| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Division of Obstetrics and Gynecology, John Hunter Hospital (W.B.G., W.A.W., T.Z.); and Discipline of Reproductive Medicine (R.F.J., W.B.G., W.A.W., T.Z.) and Mothers and Babies Research Center (R.F.J., C.M.M., W.B.G., W.A.W., T.Z.), University of Newcastle, Newcastle, New South Wales 2310, Australia
Address all correspondence and requests for reprints to: Dr. Tamas Zakar, Division of Obstetrics and Gynecology, Mothers and Babies Research Center, John Hunter Hospital, Locked Bag 1, Hunter Region Mail Center, Newcastle, New South Wales 2310, Australia. E-mail: . tzakar{at}mail.newcastle.edu.au
Abstract
Prostaglandin H synthase-2 (PGHS-2) activity and mRNA rise in the human amnion at late gestation, contributing to the increase in intrauterine PG production crucial for labor and delivery. In the present investigation we have determined the mechanism that controls amniotic PGHS-2 mRNA levels in vivo at term parturition.
Amnion membranes were collected after elective cesarean section (n = 20), and after spontaneous labor (n = 20). PGHS-2 relative gene transcription rates were determined by transcriptional run-on, and PGHS-2 mRNA and heterogeneous nuclear RNA (hnRNA) relative abundance were measured by quantitative real-time RT-PCR. The PGHS-2 mRNA degradation rate was determined by incubating amnion in the presence of the transcription inhibitor 5,6-dichlorobenzimidazole riboside. The dynamics of PGHS-2 hnRNA and mRNA abundance were characterized in 0- to 24-h tissue incubations.
The PGHS-2 relative gene transcription rate was a significant (P < 0.05) predictor of PGHS-2 hnRNA and mRNA abundance, and PGHS-2 hnRNA was also a predictor (P < 0.01) of PGHS-2 mRNA levels both before and after labor. Interestingly, even though PGHS-2 gene activity remained unchanged, PGHS-2 mRNA abundance increased with labor and displayed constitutive stability before and after labor. PGHS-2 mRNA levels spontaneously increased by 400% (P < 0.01) upon incubation for 24 h, whereas the transcription rate dropped by 95% during the first 2 h, then rebounded significantly between 6–24 h.
Thus, PGHS-2 mRNA abundance is transcriptionally controlled in term amnion. Labor does not increase PGHS-2 gene activity or mRNA stability. The PGHS-2 gene is probably induced before labor by a factor(s) originating in the amnion membrane, and the resulting stable mRNA accumulates progressively in the tissue throughout labor and delivery.
PGE2 and PGF2
produced in the gestational tissues are thought to have pivotal importance in the initiation and maintenance of labor in humans (1). Their primary roles are in the facilitation of myometrial contractions and in the stimulation of cervical ripening and membrane rupture. The involvement of PGs in human parturition is supported by a rise in the concentrations of PGE2 and PGF2 in the amniotic fluid before and during labor (2, 3, 4) and by the labor-inducing effects of PG administration at any time during gestation (5, 6). Further, drugs that inhibit PG biosynthesis prolong gestation and labor (7).
The committing and rate-limiting step of PG biosynthesis is the conversion of arachidonic acid to the endoperoxide intermediates PGG2 and PGH2. These reactions are catalyzed by the enzyme prostaglandin endoperoxide H synthase (PGHS) also known as cyclooxygenase (8). There are two PGHS isoenzymes, PGHS-1 and -2, which are the products of distinct genes. Both PGHS isoenzymes undergo rapid auto-inactivation during catalysis and must be synthesized continuously de novo to maintain a steady state level of enzyme activity (8). Therefore, catalytically active PGHS protein levels depend acutely on the abundance of the respective PGHS mRNAs in the cells.
It is well established that PGHS-2 mRNA and protein increase in the human amnion and chorion laeve at term before and during labor (9, 10, 11). This increase in PGHS-2 mRNA correlates with an increase in PGHS enzyme activity (12, 13). PGHS-1 mRNA and protein levels remain low and relatively unchanged during this time (11, 13, 14). For these reasons PGHS-2 is thought to be responsible for the increased PGHS activity that occurs in fetal membrane tissues during late pregnancy and labor, and PGHS-2 mRNA abundance appears to be the key factor in determining the level of PGHS activity in these tissues. It is also clear that the mechanisms that regulate PGHS-2 mRNA control the PG biosynthetic capacity of the fetal membranes and have a crucial role in the labor-associated increase in intrauterine PG production. These in vivo control mechanism(s), although very important, are undefined.
