During pregnancy, stretch of the uterus, imposed by the growing fetus, is an important signal for the induction of genes involved in the onset of labor. In this study, the expression of activator protein-1 (AP-1) family mRNAs in response to in vitro stretch was investigated in myometrial cells. Rat primary myometrial smooth muscle cells were plated onto collagen I-coated Flex I culture plates and subjected to 25% static stretch on day 4 of culture. Static stretch induced an increase in the expression of c-fos, fosB, fra-1, c-jun, and junB. The expression of both c-fos and junB was maximally induced at 30 min by static stretch. The peak induction for fosB and c-jun occurred at 1 h, whereas the peak of fra-1 induction occurred between 1 and 2 h after application of stretch. Treatment of myometrial cells with progesterone (100 nM, 400 nM, 1 μM) for 1 or 6 h before the application of static stretch did not affect the magnitude of the c-fos response. However, 24 h of progesterone exposure reduced the magnitude of c-fos and fosB stretch induction at both the 400 nM and 1 μM doses. These data indicate that several members of the AP-1 family are stretch-responsive genes in myometrial smooth muscle cells. This response can be attenuated by pretreatment with progesterone; however, the requirement for longer pretreatment times suggests that the inhibitory actions of progesterone do not occur through a direct action of the progesterone receptor within the promoter regions of AP-1 genes.
activator protein-1 (AP-1) family members are immediate early genes that belong to the basic leucine zipper family of transcription factors. Members of the AP-1 family (c-Jun, JunB, JunD, c-Fos, Fra-1, Fra-2, and FosB) bind to the AP-1 site (TGAG/CTCA) as either homodimers of Jun family proteins or heterodimers of Fos and Jun family proteins, but not as homodimers of Fos family proteins (see review in Ref. 10). The dimer formed by the interaction of a Fos and Jun protein is more stable than a Jun/Jun dimer, leading to higher transcriptional activation (8, 14). Furthermore, the degree of DNA binding and transcriptional activation is dependent on the specific partners that make up the dimer complex; thus differential expression of AP-1 proteins has been suggested as a major mechanism of AP-1 specificity (25). We have previously reported that the myometrial expression of fos and jun family members is increased at the end of gestation (17). With use of a unilaterally pregnant rat model, it was shown that this increase occurred only in the gravid horn, indicating a potential role for stretch imposed by the growing fetus in the induction of AP-1 gene expression (17).
Before the onset of labor, there are dramatic changes in gene expression that occur in the myometrium. These include modulation in the expression of components of the extracellular matrix (ECM) and the coordinated increase in the expression of several contraction-associated proteins (18, 20–22, 28). These changes in myometrial gene expression lead to the production of coordinated contractions during labor. Interestingly, collagen III, elastin, fibronectin, and laminin, which make up the basement membrane, contain AP-1 sites in their promoter regions and may therefore be regulated by AP-1 proteins (2, 6, 23, 32). In addition, ECM remodeling enzymes, such as matrix metalloproteinases and tissue inhibitors of metalloproteinases, contain AP-1 sites within their promoter regions (see reviews in Refs. 4, 5, 33). The promoter regions of the contraction-associated proteins (CAP) genes connexin 43 (Cx43) and oxytocin receptor (OTR) also contain consensus AP-1 sites that mediate responsiveness to the phorbol ester TPA (12-O-tetradecanoylphorbol 13-acetate) in myometrial cells and MCF7 cells, respectively (1, 7). Several components of the ECM as well as Cx43 and OTR are regulated by both endocrine and mechanical signals in pregnant myometrium, raising the possibility that AP-1 transcription factors integrate and transduce the hormonal and mechanical signals involved in the onset of labor.
