal experiments were approved by the institutional animal care committee.

Pregnant rats Mating was performed overnight with male rats, and the day of vaginal plug and sperm finding was designated as day 0 of pregnancy (dpc).

Experimental protocols of hormone treatment For studies on hormone action, rats were ovariectomized and rested for 14 days as described previously (10). Hormone regimens were initiated 2 weeks after ovariectomy. E2 and P were obtained from Sigma Chemical Co. (Deisenhofen, Germany), dissolved in benzyl benzoate, and administered sc in 200 ml sesame oil. For E2/P ratio experiments, rats were injected with either 0.1 or 1 mg E2, respectively, in combination with 0.1, 1, 4, or 10 mg P/ratzday for 3 days. E2 time-course experiments were performed in ovariectomized rats that had been pretreated with 4 mg P/day/rat for 3 days to mimic early pregnancy. One microgram of E2 in 200 ml sesame oil was applied sc, and rats were killed 1, 3, 6, and 14 h after E2 injection. The antiestrogen ICI 182 780 (0.5 mg/rat in 200 ml sesame oil (provided by A. Wakeling, Zeneca Pharmaceuticals, Cheshire, UK) was injected sc, followed by injection of 1 mg E2 1 h later. Rats were killed 24 h after E2 injection. To inhibit protein synthesis, cycloheximide (1 mg/250 mg BW; Sigma Chemical Co.) was dissolved in saline and injected ip followed by injection of 1 mg E2 30 min later. Rats were killed 4 h after E2 injection. Controls were given an equal volume of vehicle (sesame oil, 200 ml) only. Three animals were used for each experimental approach. FIG. 1. Northern blot from rat endometrial RNA from days 0 –5 of pregnancy probed for cx26 (2.5 kb), cx43 (3.0 kb), and cx32 (1.6 kb; lanes 0 –5). During preimplantation, levels of cx26 and cx43 mRNA decline from days 1–3 postcoitus and rise again during implantation from 4 dpc onward (10). In contrast, cx32 is expressed at very low levels and shows no changes during the periimplantation period. As a control, rat heart (cx43) and liver tissue (cx26 and cx32) were used (n 5 3/group).

Tissue collection

E2 and P during preimplantation and periimplantation in the rat endometrium, but not in heart (cx43) and liver (cx26) tissue of the same animals (10). Thus, the expression patterns of different connexin genes may be related to different stages of differentiation and function of epithelial and stromal cells during preimplantation. This raises the question about tissue-specific regulation properties of connexin genes. There is little information available about the mechanisms involved in the regulation of connexin gene expression in the endometrium in response to maternal hormones (10). Hormoneresponsive elements have not been clearly identified in the promoter regions of the connexin genes investigated. In the promoter of the cx43 gene, half-palindromic estrogen-responsive elements (EREs) were described (26, 27), but a functional proof for activation of these EREs is still missing. To get further insight into the mechanisms underlying the differential regulation properties of connexin genes, we characterized the sensitivities of cx26, cx32, and cx43 gene responses to P and E2 in rat endometrium.

Northern blot analysis

Animals were killed by ether. Uterine horns were removed, and a small piece of the uterus was frozen in liquid nitrogen for subsequent histochemical analysis. 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 (our unpublished results). The endometrial samples were frozen in liquid nitrogen and stored at 280 C.

Total RNA was extracted from endometrial tissue using the RNAeasy midi kit (Qiagen, Hilden, Germany). Five micrograms (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, Freiburg, Germany). Connexin-specific complementary DNA (cDNA) probes were random prime labeled with [a-32P]deoxy-CTP 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 mg/ml salmon sperm DNA. The following connexin cDNAs were used for hybridization: a 1.1-kb cDNA corresponding to part of the coding region of rat cx26 gene (28), a 1.4-kb cDNA corresponding to the coding region of rat cx43 gene (29), and a 1.5-kb cDNA corresponding to the coding region of the rat cx32 gene (30). In addition, a c-fos cDNA probe (31) and a c-jun cDNA probe (32) were used in this study. Blots were washed at 60 C in 1 3 SSC (standard saline citrate)-0.1% SDS for 1 h, in 0.5 3 SSC-0.1% SDS for 30 min, and in 0.2 3 SSC-0.1% SDS for 30 min. Exposure to Kodak XAR-5 films (Eastman Kodak Co., Rochester, NY) took place at 270 C with intensifying screens. After exposure, each blot was rehybridized with a rat actin-specific cDNA probe (33) using the same conditions of hybridization.