In principle, steady state PGHS-2 mRNA levels can be up-regulated through either an increase in PGHS-2 gene transcription rate or a decrease in the degradation rate of the mRNA (15). Both mechanisms have been shown to control PGHS-2 mRNA levels in cell culture models (16, 17, 18, 19, 20), but their contribution to the in vivo up-regulation of the enzyme in gestational tissues is unclear. Such information would be important to understand the physiology of parturition and to design strategies to influence intrauterine PG production. Therefore, our aim in the present investigation was to determine the mechanism of PGHS-2 mRNA up-regulation in vivo in the human amnion at term. We measured PGHS-2 gene activity before and after labor directly by transcriptional run-on in nuclei isolated from amnion tissue and indirectly by determining PGHS-2 heterogeneous nuclear RNA (hnRNA) abundance, a surrogate measure of the gene transcription rate, by quantitative real-time RT-PCR. We also measured the PGHS-2 hnRNA processing rate and the PGHS-2 mRNA decay rate before and after labor and characterized the changing dynamics of these RNA species in short-term tissue incubations. The results suggest that PGHS-2 mRNA in the human amnion is transcriptionally regulated and constitutively stable and that the transcriptional activity of the PGHS-2 gene is maintained by factors produced in the amnion membrane. The up-regulation of PGHS-2 in term amnion possibly involves an amnion- derived factor(s) that stimulates PGHS-2 gene activity before labor, leading to the accumulation of PGHS-2 mRNA and activity throughout the onset and progression of labor until delivery.
Materials and Methods
Materials
Cytochalasin B, spermine, spermidine, phenylmethylsulfonylfluoride (PMSF), leupeptin, pepstatin A, ATP, CTP, GTP, UTP, dithiothreitol (DTT), creatine phosphokinase, creatine phosphate, nucleotide diphosphokinase, 5,6-dichlorobenzimidazole riboside (DRB), HBSS, ribonuclease A (RNase), and formamide were obtained from Sigma (St. Louis, MO). Proteinase K, RNase-free deoxyribonuclease (DNase) and tRNA (from baker’s yeast) were obtained from Roche Molecular Biochemicals (Indianapolis, IN). RNasin was purchased from Promega Corp. (Annandale, Australia), and [-32P]UTP was purchased from Geneworks (Adelaide, Australia). All other reagents were of analytical grade or better.
Patients
Placentas with attached fetal membranes were obtained from a total of 40 uncomplicated singleton pregnancies. Of these, 20 were collected after elective cesarean section (CS) in the absence of labor, and 20 were obtained after spontaneous vaginal delivery (SL) at term (37–41 wk gestational age) at John Hunter Hospital (Newcastle, Australia). Women who had received aspirin or indomethacin within 3 d before delivery or with a history of infection, histological chorioamnionitis, severe asthma, or induced delivery were excluded from the study. The Hunter Area Health Service and the University of Newcastle human ethics committees approved the use of these tissues, and informed consent was obtained from all women participating in the study.
Tissue collection
All tissues were obtained within 30 min of delivery. The reflected amnion was peeled off of the chorio-decidua, and 1–2 g of the amnion membrane were frozen in liquid nitrogen and stored at -80 C for RNA extraction. A further 6 g amnion were rinsed in PBS and used immediately to isolate nuclei.
