INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HUMAN STUDIES INDICATE GENDER differences in the regulation of vascular reactivity that correlate with gender differences in incidence of cardiovascular disease (12, 16). It has been reported that -adrenoreceptor-mediated vasoconstriction is decreased in Caucasian females compared with age-matched males (6, 10). This may be due to enhanced endothelial production of vasodilators in females compared with males, because premenopausal women reportedly have greater endothelium-dependent vasodilation compared with age-matched men (3, 10, 14). Conversely, a decline in endothelial function in females correlates with the occurrence of menopause and increased incidence of cardiovascular disease in postmenopausal women (3, 10, 14, 16).

Estrogen has been reported to upregulate the capacity for nitric oxide (NO) biosynthesis, a major endothelium-derived vasodilator and regulator of vascular tone (11, 15), and a number of studies using experimental animals have also shown that females have enhanced endothelium-dependent vasodilation compared with males (8, 21, 22). For example, resistance in rat microvessels is lower in females compared with males, an effect attributed to a gender difference in the production of endothelium-derived dilator factors, such as NO (8). Acetylcholine-induced relaxation of isolated rat mesenteric arteries from females was found to be greater than responses in arteries from males, an effect dependent on both NO-dependent and -independent pathways (22). In addition to differences in endothelial function and NO production, there is also evidence that vasoactive cyclooxygenase metabolites of arachidonic acid may be differentially produced as a function of gender. A relatively consistent finding is enhanced cyclooxygenase-derived vasoconstrictors produced by endothelium in arteries from males compared with females (1, 9, 19, 23).

Most of the studies leading to the conclusion that the capacity for endothelium-dependent production of NO is greater in females have looked directly at endothelium-dependent vasodilators in preconstricted blood vessels. However, when considering responsiveness to a vasoconstrictor, the relative importance of this modulatory pathway can be expected to vary with respect to both stimulus and blood vessel type and is dependent on the relative complement of agonist receptors on endothelium versus smooth muscle (20, 24). Similarly, the production of vasoactive arachidonic acid metabolites by endothelium or vascular smooth muscle can be expected to vary depending on stimulus, vascular bed, and possibly gender. Because androgens may have independent effects on vascular reactivity (7), it is also important to consider male gonadal steroids as a possible source of observed gender differences in vascular regulation.

The present study was designed to test the hypothesis that net contractile responsiveness in the rat mesenteric artery may be modulated in an agonist-specific and gender-dependent manner. Gender-dependent differences in the modulation of ₁-adrenergic contractile responses were observed. Results suggest a model in which the net vasomotor response to a contractile stimulus results from a direct smooth muscle contractile effect modulated by opposing endothelial vasodilator factors and smooth muscle vasoconstrictors, which are cyclooxygenase-1 metabolites of arachidonic acid. In males, production of modulatory endothelial vasodilators is balanced by cyclooxygenase-derived vasoconstrictors from smooth muscle, the latter pathway upregulated by male gonadal steroids. Females lack production of the cyclooxygenase-derived vasoconstrictors and production of the otherwise compensatory endothelium-derived vasodilators, specifically in response to ₁-adrenergic stimuli, is suppressed by female gonadal steroids.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Tissue preparation. Sprague-Dawley rats were purchased from Taconic Farms (Germantown, NY). In the case of gonadectomized animals, surgery was performed at 5 wk of age by the vendor. Efficacy of ovariectomy was confirmed by measuring the uterus and body weight at the time of the experiment. At 13 wk of age, rats were euthanized with pentobarbital sodium (120 mg/kg ip), followed by a thoracotomy. A portion of the small intestine was removed and placed in cold physiological salt solution (PSS) of the following composition (in mmol/l): 130 NaCl, 4.7 KCl, 1.17 MgSO₄ · 7H₂O, 1.18 KH₂PO₄, 14.9 NaHCO₃, 5.5 dextrose, 0.03 NaCa₂ · EDTA, and 1.6 CaCl₂ · 2H₂O. With the use of a dissecting microscope, third- or fourth-order mesenteric resistance arteries (250- to 400-µm diameter) were isolated, cleared of fat and connective tissue, and placed in a dual-chambered arteriograph without flow at a constant intraluminal pressure of 60 mmHg (Living Systems Instrumentation; Burlington, VT) (19, 21, 22). We observed no trends in agonist sensitivity, response to endothelium disruption, or responses to cyclooxygenase inhibition as a function of blood vessel diameter in the present study. Arteries were superfused with warmed (37°C) oxygenated (95% O₂-5% CO₂) PSS, pH 7.35, at a flow rate of 20 ml/min. The outer diameter of the vessel was calculated continuously by using an image analysis system previously described (2, 19, 22). All procedures involving the use of animals were approved by the Institutional Animal Care and Use Committee and conformed to federal, state, and institutional guidelines.

