17β-Estradiol Decreases Endothelin-1 Levels in the Coronary Circulation of Postmenopausal Women With Coronary Artery Disease
Background—Estrogen reverses acetylcholine-induced coronary vasoconstriction via the possible facilitation of endothelium-derived NO. Estrogen also affects endothelium-derived constrictor factors. We therefore investigated the effects of 17β-estradiol on coronary vasomotor responses to substance P (SP), and coronary sinus endothelin-1 and NO metabolite levels in postmenopausal women with coronary heart disease.
Methods and Results—We studied 20 women; 14 received estrogen (mean age 65±2 years) and 6 served as ethanol control subjects (age 63±3 years). Intracoronary infusions of papaverine (8 mg) and SP were administered before and 20 minutes after 50 pg/min 17β-estradiol or 0.2 μL/min control. Coronary blood flow was calculated from the diameter, as measured with quantitative coronary angiography, and flow velocity, as measured with intracoronary Doppler. Coronary sinus plasma endothelin-1 and nitrite/nitrate (NO2/NO3) were measured at baseline, at peak velocity response to each infusion, and every 5 minutes during the estradiol infusion. Endothelin-1 levels were decreased after 20 minutes of estradiol (1.12±0.18 versus 0.86±0.17 pmol/L baseline2 versus estradiol, P=0.05). Endothelin-1 levels to SP decreased after 17β-estradiol (1.29±0.18 versus 1.04±0.15 and 1.3±0.16 versus 0.99±0.17 pmol/L for before versus after estradiol, 10 and 25 pmol/min SP; both P<0.05). NO2/NO3 levels did not change. There was no change in vasomotor responses to estradiol alone or to papaverine or SP before versus after estradiol.
Conclusions—Short-term intracoronary 17β-estradiol administration decreases coronary endothelin-1 levels. There was no enhancement of vasomotor responses to SP after the administration of estrogen, suggesting that the effects of estrogen on coronary acetylcholine responses may be a specific and not a generalized effect on coronary vasoreactivity.
Estrogen reverses acetylcholine (ACh)-induced vasoconstriction of the coronary arteries in animals and humans via a possible facilitation of endothelium-dependent relaxation.1 2 3 4 This may be mediated, at least in part, by an enhancement of endothelium-derived NO production via stimulation of NO synthase (NOS).5 6 7 Recent evidence suggests that this may be a rapid nongenomic but estrogen receptor-α–dependent action of estrogen.8
Alternatively, estrogen may affect the production and/or release of endothelium-derived constrictor factors such as endothelin-1. Endothelin-1 is a potent vasoconstrictor that is released from endothelial cells. Plasma endothelin-1 levels increase after ACh infusion in pigs with coronary atheroma.9 Human subjects whose coronary arteries constrict in response to intracoronary ACh have higher baseline endothelin-1 levels than those whose coronaries dilate in response to Ach.10 Subjects whose arteries constrict further increase their coronary sinus endothelin-1 levels in response to ACh, whereas in subjects whose coronary arteries dilate, the levels are not significantly changed despite there being no differences in the extent of coronary atheroma between groups. Estrogen inhibits endothelin-1–induced constriction in rabbit coronary arteries in vitro11 and may regulate both endothelial NOS and prepro–endothelin-1 at the transcriptional levels in cultured porcine endothelial cells.12 Estrogen can inhibit endothelin-1 production via negative transcriptional regulation13 and by inhibiting gene expression and peptide secretion in bovine arterial endothelial cells.14 Physiological doses of 17β-estradiol inhibit the release of endothelin-1 from cultured human umbilical endothelial cells.15 In vivo, Sudhir et al16 demonstrated attenuation of endothelin-1–induced decreases in coronary vasoreactivity via the intracoronary administration of a physiological concentration of estradiol in anesthetized female pigs. Estrogen decreases plasma endothelin-1 levels in postmenopausal women,17 18 but the effects of estrogen on endothelin-1 production in the human female coronary circulation are unknown.
The aim of the present study was to assess whether acute intracoronary estrogen enhances the production of endothelium-derived NO, as indicated by an increase in plasma nitrite/nitrate (NO2/NO3) levels, or decreases plasma endothelin-1 levels in postmenopausal women with coronary artery disease. The demonstration of a possible interrelationship between NO and endothelin-1 may provide important information with regard to the effects of estrogen in the coronary circulation of postmenopausal women with coronary heart disease.