Isolation of cell nuclei
Nuclei were isolated using modifications of the methods used by Hahn and Covault (21) and Wigler and Weinstein (22). Briefly, minced amnion was incubated in PBS with cytochalasin B (42 µmol/liter) at 30 C for 30 min. The tissue was then homogenized in 40 ml ice-cold buffer A [0.15 mmol/liter spermine, 0.5 mmol/liter spermidine, 0.5 mmol/liter EGTA, 2 mmol/liter EDTA, 15 mmol/liter HEPES (pH 7.5), 14 mmol/liter 2-mercaptoethanol, 0.1 mmol/liter PMSF, 1 µmol/liter leupeptin, and 1 µmol/liter pepstatin A] using a Polytron homogenizer (Brinkmann Instruments, Inc., Westbury, NY) on the maximal setting with five bursts of 30 sec and 30-sec cooling periods. The release of nuclei was monitored by examining aliquots of the homogenate stained with trypan blue. Nonidet P-40 was added to a final concentration of 0.5% (vol/vol), followed by homogenization with a Potter-type glass-Teflon homogenizer. The homogenate was pressed through a metal grid (380 µm pore size) and homogenized again by 40 strokes in a Dounce homogenizer (Kontes Co., Vineland, NJ; tight pestle; clearance, 50.8 µm). The final homogenate was filtered through a second metal grid (74 µm pore size) and centrifuged at 1000 x g for 10 min at 4 C. The pellet was suspended in 70 ml buffer B [1.5 mol/liter sucrose, 0.15 mmol/liter spermine, 0.5 mmol/liter spermidine, 0.1 mmol/liter EGTA, 0.1 mmol/liter EDTA, 15 mmol/liter HEPES (pH 7.5), 14 mmol/liter 2-mercaptoethanol, 0.1 mmol/liter PMSF, 1 µmol/liter leupeptin, and 1 µmol/liter pepstatin A], and purified nuclei were sedimented by centrifugation for 45 min in a JA-17 rotor (Beckman Coulter, Inc., Fullerton, CA) at 17 000 rpm and 4 C. The purified nuclei were suspended in an equal volume of nuclear storage buffer [40% (vol/vol) glycerol, 5 mmol/liter MgCl2, 0.1 mmol/liter EDTA, 5 mmol/liter DTT, 0.1 mmol/liter PMSF, and 50 mmol/liter Tris HCl, pH 8.0] and stored at -80 C until further use.
Preparation of nitrocellulose membranes
Plasmids without insert (pUC19, negative control) and with cDNA inserts for PGHS-2 (9) and ß-actin (a gift from Dr Richard Nicholson, Mothers and Babies Research Center, University of Newcastle, Newcastle, Australia) were diluted to a final concentration of 15 µg/ml in 0.1 mol/liter NaOH and boiled in a water bath for 4 min. The nitrocellulose membrane (0.45 µm, Protran, Schleicher \|[amp ]\| Schuell, Inc., Keene, NH) was prewet for 10 min in NaCl/sodium phosphate buffer (1 mol/liter sodium chloride and 0.5 mol/liter sodium phosphate, pH 7.2) and placed into a Minifold II slot blotter (Schleicher \|[amp ]\| Schuell, Inc.). Immediately before loading, the plasmid solutions were neutralized by adding 1 vol 2 x NaCl/sodium phosphate buffer. The plasmids were loaded into triplicate slots on the nitrocellulose. The slots were rinsed with NaCl/sodium phosphate buffer, and DNA was immobilized on the membrane by UV cross-linking (Spectrolinker XL-1000 UV crosslinker, Spectronics Corp., Westbury, NY) on both sides.
Nuclear run-on assay
Nuclei were labeled by a modification of a previously described procedure (21). Briefly, 100 µl nuclear suspension (containing 
50–100 µg DNA) were added to an equal volume of labeling mixture (5 mmol/liter MgCl2; 200 mmol/liter KCl; 0.1 mmol/liter EDTA; 1 mmol/liter each of ATP, CTP, and GTP; 2.5 µmol/liter UTP; 5 mmol/liter DTT; 100 U RNasin; 10 mmol/liter creatine phosphate; 12.8 U creatine phosphokinase; 0.8 U nucleotide diphosphokinase kinase; and 150 µCi [-32P]UTP) and incubated at 25 C for 30 min. Yeast tRNA (200 µg) and RNase-free DNase I (100 U) were added, and the incubation was continued for an additional 5 min. After adding 95 µl digestion mix (10% SDS, 0.5 mol/liter EDTA, and 1 mg/ml proteinase K), the nuclei were incubated at 42 C for 30 min. After digestion, 300 µl SDS/EDTA/Tris (1%; 25 and 50 mmol/liter, respectively; pH 7.4) were added, and the RNA was extracted with an equal volume of buffered phenol/chloroform/isoamyl alcohol (25:24:1). RNA in the aqueous phase was precipitated with an equal volume of 20% trichloroacetic acid/120 mmol/liter sodium pyrophosphate at 0 C, and the RNA pellet was washed four times with cold 5% trichloroacetic acid/30 mmol/liter sodium pyrophosphate. Fifty microliters of 1 mol/liter ice-cold sodium hydroxide were then added for 5 min. The solution was neutralized with 1 mol/liter HEPES acid (100 µl), followed by 15 µl sodium acetate (3 mol/liter) and 3 vol absolute ethanol. After overnight storage at -20 C, the RNA pellets were suspended in 10 mmol/liter EDTA, 0.1% (wt/vol) SDS, and 10 mmol/liter Tris, pH 7.4. Incorporated radioactivity was measured by liquid scintillation counting, and 1.5 x 106 cpm of the 32P-labeled RNA was added to 3 ml hybridization solution [0.5 mol/liter NaCl, 50 mmol/liter PIPES (pH 7.0), 0.4% SDS, 2 mmol/liter EDTA, and 33% (vol/vol) formamide]. Hybridization was performed at 45 C for 72 h. After hybridization, the membranes were washed four times in wash solution [0.3 mol/liter NaCl, 2 mmol/liter EDTA, 10 mmol/liter Tris (pH 7.5), and 0.1% SDS] at 65 C for 30 min. After one additional wash with wash solution without SDS, the membranes were incubated with 10 µg/ml RNase A for 30 min at 37 C. Four more washes with wash solution followed at 45 C for 30 min. The nitrocellulose was air-dried and exposed to Kodak Biomax MS film using a Biomax Transcreen HE intensifying screen (Eastman Kodak Co., Rochester, NY) at -70 C for 4 d. The films were analyzed by densitometry using the Scion Image program (ß Release 4.0.2) available online from the NIH web page.
