Prostaglandin E2—Mediated Relaxation of the Ductus Arteriosus
Effects of Gestational Age on G Protein-Coupled Receptor Expression, Signaling, and Vasomotor Control
Background— In the preterm newborn, a patent ductus arteriosus is in large part a result of the increased sensitivity of the immature ductus to prostaglandin E2 (PGE2). PGE2 acts through 3 G protein–coupled receptors (EP2, EP3, and EP4) that activate both adenyl cyclase and KATP channels. We explored these pathways to identify the mechanisms responsible for the increased sensitivity of the immature ductus to PGE2.
Methods and Results— We measured EP receptor content (mRNA and protein), receptor binding, cAMP production, and isometric tension in rings of ductus taken from immature (65% gestation) and mature (95% gestation) sheep and baboon fetuses. Ductus relaxation and cAMP generation were augmented in response to selective EP receptor agonists in the immature ductus. 8-Br-cAMP, a stable cAMP analogue, produced greater relaxation in the immature ductus. In the presence of a selective protein kinase A inhibitor, Rp-8-CPT cAMPS, the developmental differences in sensitivity to PGE2 could no longer be demonstrated. EP2, EP3, and EP4 receptor densities were higher in immature ductus, despite similar receptor mRNA and protein contents at the 2 gestational ages. In contrast, forskolin and NaF, direct activators of adenyl cyclase and Gs, respectively, elicited comparable increases in cAMP in both age groups. KATP channel inhibition also had similar effects on PGE2-induced relaxation in both age groups.
Conclusions— Two mechanisms explain the increased sensitivity of the immature ductus to PGE2: (1) increased cAMP production because of increased binding of PGE2 to the individual EP receptors and (2) increased potency of cAMP on protein kinase A–regulated pathways.
Received March 1, 2004; accepted May 20, 2004.
Preterm infants have a higher incidence of patent ductus arteriosus than do term infants. In large part, this is due to the increased sensitivity of the premature ductus to prostaglandin (PG) E2.1,2 The mechanisms responsible for the increased sensitivity of the immature ductus are currently unclear. In the lamb and pig fetus, PGE2 exerts its effects through 3 selective prostaglandin receptors: EP2, EP3D, and EP4.3,4 When stimulated, these G protein–coupled receptors activate adenyl cyclase and KATP channels.3
We designed the following study using sheep and baboon ductus arteriosus to examine the mechanisms responsible for the increased sensitivity of the immature ductus to PGE2. Our findings reveal that the increased sensitivity of the immature ductus is not due to differences in KATP channel activation or differences in the amount of EP receptor mRNA or protein. Rather, the increased sensitivity appears to be a result of increased cAMP production secondary to increased availability of EP receptors for PGE2 binding and increased potency of cAMP on protein kinase A–regulated relaxation.
Pregnant ewes (mixed Western breed for contractility and biochemical studies and Merino for biochemical studies) were anesthetized as previously described (either IV ketamine or IM ketamine/xylazine with lidocaine spinal anesthetic),3 and the fetal lambs were delivered by cesarean section at 2 gestational ages: Immature=96 to 107 days and Mature=135 to 144 days gestation (full term=150 days gestation).
Fetal baboons (Papio sp) were delivered by cesarean section at either 175 days (n=8) or 125 days (n=8) gestation (full term is 185 days) and euthanized with pentobarbital sodium. Their ductus arteriosus was removed for polymerase chain reaction (PCR) studies.
These procedures were approved by the Committees on Animal Research at the University of California, San Francisco, at the Department of Agriculture, Western Australia, and at the Southwest Foundation for Biomedical Research, San Antonio, Tex.