Postmenopausal female patients (not taking hormone replacement therapy) aged 35 to 70 years with angiographically proven coronary artery disease and stable angina pectoris were enrolled. All patients gave their written informed consent in accordance with the ethical requirements of the Royal Brompton Hospital Ethics Committee.
Cardiac medication and caffeine-containing beverages were withheld for at least 24 hours before cardiac catheterization. After diagnostic coronary angiography and full heparinization, a 0.014-inch Doppler wire (Cardiometrics Inc) was positioned in the proximal portion of an unobstructed coronary artery, and continuous traces of average peak blood flow velocity were recorded. A catheter was positioned into the coronary sinus via the femoral vein for blood sampling. Arterial pressure, heart rate, and ECG were displayed continuously.
An intracoronary bolus of the endothelium-independent vasodilator papaverine (8 mg) was administered, followed by two 3-minute intracoronary infusions of the endothelium-dependent vasodilator substance P (10 and 25 pmol/mL, respectively). A 20-minute intracoronary infusion of 17β-estradiol (50 pg/min; similar to that used previously3 ) or an equivalent dose of ethanol control (0.2 μL/min) was then administered, and the infusions of papaverine and substance P were repeated. Finally, an intracoronary bolus of isosorbide dinitrate (1000 μg) was infused. Coronary angiograms were performed at baseline, at peak velocity response to each vasoactive substance, and at 10 and 20 minutes during the estradiol infusion. There was a rest period of ≈1 minute between each infusion to allow all measured parameters to return to baseline.
Coronary sinus blood was sampled at baseline and then immediately before angiography after each infusion and every 5 minutes during the infusion of 17β-estradiol. The blood taken was used for measurement of plasma 17β-estradiol, NO2/NO3, and endothelin-1 levels.
Quantitative Coronary Angiography and Calculation of Flow
Coronary angiograms were acquired and analyzed with a real-time digital image acquisition and analysis system (Digitron III VACI; Siemens AG) as previously described.4 Measurements of diameter and velocity were made at baseline and at peak velocity change. Diameter was measured ≈4 mm distal to the Doppler wire tip (sample volume site) by an independent observer. Diameter changes throughout the entire artery were also measured, providing mean coronary diameter and responses at sites of defined focal narrowing. Coronary blood flow was calculated as previously described.19
Calculation of Coronary Flow Reserve and Coronary Resistance
Papaverine was infused to induce maximal hyperemic response. Coronary flow reserve (CFR) was calculated as the quotient of maximal coronary blood flow/baseline coronary blood flow. Coronary resistance was calculated as the quotient of mean arterial pressure (MAP [mm Hg]) and coronary blood flow (mL/min).
17β-Estradiol assays were performed with microparticle enzyme immunoassay (Abbott IMX system; Abbott Diagnostics Division).
Endothelin-1 measurement was performed according to a modified radioimmunoassay technique as previously described.20 Briefly, 200 μL of unextracted plasma samples was assayed in duplicate. Hormone-free plasma (200 μL), prepared through the prior charcoal adsorption of peptides, was added to all zero and standard tubes. Trasylol (Bayer) was added to the assay buffer. Antibody to endothelin-1 was obtained from immunoresponsive rabbits20 and could distinguish changes of 0.1 fmol/assay tube (0.1 pmol/L plasma) from zero concentrations with 95% CI. Plasma concentrations of immunoreactive endothelin measured showed a good correlation with those obtained according to the previously reported method with Sep-Pak extraction.20 Cross-reaction with big endothelin, endothelin-2, and endothelin-3 was 0.1%, 60%, and 70%, respectively. Intra-assay and interassay coefficients of variation were 12% (n=9) and 19% (n=7), respectively.
NO2/NO3 measurement was performed at an independent laboratory (OXONONUS) with a previously described technique.21
Each variable was analyzed with a mixed-model ANOVA with the patient as a random factor. Group (patient or control), time, and the group×time interaction were included as factors. Each intervention (infusion) was compared with the most recent baseline measurement, and papaverine and substance P responses were compared before versus after estradiol administration. We tested the assumptions of normality of residuals by both a normal plot and the Shapiro-Francia W′ test and of equality of variances in the 2 groups and in the time points by Bartlett’s test. Where these assumptions were invalid, the data were log transformed and presented as geometric mean (95% CI). All other data are presented as mean±SEM. P<0.05 was considered statistically significant.