Isolation of total RNA
Total RNA was prepared from frozen tissues using TRIzol reagent (Invitrogen, Mt. Waverley, Australia) following the manufacturer’s protocol. RNA integrity was determined by the detection of 18S and 28S rRNA using an agarose gel after ethidium bromide staining.
Extracted RNA was purified on RNeasy mini-spin columns and treated with DNase using an RNase-free DNase kit (QIAGEN, Clifton Hill, Australia) following the directions provided by the manufacturer.
Quantitative real-time RT-PCR
Two micrograms of amnion RNA were used as a template for cDNA synthesis with random hexamers as primers, following the protocol supplied with the Superscript First Strand Synthesis System for RT-PCR (Invitrogen).
Quantification of PGHS-2 mRNA and hnRNA abundance was performed by real-time PCR detection using an ABI PRISM 7700 sequence detector (PE Applied Biosystems, Foster City, CA) with SYBR green detection of amplification products. Amplification mixtures contained 12.5 µl 2 x SYBR green PCR master mix (PE Applied Biosystems, Branchburg, NJ), 2 µl cDNA synthesis mixture, 10 pmol each of the forward and reverse primers, and distilled water to a total volume of 25 µl. All primers (Table 1
) were designed using the computer program Primer Express v 1.0 (PE Applied Biosystems). Primers for PGHS-2 mRNA were directed to a sequence that spans the junction of exon 8 and 9, corresponding to open reading frame 1228–1294 (23). PGHS-2 hnRNA primers were directed to a sequence located at bases 6294–6395 within intron 9 of the PGHS-2 gene sequence as previously described (24). Primers for ß-actin mRNA were directed to a sequence located at open reading frame 878-1090 (25). The standard amplification system included cross-contamination control reagents (AmpErase UNG), and hot start capability. Parameters of thermal cycling were as set by the manufacturer (PE Applied Biosystems).
|
ß-Actin mRNA was used as the reference message for both PGHS-2 mRNA and PGHS-2 hnRNA. Accordingly, the relative abundance of PGHS-2 mRNA and hnRNA were calculated (with reference to ß-actin) using the formula: relative abundance = 2-
Ct, where CT is the difference between the sample CT (cycle threshold) and the CT observed for the reference message. Amplification curves were analyzed using software provided by the manufacturer. The identities of the PCR products were confirmed by size determination and sequencing of the amplified products.
Tissue incubations
Six grams of the amnion membrane were briefly washed in sterile PBS, blotted and cut into approximately 2-cm2 pieces. The tissue pieces were distributed into approximately 1-g portions, and each portion was incubated in 19 ml medium [HBSS, 20 mmol/liter HEPES (pH 7.5), 4.2 mmol/liter sodium hydrogen carbonate, and 40 µg/ml gentamicin] for the time periods required by the experiments. To measure RNA degradation rates, the transcription inhibitor, 5,6-dichlorobenzimidazole riboside (DRB), was added in dimethylsulfoxide at a final concentration of 80 µmol/liter, and the mixtures were incubated at 37 C for 1, 2, 4, 6, and 24 h. After incubation, the tissues were removed from the medium, blotted, and frozen in liquid nitrogen until extraction of RNA. Dimethylsulfoxide vehicle was added to the 0 h sample at a final concentration of 0.1%. Incubations to determine the dynamics of PGHS-2 hnRNA and mRNA abundance in vitro were performed in the absence of DRB for 0, 2, 4, 6, 18, and 24 h. The viability of the amnion was assessed by lactate dehydrogenase release in the medium and was found to be unimpaired during the incubations, as reported previously (26).