2511

FIG. 2. Northern blot of endometrial RNA of ovariectomized rats treated with different amounts of E2 or/and P for 3 days. Rats treated with vehicle only (C, left lane) revealed nearly no connexin transcripts. Application of 0.1 and 1 mg E2, respectively, led to a significant increase in cx26 and cx43 transcripts (P , 0.05). Transcript levels of cx26 as well as cx43 decreased markedly with increasing P concentration applied in combination with 0.1 mg E2. Treatment with 1 mg P resulted in a significant reduction of both connexin transcripts compared with the effect of 0.1 mg P (P , 0.05). Application of 4 mg P or more suppressed connexin expression to weak background levels, comparable to the situation in the preimplantation phase on day 3 dpc (see Fig. 1). A 10-fold increase in the amount of E2 (1 mg E2) abolished the suppressive effect of P on cx26 expression. Even application of 1 mg E2 and 10 mg P had no effect on E2-induced cx26 expression, which stayed at a high expression level. In contrast, cx43 gene expression was significantly suppressed by administration of 1 mg E2 and 4 mg P (P , 0.05). The expression of cx32 stayed at weak levels independent from the amount of hormones applied (n 5 3/group).

Signals detected by autoradiography were quantified using a scanning densitometer (Biometra, Go¨ttingen, Germany). Densitometric values for connexin expression were calculated relative to the b-actin level of the corresponding lane for possible differences in RNA loading. Statistical evaluation was performed using the Kruskal-Wallis test. Differences were considered significant at P # 0.05.

Immunohistochemistry Immunohistochemical staining was performed on cryostat sections (4 – 6 mm) as described previously (24), using affinity-purified rabbit antibodies (1 mg/ml) to cx26 and to cx32 from mouse liver gap junctions (34) and affinity-purified antibodies directed at a synthetic peptide representing the C-terminal 22 amino acids of rat cx43 (35). For positive controls, rat heart (cx43) and liver (cx26) were tested. For controls, rabbit preimmune serum was used instead of the primary antibody. Photo-

graphs were taken with an Axiophot microscope (Carl Zeiss, Inc., New York, NY) equipped for epifluorescence.

Results Cx26 and cx43, but not cx32, are hormonally regulated during early pregnancy

In contrast to cx26 and cx43, which have been shown to be suppressed during preimplantation, with a maximum suppression on day 3 of pregnancy (10), the very weak expression of cx32 transcript remained unchanged. At implantation, cx26 and cx43 were induced from day 4 pc onward, whereas no change in cx32 expression could be detected in Northern analysis during this period (Fig. 1). In all stages of

2512


Endo • 1999 Vol 140 • No 6

FIG. 3. Immunohistochemical staining of uterine sections of ovariectomized rats treated with different amounts of E2 and P. After sc application of 0.1 mg E2 for 3 days, a punctate reaction for cx26 was detected between the epithelial cells (A and B) and for cx43 in the surrounding stromal cells (C and D). Neither expression of cx26 (E and F) nor that of cx43 (G and H) was changed by additional administration of 0.1 mg P. After application of 0.1 mg E2 and 10 mg P, no staining of cx26 (I and J) or cx43 (K and L) could be observed. E, Luminal epithelium; S, stroma. Bar, 50 mm.

early pregnancy investigated (0 – 6 dpc), the cx32 protein was not detectable (data not shown). P antagonism of estradiol action depends on the E2/P ratio

The endometrium of ovariectomized rats revealed no expression of cx26 and a low level of cx43 transcripts. Application of the ovarian steroid hormones E2 and P for 3 days in an amount able to maintain pregnancy in rats (0.1 mg E2 and 4 mg P/day/rat) led to a suppression of cx43 in addition to cx26 comparable to the situation during preimplantation on day 3 of pregnancy (10). To investigate the role of the E2/P

ratio, ovariectomized rats were treated with different concentrations of the steroid hormones. In a first set of experiments 0.1 mg E2/ratzday was injected sc for 3 days in combination with varying concentrations of P. Northern blot analysis revealed that treatment with 0.1 mg E2 and 0.1 mg P led to a significant induction of cx26 and cx43 in rat endometrium compared with that in control animals (Fig. 2). The cx43 cDNA probe hybridized to a single transcript of approximately 3 kb, the cx26 probe hybridized to a transcript of approximately 2.5 kb. Transcript levels of cx26 as well as cx43 decreased markedly with increasing P concentration.


