| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Articles |
Department of Obstetrics, Gynecology, and Reproductive Medicine, State University of New York School of Medicine, Stony Brook, New York 11794-8091
Address all correspondence and requests for reprints to: Dr. Janet Andersen, Department of Obstetrics, Gynecology, and Reproductive Medicine, State University of New York School of Medicine, Health Science Center, Stony Brook, New York 11794-8091. E-mail: jandersen{at}mail.som.sunysb.edu
Abstract
Just previous to the onset of parturition, a number of genes such as the one that codes for connexin-43 (Cx43) gap junction protein are induced in the myometrium. We have shown previously that activation of protein kinase C in human myometrial cultured cells leads to an up-regulation of cx43 transcription through an activating protein-1 element in the 5'-flanking promoter. Analyses were now performed on extracts of term myometrial tissue to test for an association between the up-regulation of cx43 expression and the expression of transcription factors and steroid hormone receptors that might regulate cx43 expression at term.
Immunoblot analyses were performed on extracts of term myometrial tissue from women receiving elective or indicated cesarean sections to test for an association between the up-regulation of cx43 expression and the up-regulation of expression of the transcription factors c-Jun, c-Fos, and Sp1, which have cognate binding elements in the cx43 5'-flanking promoter. Immunoblot analysis, immunohistochemistry, and receptor binding assays were also performed to analyze the levels of progesterone receptors (PR) and estrogen receptors (ER) in the same term myometrial tissue, and these were compared to the levels in nonpregnancy myometrial tissue.
The levels of PR were consistently 2- to 3-fold higher in term myometrial tissue than in nonpregnancy values and did not fluctuate during the menstrual cycle as did ER levels. Surprisingly, in term myometrium, ER was barely detectable by immunoblot and had whole cell diffuse staining by immunohistochemistry. In addition, very low levels of estrogen binding were observed in the term myometrial tissue. Treatment of primary myometrial cultures containing ER with estrogen for 3 or 48 h did not result in up-regulation of c-Jun or c-Fos proteins or in trans-activation from the proximal cx43 promoter with the activating protein-1 element. In contrast, an activated form of c-Jun protein was 10- to 18-fold higher in term myometrial tissue that also had elevated cx43 expression compared to c-Jun levels in term myometrial tissue with low cx43 expression. Likewise, c-Fos and Sp1 levels were 2–4 fold higher in term myometrial tissue with elevated cx43 expression. Although c-Fos and Sp1 proteins could be detected by immunoblot in myometrial tissue from nonpregnant women, c-Jun and Cx43 proteins could not.
In summary, these results suggest that up-regulation of human myometrial cx43 gene expression at term involves induction of primarily c-jun expression through a mechanism that does not directly involve myometrial ER or the loss of PR. Peptide hormones that activate protein kinase cascades, such as the protein kinase C cascade, may be important to signal the onset of labor in humans.
PRETERM birth occurs in 10% of births in the United States and is a leading problem in obstetrics. The hormone signals initiating labor in humans are still poorly understood, so intervention to arrest preterm labor is limited. Uterine smooth muscle, called myometrium, is the principal tissue of labor. Just previous to the onset of parturition, connexin-43 (Cx43) gap junctions form in the myometrium, promote intercellular communication, and are believed to synchronize labor contractions through the passage of molecules such as Ca2+ (1-4). Early cx43 expression is associated with premature labor (5, 6, 7). The cx43 gene is just one of the labor-associated genes whose expression rises dramatically in the myometrium previous to labor (8, 9). Regulation of its expression probably reflects the regulation of other labor-associated genes.
The steroid hormones, estrogen and progesterone have been implicated in the regulation of myometrial cx43 gene expression based on studies in pregnant animals (6, 8, 9, 10, 11). Progesterone contributes to the scarcity of myometrial gap junctions during pregnancy (10, 11, 12). Progesterone suppresses cx43 gene transcription and inhibits the trafficking of Cx43 protein through the Golgi apparatus (11, 12). On the other hand, chronic administration of estrogen to pregnant animals can induce premature gap junction formation in the myometrium and lead to labor (6). An increase in estrogen serum levels committal with a sharp decrease in progesterone serum levels has been connected with up-regulation of cx43 expression in pregnant nonprimate mammals (13, 14). However, in humans, progesterone levels in serum and amniotic fluid do not decrease at the end of pregnancy (15). This suggests that the formation of myometrial gap junctions in humans and animals at term may have different regulations.
