Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Prostacyclin (PGI₂) and other vasodilator prostaglandins are important mediators of pulmonary vascular and parenchymal function in the perinatal period. PGI₂ infusion causes pulmonary vasodilation in the fetus and newborn, and the inhibition of endogenous PGI₂ synthesis leads to pulmonary vasoconstriction (1, 2). There is also evidence that PGI₂ modulates vascular cell growth in the pulmonary circulation (3). In addition, endogenous prostaglandins are important stimulators of surfactant synthesis and cell differentiation in the developing lung (4). PGI₂ is the primary prostaglandin produced in the developing pulmonary vasculature, where the main site of synthesis is the endothelium (5).

The role of PGI₂ in the developing pulmonary circulation is particularly evident in the immediate perinatal period because the inhibition of its synthesis causes marked attenuation of the fall in pulmonary vascular resistance at birth (2). This is occurring at a time when fetal plasma levels of unconjugated estrogen are rising due to marked enhancement in placental estrogen synthesis following the onset of parturition. Experiments in sheep have revealed a progressive increase in fetal plasma estrogen beginning within 12-24 h of birth, achieving 13-fold increases to levels in the range of 10⁹ M immediately before delivery (6). Studies in humans also indicate that fetal blood estrogen levels rise during the course of labor (7). Work in guinea pigs has shown that these elevated levels persist for at least 5 h after birth, returning to baseline by 24 h of age (8). Thus, the maximal effects of endogenous PGI₂ on pulmonary vasomotor tone occur when there are increasing levels of estrogen shortly before, during, and immediately after cardiopulmonary transition at birth.

In an effort to better understand the mechanisms regulating the transitional circulation, the present investigation was designed to determine the acute effects of estrogen on endothelial PGI₂ synthesis in the developing pulmonary vasculature. To avoid the cardiac and systemic effects of the hormone and to evaluate its direct effects on the pulmonary endothelium (9,10), studies were performed with isolated, early passage ovine fetal pulmonary artery endothelial cells (PAEC). We have previously demonstrated that physiologic levels of estradiol-17 (E₂) at 10¹⁰ to 10⁸ M rapidly stimulate the production of the vasodilator nitric oxide by PAEC, and that this occurs through nongenomic mechanisms which are calcium- and tyrosine kinase-mitogen-activated protein (MAP) kinase-dependent (11, 12). In addition, we have shown that prolonged E₂ exposure (48 h or longer) causes an upregulation in cyclooxygenase (COX) type I gene expression in PAEC (13). However, the acute effects of E₂ on PGI₂ production by PAEC are unknown. The hypothesis was raised that E₂ causes rapid activation of PAEC PGI₂ synthesis. In addition to testing this hypothesis, studies were performed to address the following questions: (i) are the acute effects of estrogen on PGI₂ production due to changes in COX expression?; (ii) are the rapid effects of estrogen on PGI₂ synthesis mediated by the activation of endothelial estrogen receptors (ER)?; and (iii) What are the signal transduction mechanisms by which E₂ acutely modifies PGI₂ production?

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

PAEC were obtained from mixed breed fetal lambs at 125-135 d gestation, with term being 144 ± 4 d, using methods that we have previously described (11). The PAEC were propagated in RPMI 1640 medium containing 10% iron-supplemented calf serum, 10% lamb serum, 1% L-glutamine, 1% antibiotic-antimycotic mixture, 0.15% nystatin, 0.15% gentamycin, and 0.10% tylosin, in a humidified incubator with 5% CO₂ in air at 37°C. The identity of the cells was confirmed by phenotype (cobblestone appearance and contact inhibition), by immunoflourescence studies with antibody to factor VIII-related antigen, and by examination of acetylated low-density lipoprotein uptake. The cells were studied at passage 4-6. Before study, near-confluent PAEC were placed in phenol red-free, serum-free media for 12 h to remove the effects of the estrogen-like activity of phenol red and serum-derived estrogen.