Treatment with 1 mg P resulted in a significant reduction of cx26 and cx43 expression compared with application of 0.1 mg P after 3 days of treatment. Application of 4 mg P or more suppressed cx26 expression to weak background levels comparable to the situation in the preimplantation phase on day 3 dpc (see Fig. 1). Interestingly, a 10-fold increase in the amount of E2 (1 mg E2) abolished the suppressive effect of P on cx26 expression. Even application of 10 mg P had no suppressive effect on E2-induced cx26 expression (Fig. 2). In contrast to cx26, cx43 was still suppressed by P despite the increase in E2 concentration. With increasing P concentration, a decreasing expression of cx43 could be observed; at concentrations of 1 mg E2 and 4 mg P, expression of cx43 was suppressed to a very weak level (Fig. 2). Expression of cx32 remained barely detectable independent of the concentrations of hormones applied (Fig. 2). These experiments indicated that only a hormonal profile adequate to conditions during pregnancy with high P in combination with low E2 levels was able to suppress cx26. Immunohistochemistry confirmed the results obtained by Northern blot analysis. After treatment of ovariectomized rats with 0.1 mg E2 for 3 days, staining for cx26 was detected in the epithelium (Fig. 3, A and B), and staining for cx43 was found in the surrounding stroma cells (Fig. 3, C and D). Similar results were obtained for both connexins after administration of 0.1 mg E2 and 0.1 mg P (Fig. 3, E–H), whereas no staining for either connexin could be observed after application of 0.1 mg E2 and 4 and 10 mg P, respectively (Fig. 3, I–L). Treatment with 1 mg E2 and at least 4 mg P suppressed the expression of cx43 protein, but not that of cx26. The expression of cx32 protein was absent regardless of hormonal treatment (not shown).

2513

FIG. 4. Northern blot of endometrial RNA from ovariectomized rats treated with 1 mg E2/rat for 1–14 h. After injection of E2, mRNA of cx26 was significantly increased within 3 h, and mRNA of cx43 was increased within 14 h (P , 0.05). Simultaneous application of the antiestrogen ICI 182780 for 24 h prevented induction of cx26 as well as cx43 expression. Cx32 transcripts were not induced by E2 within this time period. Application of vehicle only (C) for 14 h showed no effect on connexin expression (n 5 3/group).

Effect of E2 administration on connexin expression

To analyze the regulation of connexin expression by E2, time-course experiments were performed applying 1 mg E2 to ovariectomized rats pretreated with P. The different connexins revealed different temporal patterns of response to E2 treatment. Cx26 showed a 3-fold increase in transcript expression in rat endometrium within 3 h after E2 application compared with the control level (Fig. 4). With progressing time of E2 action, cx26 expression increased 8.7-fold within 6 h and 10-fold within 14 h after E2 injection compared with control values. In contrast, an increase in cx43 expression could not be observed until 14 h after E2 injection, when a 2.5-fold increase in the cx43 messenger RNA (mRNA) level occurred compared with the control level after normalizing to the level of b-actin (Fig. 4). The increase in cx26 and cx43 mRNA expression clearly was due to estrogen action, as induction of both connexins could be inhibited by simultaneous application of the estrogen antagonist ICI 182780 (Fig. 3). Expression of cx32 was not induced by E2 during this period (Fig. 3) and not even after 5 days of estrogen treatment (data not shown). This time course of transcript expression was accompanied by expression of the corresponding protein, as demonstrated by immunofluorescent analysis. Immunostaining of cx26 was absent in untreated ovariectomized rats (Fig. 5, A and B), but could be detected on the plasma membranes of

the uterine epithelium from 3 h after injection of E2 onward (Fig. 5, C and D); cx43 immunoreactivity was found in endometrial stromal cells within 14 h of treatment (Fig. 5, E and F). The localization of this connexin expression corresponded to the situation during cycling as well as that during pregnancy, showing cx26 in epithelial cells and cx43 in the stromal compartment (9, 24). No cx32 protein could be detected in the endometrium of estrogen-treated rats (Fig. 5, G and H). Controls treated with preimmune serum instead of the primary antibody revealed no staining (not shown). A newly synthesized transcription factor is necessary for cx26 induction