In addition to the steroid hormones, the human placenta produces a number of peptide hormones that appear to mediate pregnancy maintenance and may contribute to the onset of labor by affecting myometrial gene expression (16, 17). Peptide hormones bind to specific membrane-bound receptors and initiate events that lead to a cascade of intracellular signalings called signal transduction. The initial signals are mediated through secondary messengers that activate specific protein kinases, such as protein kinase C (PKC). The protein kinases trigger a cascade of phosphorylation events leading to the activation of a variety of transcription factors. Activated transcription factors subsequently bind cognate cis-acting elements in promoters and regulate gene expression.
The 5'-flanking promoter region of the human cx43 gene contains a canonical activating protein-1 (AP-1) site (1). The AP-1 element is the cognate binding site for the early immediate transcription factors Jun and Fos that mediate transcriptional regulation after stimulation of protein kinase cascades (16, 17, 18, 19). Activation of PKC transiently increased cx43 gene transcription in myometrial primary cultures. An AP-1 site in the proximal promoter was sufficient for the increased expression, and c-Jun and c-Fos were shown to be involved (1). This work suggested that up-regulation of cx43 expression in human myometrium at parturition may involve increased expression and activation of c-Jun and c-Fos through the PKC cascade.
To corroborate the biological relevance of the in vitro study, the present study examined the levels of factors that may regulate cx43 expression in term human myometrium. The relative levels of transcription factors c-Jun, c-Fos, and Sp1, which have cognate binding elements in the 5'-flanking human cx43 promoter (1, 20), were determined and coordinated with the levels of Cx43 protein. Also, as it had been suggested that the levels of progesterone receptor (PR) may decline in human myometrium just before the onset of labor (21), the relative levels of PR and estrogen receptors (ER) were determined in the term myometrium. The results support the concept that c-Jun transcription factor may be instrumental in the induction of cx43 expression at term. It also shows that the onset of labor in humans does not involve the loss of myometrial PR and may be independent of myometrial ER.
Materials and Methods
Tissue acquisition
Approval to use human tissue was given by Stony Brook University
Hospital’s committee on research involving human subjects (no.
95-1224). Human myometrial tissue from nonpregnant and postpartum women
was obtained from surplus pieces of surgical specimens after
hysterectomies. Uterine leiomyoma tissue and endometrial tissue were
obtained in the same manner. Nonpregnancy myometrial tissue was taken
from women at various times during their menstrual cycle at the time of
surgery (see Table 1
). After patients
gave written consent, full thickness strips of term myometrial tissue
were obtained by excision of the upper margin of the transverse uterine
incision during elective and indicated cesarean section (C-section; see
Table 2). The myometrial and other
tissues were snap-frozen and reserved in liquid N2.
|
|
Cell culture system
For cell culture experiments, myometrial tissue was processed and cultured as described by Zhao et al. (11) and Andersen et al. (23). The cultures were never passaged and were used within 2 weeks of the initial plating. The source of the tissue dictated and generally limited the number of culture dishes that could be prepared for analysis (generally 10–20 culture plates for 1 experiment).
Twenty-four hours before the experiments, the culture medium was changed to serum-free, phenol red-free (24) DMEM supplemented as previously described (25). All experiments with the primary cultures were performed using this defined serum-free medium. Ethinyl estradiol (eE2) was added to the cultures at a concentration of 10 nmol/L for 3 or 48 h. 12-o-Tetradecanoylphorbol-13-acetate (TPA; Calbiochem-Novabiochem Corp., La Jolla, CA) was added to the cultures at a concentration of 100 ng/mL. Control samples were treated with vehicle (0.1% ethanol and 0.03% Me2SO4). Cell lysates were prepared for immunoblot and analyzed as previously described (23, 25).