Incubations for PGI₂ Synthesis

PAEC grown in 24-well plates were preincubated for 15 min in a humidified incubator at 37°C with 500 µl of phenol red-free RPMI media added per well. The preincubation media was replaced with fresh RPMI media, and 5- to 30-min incubations were performed in the absence (basal) or presence of 10¹² to 10⁶ M estradiol-17 (E₂). In selected studies the cells were treated with 10⁵ M A23187 to evaluate maximal PGI₂ synthesis. Incubations were also performed in the absence or presence of actinomycin D treatment (preincubation with 25 µg/ml for 120 min) to determine if the observed processes are genomic or nongenomic. Additional experiments were done to compare the effects of estradiol-17 and estradiol-17. Estradiol-17 has marked cardiovascular effects in vivo, whereas estradiol-17 is comparatively less biologically active and therefore serves as a negative control (14). The role of ER activation was determined by the addition of the nonselective ER and ER antagonist, ICI 182,780 (10⁵ M) (15), or the ER-specific antagonist RR-tetrahydrochrysene (THC) (16). THC was the kind gift of John and Benita Katzenellenbogen (University of Illinois and University of Illinois College of Medicine, Urbana, IL). ICI 182,780 or THC alone had no effect on PGI₂ synthesis in control cells. Dependence on calcium and on tyrosine kinase-MAP kinase signal transduction mechanisms was determined in studies employing the intracellular calcium chelator 1,2-Bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA-AM, 20 µM) (17, 18), genistein (preincubation with 50 uM for 20 h), or the MAP kinase kinase (MEK) inhibitor PD98059 (preincubation with 50 µM for 45 min). The conditions employed for genistein and PD98059 were those used in our previous examination of E₂ activation of endothelial nitric oxide synthase (12). In preliminary studies, genistein and PD98059 had negligible effects on basal PGI₂ synthesis. At the end of the incubation, the media was placed into ice-cold tubes containing 100 µg of acetylsalicylic acid and stored at 20°C until the time of assay for PGI₂. Incubations in the presence of 100 µM indomethacin yielded PGI₂ levels below the limits of detection, demonstrating that under these conditions the PGI₂ measured is newly-synthesized. In all experiments, n = 4-6 for each determination, and findings were replicated in three independent experiments. Cells from three unique primary cultures were used in the different experiments.

Prostaglandin Assays

Samples of incubation media were assayed for the stable metabolite of PGI₂, 6-keto-prostaglandin F₁ (6-keto-PGF₁), by radioimmunoassay as previously reported (19). Briefly, the assay procedure used duplicate aliquouts of standard (0-1,000 pg) or samples placed into a solution containing 0.1 M phosphate-buffered saline (PBS) plus 0.1% polyvinylpyrrolidone (1:1). Antiserum (0.1 ml; 1:4,000 titer) and [³H]6-keto-PGF₁ (0.1 ml; 12,000 dpm) were added, and the tubes were incubated at 4°C for 12-18 h. Bound and free ligand were separated with dextran-coated charcoal, and bound ligand was counted by liquid scintillation spectrometry. The unknown quantities of PGI₂ were determined from the standard curves generated.

