Endothelin-1-induced vasoconstriction is not mediated by thromboxane release and action in the human fetal-placental circulation

Effects of Oxygen Tension and Normalization Pressure on Endothelin-Induced Constriction of Human Placental Chorionic Plate Arteries

OBJECTIVES

Fetoplacental blood vessels constrict in response to endothelin (ET-1) or reduced oxygen tension in the placental cotyledon perfused in vitro. In nonplacental resistance arteries, hypoxia and ET-1 induce constriction by promoting Ca2+ influx into smooth muscle through membrane ion channels, which include voltage-gated Ca2+ channels (VGCCs). We hypothesized that VGCCs are involved in ET-1-induced constriction of fetoplacental resistance vessels and that their contribution to constriction is enhanced at low oxygen tension.

METHODS

Chorionic plate small arteries from term placentas were studied using parallel wire myography. Arteries were normalized at 0.9 of L(5.1 kPa) ("low stretch" approximately 25 mm Hg; approximating physiologic vascular pressure) or 0.9 of L(13.3 kPa) ("high stretch" approximately 42 mm Hg) and experiments performed at oxygen tensions of 156, 38, and 15 mm Hg.

RESULTS

When chorionic plate arteries were normalized at low stretch, oxygen tension did not affect constriction to ET-1. Nifedipine (10(-4) M), a blocker of L-type VGCCs, inhibited ET-1 (EC80)-induced constriction to a similar extent at each oxygen tension (52% to 64% inhibition). In contrast, when arteries were normalized at high stretch, constriction to ET-1 was greater at 38 than at 156 or 15 mm Hg oxygen and nifedipine inhibition of ET-1-induced constriction was greater at 38 and 15 mm Hg than at 156 mm Hg oxygen.

CONCLUSIONS

VGCCs and nifedipine-insensitive processes underlie the contractile response of chorionic plate arteries to ET-1 and their relative contribution to vasoconstriction is modulated by oxygen tension when vessels are normalized at high stretch. However, contrary to our hypothesis, the response of chorionic plate arteries to ET-1 is not modulated by oxygen when vessels are normalized at physiologic pressure.

Vaso-active factors in pregnancy

Pregnancy results in profound physiological changes in the cardiovascular system, yet these changes are completely reversible. It is apparent that vaso-active factors, some as yet probably unidentified, which act as humoral or local autocrine or paracrine regulators of vasular resistance, play a major role in these cardio-vascular changes. This role may be heightened in pregnancy when there has to be a large increase in blood flow to the uterus and placenta while maintaining adequate flow to other vascular beds. Our knowledge of the mechanisms of action of these vaso-active factors and their interactions with each other still remains incomplete. Alterations in synthesis and action of these vaso-active factors may occur in pregnancies associated with pregnancy-induced hypertension, pre-eclampsia or intra-uterine growth retardation. Investigation of such alterations may help to elucidate the roles of vaso-active factors in both normal and pathological situations. The gestational hormones oestrogen and progesterone, are obviously prime candidates as overall regulators of the cardiovascular changes of pregnancy and as agents which alter the synthesis or action of other vaso-active factors. Currently, much attention is being focused on the role of local autocrine or paracrine vaso-active factors which may be produced by the endothelium or by the underlying vascular smooth muscle cells and alterations in their production or action in the hyptertensive disorders of pregnancy. The endothelium forms the largest endocrine organ within the body and so its importance in the mediation of vascular events should not be under-estimated. The principal objective of this review is to examine the roles of these many autocrine and paracrine vaso-active factors during pregnancy and their relation with the overall regulation of the vascular system. Changes which may occur and be involved in the aetiology of pre-eclampsia and growth retardation will also be examined.

Control of the fetoplacental circulation

This year marks the 250th anniversary of the discovery by William Hunter of the existence of two distinct circulations within the human placenta. Until relatively recently the placenta has been viewed with “respect” – a passive structure which occasionally elicited fear and anxiety if implanted either too low or too deep. More recently our understanding of perinatal physiology, blood flow regulation and epidemiological data linking prenatal events with adult disease has created renewed interest in the placenta. This review will focus on the regulation of fetal blood flow in the placenta, the possible mechanisms whereby it may be deranged and why this might be relevant to adult disease.