To further analyze whether induction of cx26 is mediated by a newly synthesized transcription factor, ovariectomized rats were treated with the protein synthase inhibitor cycloheximide in combination with estrogen. As this potent protein synthase inhibitor was applied to living animals, we did not exceed 4 h of treatment and thus restricted our investigation to the induction of the cx26 gene. Application of cycloheximide to ovariectomized rats in combination with E2 for 4 h prevented E2-induced cx26, but not c-fos expression, in rat endometrium (Fig. 6). This points to the fact that a newly synthesized transcription factor is necessary for induction of the cx26 gene. This signaling cascade maybe reg-

2514


Endo • 1999 Vol 140 • No 6

FIG. 5. Immunohistochemical staining for cx26, cx43, and cx32 in endometrium of ovariectomized rats treated with 1 mg E2. Staining for cx26 could not be detected in untreated rats (A and B), but was expressed at the cell membranes of the luminal epithelium by 3 h after E2 application (C and D). Cx43 protein was expressed between the endometrial stroma cells 14 h after the injection of E2 (E and F). Staining for cx32 could not be detected in rat endometrium even 24 h after E2 application (G and H). E, Luminal epithelium; S, stroma. Bar, 50 mm.

ulated by the immediate early genes c-fos and c-jun, as both transcription factors were up-regulated by E2 in the endometrium within 1 h, showing a peak in expression 3 h after E2 injection, whereas expression of c-myc was elevated not earlier than 14 h after E2 treatment (data not shown). Discussion

In recent studies we demonstrated that in rat endometrium expression of cx26 as well as that of cx43 are regulated by the ovarian steroid hormones P and E2. Due to maternal P, both connexins are suppressed during preimplantation, but an additional injection of E2 is able to reinduce the expression of both connexins (10). In the present study we found that this hormonal regulation of connexin expression is more sophisticated and fine tuned. Interestingly, cx26 and cx43 reveal different sensitivities to the E2/P concentration ratio. In ovariectomized rats, obviously only a hormonal profile similar to conditions during pregnancy with a high amount of P in combination with low E2 levels was able to suppress both transcripts comparable to the situation during preimplantation. With higher E2 levels, expression of cx26 was reexpressed, whereas cx43 mRNA levels remained suppressed. In contrast to our studies on adult rats, Risek and co-workers (36) observed no detectable effect of E2 on steady state cx26 mRNA levels in sexually immature rat uteri, but they documented an up-regulation of cx26 in uteri of rats treated with P in combination with very high amounts of E2, suggesting a synergistic, rather than an antagonistic, interaction of these two hormones. It seems that the cell-specific

responses of the uterus to ovarian steroid hormones differ significantly from immature to sexual mature animals. In addition, cx32 expression has been shown in the luminal epithelium of nonpregnant rats and during mid- and late gestation (25). However, in contrast to cx26 and cx43, neither cx32-mRNA nor cx32-protein is found in the uterine epithelium during the pre- and periimplantation periods on mRNA and protein level and cx32 is not induced by the different hormonal treatments. Thus, cx32 seems not to be involved in preparing the uterine epithelial cells for receptivity. Expression of cx26 and cx43 reacts to an E2 stimulus as an early gene response, with induction of cx43 within 14 h and of cx26 even within 3 h. This high plasticity to hormonal changes predestinates both genes to react very fast to changing physiological requirements during the interaction of the blastocyst with the uterine epithelium. As the presence especially of estrogen receptor-a (ERa) and, to a lesser extent, ERb could be demonstrated in the mouse uterus (37), the question arises of whether E2 directly regulates the expression of cx26 as well as that of cx43 as a result of interactions between the dimers of the ligand-bound ER and specific DNA sequences (EREs). It is known that the cx43 gene is hormonally regulated in the myometrium (21– 23, 38, 39), and the sequences of the 59-flanking region of the mouse, rat, and human cx43 genes have been reported (26, 40, 41). The putative promoter regions of the cx43 genes do not contain full EREs, but have putative ERE half-palindromic sites (26, 27). Functional promoter studies revealed that the luciferase activity of a cx43 promoter-luciferase con-