Immunoblot Analysis

Immunoblot analysis was performed using methods that generally followed those we have previously reported (19). Cells were harvested in ice-cold PBS containing 2.67 mM KCl, 1.47 mM KH₂OI₄, 138 mM NaCl, and 8.10 mM Na₂HPO₄, pelleted, resuspended in 50 mM Tris buffer (pH 7.4) containing 16 mM CHAPS, 100 mM NaCl, 0.5 mM EDTA, 0.02 mM EGTA, 0.4 mM -mercaptoethanol, 1.6 mM dithiothreitol, and 2 µg/ml each of soybean trypsin inhibitor, limabean trypsin inhibitor, antipain, and leupeptin, and ultrasonically disrupted (Branson Ultrasonics, Chicago, IL). The protein contents of the preparations were determined, sodium dodecylsulfate-polyacrylamide gel electrophoresis was performed on equivalent protein samples with 10% acrylamide, and the proteins were electrophoretically transferred to polyvinylidene difluoride membranes (Millipore Corp., Bedford, MA) over 2 h on ice. The membranes were blocked overnight at 4°C in buffer containing 137 mM NaCl and 20 mM Tris (pH 7.5) with 0.5% Tween-20 and 5% dried milk, and were incubated for 2 h at room temperature with either a 1:100 dilution of polyclonal antibody to COX-1 or with a 1:200 dilution of polyclonal antibody to COX-2 (Cayman Chemical Co., Ann Arbor, MI). Purified COX-1 or COX-2 proteins (Cayman Chemical) were used as positive controls. Additional samples were incubated with either 2 µg/ml of monoclonal antibody to ER (AER320; NeoMarkers, Freemont, CA) or 1 µg/ml of polyclonal antibody to ER (Affinity Bioreagents, Inc. Golden, CO) for 2 h at room temperature. ER- and ER-positive controls consisted of lysates of COS-7 cells transfected with either human ER or murine ER cDNA. After incubation with primary antibodies, the membranes were washed with the 137-mM NaCl buffer with Tween-20 at 0.2% and dried milk at 0.2%, and incubated for 2 h at room temperature with a 1:4,000 dilution of anti-rabbit (for polyclonal antibodies) or anti-mouse (for monoclonal antibodies) Ig antibody-horseradish peroxidase conjugate raised in goat (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The membranes were washed in the 137 mM NaCl buffer with Tween-20, and bands for COX-1, COX-2, ER, or ER were visualized by chemiluminescence (ECL Western Blotting Analysis System; Amersham Bio Science, Inc., Piscataway, NJ).

Reverse Transcription-Polymerase Chain Reaction Assays

To determine which isoforms of ER are expressed in fetal PAEC, reverse transcription-polymerase chain reaction (RT-PCR) assays were performed for ER and ER. Total cellular RNA was obtained from the PAEC by a single extraction using RNAzol B (Tel-test, Inc., Friendswood, TX) according to the manufacturer's instructions. cDNA synthesis was performed with 200 U murine Moloney leukemia virus reverse transcriptase (Superscript II; Lifetechnologies, Gaithersberg, MA) according to the manufacturer's instructions using 5 µg total RNA, 50 mM random hexamer primers, 1 mM dNTPs in buffer containing 50 mM Tris (pH 8.3), 37.5 mM KCl, 1.5 mM MgCl₂, and 10 mM DTT in a volume of 20 µl. In selected tubes, the reverse transcriptase was omitted to control for amplification from contaminating cDNA or genomic DNA. The temperature profile was: (i) annealing at room temperature for 10 min, (ii) extension at 42°C for 60 min, and (iii) termination at 99°C for 5 min.

PCR was performed on the resulting reverse transcription product using specific oligonucleotide primers designed from the sequence for sheep ER (20) and ER (Genbank accession no. AF177936). The sequence of the sense primer for ER was 5'-AGCATGGCCATGGAATCTGC-3' and that of the antisense primer was 5'-GTGTGTTTAATCATGATCGGG-3'. The sense primer for ER was 5'-ACGACCAAGTGCGGCTCTTG-3' and that of the antisense primer was 5'-TCTTGGCAATCAC CCAGACC-3' . The PCR reactions contained 1.5 mM Mg²⁺, 1 µM primers, 200 µM dNTPs, reaction buffer, and 5 µl cDNA in a final volume of 50 µl. To minimize nonspecific amplification, a "hot start" procedure was employed in which the PCR reaction tubes were placed in a thermal cycler (Perkin-Elmer Model 9,600; Perkin-Elmer, Wellesley, MA) prewarmed to 94°C. After 2 min, the temperature was lowered to 60°C and each tube was opened sequentially and 2.5 U (in 5 µL) of Platinum Taq DNA polymerase (Lifetechnologies) was added. The PCR temperature profile consisted of 35 cycles of 94°C for 30 s (denaturation), 60°C for 60 s (annealing), and 72°C for 60 s (extension) followed by an additional 5 min final extension at 72°C. The primer location, primer concentration, Mg²⁺ concentration, and annealing temperature were optimized to produce the greatest amount of a single PCR product. The PCR products were size-fractionated on a 2.5% NuSeive agarose (FMC Bioproducts, Rockland, ME) gel, and their identity was confirmed by direct double-stranded sequencing.