We have used an in vitro system to determine whether mechanical forces modulate the expression of AP-1 genes. In this system, freshly isolated primary rat myometrial cells are plated onto a flexible bottom plate and subjected to static stretch through the use of a computer-driven vacuum system. We have previously reported that in vitro mechanical stimulation results in a time- and force-dependent induction of c-fos mRNA that peaked after 30 min of static stretch (29). The magnitude of this c-fos response was dependent on the presence of specific ECM components. The greatest responses were obtained when the cells were plated onto collagen I, with laminin and elastin showing an intermediate response and pronectin, a synthetic fibronectin-like substrate, the lowest response (29). Static mechanical stretch of primary myometrial cells also stimulates the phosphorylation of the mitogen-activated protein kinases (MAPK), extracellular regulated kinase 1/2 (ERK1/2), c-Jun amino-terminal kinase (JNK), and p38. Preventing the activation or action of any of these kinases with the use of specific inhibitors was found to inhibit c-fos induction by mechanical stimulation, indicating that this induction is dependent on all three of the MAPK-signaling pathways (19).
In the pregnant rat myometrium, the expression of several of the AP-1 family mRNAs was found to be inhibited by progesterone treatment, which prevents the onset of labor in these animals (17). In nonpregnant rats, it has been shown that treatment with progesterone prevents the induction of c-fos mRNA by estrogen; however, the mechanisms that mediate this inhibition have not been characterized (12). These complex in vivo interactions of endocrine, paracrine, and mechanical signals at term in the myometrium complicate the analysis of the role of these signals in AP-1 gene expression. In this study, therefore, we have used an in vitro stretch system to further investigate the regulation of AP-1 genes by mechanical stretch in myometrial cells, the effect of progesterone pretreatment on this stretch response, and the involvement of the ERK MAPK pathway in the actions of progesterone.
MATERIALS AND METHODS
Primary myometrial smooth muscle cell isolation and culture.
Primary cultures of enriched uterine smooth muscle cells (SMCs) were prepared as previously described by us (28). Briefly, dissociated myometrial cells were collected by centrifugation (200 g for 15 min), and the cell pellet was resuspended in sterile, phenol red-free DMEM (GIBCO, Grand Island, NY) supplemented with 10% FBS, 25 mM HEPES, 100 U/ml of penicillin-streptomycin, and 2.5 μg/ml of amphotericin B. To enrich for uterine myocytes, the freshly isolated cell mixture was subjected to a differential attachment technique (11). SMC cells were plated on 6-well silicone elastomer Flex I culture plates coated with type I collagen at a density of 3 × 106 cells per well. The cells were grown to confluence in phenol red-free DMEM supplemented with 10% FBS, 25 mM HEPES, 100 U/ml of penicillin-streptomycin, and 2.5 μg/ml of amphotericin B and were stretched within 4 days. All animal experiments were approved by the Samuel Lunenfeld Research Institute animal care committee.
Stretch regimens and treatments.
Confluent myometrial SMCs were serum starved for 18–24 h starting on day 3 of culture. On day 4 of culture, cells were subjected to static mechanical deformation (25% elongation) with the Flexercell Strain Unit 4000 (Flexcell International, McKeesport, PA) for the indicated period of time. Nonstretched cells were cultured under identical conditions but were not subjected to mechanical stimulation. Cells exposed to progesterone for 1–6 h were serum starved on day 3, treated with either vehicle (EtOH) or 100 nM, 400 nM, or 1 μM progesterone (Sigma, St. Louis, MO) on day 4, and stretched on day 4 of culture after the appropriate period of exposure to progesterone. SMCs exposed to progesterone for 24 h were treated with vehicle or the appropriate amount of progesterone on day 3 of culture immediately after serum starvation and stretched 24 h later, on day 4 of culture. The doses of progesterone (100 nM and 400 nM) were chosen to mimic the peak levels of maternal serum progesterone during early and late rat pregnancy, respectively. Protein or RNA was extracted from SMCs as described in the following Western and Northern analysis sections.
Complementary DNA probes.
cDNA probes were obtained from the American Type Culture Collection for c-jun (63026), junD (95654), and fosB (63118). The following probes were provided by other research labs: rat c-fos cDNA (Dr. Curran, Roche Research Center, Nutley, NJ); mouse junB (Dr. Woodgett, Ontario Cancer Institute, Toronto, ON, Canada); 18S ribosomal protein (Dr. Denhardt, Rutgers University, Piscataway, NJ). Probes for rat fra-1 (GenBank M19651) and fra-2 (GenBank U18913) were generated by PCR using the following primers: fra-1 upper 5′ CCA GCA AGC GCA GAC ACA GAC, fra-1 lower 5′ CGG AGG AGG GGT CAC CAC TG; fra-2 upper 5′ ATC CCG GGA ACT TTG ACA CCT, fra-2 lower 5′ GGC TCT TCC CCG TAG AAA CCA.