FIG. 6. Northern blot of endometrial RNA of ovariectomized rats treated with cycloheximide (CH) and/or E2 for 4 h probed for cx26 and c-fos. Application of CH alone or in combination with E2 did not induce cx26 expression in the endometrium, whereas application of E2 alone significantly increased cx26 transcripts (P , 0.05). Expression of the transcription factor c-fos was significantly increased by E2 independent of CH application (P , 0.05).

struct was up-regulated by E2 in HeLa cells when cotransfected with the ER. However, until now no binding of the ER to the half-palindromic EREs in the promoter of the cx43 has been evidenced, and a direct effect of the ER on the cx43 gene has not been proved. Piersanti and Lye (42) demonstrated that an E2-induced increase in transcription of cx43 in the rat myometrium was mediated directly through newly synthesized trans-activating factors. Using the protein synthase inhibitor cycloheximide, we showed that the action of E2 on cx26 expression in the endometrium is also not direct, but requires newly synthesized transcription factors. It is known that E2 may act indirectly by inducing the synthesis of nuclear transcription factors and that the expression of c-jun, c-fos, and c-myc can be increased by E2 in the rat uterus (43– 45). c-jun and c-fos are products of immediate early genes and induce the expression of “later” genes that contain activating protein-1 (AP-1) sites (46). An E2-mediated induction of c-jun and c-fos was previously shown in endometrium (47, 48), and their induction in the myometrium was suggested to be involved in cx43 transcription (27, 32, 49). This was supported by the identification of AP-1 sites in the promotor of the cx43 gene of the rat (26, 49), human (27, 41), and mouse (40, 49, 50). The evidence of a functional AP-1 site in the human cx43 promoter (27) amplifies the possibility that transcriptional regulation of the cx43 gene by steroid hormones

2515

may involve the Fos/Jun transcription complex. Expression of c-jun and c-fos was increased by E2 early before the increase in cx43 mRNA and shortly before induction of cx26. This could point to an involvement of these transcription factors not only in the induction of cx43 but also in the regulation of cx26 expression. However, until now neither an AP-1binding site nor an ERE could be identified in the putative promoter region of the mouse (51) and rat (52) cx26 gene within 500 bp upstream of the exon 1. The different time courses of induction of the two genes point to different regulatory mechanisms and may reflect heterogeneity in the types of transcriptional elements present within the two genes. Different regulations of these connexins has already been observed. Orsino and co-workers (53) showed that P had opposite effects on the expression of cx43 and cx26 in the myometrium shortly before delivery. In contrast to cx43, whose expression is low throughout pregnancy but increases immediately before the onset of labor, the expression of cx26 increases during the third trimester of pregnancy and falls to low levels before the onset of labor. Corresponding to these results, cx26 expression was also elevated in the endometrial epithelium of the rat shortly before parturition (21). The different time course of E2-mediated induction of cx26 and cx43 as well as the missing AP-1 site in the putative cx26 promoter point to different regulative mechanisms for cx26 and cx43, respectively, which may be related to different functions of these connexins during periimplantation in the rat. The role of connexin expression in the rat endometrium during periimplantation is speculative at this stage. It is known that channels composed of cx43 are important for electrical coupling in myometrium (54). Cx26 channels, in contrast, are found in nonexcitable cell types, such as hepatocytes (28), cells of the chochlea (55), uterine epithelial cells (9, 21), and parts of the rodent placenta (33), where they are thought to contribute to metabolic coupling by mediating the transfer of glucose between the trophoblast layers forming the placental barrier (56). This is supported by the finding that cx26 gene-deficient mice die in utero on day 10.5 of pregnancy due to impaired glucose uptake in the placenta as a defect in feto-maternal exchange (16). The data presented here demonstrate that the expression of cx26 and cx43 in the rat endometrium is sensitively and differentially regulated by the ovarian steroid hormones E2 and P. The differences in E2-mediated regulation of the two different connexins may be related to different stages of cellular differentiation and function, i.e. the immediate responsiveness of cx26 in the uterine epithelium to a blastocyst signal (9). To investigate the cell-specific regulatory mechanisms of the cx26 gene, further promoter studies should help to identify transcription factors responsible for the regulation of this gene. Acknowledgments The authors thank Dr. Alan Wakeling (Zeneca Pharmaceuticals) for providing ICI 182780, Gabriele Luhn and Georgia Rauter for excellent technical assistance, and Dave Kittel for preparation of illustrations.