Statistical Analysis

ANOVA with Newman-Keuls post hoc testing was used to compare mean values between more than two groups. Nonparametric ANOVA was used when indicated. Single comparisons between two groups were performed with nonpaired Student's t tests. Significance was accepted at the 0.05 level of probability. All results are expressed as mean ± SEM.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

PGI₂ Synthesis

The effect of E₂ on PGI₂ synthesis is shown in Figure 1A. In incubations performed over 15 min, there was a 52% increase in PGI₂ production above basal levels with exposure to E₂. Parallel studies with the calcium ionophore A23187 revealed that maximal potential PGI₂ production was 3-fold above basal levels. The dose-response to E₂ is depicted in Figure 1B. The threshold concentration for PGI₂ stimulation was 10¹⁰ M E₂, and maximal effects were achieved at that concentration or greater. The time-course of PGI₂ stimulation by E₂ is shown in Figure 1C. The effect was detectable within 5 min of exposure to the hormone, and maximal stimulation was achieved by 15 min. Studies performed in the presence of actinomycin D to inhibit gene transcription yielded undetectable levels of PGI₂ under all conditions (data not shown).

View larger version (16K): [in this window] [in a new window]   Figure 1.   Estrogen activates PGI2 synthesis in fetal PAEC. (A) Cells were incubated under basal conditions or in the presence of estradiol-17 (E2, 108 M) or 105 M A23187 for 15 min, and PGI2 (6-keto-PGF1) synthesis was measured by RIA. (B) Dose- response to E2 in cells incubated for 15 min. (C) Time-course of the effect of 108 M E2 on PGI2 production. Results are expressed as the percent of basal production under control conditions over the same time period. Values are mean ± SEM, n = 4-6. *P < 0.05 versus basal, P < 0.05 versus E2.

COX Expression

We have previously demonstrated that acute changes in COX-1 expression in PAEC underlie oxygen-mediated changes in PGI₂ production (19). To determine if similar mechanisms play a role in the rapid effects of E₂, immunoblot analyses were performed for both COX-1 and COX-2 in cells that were exposed to the hormone for 15 min (Figure 2). The abundance of COX-1 protein was similar in control and E₂-treated cells (Figure 2, upper panel). COX-2 protein was not detected in PAEC under either control conditions or with E₂ exposure (Figure 2, lower panel). In evaluations of COX-1 and COX-2 protein abundance following actinomycin D treatment, neither enzyme protein was detected (data not shown).

View larger version (38K): [in this window] [in a new window]   Figure 2.   Brief estrogen exposure does not modify COX expression. Cells were exposed to control media (CON) or media containing 108 M estradiol-17 (E2) for 15 min, and COX-1 (upper panel) and COX-2 protein abundance (lower panel) were assessed by immunoblot analyses which included purified proteins as positive controls (+). COX-1 and COX-2 proteins were detected at 67 kD. The immunoblots shown are representative of three independent experiments.