Immunoblot analysis
The relative levels of PR (112 kDa), ER (67 kDa), c-Jun (39- and 46-kDa forms), c-Fos (61 kDa), Sp1 (106 kDa), and desmin, a muscle-specific intermediate filament protein (55 kDa) were determined in each of the tissue lysates through immunoblot analysis. Preparation of cell lysates and immunoblot analysis using enhanced chemiluminescence (Amersham Life Science, Aylesbury, UK) were performed as previously described (23, 25). Prestained marker proteins (Bio-Rad, Richmond, CA) were run in parallel. HeLa cell lysates were used as a negative control for ER and PR immunoblot analyses. The resulting autoradiographs were scanned using the Bio-Rad laser densitometer and analyzed by density volume using the Molecular Analysis Program (Bio-Rad Laboratories, Hercules, CA). The volumes of ER, PR, c-Jun, c-Fos, and Sp1 proteins were normalized to the volume of desmin for each sample.
One of two antibodies purchased from Santa Cruz Biotechnology (Santa Cruz, CA) was used to detect ER in each analysis: one mouse monoclonal antibody (mAb) recognizing amino acids within the amino-terminus of human ER (catalogue no. sc-786) and one rabbit polyclonal antibody (pAb) recognizing amino acids at the carboxyl-terminus (catalogue no. sc-543). To detect PR, a pAb purchased from Santa Cruz Biotechnology (catalogue no. sc-539) was used recognizing amino acids in the carboxyl-terminus of human PR. The antidesmin mAb (purchased from Sigma Chemical Co., St. Louis, MO; diluted 1:10,000) was used to detect desmin. The mouse mAb used to detect Cx43 protein was purchased from Zymed Laboratories (San Francisco, CA). Antibodies specific for c-Fos (sc-52), c-Jun (sc-822), Jun B (sc-73), and Sp1 (sc-59) proteins were obtained from Santa Cruz Biotechnologies. The levels of c-Jun and c-Fos were analyzed, because these forms of Fos and Jun predominate in myometrial tissue (1, 26, 27, 28). JunB protein was not detected by immunoblot of the c-section lysates (data not shown).
Immunohistochemical detection of PR and ER
Immunohistochemical detection of PR and ER was performed with the same primary polyclonal antibodies to PR (sc-539) and ER (sc-543) as those used for the immunoblot. All incubations and washes were performed at room temperature. Tissue sections were fixed in cold acetone, rinsed in phosphate-buffered saline (PBS; 0.2 g/L KCl, 2.0 g/L KH2PO4, 8.0 g/L NaCl, and 2.2 g/L Na2HPO4-7H2O, pH 7.4), and blocked for 20 min with rabbit serum diluted in PBS (1:67). After draining excess blocking solution, the sections were incubated for 30 min with primary antibody diluted 1:100 in a solution containing BSA (Sigma A-3350). The sections were subsequently washed in PBS, incubated for 30 min with a second antibody (biotin-labeled antirabbit IgG) diluted 1/100 in PBS, and washed again in PBS. Sections were then incubated for 30 min with diluted Vectastain reagent (Vector Laboratories, Burlingame, CA) according to the manufacturer’s instructions, washed in PBS, and incubated with peroxidase substrate (Vector Laboratories) for 30 min. The slides were then rinsed in PBS and incubated with diaminobenzidene-peroxide chromogen solution for 3 min. The reaction was stopped by incubation with deionized water, and the sections were drained, dehydrated, and mounted. Peroxidase activity was observed by light microscopy using a x25 objective. (Initial sections were also counterstained with light green using standard protocols.) Duplicate sections were stained with hematoxylin and eosin using standard protocols to facilitate identification of cellular elements and areas of necrosis (data not shown). Evaluation of results was performed by a comparison of staining in tissue sections that were previously determined to have ER and PR and by a comparison of staining in duplicate sections that were incubated with PBS not containing the first antibody.
Estrogen binding assay
After washing with PBS, tissue sections were incubated at room temperature for 2 h with Fluoro-Cep estrogen solution containing 17ß-estradiol-6-carboxymethyloxine-bovine serum-fluorescein isothiocyanate conjugate (Zeus Scientific, Raritan, NJ). The sections were subsequently washed twice in PBS for 30 min each time at room temperature. Coverslips were applied using 50% glycerol in PBS. Slides were stored in the dark at 4 C. Duplicate sections were also stained with hematoxylin and eosin using standard protocols (data not shown).
