| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Baker Medical Research Institute, St. Kilda Central, Melbourne 8008, Australia
Address all correspondence and requests for reprints to: Krishnankutty Sudhir, M.D., Ph.D., Associate Professor of Medicine, Stanford University, 995 East Arques Avenue, Sunnyvale, California 94085-4521. E-mail: ksudhir{at}pcyc.com.
Abstract
Estrogens influence cardiovascular function through direct and indirect effects and via genomic and nongenomic mechanisms. The pathways underlying the nongenomic mechanisms are not completely understood. Estrogen-induced responses in vascular cells have been shown to influence prostaglandins and cyclooxygenase (COX), a key enzyme in the production of prostaglandins, with two isoforms, COX-1 and COX-2. We investigated the effects of prostaglandins on the acute potentiation by 17ß-estradiol (E) of acetylcholine (ACh)-mediated vasodilation in the cutaneous vasculature. Using a double-blind placebo-controlled design, we assessed skin blood flow in 32 healthy, postmenopausal women by laser Doppler velocimetry with direct current iontophoresis of ACh and sodium nitroprusside before and after 6-wk treatment periods with aspirin (a nonspecific COX-1 and COX-2 inhibitor), diclofenac (predominantly a COX-2 inhibitor, which also inhibits COX-1), celecoxib (a specific COX-2 inhibitor), given at anti-inflammatory doses, or placebo. Blood flux values before iontophoresis of ACh did not differ between the treatment groups or after E administration, excluding a direct cutaneous vasodilator effect of the treatments or of E. Acute E administration enhanced the response to ACh after aspirin, diclofenac, and placebo; however, this effect was completely abolished with celecoxib treatment (P < 0.05). E had no effect on sodium nitroprusside-mediated vasodilation after any of the treatments. We conclude that the COX-2 pathway plays a specific role in the rapid E-induced potentiation of cholinergic vasodilation in postmenopausal women.
THE MANY ACTIONS of estrogens on the cardiovascular system include improvements in lipid profiles, effects on endothelial and vascular smooth muscle function, direct inotropic actions on the heart, effects on coagulation and the fibrinolytic system (1, 2), and modulation of the stress response (3). The rapid time course of some of these effects suggests that in some cases nongenomic mechanisms are involved. For example, within a time frame of 30 min or less, estrogen administration enhances endothelium-dependent vasodilation in forearm (4, 5) and coronary vessels of postmenopausal women (6) and in cutaneous forearm vessels of healthy young men (7), and it reverses endothelial dysfunction in atherosclerosis (4, 8, 9); in animal models and in vitro, estrogens have also been shown to act directly on smooth muscle, indicating endothelium-independent actions (10, 11, 12).
The mechanisms underlying these nongenomic actions remain uncertain. It has been shown that prostaglandins, in conjunction with nitric oxide (NO), are key mediators of endothelium-dependent vasodilation and that estrogens stimulate the production of the vasodilator prostaglandin prostacyclin (13, 14) and influence expression of cyclo-oxygenase (COX), a key enzyme in the synthesis of prostaglandins (15, 16). We therefore hypothesized that COX may have a role in mediating the rapid effects of estrogens on the vasculature and sought to test this by conducting a study of the effects of inhibitors of this enzyme on estrogen-induced modulation of vascular reactivity.
Materials and Methods
Thirty-two healthy, normotensive, postmenopausal women were recruited from the general community to participate in a double-blind, randomized, placebo-controlled, cross-over study of the effects of 6-wk treatment with the specific COX-2 inhibitor celecoxib (200 mg twice daily); the predominant COX-2 inhibitor diclofenac (50 mg twice daily), which also inhibits COX-1; the nonspecific COX-1 and COX-2 inhibitor aspirin (300 mg twice daily); and placebo. Each subject received 3 of the 4 treatments in random order, giving 24 subjects per group. Each subject underwent testing on four occasions, at baseline and after each treatment period. Women who were using hormonal therapy, aspirin, nonsteroidal anti-inflammatory agents, and COX-2 inhibitors during the 3 months before entry into the study were excluded. Ethics approval was granted by the Alfred Hospital Human Research Ethics Committee, and all subjects gave full written informed consent.