9

5

9

5

Statistical analysis. Contractile responses are expressed as a percentage of the maximal constriction [(baseline  experimental diameter)/(baseline  maximally constricted diameter) × 100]. Individual concentration-response relationships were fit by nonlinear regression to a sigmoidal curve to calculate EC₅₀ values and maximum responses. Concentration-response curves were analyzed by a three-factor ANOVA with repeated-measures design. Significant differences among groups were determined by using Newman-Keuls post hoc comparisons with a P value of 0.05 considered statistically significant (STATISTICA for Windows 4.0; StatSoft). EC₅₀ values were compared by paired or unpaired Students t-test. Only a single pair of arteries (±Endo) was obtained from an animal. Therefore, in all groups, n refers to both the number of arteries and number of animals used.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of gender on mesenteric artery contractile responses. Contractile responses to the ₁-adrenergic agonist PE, the 5-HT₂ receptor agonist 5-HT, and KCl depolarization were assessed in intact mesenteric arteries from male and female rats. There were no gender differences in overall sensitivity to PE (EC₅₀: male, 0.90 ± 0.13 µmol/l; female, 0.84 ± 0.17 µmol/l; Fig. 1), 5-HT (EC₅₀: male, 0.35 ± 0.07 µmol/l; female, 0.29 ± 0.06 µmol/l; Fig. 2), or KCl depolarization (EC₅₀: male, 21.5 ± 1.0 mmol/l; female, 21.2 ± 2.2 mmol/l; n = 6). Maximal contractile responses expressed as a percentage of resting diameter were also not significantly different between stimuli and genders. However, net mesenteric artery contractile responses may be modulated by both endothelium and vascular smooth muscle-derived vasoactive factors (2, 9, 19, 22, 24). Therefore, we considered the possibility that these modulatory influences could be gender dependent and, because different G protein-coupled contractile agonists may variably stimulate smooth muscle and endothelium-dependent pathways, agonist specific.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Effects of endothelium disruption and indomethacin (Indo) pretreatment on phenylephrine (PE)-stimulated contractile responses in rat mesenteric arteries. A: contractile responses in arteries from male rats with intact endothelium (+Endo) and mechanically disrupted endothelium (Endo) with and without Indo pretreatment (10 µmol/l, 15 min). Resting diameters were 381 ± 18 µm +Endo and 394 ± 13 µm Endo. B: contractile responses in arteries from female rats. Resting diameters were 348 ± 17 µm +Endo and 320 ± 32 µm Endo. Contractile responses are calculated as (change in diameter/maximal change in diameter) × 100. Values are reported and plotted as means ± SE; n = 6 for each group. Statistically significant changes due to endothelium disruption are indicated by +P  0.05 and ++P  0.01. Changes due to Indo pretreatment are indicated by *P  0.05 and **P  0.01.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Effects of endothelium disruption and Indo pretreatment on serotonin (5-HT)-stimulated contractile responses in rat mesenteric arteries. A: contractile responses in +Endo and Endo arteries from male rats with and without Indo pretreatment (10 µM, 15 min). Resting diameters were 417 ± 16 µm +Endo and 351 ± 24 µm Endo. B: contractile responses in arteries from female rats. Resting diameters were 339 ± 24 µm +Endo and 388 ± 22 µm Endo. Contractile responses are calculated as (change in diameter/maximal change in diameter) × 100. Values are plotted as means ± SE; n = 5-6 for each group. Statistically significant changes due to endothelium disruption are indicated by +P  0.05 and ++P  0.01. Changes due to Indo pretreatment are indicated by *P  0.05 and **P  0.01.