Fourteen postmenopausal women were enrolled (mean age 65±2 years; Table 1⇓). Nine patients had 1 significantly diseased coronary artery (stenosis >70%) and 5 patients had 2-vessel disease. We studied the left anterior descending coronary artery in 7 patients, the left circumflex artery in 2 patients, and the right coronary artery in 5 patients.
Six control subjects were enrolled (mean age 63±3 years; Table 1⇑). Four control subjects had 1-vessel disease and 2 had 2-vessel disease. The study vessel was the left circumflex artery in 2 control subjects and the right coronary artery in 4. There were no differences between patients and control subjects with respect to age, baseline plasma estradiol, or factors that might affect endothelial function, such as lipid profile, blood pressure, and coronary atherosclerosis (Table 1⇑).
Coronary Sinus Plasma 17β-Estradiol Concentrations
Coronary sinus 17β-estradiol levels were measured in 6 patients and 3 control subjects after the estradiol/control infusion. There was a significant increase after 20 minutes in the study group, equivalent to premenopausal levels,22 but not in the control subjects (127±19 versus 1754±451 pmol/L, P<0.05, and 104±19 versus 70±13 pmol/L; baseline 2 versus 20 minutes for study and control group, respectively). There is a small but nonsignificant difference between these estradiol concentrations and those presented in Table 1⇑. The latter are for all patients for samples taken from the femoral artery sheath.
Coronary Sinus Endothelin-1 Concentrations
Coronary sinus endothelin-1 levels were significantly decreased by 23% after 20 minutes of estradiol compared with baseline 2 (1.12±0.18 versus 0.86±0.17 pmol/L, baseline 2 versus 20 minutes estradiol; P=0.05; Figure 1⇓). Endothelin-1 levels after substance P infusions were not changed compared with baseline 1 (1.27±0.14 pmol/L) but were significantly decreased, by 19% and 24% respectively, after the administration of estradiol (1.29±0.18 versus 1.04±0.15 and 1.3±0.16 versus 0.99±0.17 pmol/L; before versus after estradiol and substance P at 10 and 25 pmol/mL; both P<0.05; Figure 2⇓). There was no effect of ethanol control on any of these parameters (Figure 2⇓).
Coronary Sinus NO Concentrations
Baseline levels of NO2, NO3, and NO2+NO3 were 0.77±0.06, 45.5±6.4, and 46.3±6.4 μmol/L, respectively, in the study group. There was a borderline increase in coronary sinus NO2 levels in response to 10 pmol/min substance P (0.88± 0.08 μmol/L; P=0.068) compared with baseline; however, there was no change in NO metabolite levels after any other intervention. Baseline levels of NO2, NO3, and NO2+NO3 in control subjects were 0.74±0.09, 38.7±2.2, and 39.4± 2.2 μmol/L, respectively. There was no significant change in these levels after any intervention.
Coronary Artery Responses to 17β-Estradiol
Estradiol/ethanol had no significant effect on coronary artery diameter, blood flow velocity, or volume or on coronary resistance compared with baseline 2 in the study (Table 2⇓) and control (Table 3⇓) groups.
Coronary Artery Responses to Substance P Before and After 17β-Estradiol and Ethanol Control
Substance P (10 and 25 pmol/min, respectively) increased coronary diameter by 8% (P=0.08) and 12% (P=0.003) in study patients and by 17% (P=0.006) and 17% (P=0.015) in control subjects; increased blood flow velocity by 24% and 40% in patients (both P<0.001) and by 55% and 75% (both P<0.001) in control subjects; and increased volume blood flow by 40% and 72% in patients (both P<0.001) and by 124% and 120% (both P<0.001) in control subjects compared with baseline 1 (Tables 2⇑ and 3⇑, study patients and control subjects, respectively). However, there was no change in coronary diameter or blood flow responses to substance P after estradiol or ethanol. Flow velocity response to substance P (10 pmol/min) was significantly decreased by 17% after estradiol (P<0.01; Table 2⇑). Coronary resistance decreased by 44% after 25 pmol/min substance P compared with baseline 1 in study patients (P=0.007); however, there was no change in this response after estradiol (Table 2⇑). In control subjects, coronary resistance decreased by 63% in response to 10 and 25 pmol/min substance P compared with baseline 1 (both P<0.001; Table 3⇑); however, this response did not change after ethanol.