Neuroimaging of delayed eclampsia. Report of 3 cases and review of the literature

Computed tomography and magnetic resonance imaging findings consistent with eclampsia were recently encountered in 3 patients who developed severe headache days to weeks after uncomplicated delivery. The neurologic presentation was nonspecific, and pre-eclamptic symptoms were not present, including significant hypertension. Variable expression of pre-eclampsia occurred during the course of their observation. Conventional angiography obtained in 2 patients because of a concern for aneurysm demonstrated central and peripheral vasospasm. Magnetic resonance angiography demonstrated central vasospasm in the third patient.

Heterogeneity of microvascular endothelial cells isolated from human term placenta and macrovascular umbilical vein endothelial cells

The present study compares some phenotypic and physiologic characteristics of microvascular and macrovascular endothelial cells from within one human organ. To this end microvascular endothelial cells from human full-term placenta (PLEC) were isolated using a new method and compared with macrovascular human umbilical vein endothelial cells (HUVEC) and an SV40-transformed placental venous endothelial cell line (HPEC-A2). PLEC were isolated by enzymatic perfusion of small placental vessels, purified on a density gradient and cultured subsequently. Histological sections of the enzyme-treated vessels showed a selective removal of the endothelial lining in the perfused placental cotyledons. The endothelial identity of the cells was confirmed by staining with the endothelial markers anti-von Willebrand factor, Ulex europaeus lectin and anti-QBEND10. The cells internalized acetylated low-density lipoprotein and did not show immunoreactivity with markers for macrophages, smooth muscle cells and fibroblasts. The spindle-shaped PLEC grew in swirling patterns similar to that described for venous placental endothelial cells. However, scanning electron microscopic examination clearly showed that PLEC remained elongated at the confluent state, in contrast to the more polygonal phenotype of HPEC-A2 and HUVEC that were studied in parallel. The amount of vasoactive substances (endothelin-1,2, thromboxane, angiotensin II, prostacyclin) released into the culture medium and the proliferative response to cytokines was more similar to human dermal microvessels (MIEC) derived from non-fetal tissue than to HUVEC. Potent mitogens such as vascular endothelial growth factors (VEGF121, VEGF165) and basic fibroblast growth factor (FGF-2) induced proliferation of all endothelial cell types. Placental growth factors PIGF-1 and PIGF-2 effectively stimulated cell proliferation on PLEC (142 +/- 7% and 173 +/- 10%) and MIEC (160 +/- 20% and 143 +/- 28%) in contrast to HUVEC (9 +/- 8% and 15 +/- 20%) and HPEC-A2 (15 +/- 7% and 24 +/- 6%) after 48 h incubation time under serum-free conditions. These data support evidence for (1) the microvascular identity of the isolated PLEC described in this study, and (2) the phenotypic and physiologic heterogeneity of micro- and macrovascular endothelial cells within one human organ.

Role of placenta in preeclampsia

Preeclampsia, which manifests itself as hypertension, proteinuria, and edema in pregnancy, requires the presence of trophoblast tissue but not a fetus. It is characterized by abnormal trophoblast invasion of the spiral arteries of the decidua and myometrium leading to a failure to establish an adequate uteroplacental blood flow and, therefore, is thought to give rise to relatively hypoxic trophoblast tissue. This, in turn, may promote an exaggerated state of oxidative stress in the placenta. This hypoxia/oxidative stress may then further attenuate trophoblast invasion but also alters placental villous angiogenesis leading to a poorly developed fetoplacental vasculature with abnormal reactivity. Oxidative stress per se may also affect vascular reactivity, blood flow, and oxygen and nutrient delivery to the fetus, which ultimately may be compromised. The synthetic and transport functions of the syncytiotrophoblast may also be altered, and there is an increased rate of trophoblast apoptosis. The linkage among abnormal trophoblast invasion, trophoblast dysfunction, and the maternal disease remains unidentified. The presumptive humoral factor that is released by the preeclamptic placenta to cause maternal disease remains elusive. Current therapies to prevent preeclampsia aim toward preventing the maternal syndrome, not preventing the primary pathophysiology.