Endothelium-dependent modulation of contractile responses. To assess the contribution of endothelium-dependent factors on net contractile responsiveness, the effect of mechanical disruption of endothelium was compared in mesenteric arteries from males and females. As shown in Fig. 1A, in arteries from males, there was a significant increase in contractile responsiveness to PE after disruption of endothelium, whereas in females (Fig. 1B) this damage resulted in no effect on PE responsiveness. Thus it appears that in arteries from males, ₁-adrenoreceptor activation couples to release of an endothelium-derived vasodilator that attenuates the direct (and indirect) contractile affects of PE in vascular smooth muscle. In contrast to the results using PE as an agonist, damage to the endothelium produced a significant increase in contractile reactivity to 5-HT in arteries from both male (Fig. 2A) and female (Fig. 2B) rats. We conclude from these results that the variable contribution of endothelium-derived vasoactive factors to mesenteric artery responsiveness is both gender and agonist dependent.

Cyclooxygenase-dependent modulation of contractile responses. To assess potential gender differences in cyclooxygenase-dependent modulatory pathways, we compared contractile responses in indomethacin-pretreated arteries from males and females. Indo pretreatment (10 µmol/l, 15 min) resulted in a significant decrease in net PE contractile responsiveness in +Endo arteries from males (Fig. 1A) but not females (Fig. 1B). Indo pretreatment resulted in the same effect in Endo arteries from males (Fig. 1A), suggesting that vascular smooth muscle is the predominant source of modulatory cyclooxygenase metabolites produced in response to ₁-adrenergic stimuli. Indo pretreatment also produced a small decrease in net PE contractile responsiveness in Endo arteries from females (Fig. 1B). After both endothelium disruption and Indo pretreatment, there was no gender difference in sensitivity to PE (EC₅₀: male, 1.26 ± 0.15 µmol/l; female, 1.58 ± 0.42 µmol/l).

Effect of gonadectomy on modulation of ₁-adrenergic responses. Two gender-dependent modulators of net ₁-adrenergic contractile responses were identified as described above: 1) an endothelium-derived vasodilator (noncyclooxygenase dependent) expressed predominantly in males and 2) a smooth muscle-derived cyclooxygenase-1-dependent vasoconstrictor expressed to a greater extent in males compared with females. To assess the role of gonadal steroids on the gender-dependent expression of these modulatory pathways, PE dose-response curves were assessed in mesenteric arteries from ovariectomized (Ovx) and orchiectomized (Orch) rats. Net contractile responsiveness to PE in genetic females was unaffected by ovariectomy, whereas responsiveness to PE in arteries from genetic males was decreased by orchiectomy (Fig. 3).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   PE-stimulated contractile responses in +Endo mesenteric arteries from control male and female rats, orchiectomized (Orch) males, and ovarectomized (Ovx) females. Significance between gender differences in vasoconstriction are indicated by +P  0.05 and ++P  0.01. Within-gender differences are denoted by **P  0.01. Values represent means ± SE; n = 6.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Effects of endothelium disruption and Indo pretreatment on PE-stimulated contractile responses in mesenteric arteries from gonadectomized rats. A: contractile responses in +Endo and Endo arteries from Orch male rats with and without Indo pretreatment (10 µM, 15 min). Resting diameters were 367 ± 17 µm +Endo and 406 ± 14 µm Endo. B: contractile responses in arteries from Ovx female rats. Resting diameters were 381 ± 21 µm +Endo and 351 ± 38 µm Endo. Contractile responses are calculated as (change in diameter/maximal change in diameter) × 100. Values are plotted as means ± SE; n = 6. Statistically significant changes due to endothelium disruption are indicated by +P  0.05 and ++P  0.01. Changes due to Indo pretreatment are indicated by *P  0.05 and **P  0.01.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study was designed to examine gender differences in the control of mesenteric artery vasoconstriction. Specifically, we examined the modulatory function of the mesenteric artery endothelium and pathways related to the production of prostaglandins. Comparison of the net contractile responses to PE, 5-HT, and potassium chloride in intact mesenteric-resistance arteries from male and female rats indicated no gender differences in overall sensitivity to any of these stimuli. However, the study showed that the net contractile responses to G protein receptor-coupled agonists variably depend on both positive (smooth muscle-derived cyclooxygenase products) and negative (endothelium-dependent) modulatory influences that are both agonist and gender dependent. Gonadal steroids appear to underlie the variable expression or coupling of ₁-adrenergic stimuli to these modulatory pathways.