Coronary Artery Responses to Isosorbide Dinitrate
Compared with baseline 3, isosorbide significantly increased coronary artery diameter by 4% (P=0.032) and 8% (P=0.019), velocity by 143% (P<0.001) and 132% (P<0.001), and blood flow by 128% (P<0.001) and 214% (P<0.001) and decreased coronary resistance by 54% (P=0.002) and 68% (P<0.001) in the study group and control subjects, respectively (Tables 2⇑ and 3⇑).
Mean Coronary Artery Diameter Changes
There was no change in mean coronary diameter after any intervention in the study group. In 5 areas of focal narrowing (mean severity 33%), the diameter response to substance P before and after estradiol was unchanged (2.15±0.14 and 2.08±0.12 mm versus 2.27±0.06 and 2.08±0.08 mm; substance P 10 and 25 pmol/min, before versus after estradiol, respectively), as was the response after 20 minutes of estradiol alone (2.03±0.11 versus 2±0.08 mm, baseline 2 versus estradiol).
Coronary Artery Responses to Papaverine and CFR
Papaverine significantly increased coronary artery diameter by 12% (P=0.002) and 68% (P<0.001), velocity by 110% (P<0.001) and 215% (P<0.001), and blood flow by 149% (P<0.001) and 280% (P<0.001) and decreased coronary resistance by 56% (P<0.001) and 75% (P<0.001) for study and control groups, respectively, compared with baseline 1 (Tables 2⇑ and 3⇑). However, these responses were not different before compared with after estradiol or ethanol. Mean CFR was 2.6±0.2 (range 1.1 to 3.7) and 3.8±0.2 (range 3 to 5.2) for study and control patients, respectively.
MAP and heart rate did not change significantly throughout the procedure in either the study patients (114±4 mm Hg and 71±4 bpm, respectively) or control subjects (89±7 mm Hg and 71±4 bpm, respectively).
We demonstrated an estradiol-induced decrease in coronary sinus endothelin-1 levels after a 20-minute infusion of 17β-estradiol, the first report of such an effect in atherosclerotic human coronary arteries. Previous studies have described estrogen-induced decreases in plasma endothelin-1 levels in cultured endothelial cells15 and an effect of sex (and possibly ovarian hormones) on the expression of prepro–endothelin-1 and endothelial NOS.12 Acute intracoronary exposure of pig coronary arteries to estradiol attenuated endothelin-1–induced epicardial vasoconstriction and decreases in coronary blood flow.16 In postmenopausal women, chronic estrogen replacement decreased plasma endothelin-1 levels in the systemic circulation, associated with an increased NO–to–endothelin-1 ratio.23 Estrogen-induced decreases in plasma endothelin-1 levels may possibly relate to the anti-ischemic and antiatherogenic effects of estrogen.24 25 26 It is also possible that estradiol may affect the degradation and clearance of endothelin-1.27
Interestingly, but not surprisingly, there was no change in coronary vasomotor or flow responses associated with the decrease in endothelin-1 levels to estradiol alone, similar to previous studies.3 4 It is possible that the decrease in endothelin-1 level was not great enough to elicit an additional vasomotor response. In a study of coronary endothelial dysfunction and early atherosclerosis in humans, patients who constricted in response to intracoronary ACh increased their endothelin-1 levels in response to ACh by ≈150% (P<0.01), whereas in subjects whose coronary arteries dilated, the levels were not significantly changed from baseline.10 In the present study, we demonstrated a decrease in endothelin-1 level of 23% (P=0.05). However, endothelin-1 may be released abluminally as well as into the lumen of the vessel; therefore, plasma levels may be only an indicator of endothelin release.