2516

DIFFERENT HORMONAL RESPONSE OF cx26 AND cx43 References

1. Wilcox AJ, Weinberg CR, O’Connor JF, Baird DD, Schlatterer JP, Canfield RE, Armstrong EG, Nisula BC 1988 Incidence of early loss of pregnancy. N Engl J Med 28:189 –194 2. Roberts RM, Cross JC, Leaman DW 1992 Interferons as hormones of pregnancy. Endocr Rev 13:432– 452 3. Psychoyos A 1973 Endocrine control of egg implantation. In: Greep RO, Astwood EB, Geiger SR (eds) Handbook of Physiology. American Physiological Society, Washington DC, pp 187–215 4. Glasser SR, Clark JH 1975 A determinant role for progesterone in the development of uterine sensitivity to decidualization and ovo-impantation. In: Merkert CL, Papaconstantinos J (eds) The 33rd Symposium of the Society for Developmental Biology: The Developmental Biology of Reproduction. Academic Press, New York, pp 311–354 5. Psychoyos A, Prapas I 1987 Inhibition of egg development and implantation in rats after post-coital administration of the progesterone antaginist RU 486. J Reprod Fertil 80:487– 491 6. Barkai U, Kidron T, Kraicer PF 1992 Inhibition of decidual induction in rats by clomiphene and tamoxifen. Biol Reprod 46:733–739 7. Lessey BA, Castelbaum AJ, Sawin SJ, Sun J 1995 Integrins as markers of uterine receptivity in women with primary unexplained infertility. Fertil Steril 63:535–542 8. Stewart CL, Kaspar P, Brunet LJ, Bhatt H, Gadi I, Ko¨ntgen F, Abbondanzo SJ 1992 Blastocyst implantation depends on maternal expression of leukaemia inhibitory factor. Nature 359:76 –79 9. Winterhager E, Gru¨mmer R, Jahn E, Willecke K, Traub O 1993 Spatial and temporal expression of connexin 26 and connexin 43 in rat endometrium during trophoblast invasion. Dev Biol 157:399 – 409 10. Gru¨mmer R, Chwalisz K, Mulholland J, Traub O, Winterhager E 1994 Regulation of connexin26 and connexin43 in rat endometrium by ovarian steroid hormones. Biol Reprod 51:1109 –1116 11. Jahn E, Classen-Linke I, Kusche M, Beier HM, Traub O, Gru¨mmer R, Winterhager E 1995 Expression of gap junction connexins in the human endometrium throughout the menstrual cycle. Hum Reprod 10:2666 –2670 12. Bruzzone R, White TW, Paul DL 1996 Connections with connexins: the molecular basis of direct intercellular signaling. Eur J Biochem 238:1–27 13. Kumar NM, Gilula NB 1996 The gap junction communication channel. Cell 84:381–388 14. Willecke K 1993 The mouse connexin gene family. In: Hall JE, Zampighi GA, Davis RM (eds) Progress in Cell Research 3. Elsevier, Amsterdam, pp 33–37 15. Reaume AG, de Sousa PA, Kulkarni S, Langille BL, Zhu D, Davies TC, Juneja SC, Kidder GM, Rossant J 1995 Cardiac malformation in neonatal mice lacking connexin43. Science 267:1831–1834 16. Gabriel HD, Jung D, Bu¨tzler C, Temme A, Traub O, Winterhager E, Willecke K 1998 Transplacental uptake of glucose is decreased in embryonic lethal connexin26 deficient mice. J Cell Biol 140:1453–1461 17. Nelles E, Bu¨tzler C, Jung D, Temme A, Gabriel HD, Dahl U, Traub O, Stu¨mpel F, Jungermann K, Zielasek J, Toyka KV, Dermietzel R, Willecke K 1996 Defective propagation of signals generated by sympathetic nerve stimulation in the liver of connexin32-deficient mice. Proc Natl Acad Sci USA 93:9565–9570 18. Simon AM, Goodenough DA, Li E, Paul DL 1997 Female infertility in mice lacking connexin 37. Nature 385:525–529 19. Gong X, Li E, Klier G, Huang Q, Wu Y, Lei H, Kumar NM, Horwizt J, Gilula NB 1997 Disruption of a3 connexin gene leads to proteolysis and cataractogenesis in mice. Cell 91:833– 843 20. MacKenzie LW, Garfield RE 1985 Hormonal control of gap junctions in the myometrium. Am J Physiol 248:C296 –C308 21. Risek B, Guthrie S, Kumar N, Gilula NB 1990 Modulation of gap junction transcripts and protein expression during pregnancy in the rat. J Cell Biol 110:269 –282 22. Petrocelli T, Lye S 1993 Regulation of transcripts encoding the myometrial gap junction protein, connexin-43, by estrogen and progesterone. Endocrinology 133:284 –290 23. Lye SJ, Nicholson BJ, Mascarenhas M, MacKenzie L, Petrocelli T 1993 Increased expression on connexin-43 in the rat myometrium during labour is associated with an increase in plasma estrogen-progesterone ratio. Endocrinology 132:2380 –2386 24. Winterhager E, Stutenkemper R, Traub O, Beyer E, Willecke K 1991 Expression of different connexin genes in rat uterus during decidualization and at term. Eur J Cell Biol 55:133–142 25. Risek B, Gilula NB 1991 Spatiotemporal expression of three gap junction gene products involved in fetomaternal communication during rat pregnancy. Development 113:165–182 26. Yu W, Dahl G, Werner R 1994 The connexin43 gene is responsive to estrogen. Proc R Soc Lond ]B[ 255:125–132 27. Geimonen E, Jiang W, Ali M, Fishman GI, Garfield RE, Andersen J 1996 Activation of protein kinase C in human uterine smooth muscle induces connexin-43 gene transcription through an AP-1 site in the promoter sequence. J Biol Chem 271:23667–23674 28. Zhang JT, Nicholson B 1989 Sequence and tissue distribution of a second