A number of studies have demonstrated enhanced endothelium-dependent vasodilation in females compared with males (8, 21, 22) and estrogen-dependent regulation of NO synthase expression and NO production (11, 15). Therefore, the observed gender difference in the contribution of endothelium-derived vasodilators to net ₁-adrenergic-stimulated contractions that was greater in males compared with females was unexpected. Our results showing comparable increases in contractile sensitivity to 5-HT in arteries from both males and females upon endothelium disruption suggest no intrinsic deficit in endothelium function in arteries from females. Assuming equal or even greater capacity for endothelium-dependent vasodilation in females, a lack of effect of endothelium disruption on net PE contractile responsiveness in arteries from females implies either a lack of endothelial cell ₁-adrenergic receptors or a failure of endothelial ₁-adrenergic receptors to couple to production and release of vasodilators. Consistent with this idea, ovariectomy enhanced or unmasked ₁-adrenergic receptor coupling to endothelial vasodilators in genetic females, suggesting estrogen-dependent suppression of this pathway in control animals as modeled in Fig. 5. Disruption of endothelium has also been shown to differentially potentiate contractile responses to vasopressin in thoracic aortic rings from male rats compared with females (18).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Hypothesized effect of gonadal sex steroids, estrogen, and testosterone on endothelium-dependent and Indo-sensitive pathways affecting PE-induced contractile responses in mesenteric arteries from male and female rats. 1, 1-Adrenergic receptor; AA, arachidonic acid; COX-1, cyclooxygenase-1; PG, prostaglandin.

Previous studies using rat cremaster arterioles have demonstrated that ₁-adrenergic contractile responsiveness is modulated by the release of NO from endothelium (20). It seems likely that NO also contributes to the modulation of PE and 5-HT responsiveness observed in the present study, although this has not yet been tested directly. However, our results differ from previously reported studies of contractile responses in mesenteric artery rings from female rats that reported a significant increase in sensitivity to an ₁-adrenergic stimulus after disruption of endothelium or administration of an NO synthase inhibitor (4). In that study, net sensitivity to PE in intact rings was somewhat less than we observed here (EC₅₀ 1.48 vs. 0.84 µmol/l in the present study), whereas the sensitivity in Endo arteries appeared similar in the two studies (EC₅₀ 0.74 vs. 0.92 µmol/l in the present study). The reasons for these differences are not clear. Although age of the animals appeared similar in the two studies, given the different preparations used (isometric contraction in rings versus isotonic contractions in intact vessel segments), it is possible that basal production of endothelium-dependent vasodilators was greater in the former study, accounting for the decreased sensitivity to PE in control vessels compared with the present results.