Within 24 h, the results were viewed using a x25 objective and photographed using a confocal fluorescence microscopy equipped with the appropriate filters for excitation of fluorescein. Positive staining was determined by the intensity of fluorescence relative to the intensity of the proliferative endometrial section used as a positive control. Photographic settings were determined for the endometrial section and used for all samples.
Transient expression assays
The primary myometrial cultures were transfected with 20 µg reporter plasmid and 20 µg herring sperm DNA/10-cm plate using the calcium phosphate precipitation method as previously described (25). A reporter plasmid, pCXHB-chloramphenicol acetyltransferase3 (CAT3), containing a portion of the 5'-flanking cx43 promoter with the proximal AP-1 site upstream of the bacterial CAT gene (pBL-CAT3) (1) was used in transient expression assays. A cryptic AP-1 site in the backbone of vectors was deleted previously (29). Parallel cultures were transfected with a plasmid containing an estrogen response element (ERE) and thymidine kinase (TK) promoter elements upstream of the CAT reporter gene, pERE/TK/CAT (30), as a control to measure the transcriptional response of the cultures to estrogen. In some experiments, cultures were cotransfected with one of two expression vectors for human ER, pHER-wt (31) or pHEO (32), at 0.5 µg/10-cm plate. Transfected cells were treated with either 10 nmol/L eE2 or hormone vehicle for 48 h before obtaining cell lysates. Lysates were normalized for protein concentration and analyzed for CAT activity as described previously (25).
Statistical analysis
Statistical analysis of the results was made with GraphPad’s Instat computer software for small sample sets using two-tailed Student’s t test.
Results
Levels of PR and ER in human term myometrium
Estrogen and progesterone serum levels rise during human
pregnancy, and both remain elevated at term. This raises questions
concerning the levels of ER and PR in human myometrium at term.
Immunoblot analysis was used to determine the levels of PR in
myometrial tissue from nonpregnant women (see Table 1
for nonpregnant
patient profiles) and from woman undergoing elective and indicated
C-section (see Table 2 for pregnant patient profiles). We have
previously shown that in human myometrium, ER levels rise during the
follicular phase until the midluteal phase and then fall during the
luteal phase (25). Table 1 lists the relative levels of ER reported for
each tissue. The same tissue was analyzed for PR levels using
immunoblot analysis (Fig. 1A) and, for
each sample, normalized with the levels of desmin protein. In contrast
to ER, myometrial PR levels remained constant during the menstrual
cycle (Fig. 1A).
|
). Compared to the relative
levels in nonpregnancy myometrial tissue, the levels of PR were 2- to
3-fold higher in the term myometrial tissue after normalization to
protein concentrations (Fig. 1B, top panel). In contrast, ER
levels were barely detectable by immunoblot in the term myometrial
tissue (Fig. 1B, bottom panel). The results were the same
using a pAb to the carboxyl-terminus of ER and a pAb to amino-terminus
of ER (data not shown). ER levels were barely detectable in all term
myometrial tissue tested (see Table 2; not all data are shown).
Because ER was barely detectable in the term myometrial tissue by
immunoblot, an immunohistochemical method was used as an alternative
method to detect ER in the same tissue (see Table 2 and Fig. 2). The tissue was incubated with the
same anti-ER pAb as that used in the immunoblot analyses. Clear nuclear
staining was observed in the nonpregnancy myometrial tissue and the
uterine leiomyoma tissue (Fig. 2, A, C, and D). However, all term
myometrial tissue displayed light, whole cell, diffuse staining (Fig. 2, F–H), which was darker than the no first antibody control (Fig. 2E). In contrast, when the tissue was incubated with an antibody
against PR, both nonpregnancy (Fig. 3C)
and term myometrial tissue displayed nuclear staining (Fig. 3D, see
arrows). The nuclear staining observed for PR in the term
myometrial tissue also demonstrates that the diffuse whole cell
staining that was observed for ER was not caused by a lack of tissue
section integrity.
|
|
). Proliferative endometrium (35) was
included as a control to calibrate positive ER staining (Fig. 4A).