29. 30. 31. 32. 33.

34. 35. 36. 37.

38. 39. 40. 41. 42.

43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56.

Endo • 1999 Vol 140 • No 6

protein of hepatic gap junctions, cx26, as deduced from its cDNA. J Cell Biol 109:3391–3401 Beyer EC, Paul DL, Goodenough DA 1987 Connexin43: a protein from rat heart homologous to a gap junction protein from liver. J Cell Biol 105:2621–2629 Paul DL 1986 Molecular cloning of cDNA for rat liver gap junction protein. J Cell Biol 103:123–134 Loose-Mitchell DS, Chiappetta C, Stancel GM 1988 Estrogen regulation of c-fos messenger ribonucleic acid. Mol Endocrinol 2:946 –951 Chiappetta C, Kirkland JL, Loose-Mitchell DS, Murthy L, Stancel GM 1992 Estrogen regulates expression of the jun family of protooncogenes in the uterus. J Steroid Biochem Mol Biol 41:113–123 Reuss B, Hellmann P, Dahl E, Traub O, Butterweck A, Gru¨mmer R, Winterhager E 1996 Connexins and E-cadherin are differentially expressed during trophoblast invasion and placenta differentiation in the rat. Dev Dynam 205:172–182 Traub O, Look J, Dermietzel R, Bru¨mmer F, Hu¨lser D, Willecke K 1989 Comparative characterization of the 21-kD and 26-kD gap junction proteins in murine liver and cultured hepatocytes. J Cell Biol 108:1039 –1051 Traub O, Willecke K, Breuer I, Stachewsky M 1992 Changes in expression of three different connexins in organs and tissue during mouse embryonic development. Eur J Cell Biol 57:36 – 81 Risek B, Klier FG, Phillips A, Hahn DW, Gilula NB 1995 Gap junction regulation in the uterus and ovaries of immature rats by estrogen and progesterone. J Cell Sci 108:1017–1032 Couse JF, Lindzey J, Grandien K, Gustafsson JA, Korach KS 1997 Tissue distribution and quantitative analysis of estrogen receptor-a (ERa) and estrogen receptor-b (ERb) messenger ribonucleic acid in the wild type and ERalphaknockout mouse. Endocrinology 138:4613– 4621 Garfield RE, Blennerhassett MG, Miller SM 1988 Control of myometrial contractility: role and regulation of gap junctions. Oxf Rev Reprod Biol 10:436 – 490 Hendrix EM, Myatt L, Sellers S, Russell PT, Larsen WJ 1995 Steroid hormone regulation of rat myometrial gap junction formation: effects of cx43 levels and trafficking. Biol Reprod 52:547–560 Sullivan R, Ruangvoravat C, Joo D, Morgan J, Wang BL, Wang XK, Lo CW 1993 Structure, sequence and expression of the mouse cx43 gene encoding connexin43. Gene 130:191–199 DeLeon JR, Buttrick PM, Fishman GI 1994 Functional analysis of the connexin43 gene promoter in vivo and in vitro. J Mol Cell Cardiol 26:101–111 Piersanti M, Lye SJ 1995 Increase in messenger ribonucleic acid encoding the myometrial gap junction