Mechanisms of endothelin-1-induced contraction in isolated placental veins from normal full-term and preterm pregnancies

This study characterizes the reactivity of human chorionic plate vein in full-term (39.4+/-0.3 weeks of gestation) and preterm (34.4+/-0.6 weeks of gestation) pregnancy to endothelin-1 (ET-1) and attempts to characterize ET-1 receptor subtype, and the contribution of nitric oxide and cyclooxygenase products in these responses. In placental veins from full-term and preterm pregnant women, cumulative addition of ET-1 (10(-10)-10(-6) M) caused marked and long-lasting concentration-dependent contractile responses. The mean EC(50) and E(max) values for ET-1-induced venoconstriction did not differ between the full-term and preterm pregnancy groups. In the veins from preterm placental preparations, the ET(A) receptor-selective antagonist cyclo(D-alpha-aspartyl-L-propyl-D-valyl-L-leucyl-D-tryptophyl (BQ123) reduced the ET-1-induced contraction by 28.6+/-2.4%, compared to a decline in tension of 51.2+/-4.2% in the full-term placental vessels. The ET(B) receptor-selective antagonist, N-[N-[N-[2,6-dimethyl-1piperidinyl)carbonyl]-4-methyl-L-leucyl]-1-(methoxycarbonyl)-D-tryptophyl]-D-norleucinemonosodium (BQ788), did not influence ET-1-induced contraction in placental vein from both pregnancy groups in terms of maximal contraction and sensitivity. Pretreatment with the cyclooxygenase inhibitor, indomethacin (1 microM) and the nitric oxide synthase inhibitor N(w)-nitro-L-arginine (NOLA, 100 microM) did not significantly affect either the EC(50) or the maximum contraction to ET-1 in veins from normal full-term and preterm preparations. The results of this study suggest that there is no correlation between ET-1-induced vasoconstriction and gestational age and that this vasoconstriction is mediated predominantly via ET(A) receptor subtype in both groups of pregnant women, independent of NO and eicosanoids.

Autacoid interactions in the regulation of blood flow in the human placenta

1. Interactions between autacoids may play important roles in the regulation of blood flow in the foetal placenta. In order to investigate this aspect of placental haemodynamics, human normal-term placentae were perfused in vitro and the responses of the foetal vessels to various combinations of vasoactive agents were determined. 2. Vasoconstriction responses to 5-hydroxytryptamine (5-HT) were potentiated in the presence of endothelin-1 (ET-1), the thromboxane A2-mimetic U46619 and a nitric oxide synthase inhibitor, N-nitro-L-arginine (NOLA), but not in the presence of angiotensin II. 3. N-Nitro-L-arginine caused vasoconstriction of the perfused placenta and indomethacin attenuated this effect and blocked the potentiation of the 5-HT response by NOLA. 4. Indomethacin did not affect ET-1-induced pressure increases and infusion of U46619 had no effect on release of ET-like immunoreactivity into the foetal placental circulation. 5. The present study provides evidence of interactions between several autacoids in human perfused placentae in vitro. These interactions may play important roles in foetal placental haemodynamics in normal or pathological situations.