Role of Estrogen Receptors

Comparison of the effects of estradiol-17 and estradiol-17 is provided in Figure 3A. In contrast to the 64% increase in PGI₂ synthesis observed with 10⁸ M estradiol-17, a similar concentration of estradiol-17 had no effect. There was also no stimulation of PGI₂ production with estradiol-17 at 10⁶ M. The effects of ER antagonism with ICI 182,780 are shown in Figure 3B. In contrast to the stimulation in PGI₂ production seen with E₂ alone at concentrations of 10⁸ or 10⁶ M, the concomitant addition of ICI 182,780 completely blocked the response.

View larger version (22K): [in this window] [in a new window]   Figure 3.   Rapid activation of PGI2 synthesis by estrogen is estrogen receptor-mediated. (A) Cells were incubated under basal conditions or in the presence of estradiol-17 (E2, 108 M) or estradiol-17 (E2, 108 M or 106 M) for 15 min, and PGI2 (6-keto-PGF1) synthesis was measured by RIA. (B) Cells were incubated for 15 min under basal conditions or in the presence of estradiol-17 (E2, 108 M or 106 M), with or without the addition of the ER antagonist ICI 182,780 (105 M), and PGI2 synthesis was measured. Values are mean ± SEM, n = 4-6. *P < 0.05 versus basal, P < 0.05 versus E2 alone.

Immunoblot analyses were performed to determine which ER isoforms are expressed in PAEC. Antiserum specific to ER detected a single protein species of 67 kD in PAEC lysates (Figure 4A, upper panel). In addition, antiserum specific to ER detected a single protein species of 54 kD (Figure 4A, lower panel). To confirm the identity of the ER isoforms expressed in PAEC, RT-PCR studies were done with primer pairs directed against either ER or ER. Figure 4B shows a reverse image of a typical ethidium bromide-stained agarose gels of the RT-PCR products for ER in the PAEC. Single PCR products of the correct predicted sizes were obtained for both ER and ER. The identities of the PCR products were confirmed by direct sequencing. PCR product was not obtained when the reverse transcriptase enzyme was omitted from the RT step.

View larger version (33K): [in this window] [in a new window]   Figure 4.   Estrogen activation of PGI2 synthesis is mediated by ER, which are are expressed in fetal PAEC. (A) Immunoblot analysis for ER (upper panel) or ER (lower panel) in transfected COS-7 cell positive controls or ovine fetal PAEC (EC) whole cell lysates. ER and ER proteins were detected at 67 and 54 kD, respectively. (B) Negative image of ethidium bromide-stained gel displaying RT-PCR products for ER and ER from RNA from ovine fetal PAEC. The RT step was performed in the presence (+) or absence () of reverse transcriptase. M = marker. Results are representative of three independent experiments. (C) Role of ER in estrogen-induced PGI2 synthesis. Cells were incubated for 15 min under basal conditions or in the presence of estradiol-17 (E2, 108 M), with or without the addition of the ER-selective antagonist THC (107 M), and PGI2 synthesis was measured. Values are mean ± SEM, n = 4. *P < 0.05 versus basal, P < 0.05 versus E2 alone.

To delineate the roles of ER and ER in E₂ stimulation of PGI₂ production, experiments were performed with the ER-specific antagonist THC (Figure 4C). With E₂ alone (10⁸ M for 15 min), PGI₂ synthesis was increased by 110% above basal levels. In contrast, concomitant treatment with THC completely prevented E₂ stimulation of PGI₂ production.

Signaling Mechanisms

Nongenomic effects of E₂ on endothelial nitric oxide synthase are calcium- and tyrosine kinase-MAP kinase-dependent (11, 12). The roles of these signaling mechanisms in E₂ stimulation of PGI₂ production were evaluated using the calcium chelating agent BAPTA-AM, the tyrosine kinase inhibitor genistein, and the MEK inhibitor PD98059. BAPTA- AM caused complete inhibition of E₂-induced PGI₂ synthesis (Figure 5A). In contrast, neither tyrosine kinase or MEK antagonism had any effect (Figure 5B).