protein, cx43, requires protein synthesis and is associated with increased expression of the activator protein-1, c-fos. Endocrinology 136:3571–3578 Murphy LJ, Murphy LC, Friesen HG 1987 Estrogen induction of N-myc and c-myc proto-oncogene expression in the rat uterus. Endocrinology 120:1882–1888 Weisz A, Bresciani F 1988 Estrogen induces expression of c-fos and c-myc protooncogenes in rat uterus. Mol Endocrinol 2:816 – 824 Webb DK, Moulton BC, Khan SA 1990 Estrogen induced expression of the c-jun proto-oncogene in the immature and mature rat uterus. Biochem Biophys Res Commun 168:721–726 Karin M 1995 The regulation of AP-1 activity by mitogen-activated protein kinases. J Biol Chem 270:16483–16486 Boettger-Tong HL, Murthy L, Stancel GM 1995 Cellular pattern of c-fos induction by estradiol in the immature rat uterus. Biol Reprod 53:1398 –1406 Morishita S, Niwa K, Ichigo S, Hori M, Murase T, Fujimoto J, Tamaya T 1995 Overexpressions of c-fos/jun mRNA and their oncoproteins (Fos/Jun) in the mouse uterus treated with three natural estrogens. Cancer Lett 97:225–231 Lefebvre DL, Piersanti M, Bai XH, Chen ZQ, Lye SJ 1995 Myometrial transcriptional regulation of the gap junction gene, connexin-43. Reprod Fertil Dev 7:603– 611 Chen ZQ, Lefebvre D, Bai XH, Reaume A, Rossant J, Lye SJ 1995 Identification of two regulatory elements within the promoter region of the mouse connexin 43 gene. J Biol Chem 270:3863–3868 Hennemann H, Kozjek G, Dahl E, Nicholson B, Willecke K 1992 Molecular cloning of mouse connexins26 and -32: similar genomic organization but distinct promoter sequences of two gap junction genes. Eur J Cell Biol 58:81– 89 Gru¨mmer R, Stro¨bl B, Traub O, Winterhager E 1998 Regulation of cx26 and cx43 transcripts by progesterone and 17-b-estradiol. In: Werner R (ed) Gap Junctions. IOS Press, Amsterdam, pp 326 –329 Orsino A, Taylor CV, Lye SJ 1996 Connexin-26 and connexin-43 are differentially expressed and regulated in the rat myometrium throughout late pregnancy and with the onset of labor. Endocrinology 137:1545–1553 Miyoshi H, Boyle MB, MacKay LB, Garfield RE 1996 Voltage-clamp studies of gap junctions between uterine muscle cells during term and preterm labor. Biophys J 71:1324 –1334 Kikuchi T, Kimura RS, Paul DL, Adams JC 1995 Gap junctions in the rat cochlea: immunohistochemical and ultrastructural analysis. Anat Embryol 191:101–118 Takata K, Kasahara T, Kasahara M, Ezaki O, Hirano H 1994 Immunolocalization of glucose transporter GLUT1 in the rat placental barrier: possible role of GLUT1 and the gap junction in the transport of glucose across placental barrier. Cell Tissue Res 276:411– 418