Effects of continuous infusion of endothelin-1 in pregnant sheep

Plasma concentration of endothelin-1, a potent vasoconstrictor produced by the vascular endothelium, has been observed to be significantly increased in a number of pathophysiological states, including preeclampsia. In the present study we have evaluated the effects of elevated plasma endothelin-1 in pregnant sheep by continuous exogenous endothelin-1 administration. Nine pregnant ewes (110+/-5 days' gestation) were instrumented for measurements of maternal mean arterial pressure, renal blood flow, and uterine blood flow. After recovery, endothelin-1 was infused intravenously for 4 hours at a dose that was adjusted to raise mean arterial pressure by approximately 20 mm Hg by the end of the first hour (range 5 to 20 ng/kg per minute). Mean arterial pressure, renal blood flow, uterine blood flow, urinary protein excretion, hematocrit, and plasma endothelin-1 concentration were measured hourly, and renal and uterine vascular resistances were calculated. Endothelin-1 produced significant increases (% change from baseline at t=4 hours) in mean arterial pressure (45+/-8%), renal vascular resistance (353+/-66 %), and uterine vascular resistance (59+/-21%). Endothelin-1 also increased microvascular permeability both systemically and within the kidney, as suggested by marked increases in hematocrit (0.27+/-0.01 to 0.32+/-0.01) and urinary protein concentration (0.95+/-0.1 to 7.9+/-3.2 mg/mL per mg creatinine). There was a highly significant correlation (P<.0001) between plasma endothelin-1 and mean arterial pressure, renal vascular resistance, uterine vascular resistance, hematocrit, and urinary protein content in all sheep studied. In addition, plasma endothelin-1 corresponded well with the time course of the changes in cardiovascular parameters and urinary protein excretion observed. These results provide evidence to suggest that elevation of circulating endothelin-1 in pregnant sheep can produce cardiovascular and hemodynamic changes that in many ways resemble the human disease preeclampsia. This supports the hypothesis that endothelial cell damage and/or dysfunction that is associated with increased production of endothelin-1 could directly contribute to the progression of preeclampsia.

Elevated endothelin levels are associated with increased placental resistance

OBJECTIVE

The purpose of this study was to measure cord blood endothelin-1,2 concentrations in growth-restricted infants with abnormal flow velocity waveforms.

STUDY DESIGN

Endothelin-1,2 concentrations were measured by radioimmunoassay in the cord blood of 16 growth-restricted infants with abnormal flow velocity waveforms before delivery, 16 growth-restricted infants with normal flow velocity waveforms before delivery, and 44 appropriately grown infants. Clinical data regarding pregnancy complications and neonatal outcome were collected.

RESULTS

The mean endothelin-1,2 concentration in growth-restricted infants with abnormal flow velocity waveforms (50.2 +/- 16.4 pg/ml) was significantly higher than in growth-restricted infants with normal flow velocity waveforms (33.3 +/- 14.2 pg/ml, p < 0.05) or in appropriately grown infants (25.8 +/- 9.7 pg/ml, p < 0.05). Oligohydramnios was also associated with elevated endothelin levels.

CONCLUSION

We conclude that endothelin-1,2 concentrations are elevated in growth-restricted infants with abnormal flow velocity waveforms and may play a role in the development of abnormal fetoplacental resistance.

Endothelin-1-like immunoreactivity and endothelin receptors in the human placenta from normotensive and hypertensive pregnancies

The levels of endothelin-1-like immunoreactivity (ET-1-LI) and characteristics of endothelin receptors in the chorionic villous tissue of human placenta were determined. The ET-1-LI level in chorionic villous tissue obtained from normal term placenta was 2,450 +/- 940 pg/g wet weight (mean +/- SD, n = 4). Further analysis using gel permeation chromatography and reverse-phase high performance liquid chromatography showed that the main ET-1-LI constituent of ET-1-LI in this tissue was ET-1. Scatchard analysis of [125I]ET-1 binding to the membrane fraction of chorionic villous tissue obtained from term placenta showed high affinity receptor sites with an apparent dissociation constant (Kd) of 23.6 +/- 11.1 pM and a Bmax value of 388 +/- 238 fmol/mg protein (n = 5). The same binding study with [125I]ET 3 showed a Kd of 13.9 +/- 3.8 pM and a Bmax value of 176 +/- 78 fmol/mg protein (n = 5). These results suggest that both ET-A and ET-B receptors (ET-AR and ET-BR) are expressed in chorionic villous tissue. This finding was further confirmed by Northern blot analysis showing the expression of both ET-AR and ET-BR mRNAs in this tissue. ET-1-LI in the umbilical venous plasma of the newborns from women with pregnancy-induced hypertension (PIH) (38.3 +/- 10.4 pg/mL, n = 5) was significantly (P < 0.05) higher than that in the normal newborns from normotensive pregnant women (26.3 +/- 5.2 pg/mL, n = 12). However, in placental chorionic villous tissue obtained from PIH women, both ET-1-LI level and ET binding profile were not different from those in chorionic villous tissue from normotensive pregnant women. These results suggest that the abundant ET-ET receptor system is present in the placental chorionic villous tissue and that this system is not the major factor of the pathogenesis of placental dysfunction occurring in PIH because these systems are similar in normotensive and hypertensive pregnancies.