RNA was extracted from SMCs in 3 wells of a 6-well plate with 1 ml of TRIzol (GIBCO-BRL, Grand Island, NY) according to the manufacturer's specifications. Purified RNA (5–10 μg) was then analyzed by Northern analysis. RNA was separated on 1% (wt/vol) agarose (GIBCO-BRL) gel, containing 3.7% (vol/vol) formaldehyde (J. T. Baker, Phillipsburg, NJ) in MOPS (Sigma), transferred in 0.1 M sodium phosphate (NaP, Sigma) onto a nylon membrane (GeneScreen, DuPont, NEN Research Products, Boston, MA), and cross-linked by UV irradiation. cDNA probes were labeled with [α-32P]dCTP (NEN Research Products) with the multiprime DNA labeling system (Amersham Biosciences, Little Chalfont, UK). Hybridization was conducted at 55°C in 30% formamide for 20 h according to the method described in Bio-Rad Bulletin 1110 (Bio-Rad Laboratories, Richmond, CA). Subsequently, the membrane was washed to a final stringency of 30 mM NaP-0.1% SDS (EM Science, Darmstadt, Germany). All RNA isolation and analysis were carried out in diethyl pyrocarbonate (DEPC; Sigma) water. Probed membranes were exposed to X-ray film (Kodak XAR, Eastman Kodak, Rochester, NY) with an intensifying screen at −70°C, normalized to 18S rRNA levels, and expressed as relative optical density (ROD) units analyzed by densitometry.
Protein was extracted from SMCs in 3 wells of a 6-well plate using 100 μl of RIPA lysis buffer [50 mM Tris·HCl, pH 7.5, 150 mM NaCl, 1% (vol/vol) Triton X-100, 1% (vol/vol) sodium deoxycholate, and 0.1% (wt/vol) SDS, supplemented with 100 μM sodium orthovanadate and protease inhibitor cocktail tablets (Complete Mini; Roche, Quebec, Canada)]. Samples were spun at 12,000 g for 15 min at 4°C, and the supernatant was transferred to a fresh tube to obtain a crude protein lysate. Protein concentrations were determined using the Bio-Rad protein assay buffer (Bio-Rad, Hercules, CA). Protein samples (15–40 μg) were suspended in Laemmli buffer, heated at 95°C for 5 min, and resolved by electrophoresis on a 10% SDS-polyacrylamide gel. Proteins were transferred onto polyvinylidene difluoride (PVDF) membrane (Millipore, Bedford, MA) in 48 mM Tris·HCl, 39 mM glycine, and 0.037% (wt/vol) SDS, pH 8.3, for 1.5 h at 300 mA at 4°C. Activation of MAPK pathways was measured by Western analysis for phosphorylated ERK (p-ERK), as described by Oldenhof et al. (19), by use of the phosphospecific ERK1/2 primary antibody (Promega, Madison, WI), and normalized to total ERK1/2 using the anti-ERK1/2 primary antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Probed membranes were exposed to X-ray film (Kodak XAR, Eastman Kodak) and analyzed by densitometry.
Stretch induction profiles were subjected to a one-way analysis of variance (ANOVA) followed by pairwise multiple comparison procedures (Bonferroni t-test) to determine differences between the control and each time point. Where required, the data were transformed to obtain a normal distribution. The progesterone pretreatment data were subjected to a two-way ANOVA followed by multiple comparison procedures (Student-Newman-Keuls method) to determine differences between groups. Statistical analysis was carried out using SigmaStat version 2.03 (Jandel, San Rafael, CA), with the level of significance for comparison set at P < 0.05.
We have previously shown that the uterine SMCs generated as described above maintain a smooth muscle phenotype beyond 4 days in culture (29). Therefore, all stretch experiments conducted in this study were carried out on day 4.