Isometric Tension In Vitro
The relaxing effects of prostaglandins were measured in rings of ductus arteriosus that were precontracted with oxygen and in which endogenous prostaglandin and nitric oxide production were maximally inhibited.3 Rings were suspended at 38°C in a modified Krebs solution (in mmol/L: 118 NaCl, 4.7 KCl, 2.5 CaCl2, 0.9 MgSO4, 1 KH2PO4, 11.1 glucose, 23 NaHCO3 [pH 7.4], equilibrated with 5% CO2/30% O2/65% N2) containing indomethacin 5.6×10−6 mmol/L and NG-nitro-l-arginine methyl ester (L-NAME) 10−4 mmol/L3). The bath solution was changed every 20 minutes. Each of the rings was stretched to an initial length that resulted in a maximal isometric contractile response to increases in oxygen tension.3 After the rings reached a steady-state tension (≈100 to 120 minutes), potassium (K+)-Krebs solution (containing KCl 10−1 mmol/L substituted for an equimolar amount of NaCl) was used to measure the maximal contraction that could be developed by the ductus (maximal contraction).
The (K+)-Krebs solution was washed out of the bath (prerelaxation tension), and cumulative relaxation dose-response curves were constructed for prostaglandin E2 (PGE2), Butaprost (an EP2 agonist), M&B 28767 (EP3 agonist), and 8-bromoadenosine 3′,5′ cAMP (8-Br-cAMP, a nonhydrolyzable cAMP mimetic). The concentrations that produce 50% of the maximal response to the drug (EC50 values) were determined from each dose-response curve. In some experiments, a cumulative dose-response curve was performed to a relaxing agent after the tissue had equilibrated with either an EP4 receptor antagonist (AH23848B,3), a selective inhibitor of KATP channels (glibenclamide), or a selective inhibitor of protein kinase A, 8-(4-chlorophenylthio)adenosine-3′,5′-cyclic monophosphorothioate, Rp isomer (Rp-8-CPT-cAMPS).5 Sodium nitroprusside (10−4 mmol/L) was added to each ring at the end of the experiment to determine its minimal tension.3 The difference in tensions between the prerelaxation tension and the tension after sodium nitroprusside was considered the net tension (Immature=273±139 g/cm2; Mature=324±207 g/cm2). Tensions are expressed as a percentage of the net tension. After the experiment, the tissues were removed from the baths and blotted dry, and their wet weights were determined.
AH23848B was generously provided by Dr Simon Lister (Glaxo-Wellcome); M&B28767 by Dr Jean Hough (Rhone-Poulenc Rorer). PGE2, butaprost (Cayman Chemical), and Rp-8-CPT-cAMPS (Biolog, Life Science Institute) were purchased. All other chemicals were from Sigma Chemical Co.
EP Receptor Binding, cAMP Generation, and Immunoblotting of EP Receptors
Assays of EP receptor binding, cAMP generation, and EP receptor immunoblotting have been published in detail previously.3,4
[3H]PGE2 binding and displacement studies with 16,16-dimethyl-PGE2 (a nonselective EP agonist), butaprost, M&B28767, and AH23848B were performed on ductus membranes; total and individual EP receptor densities (Bmax) were determined.3,4
For cAMP generation, ductus homogenates were incubated for 10 minutes (in mmol/L: 1 ATP, 7.5 MgCl2, 15 creatine phosphate, 0.5 EGTA, 0.5 IBMX, 1 dithiothreitol, 1 benzamidine, and 0.1 PMSF, and 185 U/mL creatine phosphokinase, 200 μg/mL aspirin, and 100 μg/mL soybean trypsin inhibitor), and cAMP was measured by radioimmunoassay.3,4
Immunoblots of EP2 and EP4 were performed on ductus arteriosus membranes using a monoclonal antibody against EP2 (3E6, from Dr J. Castracane, Exalpha) and a polyclonal anti-EP4 antibody.3,4 No antibody is currently available for EP3D.
EP Receptor Expression: Preparation of Total RNA, Reverse Transcription, and Quantitative PCR
Total RNA was isolated from fetal sheep and baboon ductus arteriosus as described elsewhere.3 We used the TaqMan Universal PCR master mix of PE Applied Biosystems to quantify the expression of EP receptors. TaqMan probes were designed using the Primer Express program and labeled with fluorophores FAM (6-carboxy-fluorescein) and TAMRA (6 carboxy-tetramethyl-rhodamine) as reporter and quencher dyes, respectively. An ABI PRISM 7700 Sequence Detection system was used to determine number of PCR cycles required for product detection (the cycle threshold [CT] value). Reactions were performed in triplicate. The smaller the number of starting copies of a gene, the higher the CT value required for product detection. All reactions were repeated on at least 3 separate days. Data were analyzed using the Sequence Detector version 1.6.3 program.