Statistical analysis
Relative PGHS-2 hnRNA and mRNA abundance values and gene transcription rates were logarithmically transformed before parametric tests (regression, t tests, and simple correlation) to approximate normal distribution and homogeneity of variance. After transformation, relative abundance values more than 2.3SD from the mean (beyond 0.01 of the normal deviate) were considered outliers. There was one outlier in each of the CS and SL groups. Linear regression was used to test relationships. Slopes of regression lines were compared by t statistics using the formula: t = b1 - b2/sb1 - b2 where b1 and b2 are the slopes (regression coefficients) of the respective regression lines and sb1 - b2 is theSE of the difference between the regression coefficients (27). The intersection point of regression lines was calculated using the formula: XI = a2 - a1/b1 - b2, where XI is the x-coordinate of the intersection point, and a1 and a2 are the y-intercepts of the regression lines (27). Nonparametric group comparisons and correlation analyses were made using the Mann-Whitney U test and the Spearman rank correlation procedure, respectively.
mRNA and hnRNA half-lives before and after labor were compared by t test. Dynamics of mRNA and hnRNA abundance were analyzed by two-way ANOVA (mixed model, with repeated measures), followed by Tukey’s test for multiple comparisons if a significant F value was obtained. In all statistical analyses, P value less than 0.05 was considered significant.
Results
Influence of PGHS-2 gene transcription rate on PGHS-2 mRNA level
PGHS-2 relative gene transcription rates were determined by transcriptional run-on in nuclei isolated from amnion tissues immediately after delivery. As shown in Fig. 1, labeled PGHS-2 and ß-actin transcripts were detected after hybridization in triplicate, whereas there was negligible background from plasmid DNA lacking a cDNA insert (pUC19). Transcription of the PGHS-2 gene was detected in all tissues studied (n = 10). The abundance of PGHS-2 mRNA was measured in the same tissues by real-time quantitative RT-PCR. The dependence of PGHS-2 mRNA levels on the gene transcription rate was evaluated by regression analysis where the PGHS-2 gene transcription rate (relative to ß-actin gene transcription rate) was considered the independent variable, and PGHS-2 mRNA abundance (relative to ß-actin mRNA abundance) was considered the dependent variable. The data in Fig. 2A demonstrate that regression was significant (F = 12.62; P < 0.01, by ANOVA), indicating that the gene transcription rate was a major determinant of PGHS-2 mRNA levels. This relationship was reflected by the significant correlation, by both parametric (r = 0.8110; P < 0.01) and nonparametric (Spearman r = 0.8181; P < 0.01) analyses, between the transcription rates and mRNA levels in individual tissues.
|
|
Tissue levels of hnRNA, the unprocessed precursor of the mature, functional mRNA, have been proposed as a surrogate for nuclear run-on assays to determine gene transcription rates (28). To develop a relatively simple procedure to measure PGHS-2 gene transcription by this approach, we have designed intron-specific oligonucleotide primers and quantified PGHS-2 hnRNA abundance in amnion tissues using real-time quantitative RT-PCR. PGHS-2 hnRNA was detected in all tissues, and the PGHS-2 gene transcription rate, determined by nuclear run-on in the same tissues, was a significant predictor of its level (Fig. 2B
; F = 9.97; P < 0.05, by ANOVA). Correlation between transcription rates and hnRNA levels was also significant (r = 0.7561; P < 0.01; and Spearman r = 0.6848; P < 0.05). These results validated the use of PGHS-2 hnRNA as an estimate of PGHS-2 gene activity in the amnion membrane.