View larger version (25K): [in this window] [in a new window]   Figure 5.   Role of calcium (A) and tyrosine kinase-MAP kinase signaling (B) in estrogen activation of PGI2 synthesis. Cells were incubated for 15 min under basal conditions or in the presence of estradiol-17 (E2, 108 M), with or without treatment with the calcium chelator BAPTA-AM (20 µM), the tyrosine kinase inhibitor genistein (TKI, preincubation with 50 µM for 20 h), or the MEK inhibitor PD98059 (MEKI, preincubation with 50 µM for 45 min), and PGI2 synthesis was measured. Values are mean ± SEM, n = 4. *P < 0.05 versus basal, P < 0.05 versus E2 alone.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we have evaluated the acute effects of E₂ on PGI₂ synthesis in cultured ovine fetal PAEC, enabling us to examine the direct effects of the hormone on the fetal pulmonary endothelium. We have demonstrated that E₂ causes rapid stimulation of PGI₂ synthesis (within 5 min) in the ovine fetal PAEC. To our knowledge, this is the first demonstration of a rapid effect of estrogen on PGI₂ synthesis in a pulmonary cell type.

We have shown that the response to E₂ is evident at a threshold concentration of 10¹⁰ M. This indicates that the effect of the hormone occurs at levels that are achieved in the fetal plasma in a variety of species during late gestation (6, 21, 22). For example, plasma levels of unconjugated E₂ in fetal sheep increase 5-fold from 80 to 140 d gestation (term = 144 d), reaching concentrations in the range of 10⁹ M (21). In addition, the rapidity of E₂ activation of PGI₂ production suggests that the process is nongenomic. However, attempts to differentiate genomic and nongenomic mechanisms using actinomycin to prevent gene transcription yielded no detectable PGI₂ synthesis and a loss of COX-1 expression. This observation suggests that the half-life of COX-1 protein in fetal pulmonary endothelium is within the realm of minutes, consistent with the features of the enzyme as a suicide protein (23).

To determine whether the rapid effects of E₂ on PGI₂ synthesis are related to changes in the abundance of the rate-limiting enzyme COX, studies of COX-1 and COX-2 protein expression were performed. COX-1 was constitutively expressed in the PAEC and its abundance did not change with rapid E₂ exposure, and COX-2 protein was not detected under any circumstances. Thus, in contrast to the rapid upregulation in PAEC COX-1 expression that occurs in response to increased oxygenation (19), the effects of E₂ most likely entail the activation of existing components in the prostaglandin synthetic pathway.

Because we have demonstrated that estrogen receptors (ER) are critically involved in the acute activation of endothelial nitric oxide synthase by E₂ (11, 12), the role of ER in the rapid stimulation of PGI₂ production was evaluated. In contrast to estradiol-17, estradiol-17 had no effect on prostacyclin production, and such an observation is consistent with receptor-mediated responses to estradiol-17. More importantly, rapid PGI₂ responses to estradiol-17 were completely prevented by ER antagonism with ICI 182,780, providing strong evidence that the process is mediated by nongenomic actions of ER. However, because the ICI compound is not selective for ER versus ER, the findings do not distinguish the receptor subtype that is involved. We then determined which ER isoforms are expressed in PAEC. Mimicking previous findings, ER protein was readily detectable in PAEC, and ER mRNA was also demonstrated by RT-PCR (11,13). In addition, both ER protein and mRNA were found by immunoblot analysis and RT-PCR, respectively. Whereas numerous studies by us and others have shown that endothelial cells express ER, ER expression in endothelium has thus far only been demonstrated in mature vasculature following injury (24, 25). The present findings suggest that ER may be normally expressed in vascular endothelium during specific developmental stages such as during fetal life. We also found that rapid E₂-stimulated PGI₂ production was fully prevented by simultaneous treatment with THC, which is an antagonist of ER and an activator of ER-mediated gene transcription (16). In recent work examining nitric oxide synthase stimulation in COS-7 cells expressing the enzyme and either ER or ER exclusively, we demonstrated that THC causes selective antagonism of nongenomic ER action (26). As such, the present findings provide convincing evidence that the rapid stimulation of PGI₂ production by E₂ is primarily mediated by ER.