Although substance P did not change endothelin-1 levels before exposure to estrogen, there was a significant decrease in endothelin-1 levels in response to substance P after estradiol. This may be due to a carryover effect of estradiol on endothelin-1 levels; however, the lack of a significant effect on the endothelin response to papaverine, before versus after estradiol, would argue against this. Substance P is known to elicit coronary dilatation in vivo, at least in part via NO,28 29 but its effects on endothelin are not well documented. In mouse 3T3 cells, substance P acts as an endothelin-1 antagonist, competitively inhibiting receptor binding and blocking increases in cytosolic calcium.30 In rat cardiac membranes in vitro, specific binding of endothelin-1 can be inhibited by substance P.31 Alternatively, substance P–stimulated release of NO from the coronary endothelium, which may be enhanced locally at the coronary endothelium by exposure to estradiol, may inhibit the production or release of endothelin-1. In human internal mammary artery in vitro, endothelin-induced constriction can be completely inhibited by stimulation of endothelium-derived relaxing factor with ACh and by exposure to exogenous NO.32 Others have shown that the activation of endothelin-3–selective receptors leads to the release of endothelium-derived NO, which in turn reduces the release of endothelin-1.33
The present results differ from those of previous studies in that no change was seen in endothelium-dependent vasomotor responses to substance P after the administration of estradiol. Substance P is a known endothelium-dependent vasodilator and was used in the present study to ensure that any effect, and any change of this effect by estrogen, was truly endothelium dependent. ACh, which was used in previous intracoronary studies of estrogen,2 3 4 has actions both via the endothelium (causing vasodilatation) and via vascular smooth muscle (causing constriction). ACh-induced vasoconstriction was attenuated or reversed after estrogen, which could be explained by an enhancement of NOS but also by an attenuation of ACh-induced stimulation of endothelin-1. Interestingly, a beneficial effect of estrogen on ACh-induced coronary reactivity has been demonstrated only when ACh caused vasoconstrictor responses.2 Substance P may maximally stimulate NOS even in the presence of atheroma, and a concomitant decrease in endothelin-1 may not be detectable as a further increase in diameter. Indeed, coronary diameter after substance P was similar to that achieved after papaverine and isosorbide dinitrate (Table 2⇑). The blood flow responses were smaller with substance P than with papaverine and isosorbide, however, confirming that substance P has more of an epicardial endothelium-dependent effect than that in the microvascular vessels.29
Failure to show an enhancement of substance P–induced dilatation by estrogen emphasizes subtle differences of the actions of estrogen dependent on the agent used to test endothelial function. Beneficial effects of estrogen may be more pronounced in relative states of vasoconstriction, which occur in the presence of coronary heart disease and circulating vasoconstrictor substances such as epinephrine, norepinephrine, and angiotensin II. In this situation, a reduction in plasma endothelin-1 concentrations may result in reduction in coronary artery vasoconstriction, which is associated with cardiac events.34
NO metabolite levels induced by estradiol did not change, nor did levels in response to papaverine or substance P after the estradiol infusion. This is in contrast to a recent study in postmenopausal women with, or with risk factors for, coronary artery disease, which demonstrated inhibition of estradiol-induced potentiation of coronary blood flow and decreases in coronary resistance responses to ACh after intracoronary infusion of NG-monomethyl-l-arginine (a competitive inhibitor of NO synthesis).7 Biological activity of NO at the endothelium–vascular smooth muscle interface may not be detectable in the plasma due to a dilutional effect and may explain the null finding in the present study. Interestingly, all of the claims for an effect by estrogen on stimulated endothelium-dependent responses in the coronary circulation are based on a reversal of ACh-induced vasoconstriction in atherosclerotic coronary arteries.2 3 4 7 It remains to be proved whether the results with ACh can be extrapolated to a general improvement in coronary arterial endothelial function via NO; the results of the present study with substance P suggest that it cannot.
In a randomized trial of estrogen and progestins for secondary prevention of coronary heart disease (Heart and Estrogen/Progestin Replacement Study [HERS]), a null effect was demonstrated on myocardial infarction and coronary heart disease death in postmenopausal women with coronary heart disease.35 Our study reinforces the fact that there may be subtle differences between different estrogens with regard to cardiovascular actions and that differences may exist when estrogen is combined with a progestin. These points will require further studies.