Women had to be free of periods for 12 months before the study and to have an estradiol level of less than 120 pmol/liter, or FSH greater than 30 IU/liter, or both. Blood samples were taken at baseline to confirm menopausal status and to assess lipid levels. At each time point, cutaneous vascular reactivity to the vasodilator substances acetylcholine (ACh) and sodium nitroprusside (SNP) was assessed using laser Doppler velocimetry with direct current iontophoresis, as described elsewhere (7). Briefly, blood flow was measured with a Moor laser Doppler imager (Moor Instruments, Devon, UK) via a 633-nm (helium-neon) infrared light that detects blood flow via a frequency shift produced by scatter of photons from moving erythrocytes 1–2 mm below the surface of the skin. Polyvinylchloride chambers containing platinum electrodes were fixed to the forearm, and solutions of either ACh (BDH Chemicals, Dorset, UK) or SNP (David Bull Laboratories, Inc., Mulgrave, Victoria, Australia) mixed in methylcellulose gel (10% wt/vol) at a concentration of 10 mg/ml were placed in the chambers. A current of 0.1 mAmp was then administered for 30 sec, with SNP administered via a cathodal charge and ACh via an anodal charge, and perfusion was recorded at baseline and for 6 min after the start of infusion using Moor Instruments laser Doppler perfusion measurement package V3.01. The coefficient of variation for blood flux assessed using this method is 0.15 ± 0.05 (17). After measurement of baseline responses to ACh and SNP in duplicate, 2 mg of 17ß-estradiol (E; Estrace, Mead-Johnson Co., a division of Bristol-Myers Squibb Co., Deeside, UK) was administered sublingually; the response to ACh was measured after a further 6, 12, 18, and 34 min, and the response to SNP was measured at 26 min after administration, as previously described (7).
Blood flux responses were analyzed using Moor Instruments laser Doppler imager analysis package V3.01 and subsequent estimation of area under the response curves. Baseline data are presented in arbitrary perfusion units as treatment mean ± SEM, and the effects of E are presented as the ratio of response to baseline. Results were analyzed using within-subject ANOVA, with the SYSTAT v.9.0 (SPSS, Inc., Chicago, IL) computer program; P value no greater than 0.05 was regarded as statistically significant.
Results
The average age of subjects was 62 ± 1.5 yr, and body mass index was 25.8 ± 1.9 kg/m2. The mean total cholesterol level was 5.7 ± 0.2 mmol/liter, high-density lipoprotein cholesterol was 1.9 ± 0.1 mmol/liter, low-density lipoprotein was 3.5 ± 0.2 mmol/liter, triglycerides were 1.0 ± 0.1 mmol/liter, and estradiol was 86 ± 6 mmol/liter. Average blood pressure was 119/67 ± 3/1 mm Hg. Full blood count and liver function tests were in the normal range at all time points in all cases. None of these variables changed significantly during the course of the study.
The vasodilator response to ACh, before E administration, was significantly greater after celecoxib than after placebo (2679 ± 273 to 1373 ± 212 arbitrary perfusion units; P = 0.0004) but was unaffected by aspirin and diclofenac. The blood flow response to SNP was not altered by placebo or any of the treatments (P = 0.261)
After placebo, aspirin, and diclofenac, acute E administration significantly enhanced the vasodilator response to ACh over the 34 min time course. However, after celecoxib this potentiation of ACh-mediated vasodilation by E was completely abolished (P < 0.05; Fig. 1
). Acute E administration had no significant effect on SNP-induced vasodilation after any of the treatments (P > 0.05; Fig. 2).
|
|
).
|
This study has shown that acute administration of sublingual E in healthy menopausal women potentiates endothelium-dependent vasodilation induced by ACh after placebo or after treatment with the nonselective COX-1 and COX-2 inhibitor aspirin and the predominant COX-2 inhibitor diclofenac, which also inhibits COX-1, but not after treatment with the selective COX-2 inhibitor celecoxib. E has no effect on endothelium-independent vasodilation induced by SNP after placebo or any of the other treatments. These results suggest that the nongenomic action of E on endothelial cells shares a common pathway with the COX-2 and prostaglandin systems. Baseline blood flux values before commencement of iontophoresis of ACh did not differ between the treatment groups before administration of E or at the various time points after E administration; this excludes a direct, cutaneous vasodilator effect of either the treatments or E.
In a preliminary report, we have previously shown that COX-2 inhibition significantly enhances the vasodilator response to ACh in the cutaneous microvasculature (18), an effect most likely to be due to an alteration in the balance between prostanoid vasoconstrictors and vasodilators, favoring dilators. It is unclear why diclofenac did not attenuate the response to ACh in the present study, but in that previous report (18), the value for diclofenac, a partially selective COX-1/COX-2 inhibitor, was midway between aspirin and celecoxib, suggesting a possible dose response in relation to COX activity. We have also shown that in healthy young men, acute administration of E increases the response to ACh in a rapid time frame that suggests a nongenomic mechanism of action (7), an effect also observed in many other settings (4, 5, 6, 8, 9). The mechanisms underlying these nongenomic effects of E on the vasculature are presently uncertain, although there is evidence that receptors located in the endothelial cell plasma membrane, which may be related to the classical estrogen receptors, are involved (19, 20), or that E acts either on the vascular endothelium (4, 5, 9) or directly on vascular smooth muscle cells (10, 11, 21, 22).