In preliminary experiments, we found that the CT values of baboon and sheep malate dehydrogenase (MDH) and GAPDH genes were constant throughout gestation: baboon MDH (CT [125 days]=27.9±0.6 cycles, n=8; CT [175 days]=27.9±0.6, n=8), sheep MDH (CT [100 days]=25.4±0.3, n=7; CT [138 days]=25.4±0.4, n=7), and sheep GAPDH (CT [100 days]=24.0±0.2, n=7; CT [138 days]=23.7±0.4, n=7). Therefore, MDH and GAPDH were used as internal controls to normalize the degree of expression of the EP receptors using the relative gene expression method.
Statistical analyses of unpaired and paired data were performed with the appropriate t test or Mann-Whitney test. When >1 comparison was made, the Bonferroni correction was used. Values are expressed as mean±SD. Drug doses refer to their final molar concentration in the bath. A probability value of P<0.05 was considered significant.
We first looked at the effect of gestation on the ductus’ relaxation response to PGE2. In the presence of 30% oxygen, indomethacin, and L-NAME, the immature fetal ductus develops a net tension (75±14% of maximal active tension) similar to that of the mature fetal ductus (78±9% of maximal active tension). The immature ductus is more sensitive to the relaxing effects of both nonselective (PGE2) and selective (EP2=butaprost, EP3=M&B 28767) EP receptor stimulation (Figure 1). The EP4 receptor antagonist AH23848B shifts the EC50 of PGE2 to the right. AH23848B tends to have a greater effect on the PGE2 EC50 in the immature (5.8±4-fold increase) than in the mature (2±1-fold increase, P<0.10) ductus (Figure 1).
To determine whether the increased sensitivity of the immature ductus to PGE2 could be a result of differences in signaling through KATP channels (Figure 2) or protein kinase A (Figure 3), we used a selective KATP channel inhibitor, glibenclamide (10−5 mmol/L) and a selective protein kinase A inhibitor, Rp-8-CPT cAMPS. Glibenclamide decreased the sensitivity of the ductus to PGE2 at both gestational ages. However, it did not eliminate the significant difference in PGE2 sensitivity between the 2 age groups (Figure 2).
Rp-8-CPT cAMPS (250 μmol/L) blocked the relaxation caused by 8-Br-cAMP and decreased the sensitivity of the ductus to both nonselective and selective EP agonists (Figure 3). Rp-8-CPT cAMPS had a significantly greater effect on the PGE2 EC50 in the immature (6.1±1.5-fold increase) than in the mature (2.2±0.8-fold increase, P<0.01) ductus. In the presence of Rp-8-CPT cAMPS, the statistically significant difference in PGE2 sensitivity between the 2 age groups could no longer be demonstrated. This suggests that the increased sensitivity of the immature ductus to PGE2 may be due to increased signaling through the protein kinase A pathway.
Therefore, we examined whether the increased sensitivity of the immature ductus to PGE2 might be due to increased cAMP production after PGE2 stimulation. Both the nonselective EP receptor agonist PGE2 and the selective EP receptor agonists butaprost (EP2) and M&B 28767 (EP3) increased cAMP production in the lamb ductus arteriosus (Figure 4). To test the role of EP4, we used the EP4 antagonist AH23848B in the presence of PGE2. AH23848B decreased PGE2-induced cAMP formation (Figure 4). This suggests that some of the cAMP formation that results from PGE2 stimulation is due to activation of EP4 receptors.
Both the selective and nonselective EP receptor agonists produced a greater increase in cAMP production in the immature ductus than in the mature ductus. Similarly, the EP4 receptor antagonist AH23848B caused a greater inhibition of cAMP production in the immature ductus (Figure 4).