Next, PGHS-2 mRNA and PGHS-2 hnRNA levels were determined in a total of 40 term amnion samples. As illustrated by Fig. 3A, PGHS-2 mRNA abundance showed significant regression on PGHS-2 hnRNA abundance (F = 47.78; P < 0.001), and the correlation of the values was tight (r = 0.7552; P < 0.001; and Spearman r = 0.7781; P < 0.001). This further confirms that PGHS-2 mRNA levels are controlled at a step upstream of its precursor by the transcriptional generation of hnRNA.
|
Of the 40 amnion samples, 20 were obtained after term elective CS and 20 after SL (Table 2). In agreement with previous reports (10, 11, 13, 14), the mean and median values for PGHS-2 mRNA abundance were significantly higher in the SL group than in the CS group. The mean and median of PGHS-2 hnRNA abundance, on the other hand, were not significantly different in the 2 groups of women, suggesting that PGHS-2 mRNA levels increased with labor without a significant rise in PGHS-2 gene activity.
|
, B and C, and Table 2, PGHS-2 gene transcription (represented by hnRNA levels) was a significant predictor of PGHS-2 mRNA levels in both groups. This indicated that gene activity was a determinant of mRNA abundance regardless of labor status. Statistical comparison of the equations showed, however, that the slope of the regression line was significantly (P < 0.001) higher in the SL group than in the CS group. Thus, the two groups were distinct because any level of gene activity was associated with higher PGHS-2 mRNA levels after labor (SL patients) than before labor (CS patients). The point of intersection of the two regression lines was calculated to be at the minimum of the distribution of the hnRNA relative abundance values (0.00199; marked by arrows on Fig. 3, B and C), which indicated that the above relationship applied to the entire patient sample and possibly to the populations sampled. Influence of labor on PGHS-2 mRNA stability
Increased mRNA abundance without a rise of gene transcription rate suggested that labor might be associated with the stabilization of the PGHS-2 mRNA. To test this possibility, we determined the rate of PGHS-2 mRNA degradation in amnion tissues before and after labor. PGHS-2 mRNA in amnion delivered by CS (Fig. 4A), showed virtually no decrease in relative abundance within 24 h. After labor, a tendency for increased degradation was detected, but the difference between the half-lives did not reach statistical significance. PGHS-2 hnRNA processing rates were equally fast in the two groups (t1/2, 20–22 min; Table 2). Thus, PGHS-2 mRNA decayed at a rate close or equal to that of the constitutive reference mRNA (ß-actin) before labor, and no stabilization occurred with labor. Furthermore, no increase in the hnRNA processing rate was detected with labor, ruling out another possible cause of the rise in amniotic PGHS-2 mRNA levels in the SL group.
|
The high stability of PGHS-2 mRNA raised the possibility that it may accumulate in the amnion without an increase in the gene transcription rate. This was tested by incubating amnion tissues for 24 h and measuring PGHS-2 mRNA and hnRNA abundance at various time points of the incubation (Fig. 5). The level of PGHS-2 mRNA remained unchanged for 6 h, then gradually increased to 400% of the 0 h value by 24 h (P < 0.05). PGHS-2 hnRNA abundance dropped to about 5% of its initial value by 2 h and remained low until 6 h of incubation. Subsequently, PGHS-2 hnRNA levels rose, reaching significantly greater abundance (P < 0.05) by 18 h and remaining elevated until the end of the incubation at 24 h. This experiment demonstrated that 1) PGHS-2 mRNA levels were stable despite a spontaneous fall in gene transcription; and 2) a rebound of transcriptional activity close to, but not exceeding, the in vivo rate was sufficient to cause a cumulative increase in PGHS-2 message within 18 h (corresponding to the period between 6–24 h of incubation).
|
Up-regulation of PG production in gestational tissues is a crucial element in the cascade of events leading to birth. Several studies (10, 11, 12) have demonstrated that in the amnion membrane, a key step of this stimulation is the increase in PGHS-2 mRNA abundance, resulting in elevated PGHS activity and an increase in PG biosynthetic capacity. The purpose of the present study was to determine the in vivo mechanisms that bring about the increase in PGHS-2 mRNA expression in the amnion at term. As the activity of the PGHS-2 gene is stimulated by agonists in a variety of cell culture models, we hypothesized that the in vivo regulation of PGHS-2 mRNA in the amnion is transcriptional. Indeed, the measurement of PGHS-2 gene transcription rates by nuclear run-on immediately after delivery, and the quantification of PGHS-2 mRNA levels by real-time RT-PCR in the same tissues demonstrated that the activity of the PGHS-2 gene is a significant determinant of the level of its cognate mRNA. Furthermore, a similar relationship was found between PGHS-2 gene transcription rates and the level of PGHS-2 hnRNA, the primary product of gene transcription and the precursor of the mature, functional mRNA. This result is in agreement with those reported by Elferink et al. (28), who established, using cultured cells, that hnRNA abundance provides a reliable estimate of gene transcription rate and may serve as a surrogate to the complicated nuclear run-on assay.