Nonpregnancy myometrial tissue showed positive staining throughout the
section (Fig. 4B). C-section and postpartum myometrial tissue showed
spotty light staining (Fig. 4, C and D, respectively), indicating low
estrogen binding in the term myometrial tissue. The areas of
fluorescence observed in the term myometrial tissue correspond to the
areas of darkness observed in the term myometrial tissue in the
immunohistochemical analysis (compare Fig. 4, C and D, with
corresponding tissue in Fig. 3, F and I). These results again confirm
the results from the immunoblot and immunohistochemical staining
showing that term myometrial tissue has low levels of ER and provide
evidence that an alternative type of ER is not expressed at high levels
in term myometrium.
|
Our previous work (1) suggested that induction of myometrial
cx43 expression at parturition involved induction and
activation of the transcription factors c-Jun and c-Fos that would bind
to a functional AP-1 site in the 5'-flanking cx43 promoter.
The 5'-flanking cx43 promoter also has a trimer of putative
Sp1 elements (1). Immunoblot analysis was used to measure the relative
levels of Cx43, c-Jun, c-Fos, Sp1, and desmin proteins in the same term
myometrial tissue samples used to analyze ER and PR levels. Figure 5A shows representative results after
incubation with antibodies against Cx43, c-Jun, c-Fos, and Sp1 proteins
(results for desmin are not shown). Two bands are detected for c-Jun
protein, 39-kDa (Fig. 5, bottom) and 46-kDa (top)
bands, which are two phosphorylated forms of the protein (36). The
numbers at the top of each lane designate the C-section
number listed in Table 2. After the autoradiographs were scanned, the
densities of the Cx43, c-Jun, c-Fos, and Sp1 bands were normalized to
the density of the desmin band for each sample (Fig. 5B). The results
were ordered by rising Cx43 protein levels. The samples that had the
highest levels of Cx43 protein also had elevated levels of c-Jun,
c-Fos, and Sp1 proteins. Of note, c-Jun levels were elevated 10- to
18-fold in samples that had elevated levels of Cx43 protein and c-Fos
and Sp1 proteins were 2- to 4-fold higher. In addition, in the
postpartum myometrial tissue (Fig. 5B, bars marked Pp), both
Cx43 and c-Jun protein levels were reduced from the those observed in
the C-section tissue from women in labor (Fig. 5B, bars
marked 5 and 7). These results suggest that the c-Jun transcription
factor may be a major player in the induction of cx43
expression at term in human myometrium.
|
Immunoblot analysis was used to determine whether estrogen would
induce the expression of Cx43, c-Jun, and c-Fos in primary cultures of
human myometrial cells and to analyze its effects on induction of
c-jun and c-fos expression after activation of
PKC (Fig. 6). The myometrial tissue
chosen to make the cultures was from women who were in the luteal phase
at the time of surgery for the analysis, as previous work established
that the primary cultures would contain functional ER (25). Cultures
were transferred to serum-free, phenol red-free DMEM and treated with
either hormone vehicle or 10 nmol/L eE2 for 2 days (Fig. 6, lanes marked 0 and eE2, respectively). Then half of the
cultures treated for 2 days with eE2 were also treated with
100 ng/mL TPA for 3 h (Fig. 6, lanes marked Both). Also, half the
vehicle-treated cultures were treated with 100 ng/mL TPA for 3 h
(Fig. 6, lanes marked TPA). Each condition was performed in duplicate
or triplicate for each set of cultures. The incubation for TPA was
chosen because in previous studies c-Jun and c-Fos protein levels rose
significantly within 3 h, and there was an observable rise in Cx43
protein levels. Two incubation times for eE2 were chosen.
The short term (3 h) pulse of eE2 was chosen to observe
estrogen’s effects on the expression especially of c-jun
and c-fos, whereas the long term eE2 treatment
(2 days) was chosen to observe estrogen’s effects on cx43
expression. The long term treatment was believed to reflect more
physiological conditions during pregnancy. The cells were harvested at
designated times, and the levels of Cx43, c-Jun, c-Fos, and desmin
proteins were analyzed by immunoblot (Fig. 6).
|
, compare lanes marked 0 with lanes marked TPA).