We used forskolin and NaF to determine whether differences in maximal adenyl cyclase activity could account for the increased PGE2-induced cAMP formation in the immature ductus. NaF maximally activates G protein–coupled adenyl cyclase6,7; forskolin maximally activates adenyl cyclase through a mechanism that is independent of G proteins.8,9 There was no difference between the immature and mature ductus in the amount of forskolin- or NaF-induced cAMP production (Figure 5).
In addition to the increased production of cAMP after PGE2-stimulation, the immature ductus is more sensitive to the relaxing effects of cAMP itself. Both 8-Br-cAMP, the nonhydrolyzable analogue of cAMP, and forskolin have twice the potency in the immature ductus (Figure 1).
We next looked for differences in EP receptor density. Maximal specific binding of [3H]PGE2 to immature ductus membranes was 2.6-fold greater than that to mature ductus membranes (Figure 6). The Bmax for all 3 EP receptors was increased in the immature ductus. The increased receptor binding in the immature ductus was not due to increased expression of EP receptor mRNA (Figure 7) or protein (Figure 8). We examined the changes in mRNA expression in 2 separate species; although the relative expression of individual EP receptors appeared to differ between baboon and sheep, the immature ductus in both species had either the same amounts of or less mRNA than the mature ductus.
The present studies confirm our previous findings that the immature fetal ductus arteriosus is more sensitive than the late-gestation fetal ductus to the relaxing effects of PGE2.1,10 The immature ductus is also more sensitive to stimulation of each of the individual EP receptors than is the ductus near term (Figure 1). EP receptor stimulation relaxes the fetal lamb ductus arteriosus by increasing cAMP and activating KATP channels (Figures 2 to 4⇑⇑).3 We found that the increased sensitivity of the immature ductus to EP receptor stimulation can be explained by differences in cAMP signaling (rather than by differences in KATP channel activity). Stimulation of EP receptors in the immature ductus produces a greater degree of cAMP production (Figure 4) and protein kinase A–mediated relaxation (Figure 3) than is found in the mature ductus. When protein kinase A activity is inhibited, the immature ductus loses its increased sensitivity to PGE2 and behaves like the mature ductus (Figure 3).
Our data suggest that the increased cAMP-mediated relaxation in the immature ductus is in part a result of both increased binding of PGE2 to EP receptors and increased potency of cAMP on protein kinase A–regulated pathways. There does not seem to be a difference in the amount of EP receptor mRNA or protein between the 2 gestational ages studied, nor is there a difference in the maximal activity of adenyl cyclase after forskolin or NaF stimulation.
Our findings differ somewhat from those of Smith et al.11 These authors did not detect EP2 receptors in the sheep ductus arteriosus. We detected the presence of EP2 receptor protein and mRNA in all of our ductus samples and demonstrated that the highly selective EP2 agonist butaprost relaxed the fetal lamb ductus (Figures 1, 7, and 8⇑⇑). The reduction in EP4 expression observed by Smith et al is in some ways consistent with the decreased binding that we observed with advancing gestation. However, in contrast with Smith et al, we found that although binding decreased with advancing gestation, EP receptor expression (mRNA and protein) was unchanged. This suggests that in the mature ductus, the availability of EP receptors for PGE2 is decreased. Changes in receptor affinity (altered Gs protein binding–induced conformational changes12) and/or differences in receptor phosphorylation pattern (affecting ligand binding13 and receptor internalization14) could be responsible for the decrease in Bmax with advancing gestation. Elevated PGE2 levels have been associated with EP receptor internalization and decreased PGE2 binding in other fetal vessels.15–17 Because PGE2 production in the lamb ductus increases with advancing gestation,10 we hypothesize that elevated PGE2 concentrations also may play a role in receptor desensitization in this vessel.
We observed no gestational difference in NaF-induced cAMP production. Although this rules out differences in total Gs protein–coupled adenyl cyclase activation between immature and mature ductus, it still is possible that a specific Gs protein that couples to EP receptors might be different between the 2 age groups.
Our studies do not address the possibility that decreased phosphodiesterase activity may play a role in the increased sensitivity of the immature ductus to EP receptor–mediated relaxation. Our experiments were designed to measure cAMP production; therefore, in our cAMP generation assays, ductus membranes were treated with nonselective phosphodiesterase inhibitors. In preliminary studies, we have found that phosphodiesterase inhibitors have a more potent effect on ductus relaxation in the immature fetus than in the late gestation ductus (unpublished results).