In agreement with the former results, hnRNA abundance was a predictor of mRNA levels, and the two exhibited a significant correlation. This is strong further support to our hypothesis that the step controlling the abundance of PGHS-2 mRNA is upstream of its precursor, i.e. at the generation of hnRNA by transcription.
It has been reported by several investigators (9, 11, 13, 14) that term labor is associated with elevated PGHS-2 mRNA, protein, and activity in the amnion membrane. Our results confirm these observations. The abundance of PGHS-2 hnRNA, however, did not increase significantly in the SL group of tissues, which suggests that labor itself, surprisingly, had little or no effect on PGHS-2 gene activity.
We have also examined the dependence of mRNA abundance on the transcription rate in individual samples in the CS and SL groups and found that gene transcription rate was a significant determinant of PGHS-2 mRNA abundance both before and after labor. The regression parameters indicated, however, that at any level of gene activity, PGHS-2 mRNA abundance was higher after labor than before. This raised the possibility of message stabilization during labor.
Direct determination of PGHS-2 mRNA degradation rates ruled out that labor had a stabilizing effect on PGHS-2 mRNA. Unexpectedly, PGHS-2 mRNA relative abundance did not fall in CS tissues incubated for 24 h in the presence of a transcription inhibitor, and labor was associated with a tendency for decreasing PGHS-2 mRNA stability. Moreover, PGHS-2 hnRNA processing rates were unaffected by labor, ruling out regulation at this step. Thus, PGHS-2 mRNA was already stable before labor to an extent equal to that of the constitutive reference message. This is an important in vivo difference in regulation compared with most cell culture models where PGHS-2 mRNA usually decays with a half-life between 1 and 2 h (16, 19). Furthermore, this finding has potential physiological significance, because stable PGHS-2 mRNA accumulating in the amnion may cause a steady up-regulation of PG synthesis, contributing to the clinical irreversibility of the parturition process.
Collectively, our data demonstrate that PGHS-2 mRNA levels rise in the amnion at labor without a significant increase in gene activity, hnRNA processing rate, and message stability. The data do not support the possibility (29) that the labor-associated increase in PGHS-2 expression is a consequence or after-effect of labor. We propose that the up- regulation of PGHS-2 gene transcription occurs before labor, generating stable mRNA that progressively accumulates in the tissue until the onset of labor and possibly during labor. As a consequence, the amnion contains higher levels of PGHS-2 mRNA and activity after labor than before. In our patient sample the mean gestational age in the term SL group was slightly, but significantly, higher than that in the term CS group. This difference, although incidental because the patients were recruited randomly, is consistent with the possibility that the increased PGHS-2 mRNA levels in the SL group are the result of a more prolonged period of message accumulation. Support for this explanation was provided by the tissue incubation experiments. When amnion membranes were maintained in culture, PGHS-2 mRNA levels remained steady for 6 h despite a 95% drop in the gene transcription rate, as indicated by PGHS-2 hnRNA abundance. Thus, PGHS-2 mRNA, once produced, persists in the tissue for prolonged periods. After 6 h, PGHS-2 mRNA levels rose to about 400% of the 0 h value by 24 h, and the increase coincided with a rise in hnRNA levels, indicating a rebound of transcriptional activity. This shows that transcription- dependent accumulation of PGHS-2 mRNA may occur in the amnion on a time scale spanning many hours, possibly even days.
The dynamics of PGHS-2 gene transcription during the amnion tissue incubations suggest intriguing possibilities about the regulation of PGHS-2 gene activity in situ. The precipitous drop in transcriptional activity within the first 2 h and the rebound after 6 h are best explained by the involvement of an amnion-derived factor(s) that maintains or stimulates PGHS-2 gene transcription in vivo. The endogenous factors are removed when the tissue is isolated and cleaned, but reaccumulate after a few hours and restore transcriptional activity. To validate this mechanism, however, the hypothetical factor(s) needs to be identified and its physiological role clarified in further studies.
In summary, the results of this study demonstrate that PGHS-2 gene expression is transcriptionally controlled in the human amnion at term. Labor is not associated with a significant rise in PGHS-2 gene activity. PGHS-2 mRNA is constitutively stable and may accumulate in the tissue, causing a steady rise in PG biosynthetic capacity before and during labor. It is likely that a factor(s) originating in the amnion membrane stimulates the transcriptional activity of the PGHS-2 gene during the parturition process.