In most paradigms the capacity for PGI₂ production is determined by the rate of mobilization of the COX substrate arachidonic acid from membrane phospholipids via the activation of phospholipase A₂, and the abundance of the COX enzyme (27). Having observed that COX expression is not modified by short-term E₂ exposure, it is most likely that E₂ causes phospholipase A₂ activation and arachidonate mobilization. There are many members of the growing PLA₂ enzyme family, and the form of PLA₂ that has been primarily implicated in agonist-stimulated prostanoid synthesis in endothelial cells is the cytosolic 85-kD form (cPLA₂). The weight of evidence is that cPLA₂ activation is a calcium-dependent and MAP kinase-dependent process, but in certain model systems the requirement for cPLA₂ phosphorylation via MAP kinase is in question (28). In the present study, we observed that E₂ activation of PGI₂ production is calcium-dependent, but tyrosine kinase-MAP kinase signaling was not required. These findings suggest that the process does indeed entail the activation of arachidonic acid mobilization by PLA₂, but it does not involve PLA₂ phosphorylation via the tyrosine kinase-MAP kinase pathway. Detailed experiments are now warranted to identify the form of PLA₂ mediating the E₂ response, and the proximal signaling mechanisms that are involved. It is important to also note that the present findings contrast with those obtained in our prior investigation of nitric oxide synthase activation by E₂, which is both a calcium-dependent and a tyrosine kinase-MAP kinase-dependent process (11, 12). As such, the signaling events which are activated by E₂ in endothelial cells are both diverse and complex.

The present observation of rapid E₂-stimulated PGI₂ production in ovine fetal PAEC is consistent with previous results obtained during 60-min incubations of cultured mature rat aortic endothelial cells (31). Similar to our findings, the rise in PGI₂ production in rat aortic endothelium was unrelated to increases in prostaglandin synthetic enzyme expression. However, the present observations provide multiple important new dimensions to our understanding of E₂ activation of PGI₂ synthesis. As outlined above, we have gained new knowledge about the signal transduction events which are requisite to this process. In addition, we have now documented the rapidity of the response (within 5 min of hormone exposure) and the obligate role of ER. Furthermore, we have specifically implicated ER in the process, thereby demonstrating that the  isoform is capable of mediating rapid E₂ responses when the receptor is present at endogenous levels.

We have previously demonstrated that long-term (48 h) E₂ exposure causes an increase in COX-1 mRNA and protein expression and at least a 64% enhancement in PGI₂ synthesis in the fetal PAEC (13). When these prior observations are combined with the present findings, it is apparent that E₂ has both nongenomic and genomic effects which enhance PGI₂ production by the developing pulmonary endothelium. We postulate that these collective mechanisms may play important roles in normal PGI₂-mediated pulmonary vasodilation in the perinatal period. In addition, there are potential pathophysiologic implications of the previous and present findings in the fetal lung. In a model of intrauterine infection induced with group B streptococcus in fetal rhesus monkeys, the normal rise in plasma estrogen which occurs before birth is absent (32). Estrogen levels are also greatly reduced in the cord blood of postmature human newborns (33). These observations suggest that in pregnancies complicated by placental dysfunction, such as that associated with intrauterine infection or postmaturity, diminished estrogen synthesis by the placenta may lead to diminished pulmonary PGI₂ synthesis, thereby contributing to the pathogenesis of persistent pulmonary hypertension of the newborn. Further studies of estrogen-mediated regulation of PGI₂ synthesis in fetal PAEC will advance both our specific knowledge of the mechanisms regulating pulmonary vascular function in the perinatal period, and our broader understanding of the unique capacity of ER to mediate nongenomic steroid hormone actions.