Scans of the autoradiograph confirmed the increases in the levels of
these proteins with TPA treatment to be significant (P
< 0.01 in each case using two-sided, unpaired, Student’s t
tests). Of importance, in cultures treated for 2 days with estrogen,
Cx43 protein levels increased significantly (6-fold; P
= 0.0154), but without any apparent increases in c-Fos and c-Jun
levels. In cultures that were treated for 3 h with
eE2, Cx43 proteins increased about 3-fold
(P = 0.0367; Fig. 6B), also without an increase in
c-Jun or c-Fos levels (data not shown). These results indicate that in
primary cultures of human myometrial cells, cx43 expression
was increased with the addition of estrogen, and this increase was
independent of the induction of c-Fos protein and c-Jun protein (39-
and 46-kDa forms). Sp1 protein levels also were not elevated (data not
shown).
Interestingly, in estrogen-treated cultures treated with TPA for 3
h, Cx43 protein levels did not increase. Also, although c-Jun protein
levels increased with the double treatment to the same level as in the
TPA-treated samples, c-Fos protein levels were significantly less (see
Fig. 6A, compare lanes marked TPA with lanes marked Both). The
estrogen- and TPA-induced increases in Cx43 protein levels were not
additive and appeared to be antagonistic.
Transient expression assays
Transient expression assays were performed to test whether the proximal AP-1 site in the 5'-flanking human cx43 promoter was involved in the up-regulation of myometrial cx43 expression after treatment with estrogen. Five sets of primary myometrial cultures and one set of leiomyoma cultures were transfected with pCXHB-CAT3 (1), which contains 406 bp of the 5-'-flanking cx43 promoter with the proximal AP-1 site. The transfected cultures were treated with estrogen (10 nmol/L) or 0.1% ethanol for 2 days. Cultures were cotransfected with an expression vector for ER [pHER-wt (31) or pHEO (32)] if they were not luteal cultures with endogenous ER (25). Treatment with estrogen did not increase transcription, as evidenced from the CAT activities in the cell lysates (data not shown). In contrast, when the same cultures were transfected with ERE/TK/CAT under the same conditions, active transcription was evident from CAT activity. We conclude that the sequences in the cx43 5'-flanking promoter that includes the proximal AP-1 site are not sufficient for transcriptional regulation by estrogen in our primary myometrial cultures. This also suggests that other members of Jun/Fos family members are not involved in the induction of cx43 expression by estrogen.
Discussion
Labor is the moment of glory for uterine smooth muscle. Cx43 gap junctions play a crucial role in the function of the myometrium during labor by propagating action potentials during muscle contractions. The cx43 gene is one of several genes that are induced just previous to labor. The present study approached the regulation of parturition onset in humans by examining the levels of the steroid hormone receptors and transcription factors that may regulate cx43 expression in human term myometrium and coordinating these with expression of the cx43 gene. The results do not support commonly held ideas concerning regulation of the initiation of labor in humans. These are discussed below.
Progesterone is necessary for the maintenance of pregnancy. For example, progesterone down-regulates cx43 transcription in human myometrial cells during pregnancy even in the presence of estrogen (11). The onset of labor in animals is associated with a decrease in serum progesterone levels (13, 14). As humans do not experience a decrease in serum progesterone as do animals (15), one possibility is that PR levels become limiting in human myometrium at the end of pregnancy. An examination of PR levels in term myometrium revealed, however, that PR levels remained elevated in term myometrium that also has high cx43 expression. A decrease in PR levels did not coordinate with an increase in cx43 expression. These data do not support the idea that progesterone signaling is lost in human myometrium just before labor through down-regulation of PR expression.
Although it is known that progesterone suppresses transcription of the human cx43 gene (11), the mechanism by which estrogen controls human cx43 expression has remained elusive. There has been much speculation about the role of estrogen in regulation of cx43 expression in human myometrium based on studies performed in pregnant animals (6, 13, 14, 37, 38). The generally held belief is that estrogen induces myometrial cx43 expression just before labor. An examination of ER levels in term myometrium using three different methods revealed that there are apparently very low levels of ER at term. In addition, low estrogen binding in term myometrium demonstrated that an alternate ER is not expressed at high levels. These results call to question the idea that myometrial ER plays a direct role in the induction of human cx43 expression at term. It does not rule out the possibility that estrogen plays a paracrine role, however, by influencing neighboring tissues to produce other labor-inducing factors or hormones.