The cDNA probes used for Northern analysis hybridized to transcripts of the appropriate sizes for all fos and jun genes. Major signals for the Fos family were detected at 2.2 kb (c-fos), 4.2 kb (fosB), 1.6 kb (fra-1), and 6.2 kb (fra-2). The c-jun probe detected a major signal at 2.7 kb and a weaker signal at 3.2 kb, and the junD and junB probes detected signals at 1.7 and 2.1 kb, respectively. The 18S band was detected at 1.8 kb.
Stretch induction of AP-1 family mRNAs.
Northern analysis of the fos family revealed that the mRNA levels of c-fos, fosB, and fra-1 were increased in response to static mechanical stretch in primary uterine SMCs, whereas the levels of fra-2 mRNA were high under basal conditions and remained constant throughout the stretch experiment (Fig. 1A). Densitometric analysis revealed significant differences from the control group for c-fos, fosB, and fra-1 (Fig. 1B). Specifically, c-fos mRNA was induced maximally after 30 min of static stretch (P < 0.001) at 14-fold over the control levels. The levels of fosB mRNA were increased at 30 min (P < 0.001), peaked at 1 h (26-fold, P < 0.001), and began to return to control levels by 2 h (P < 0.05). Stretch-induced fra-1 mRNA expression was delayed and was more sustained than either c-fos or fosB, with the peak of induction between 1 and 2 h (2.4-fold, P < 0.05).
Northern analysis of the jun family revealed that the mRNA levels of c-jun and junB were increased in response to static mechanical stretch in primary uterine SMCs, whereas the levels of junD mRNA remained low and were relatively constant throughout the stretch experiment (Fig. 2A). Densitometric analysis revealed significant differences from the control group for both c-jun and junB (Fig. 2B). A stretch-induced increase in c-jun mRNA was detected at 30 min (P < 0.05), peaked at 1 h (4-fold, P < 0.005), and returned to control levels by 2 h. The levels of junB mRNA peaked at 30 min (3.1-fold, P < 0.001), remained elevated at 1 h, and returned to control levels by 2 h.
Progesterone inhibition of the stretch response.
Our in vivo experiments revealed that treatment of pregnant rats with progesterone prevents the onset of labor and the induction of both CAP genes and several members of the AP-1 family of genes in the myometrium. Cultured uterine SMCs were treated with progesterone before mechanical stimulation to determine whether the inhibitory actions of progesterone on AP-1 gene expression occur directly on uterine SMCs. Pretreatment with progesterone for only 1 or 6 h did not affect the peak (30-min) stretch induction of c-fos mRNA (Fig. 3). In contrast, pretreatment of uterine SMCs in culture with progesterone for 24 h reduced the magnitude of c-fos and fosB mRNA induction because of static mechanical stimulation (Fig. 4A). Densitometric analysis revealed that pretreatment with 400 nM progesterone led to a 58% reduction in the magnitude of the stretch-induced c-fos response at 30 min (Fig. 4B). This was not statistically different, however, from the vehicle-treated sample at 30 min. At the 1 μM dose, progesterone pretreatment led to a significant 64% reduction in the stretch-induced c-fos response (P < 0.05). The stretch induction of fosB mRNA after 1 h was slightly stimulated by treatment with the lower 100 nM dose of progesterone, whereas higher doses of progesterone reduced the magnitude of this stretch response. Pretreatment with both 400 nM and 1 μM progesterone completely blocked the stretch induction of fosB mRNA at 30 min, with no significant differences detected from the control, nonstretched sample. These higher doses of progesterone had less of an effect on the stretch induction of fosB mRNA at 1 h; however, there was a significant reduction in the stretch response at 1 h for the 400 nM and 1 μM doses compared with the 100 nM dose. Taken together, these data suggest that the higher doses of progesterone, mimicking the levels present during late pregnancy, both delayed and reduced the stretch induction of fosB mRNA compared with the progesterone levels of early pregnancy (Fig. 4B). The stretch induction of fra-1 mRNA after 1 h was slightly stimulated by progesterone treatment at all doses (Fig. 4B). The induction of c-jun mRNA by stretch also appeared to be stimulated by progesterone pretreatment (Fig. 5A). Densitometric analysis revealed that there was a significant 50% increase in the magnitude of the stretch induction of c-jun mRNA when cells were pretreated with 100 nM progesterone (Fig. 5B). This increase, however, was not observed at the higher doses of progesterone.