Acknowledgments
Footnotes
This work was supported by the research management committee, University of Newcastle; the Mothers and Babies Research Center (Newcastle, Australia); and New South Wales Health through the Hunter Medical Research Institute. The scholarship for R.F.J. was provided by Research Higher degrees, University of Newcastle, and contributed by Profs. Roger Smith and David Smith.
Abbreviations: CS, Cesarean section; DNase, deoxyribonuclease; DRB, 5,6-dichlorobenzimidazole riboside; DTT, dithiothreitol; hnRNA, heterogeneous nuclear RNA; PGHS-2, prostaglandin H synthase-2; PMSF, phenylmethylsulfonylfluoride; RNase, ribonuclease; SL, spontaneous labor.
Received October 16, 2001.
Accepted February 9, 2002.
References
. J Obstet Gynaecol Br Commonw 76:769–782[Medline]
stimulates increased expression of prostaglandin endoperoxide H synthase type 2 mRNA in amnion-derived WISH cells. J Mol Endocrinol 20:221–231[Abstract]
B and AP-1 are required for cyclo-oxygenase 2 gene expression in amnion epithelial cell line (WISH). Mol Hum Reprod 6:561–565B activity which mediates cyclo-oxygenase-2 expression and is involved with the ‘functional progesterone withdrawal.’ Mol Hum Reprod 7:581–586. Evidence for post-transcriptional regulation. J Biol Chem 269:11769–11775 in the initiation of human parturition. J Clin Endocrinol Metab 76:1332–1339[Abstract]This article has been cited by other articles:
 |
|
 |
|
  C.M. Mitchell, R.F. Johnson, W.B. Giles, and T. Zakar Prostaglandin H synthase-2 gene regulation in the amnion at labour: histone acetylation and nuclear factor kappa B binding to the promoter in vivo Mol. Hum. Reprod., January 1, 2008; 14(1): 53 - 59. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Mohan, S. R. Sooranna, T. M. Lindstrom, M. R. Johnson, and P. R. Bennett The Effect of Mechanical Stretch on Cyclooxygenase Type 2 Expression and Activator Protein-1 and Nuclear Factor-{kappa}B Activity in Human Amnion Cells Endocrinology, April 1, 2007; 148(4): 1850 - 1857. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F Johnson, C. M Mitchell, W. B Giles, A. Bisits, and T. Zakar Mechanisms regulating prostaglandin H2 synthase-2 mRNA level in the amnion and chorion during pregnancy. J. Endocrinol., March 1, 2006; 188(3): 603 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M Lindstrom and P. R Bennett 15-Deoxy-{Delta}12,14-Prostaglandin J2 Inhibits Interleukin-1{beta}-Induced Nuclear Factor-{kappa}B in Human Amnion and Myometrial Cells: Mechanisms and Implications J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3534 - 3543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. E. Ackerman IV, B. H. Rovin, and D. A. Kniss Epidermal Growth Factor and Interleukin-1{beta} Utilize Divergent Signaling Pathways to Synergistically Upregulate Cyclooxygenase-2 Gene Expression in Human Amnion-Derived WISH Cells Biol Reprod, December 1, 2004; 71(6): 2079 - 2086. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. F. Johnson, C. M. Mitchell, V. Clifton, and T. Zakar Regulation of 15-Hydroxyprostaglandin Dehydrogenase (PGDH) Gene Activity, Messenger Ribonucleic Acid Processing, and Protein Abundance in the Human Chorion in Late Gestation and Labor J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5639 - 5648. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Johnson, C. M. Mitchell, W. B. Giles, W. A. Walters, and T. Zakar The Control of Prostaglandin Endoperoxide H-Synthase-2 Expression in the Human Chorion Laeve at Term Reproductive Sciences, May 1, 2003; 10(4): 222 - 230. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
) were determined by quantitative real-time RT-PCR. Results were analyzed by two-way ANOVA and Tukey’s post-hoc test for multiple comparisons. PGHS-2 mRNA levels remained stable for 6 h before increasing to 400% over the 0 h level (note the logarithmic scale on the y-axis). PGHS-2 hnRNA levels dropped to 5–10% of the control value within the first 2 h and remained low before rebounding to approximately 30% of the control level by 24 h of incubation. *, Significantly higher PGHS-2 hnRNA levels than the 2, 4, and 6 h values, (P < 0.01); 
, significantly higher PGHS-2 mRNA level than the 0 h value (P < 0.05).