Role of thromboxane A2 in the vasoconstrictor response to endothelin-1, angiotensin II and 5-hydroxytryptamine in human placental vessels

The possibility that the vasoconstrictor agents present in the circulation may mediate their effect on placental blood flow by stimulating local prostanoid production was investigated. Placental cotyledons obtained at term from normal pregnancies were perfused in vitro. The dose-related vasoconstrictor effects of endothelin-1 (ET-1), angiotensin II (A II) and 5-hydroxytryptamine (5 HT) were reduced by graded concentrations of the thromboxane A2 (TXA2) receptor antagonist GR32191 (10(-7)-10(-4)M), aspirin (10(-5)-10(-4)M) and indomethacin (10(-5)M). The effect of the TXA2 receptor agonist U46619 was totally abolished by GR32191 (10(-7)M) but unaffected by the prostanoid synthesis inhibitors, aspirin and indomethacin although they prevented a self-priming effect of U46619. When effluents were collected from perfused placentae, 5-10 min after administration of sub-maximal doses of ET-1 (20 pmol), A II (500 pmol) and 5 HT (15 pmol), there was a significant increase in TXB2, the stable metabolite of TXA2, indicating the vasoconstrictors had induced an increase in local prostanoid production. These findings indicate that the vasoconstrictor effects of ET-1, A II and 5 HT in the placental vascular bed are mediated, at least in part, by vasoconstrictor prostanoids including TXA2.

The role of endothelium in phenylephrine- and potassium-induced contractions of the rat aorta during pregnancy

The involvement of endothelium in the contractile responses of rat aortic rings to phenylephrine and potassium chloride in pregnancy was examined. Contractions in response to both agents were significantly greater in rings from non-pregnant rats than in rings from pregnant rats, and they were unaltered by treatment of the rings with indomethacin. De-endothelialization potentiated the contractions of rings from pregnant rats in response to phenylephrine, but had no significant effect on similar rings contracted with potassium chloride. Whereas de-endothelialization had no significant effect on the contractions to phenylephrine in rings from non-pregnant rats, it decreased those of rings from the same type of rats, contracted with potassium chloride. Pregnancy significantly inhibited contractions in response to calcium chloride of rings treated with phenylephrine or potassium chloride. The effect of endothelium removal on contractions to calcium chloride in rings from pregnant and non-pregnant rats treated with phenylephrine or potassium chloride was similar to that observed for phenylephrine-induced and potassium chloride-induced contractions, respectively. Contractions of intact aortic rings from pregnant and non-pregnant rats to phenylephrine in calcium-free medium were similar. Results of the study suggest that the effect of pregnancy on the contractions of the rat aorta in response to phenylephrine and potassium chloride is at least partly mediated by the endothelium and is independent of prostaglandin synthesis. The endothelial factor involved in this effect appears to modulate contractions by interfering with calcium influx through the receptor-operated calcium channels and the voltage-operated calcium channels.

Contractile effect of endothelin in human placental veins: role of endothelium prostaglandins and thromboxane

OBJECTIVE

The aim was to study the effects of endothelin-1 on human placental veins and the role of cyclooxygenase products as mediators of these effects.

STUDY DESIGN

Rings of placental veins with and without endothelium were suspended in organ chambers filled with physiologic salt solution. After a period of stabilization at optimal basal tension, isometric tensions in the control group were recorded at increasing concentrations of endothelin-1 (10(-10) to 10(-7) mol/L). Rings in the experimental groups were treated with either indomethacin (cyclooxygenase inhibitor, 10(-5) mol/L), dazoxiben (thromboxane synthetase inhibitor, 10(-4) mol/L), or SQ29548 (thromboxane receptor antagonist, 10(-6) mol/L) before addition of endothelin-1. To demonstrate the presence of functional thromboxane receptors in the rings, contractile responses to U-46619 (10(-9) to 10(-6) mol/L), a thromboxane A2 analog were measured. The effectiveness of SQ29548 blockade was tested in rings treated with SQ29548 (10(-6) mol/L) before addition of U-46619. The concentration-response curves of the treated and control groups were compared with the Student paired t test.