We have previously reported that in vitro stretch of uterine SMCs results in a rapid activation of intracellular signaling pathways, resulting in increased phosphorylation of the MAPKs (ERK, JNK, and p38), all three of which are required for maximal c-fos mRNA expression (19). As previously demonstrated, vehicle-treated uterine SMCs subjected to static stretch exhibited a rapid activation (phosphorylation) of ERK1/2. SMCs pretreated with 1 μM progesterone for 24 h, a dose and duration that inhibited stretch-induced c-fos and fosB mRNA expression, exhibited a similar time course and level of ERK1/2 phosphorylation as vehicle-treated cells (Fig. 6A). Furthermore, densitometric analysis revealed that there was no difference between the stretch-induced increase in ERK1/2 phosphorylation at any time point in vehicle and progesterone-treated SMCs (P = 0.842, Fig. 6B).
Several studies have shown that the expression of c-fos is increased by mechanical stimulation in vascular SMCs. Previously, we have shown that the expression of c-fos is also increased by stretch in myometrial SMCs (28). Furthermore, experiments examining the expression of all seven members of the AP-1 family in the myometrium of unilaterally pregnant rats suggested that several members of this immediate early gene family are subject to regulation by mechanical signals in vivo (17). The in vitro stretch experiments conducted in this study have indicated that, in addition to c-fos, several other members of the AP-1 family are stretch-responsive genes in myometrial SMCs. In addition, progesterone, which inhibits the expression of AP-1 family genes in vivo in the myometrium, was shown to inhibit the stretch induction of specific AP-1 family genes in vitro, revealing the ability of progesterone to act directly on myometrial SMCs.
We found that several members of the AP-1 family were responsive to stimulation by static mechanical stretch in vitro. An increase in the levels of c-fos, fosB, fra-1, junB, and c-jun was detected with the application of static stretch, although the time scale and magnitudes differed for each. The levels of junD mRNA were not altered by stretch in the in vitro model, a result that correlates with the observations made during pregnancy. Stretch in the in vitro system did not alter the levels of fra-2 mRNA, although in vivo an increase in fra-2 expression was detected only in the gravid horn. It is possible that the induction of fra-2 in the gravid horn occurs through the presence of a paracrine factor released from the feto-placental unit, rather than through mechanical stimulation. In primary culture cells, high levels of fra-2 were detected under basal conditions that were not observed in the myometrium during early pregnancy. It is possible that the expression of this gene increases as a result of the primary culture process and is therefore no longer subject to the same modes of regulation as in vivo.
It appears from these in vitro experiments that several members of the AP-1 gene family are stretch responsive and may therefore be regulated by similar intracellular signaling pathways. Our previous studies revealed that activation of the MAPK-signaling pathways (ERK, JNK, p38) occurs in response to in vitro stretch and is required for the induction of c-fos in response to this mechanical stimulation (19). The response of c-fos to MAPK signaling has been shown to occur through the serum response element (SRE) in the c-fos promoter (34). Both the fosB and junB promoters contain SRE sequences resembling the c-fos SRE, suggesting that stretch induction of these genes also requires activation of MAPK pathways (13, 15). The first intron of the fra-1 gene contains three AP-1-binding sites, whereas the promoter region of fra-1 contains only a specificity protein 1 site (3). The expression of fra-1 in response to serum stimulation has been shown to be dependent on the presence of c-Fos protein, and increased expression of fra-1 is therefore delayed compared with the increase in c-fos expression (27). A similar profile was observed in this study in response to in vitro stretch, with the peak induction of c-fos preceding the induction of fra-1, suggesting that fra-1 expression in response to mechanical stimulation may require the expression of c-fos. In the c-jun promoter, signal-induced phosphorylation of bound transcription factors, including the c-Jun protein, is thought to be the major stimulus that induces the expression of c-jun (9, 24). JNK, a member of the MAPK family, is activated in response to mechanical stimulation in myometrial SMCs and phosphorylates the c-Jun protein, leading to transcriptional activation (19, 26, 30). Therefore, in cultured myometrial SMCs, several AP-1 genes may be regulated by the MAPK-signaling pathways.