It is evident from our work and that of others that in addition to the steroid hormones, other factors are involved in the regulation of myometrial cx43 gene expression (1, 38). These factors may be peptide hormones that activate myometrial protein kinase pathways. Activation of myometrial PKC was shown to increase the levels of c-Fos and c-Jun proteins, the factors that bind to AP-1 sites. Results from transient expression assays clearly demonstrate that the proximal AP-1 site in the human cx43 promoter is functional in inducing cx43 transcription in primary cultures of uterine smooth muscle cells upon activation of PKC. This work supports the novel concept that up-regulation of cx43 expression in human myometrium at parturition is in part induced by c-Jun and c-Fos binding to the AP-1 site after activation of PKC.
Expression of c-jun and c-fos is negligible in myometrial tissue from nonpregnant women (this study and Ref.39) and is further suppressed in the myometrium of women in late pregnancy (39). We are able to detect elevated levels of c-Jun and c-Fos proteins in term myometrial tissue that also have elevated levels of Cx43 protein. The increase in the levels of c-Jun protein is particularly impressive; it is 10- to 18-fold higher in term myometrial tissue with elevated levels of Cx43 protein than in term myometrial tissue with low levels of Cx43 protein. This supports the concept that c-Jun may be a major player in the induction of cx43 expression at the onset of parturition. This opens the possibility that the c-Jun transcription factor may have a role in the induction of the other labor-associated genes also.
In steroid-responsive tissues, cross-talk has been shown to occur between steroid hormone regulation and the regulation of AP-1 activity (40, 41, 42, 43, 44, 45, 46, 47, 48, 49). For example, expressions of c-fos and c-jun are regulated by estrogen in several estrogen-responsive tissues (27, 40, 41, 42). Transcription of the gene that codes for insulin-like growth factor I is regulated by estrogen through the AP-1 element (43). Additionally, the ovalbumin gene is coactivated by ER and the Jun/Fos complex (44). The presence of the AP-1 element in the human cx43 promoter suggested the possibility that estrogen induces cx43 expression by up-regulating the expression of c-jun and c-fos in human myometrium at term. This attractive hypothesis was addressed in our study, but the results do not support the concept. First, very low levels of ER are detected in term myometrium using three methods. Our study confirms the results of Pulkkinen and Hämäläinen (45) and Perrot-Applanat et al. (46), who also found low estrogen binding in human term myometrium, but high estrogen binding in hamster term myometrium (45). Second, trans-activation of the 5'-flanking promoter containing the proximal AP-1 element did not result after transfected human myometrial primary cells were treated with estrogen with and without cotransfection with an expression vector for ER (data not shown). Finally, Cx43 protein levels are increased if the primary myometrial cells were treated with estrogen or TPA, a reagent that activates PKC. We can account for the increase in Cx43 protein after activation of PKC by increased cx43 transcription through an AP-1 site in the promoter. The mechanism by which estrogen increases Cx43 protein levels is unknown. Nonetheless, when cells are treated with both estrogen and TPA, Cx43 protein levels do not increase. Estrogen induction of myometrial cx43 expression and induction after activation of PKC appear to be antagonistic to each other. One possibility is that ligand-bound ER and c-Fos proteins interact under these conditions and interfere with their respective functions in the up-regulation of cx43 expression (see below).
Jun and Fos have been shown to antagonize the function of two steroid receptors in other systems. c-Jun protein interferes with the transcriptional activity of the glucocorticoid receptor (GR) through a mechanism that is independent of DNA binding, suggesting a mechanism of protein-protein interaction (47, 48). GR loses the ability to bind to DNA and activate transcription. Also functional antagonism exists between ER and c-Fos through protein-protein interactions (49). The reciprocal antagonism leads to negative regulation of c-fos expression, as the gene contains both an AP-1 site and an ERE. Our results support this observation for the myometrium, as the elevation of c-Fos protein levels is partially inhibited after activation of PKC if the myometrial primary cells are treated with estrogen. Estrogen may interfere with TPA induction of cx43 expression through interfering with c-fos induction and c-Fos protein function. Perhaps low levels of ER are necessary in human term myometrium to prevent interference with the functions of c-Jun and c-Fos transcription factors after activation of myometrial PKC.