RESULTS

Endothelin-1 in doses of 10(-10) to 10(-7) mol/L caused concentration-dependent contraction of placental veins. Indomethacin significantly reduced the response of veins with endothelium to low endothelin-1 concentrations (10(-9.5) to 10(-9) mol/L), (p < 0.05). However, it had no effect at higher endothelin-1 concentrations or in vessels without endothelium. The presence of functional thromboxane A2 receptors was confirmed by the vasoconstrictor effect of U-46619 and its blockade by treatment with SQ29548. Neither SQ29548 nor the thromboxane A2 synthesis inhibitor dazoxiben significantly influenced the response to endothelin-1.

CONCLUSIONS

These results demonstrated that endothelin-1 is a potent vasoconstrictor in the human placental vein. Although functional thromboxane A2 receptors exist in this vessel, endothelin-1's action is independent of thromboxane A2. Prostaglandins may mediate part of the endothelin-1-induced placental vasoconstriction. However, endothelin-1 acts primarily by a direct effect on vascular smooth muscle cells.

The comparative effects of big endothelin-1, endothelin-1, and endothelin-3 in the human fetal-placental circulation

OBJECTIVE

This study compared the effects of big endothelin-1, endothelin-1, and endothelin-3 and whether endothelin-converting enzyme was present in the human fetal-placental circulation.

STUDY DESIGN

Single cotyledons of term placentas were dually perfused in vitro, and increases in fetal-placental perfusion pressure to bolus injections of big endothelin-1, endothelin-1, and endothelin-3 (8 x 10(-10) to 1 x 10(-7) mol/L) were recorded. Responses to big endothelin-1 (10(-7) mol/L) were measured in the same placenta before and after perfusion of the fetal-placental circulation with the neutral metalloprotease inhibitor phosphoramidon (10(-5) mol/L), which acts as an endothelin-converting enzyme inhibitor. All experiments were performed in at least five separate placentas.

RESULTS

Significant concentration-dependent increases in fetal-placental perfusion pressure were seen with endothelin-1 (p < 0.0005), endothelin-3 (p < 0.0256), and big endothelin-1 (p < 0.0034, analysis of variance). Big endothelin-1 always elicited transient vasodilatation before constriction. Phosphoramidon significantly inhibited the vasoconstrictor effect of big endothelin-1 (p < 0.039, paired t test).

CONCLUSION

The three endothelins tested are vasoconstrictors, and endothelin-converting enzyme is present in the fetal-placental circulation.

The role of prostaglandins in obstetrical disorders

All pregnancy-associated tissues are capable of producing prostaglandins including PGI2 and TXA2. In normal pregnancy there is a dominance of PGI2 over TXA2 which may contribute to the maternal circulatory adaptation to pregnancy. Furthermore, both fetoplacental PGI2 and TXA2 production are important regulators of the fetal blood supply. It has been clearly established that in pre-eclampsia PGI2 production decreases in the fetoplacental tissues and quite probably also in the maternal tissues. The effect of this change may be further exaggerated by the simultaneous stimulation in pre-eclampsia of TXA2 production. The reason for PGI2 deficiency is not known. Other vasoactive agents, such as endothelin, may act in concert with prostaglandins. Relative PGI2 deficiency is likely to exist also in IUGR and lupus anticoagulant syndrome of pregnancy. In the latter, lupus anticoagulant may directly inhibit the synthesis of PGI2. One study suggests PGI2 deficiency also in early pregnancies of women with a history of repeated abortions. Prostaglandin production increases during full-term labour, and similar but smaller changes also occur in preterm labour. A silent bacterial infection may trigger the onset of preterm labour through cytokine-stimulated increase of prostaglandin production. No data were found on prostaglandin production in post-term pregnancies. That oligo-polyhydramnios is possibly prostaglandin mediated is suggested by the control of polyhydramnios by indomethacin treatment. Smoking decreases the production of PGI2 and possibly increases that of TXA2, which may lead to decreased blood flow and IUGR. Which constituent of cigarette smoke exerts this effect is not known. Ethanol consumption causes aberrations in prostaglandin metabolism which cannot be directly connected with fetal alcohol effects.