We have previously shown that stimulation of the EP3D receptor relaxes the lamb ductus by activating glibenclamide-sensitive KATP channels.3 In the present study, we also found that stimulation of the EP3D receptor relaxes the ductus by increasing cAMP (Figures 3 and 4⇑). Although stimulation of EP3 receptors caused a small constriction of the rabbit ductus arteriosus,18 its major effect was relaxation consistent with our observations. In our hands, cumulative dose-response curves to 2 selective EP3 agonists (M&B 28767 [10−9 to 10−6 mmol/L] and GR63799X [10−8 to 10−5 mmol/L]) elicited relaxation even in the presence of forskolin, indomethacin, L-NAME, and K-channel inhibitors (n=3, data not shown). It is interesting to note that the protein kinase A inhibitor Rp-8-CPT cAMPS affected the efficacy of EP3-induced relaxation to a much smaller degree than the efficacy of either butaprost or 8-Br-cAMP (which are mediated exclusively through protein kinase A–dependent pathways) (Figure 3). This finding is consistent with our other observations that signaling through EP3 involves both protein kinase A–dependent and protein kinase A–independent pathways.
This study was supported by grants from the US Public Health Service (HL-46691 and HL-56061), the Medical Research Council of Canada, the Heart and Stroke Foundation of Quebec, the Fonds de la Recherche en Santé du Québec, and a gift from the Perinatal Associates Research Foundation. NHLBI grant HL-52636 funded the BPD Resource Center. The authors thank Drs Alan Jobe, Machiko Ikegami, and John Newnham for their invaluable help in obtaining sheep ductus and Hendrika Fernandez and Francoise Mauray for their expert technical assistance.
↵*The first 2 authors contributed equally to this work.
Bouayad A, Kajino H, Waleh N, et al. Characterization of PGE2 receptors in fetal and newborn lamb ductus arteriosus. Am J Physiol. 2001; 280: H2342–H2349.
Bhattacharya M, Asselin P, Hardy P, et al. Developmental changes in prostaglandin E2 receptor subtypes in porcine ductus arteriosus: possible contribution in altered responsiveness to prostaglandin E2. Circulation. 1999; 100: 1751–1756.
Gjertsen BT, Mellgren G, Otten A, et al. Novel (Rp)-cAMPS analogs as tools for inhibition of cAMP-kinase in cell culture: basal cAMP-kinase activity modulates interleukin-1 beta action. J Biol Chem. 1995; 270: 20599–20607.
Kanaho Y, Moss J, Vaughan M. Mechanism of inhibition of transducin GTPase activity by fluoride and aluminum. J Biol Chem. 1985; 260: 11493–11497.
Seamon KB, Padgett W, Daly JW. Forskolin: unique diterpene activator of adenylate cyclase in membranes and in intact cells. Proc Natl Acad Sci U S A. 1981; 78: 3363–3367.
Negishi M, Irie A, Sugimoto Y, et al. Selective coupling of prostaglandin E receptor EP3D to Gi and Gs through interaction of alpha-carboxylic acid of agonist and arginine residue of seventh transmembrane domain. J Biol Chem. 1995; 270: 16122–16127.
Murray R, FitzGerald GA. Regulation of thromboxane receptor activation in human platelets. Proc Natl Acad Sci U S A. 1989; 86: 124–128.
Ferguson SS. Evolving concepts in G protein–coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev. 2001; 53: 1–24.
Li DY, Varma DR, Chemtob S. Up-regulation of brain PGE2 and PGF2 alpha receptors and receptor-coupled second messengers by cyclooxygenase inhibition in newborn pigs. J Pharmacol Exp Ther. 1995; 272: 15–19.
Li DY, Abran D, Peri KG, et al. Inhibition of prostaglandin synthesis in newborn pigs increases cerebral microvessel prostaglandin F2 alpha and prostaglandin E2 receptors, their second messengers and vasoconstrictor response to adult levels. J Pharmacol Exp Ther. 1996; 278: 370–377.