A number of peptide hormones are produced by the placenta, such as relaxin, GH, ß-endorphin, PRL, human placental lactogen, hCG, human chorionic corticotropin, and human chorionic TSH (16, 17). Many of these attain high serum levels during pregnancy and remain high to the end of pregnancy (17). These hormones direct the physiological changes in the uterus and the breast that are associated with pregnancy, stimulating growth and differentiation of these tissues. Progesterone, as the "pro-gestation" hormone (50), may maintain pregnancy by monitoring the stimulating effects of these and other hormones in the myometrium and by keeping the smooth muscle quiescent (50). Progesterone probably also plays a role in suppressing the physiological effects of stretch (38) on uterine smooth muscle that occurs especially during the third trimester. At the end of pregnancy, an internal switch may occur in the myometrial cells that alters progesterone’s ability to regulate myometrial quiescence. Our results indicate that the switch does not involve a loss of myometrial PR. Nonetheless, a switch is necessary to allow for activation of appropriate protein kinase cascades. Perhaps a threshold needs to be reached through synergy. Once the threshold is reached, the production of transcription factors, such as c-Jun protein, is stimulated and subsequently labor-associated genes, such as the cx43 gene, are induced. Labor may then follow. More work is needed to understand the interaction of progesterone in the protein kinase cascades in human myometrium.
Acknowledgments
We thank Dr. Ronald Jaekle for his initial help with setting up the study. We thank Judy Howell for help with the immunoblot analysis. We also thank David Colflesh for his excellent microscopy imaging.
Footnotes
1 This work was supported by NIH Grant HD-30482 (to J.A.). 
Received October 1, 1997.
Revised December 12, 1997.
Accepted December 18, 1997.
References
This article has been cited by other articles:
 |
|
 |
|
  B. F. Mitchell, B. Zielnik, S. Wong, C. D. Roberts, and J. M. Mitchell Intraperitoneal infusion of proinflammatory cytokines does not cause activation of the rat uterus during late gestation Am J Physiol Endocrinol Metab, October 1, 2005; 289(4): E658 - E664. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. E. Bethin, Y. Nagai, R. Sladek, M. Asada, Y. Sadovsky, T. J. Hudson, and L. J. Muglia Microarray Analysis of Uterine Gene Expression in Mouse and Human Pregnancy Mol. Endocrinol., August 1, 2003; 17(8): 1454 - 1469. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Cesen-Cummings, K. D. Houston, J. A. Copland, V. J. Moorman, C. L. Walker, and B. J. Davis Uterine Leiomyomas Express Myometrial Contractile-Associated Proteins Involved in Pregnancy-Related Hormone Signaling Reproductive Sciences, January 1, 2003; 10(1): 11 - 20. [Abstract] [PDF]
|
|
|
|
 |
|
 J. Condon, S. Yin, B. Mayhew, R. A. Word, W.E. Wright, J.W. Shay, and W. E. Rainey Telomerase Immortalization of Human Myometrial Cells Biol Reprod, August 1, 2002; 67(2): 506 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Knock, R. M. Tribe, A. A. Hassoni, and P. I. Aaronson Modulation of Potassium Current Characteristics in Human Myometrial Smooth Muscle by 17{beta}-Estradiol and Progesterone Biol Reprod, May 1, 2001; 64(5): 1526 - 1534. [Abstract] [Full Text]
|
|
|
|
|
|
J. A. Lenhart, P. L. Ryan, K. M. Ohleth, and C. A. Bagnell Expression of Connexin-26, -32, and -43 Gap Junction Proteins in the Porcine Cervix and Uterus During Pregnancy and Relaxin-Induced Growth Biol Reprod, December 1, 1999; 61(6): 1452 - 1459. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