Our data show that short-term 17β-estradiol administration into the coronary circulation decreases coronary blood endothelin-1 levels. The results differ from previous studies in that no change in endothelium-dependent coronary dilatation or blood flow was seen in response to substance P after exposure to estrogen, suggesting that in previous studies a specific effect of estrogen on ACh responses was seen and not a generalized effect on coronary vasoreactivity and blood flow.
This work was supported in part by Novo Nordisk Pharmaceuticals Ltd. Dr Webb was supported by the British Heart Foundation. We would like to thank the patients who participated in this study, the staff in the daycase unit and the cardiac catheterization laboratory of the Royal Brompton Hospital, and Amy Hider for statistical assistance.
Guest Editor for this article was Suzanne Oparil, MD, University of Alabama, Birmingham.
- Received February 14, 2000.
- Revision received April 20, 2000.
- Accepted May 8, 2000.
- Copyright © 2000 by American Heart Association
Reis SE, Gloth ST, Blumenthal RS, et al. Ethinyl estradiol acutely attenuates abnormal coronary vasomotor responses to acetylcholine in postmenopausal women. Circulation. 1994;89:52–60.
Gilligan DM, Quyyumi AA, Cannon RO III. Effects of physiological levels of estrogen on coronary vasomotor function in postmenopausal women. Circulation. 1994;89:2545–2551.
Collins P, Rosano GMC, Sarrel PM, et al. Estradiol-17β attenuates acetylcholine-induced coronary arterial constriction in women but not men with coronary heart disease. Circulation. 1995;92:24–30.
Weiner CP, Lizasoain I, Baylis SA, et al. Induction of calcium-dependent nitric oxide synthases by sex hormones. Proc Natl Acad Sci U S A. 1994;91:5212–5216.
Guetta V, Quyyumi AA, Prasad A, et al. The role of nitric oxide in coronary vascular effects of estrogen in postmenopausal women. Circulation. 1997;96:2795–2801.
Lerman A, Webster MWI, Chesebro JH, et al. Circulating and tissue endothelin immunoreactivity in hypercholesterolemic pigs. Circulation. 1993;88:2923–2928.
Lerman A, Holmes DRJ, Bell MR, et al. Endothelin in coronary endothelial dysfunction and early atherosclerosis in humans. Circulation. 1995;92:2426–2431.
Jiang C, Sarrel PM, Poole-Wilson PA, et al. Acute effect of 17β-estradiol on rabbit coronary artery contractile responses to endothelin-1. Am J Physiol. 1992;263:H271–H275.
Wang X, Barber DA, Lewis DA, et al. Gender and transcriptional regulation of NO synthase and ET-1 in porcine aortic endothelial cells. Am J Physiol. 1997;273:H1962–H1967.
Morey AK, Razandi M, Pedram A, et al. Oestrogen and progesterone inhibit the stimulated production of endothelin-1. Biochem J. 1998;330:1097–1105.
Wingrove CS, Stevenson JC. 17 Beta-oestradiol inhibits stimulated endothelin release in human vascular endothelial cells. Eur J Endocrinol. 1997;137:205–208.
Sudhir K, Ko E, Zellner C, et al. Physiological concentrations of estradiol attenuate endothelin 1–induced coronary vasoconstriction in vivo. Circulation. 1997;96:3626–3632.
Webb CM, McNeill JG, Hayward CS, et al. Effects of testosterone on coronary vasomotor regulation in men with coronary artery disease. Circulation. 1999;100:1690–1696.
Chamberlain G, Broughton-Pipkin F, eds. Clinical Physiology in Obstetrics. London, UK: Blackwell Scientific Publications; 1991.
Sullivan JM, Vander Zwaag R, Lemp GF, et al. Postmenopausal estrogen use and coronary atherosclerosis. Ann Intern Med. 1988;108:358–363.
Quyyumi AA, Mulcahy D, Andrews NP, et al. Coronary vascular nitric oxide activity in hypertension and hypercholesterolemia: comparison of acetylcholine and substance P. Circulation. 1997;95:104–110.
Luscher TF, Yang Z, Tschudi M, et al. Interaction between endothelin-1 and endothelium-derived relaxing factor in human arteries and veins. Circ Res. 1990;66:1088–1094.
Warner TD, Schmidt HHHW, Murad F. Interactions of endothelins and EDRF in bovine native endothelial cells: selective effect of endothelin-3. Am J Physiol. 1999;31:H1600–H1605.