The present study shows that the increase in endothelium-dependent vasodilation that is seen after COX-2 inhibition is not subject to further enhancement with acute E treatment. It therefore appears likely that E produces its acute effects, at least in part, by suppressing vasoconstrictor or enhancing vasodilator production mediated by COX-2. This is consistent with findings in animal studies in which COX-dependent pathways have been shown to modify endothelium-dependent relaxation in isolated porcine coronary arteries (23), and chronic E replacement therapy in ovariectomized rats has been shown to enhance ACh-mediated dilation by suppression of COX-dependent vasoconstrictor production (24). There is substantial additional evidence for an overlap between the actions of estrogens and those of prostaglandins. For example, E alters prostacyclin production in endothelial (13) and smooth muscle (14) cells; urinary excretion of prostacyclin increases in postmenopausal women taking hormonal therapy (25, 26); and E up-regulates COX-2 expression in some tissues (15, 27, 28, 29), down-regulates it in others (16, 30), and has varying effects on COX-1 (31, 32, 33).
The relationship between E and prostaglandins on Ach-mediated vasodilation may be mediated by other factors, such as NO. In fact, E potentiation of ACh-mediated dilation is abolished with N-monomethyl-L-arginine (5), and NO has been shown to activate COX both in vitro (34) and in vivo (35), and in cultured endothelial (36) and smooth muscle (37) cells. Furthermore, aspirin has been shown to enhance NO production by neutrophils (38) and in smooth muscle cells (39, 40) and to inhibit NO release from vascular smooth muscle cells (41, 42).
It is possible that the effects observed may vary between populations. Although COX-1 is widely expressed in vascular smooth muscle and endothelial cells, COX-2 is generally expressed at low levels only. However, COX-2 is up-regulated in atherosclerosis (43, 44) and aging (45, 46, 47, 48). It therefore cannot be assumed that actions demonstrated in postmenopausal women with mild elevations of cholesterol can be generalized to other cohorts.
In conclusion, we have shown that the rapid, nongenomic action of E on the cutaneous vascular endothelium shares a common pathway with prostaglandins, probably by reduction in COX-mediated vasoconstriction or enhancement of vasodilation, via inhibition of COX-2. The clinical implications of this finding remain to be elucidated, but the present study offers a possible mechanism by which E exerts its rapid vascular effects.
Acknowledgments
We acknowledge Pharmacia for supporting this study and for the supply of the celecoxib, and Roche Pharmaceuticals for the donation of aspirin. We also acknowledge Prof. John Ludbook for his assistance with statistical analysis.
Footnotes
K.S. and P.A.K. made equal contributions to this paper.
Abbreviations: ACh, Acetylcholine; COX, cyclooxygenase; E, 17ß-estradiol; NO, nitric oxide; SNP, sodium nitroprusside.
Received January 17, 2002.
Accepted August 7, 2002.
References
mediates the nongenomic activation of endothelial nitric oxide synthase by estrogen. J Clin Invest 103:401–406[Medline]
, tumor necrosis factor-, glucocorticoids, and 17ß-estradiol. J Rheumatol 25:1146–1153[Medline]
secretion in cultured endometrial cells. Endocrinology 138:237–247
in bovine endometrial cells. Endocrinology 139:2293–2299
release by cultured smooth muscle cells. Eur J Clin Invest 29:93–99
B activation and IB degradation. Free Radic Biol Med 28:683–692
This article has been cited by other articles:
 |
|
 |
|
  D. Z.I. Cherney, J. A. Miller, J. W. Scholey, T. J. Bradley, C. Slorach, J. R. Curtis, M. G. Dekker, R. Nasrallah, R. L. Hebert, and E. B. Sochett The Effect of Cyclooxygenase-2 Inhibition on Renal Hemodynamic Function in Humans With Type 1 Diabetes Diabetes, March 1, 2008; 57(3): 688 - 695. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. de Pascual-Teresa, K. L. Johnston, M. S. DuPont, K. A. O'Leary, P. W. Needs, L. M. Morgan, M. N. Clifford, Y. Bao, and G. Williamson Quercetin Metabolites Downregulate Cyclooxygenase-2 Transcription in Human Lymphocytes Ex Vivo but Not In Vivo J. Nutr., March 1, 2004; 134(3): 552 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Orshal and R. A. Khalil Gender, sex hormones, and vascular tone Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2004; 286(2): R233 - R249. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