During pregnancy, the increase in c-fos, fosB, fra-1, fra-2, c-jun, and junB mRNA is blocked by the administration of progesterone (17). Progesterone pretreatment of primary myometrial cells in vitro also reduced the magnitude of the stretch induction for c-fos and fosB, indicating that progesterone can act directly on myometrial SMCs to inhibit AP-1 gene expression. A reduction in the stretch response of these two genes was observed when the cells were pretreated with progesterone for 24 h, whereas pretreatment for only 1 or 6 h did not have an effect on the induction of c-fos by mechanical stretch. The direct genomic actions of steroid receptors generally require from 10 to 30 min for an effect on the transcription of target genes. For example, downregulation of the chicken c-jun gene in the oviduct by progesterone occurs at the level of transcription within 15 min and requires only 30 min to decrease the levels of mRNA (31). Interestingly, in nonpregnant rats, progesterone administration blocks the estradiol-induced increase in c-fos mRNA within 1 h, consistent with a mechanism in which progesterone inhibits transcriptional activation by the estrogen receptor, but it has no effect on the TPA-induced increase in c-fos mRNA after 3 h, although longer time points were not investigated (12). These results indicate the diverse signaling pathways through which c-fos induction can occur in myometrial cells. The requirement for 24 h of pretreatment with progesterone to affect the stretch induction of c-fos suggests that the mechanisms of progesterone action in this case are unlikely to be due to a direct effect of the progesterone receptor on the transcription of AP-1 genes.
There are several other mechanisms through which progesterone could inhibit the expression of AP-1 genes. It has been suggested that a major function of progesterone throughout pregnancy is to support uterine growth (myocyte hypertrophy) and thus prevent the development of uterine tension, which would inhibit the induction of stretch-responsive genes (16). The ligand-activated progesterone receptor could alter the ability of myometrial SMCs to sense and respond to mechanical stimulation by causing a profound change in gene expression within the cell. For example, we have shown that mechanical stimuli lead to increased expression of ECM components (including laminin and fibronectin) in the myometrium during late pregnancy, an effect that is blocked by progesterone (28). Furthermore, it was shown that the induction of c-fos expression by stretch in vitro varied when primary myometrial SMCs were plated onto different matrix components (19). Taken together, these data raise the possibility that modulation in ECM production might attenuate the ability of myometrial SMCs to respond to stretch.
In an attempt to address this issue, we investigated whether the inhibition of stretch-induced c-fos and fosB mRNA expression in the presence of progesterone was the result of a failure in uterine SMCs to activate ERK1/2 in response to stretch. Because the cells exhibited no attenuation of stretch-induced phosphorylation of ERK1/2, it would seem likely that progesterone pretreatment did not affect the ability of these SMCs to sense and respond to mechanical stimulation. Although the specific mechanisms remain to be determined, the actions of progesterone would appear to be through modifying the expression or activity of other downstream effectors, which themselves inhibit the stretch-induced expression of AP-1 genes, although the possibility remains that progesterone affects a parallel pathway involved in the stretch-induced activation of AP-1 gene expression.
In conclusion, many of the AP-1 family members are positively regulated by mechanical stretch in vitro. Progesterone acts directly on myometrial SMCs to inhibit the stretch induction of specific members of the AP-1 family. Collectively, this family of transcription factors may regulate several downstream stretch-responsive genes in myometrial SMCs. Furthermore, because different dimers can have different binding and transactivation properties, they may regulate different downstream target genes.
This study was supported by National Institute of Child Health and Human Development Grant HD-37942. The Canadian Institutes of Health Research provided a Doctoral Research Award to J. Mitchell.
O. Shynlova and J. Mitchell contributed equally to these experiments and to the preparation of this manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 by American Physiological Society