The second gender-specific modulatory pathway found in the present study resulted from the production of vasoconstrictor prostaglandins, as revealed by the desensitization of PE contractile responses after treatment with Indo in arteries from males but not females. Indo pretreatment of Endo arteries produced results similar to those from +Endo arteries, clearly indicating that the source of the cyclooxygenase products was nonendothelial and most likely smooth muscle (Fig. 5). We found that this pathway for vasoconstrictor cyclooxygenase metabolite production in vascular smooth muscle was significantly diminished on orchiectomy, suggesting that it is positively regulated by testosterone in genetic males. The findings that pretreatment with the cyclooxygenase-2 inhibitor celecoxib had no effect on net PE contractile responses, whereas the cyclooxygenase-1 inhibitor SC-560 decreased PE responsiveness, indicate that cyclooxygenase-1 is the relevant isozyme in smooth muscle with respect to production of modulatory vasoconstrictors. Results of experiments using 5-HT and KCl as contactile stimuli indicate that stimulation of the smooth muscle cyclooxygenase pathway appears to be restricted to agonists that activate phospholipases via G protein-coupled receptors.

In general, these results are similar to previous studies indicating enhanced cyclooxygenase-derived vasoconstricting factors produced in male rats, although in the previous studies, this pathway appeared to be endothelium dependent (1, 9, 19, 23). However, the experimental design of those studies, which assessed endothelium-dependent vasodilation in preconstricted blood vessels, would not readily provide insight into the status of cyclooxgenase factors produced by smooth muscle. Our study also confirm previous studies showing no effect of cyclooxygenase inhibition on ₁-adrenergic contractile responses in mesenteric artery rings from female rats (4).

Given what is known concerning the diversity of endothelium-derived vasoconstrictors and vasodilators and the importance of endothelial and vascular smooth muscle cell-to-cell interaction, the model presented in Fig. 5 is probably an oversimplification. For example, to account for the gender specificity of the endothelial effect with respect to ₁-adrenergic stimulation, the model proposes that ₁-adrenergic receptor number or coupling in endothelium is suppressed by estrogen in control females. According to this simplified view, estrogen depletion would be predicted to enhance production of endothelium-derived vasodilator and decrease net PE contractile responses in females. Yet, we observed no significant effect of ovariectomy on net PE contractile responsiveness in arteries from females. One possible explanation is that estrogen may also suppress ₁-adrenergic receptor expression or coupling in vascular smooth muscle in control females. If so, ovariectomy would be expected to result in both an enhanced ₁-adrenergic-stimulated production of endothelial vasodilator and enhanced ₁-adrenergic-mediated contractile activity at the level of smooth muscle, opposing effects that might result in no net effect on contractile sensitivity. This explanation is supported by previous studies that have documented decreased ₁-adrenergic receptor expression and desensitization of net PE-stimulated contractile responses in mesenteric arteries after estrogen replacement in Ovx female rats (24). Also, not reflected in the model is the possibility that nerve tissue or adventitial fibroblasts may also be the source of vasoactive substances, including cyclooxygenase-1-dependent arachidonic acid metabolites, capable of modulating tone. A potential limitation of the present study is that the experiments were performed under conditions of no flow. Considering that flow may differentially regulate endothelium-derived vasoactive factors in a gender-specific manner (8), it will be important to confirm whether the gender differences in modulatory pathways observed in the present study contribute to potential differences in ₁-adrenergic responsiveness in vivo or under experimental conditions with flow.

On the basis of epidemiological data, endothelial damage occurs approximately a decade earlier in men compared with women (3). An extrapolation of results from the present study is that such endothelial damage in males could result in a significant increase in ₁-adrenergic agonist-mediated vascular reactivity, whereas females are protected from this effect, in part in an estrogen-dependent manner. Such a gender difference in modulatory pathways could contribute toward a predisposition to hypertension and cardiovascular disease between age-matched males and premenopausal women. Because the level of estrogen is significantly diminished at menopause, leading to an increased susceptibility to endothelial damage, this mechanism could also contribute to the increased vascular reactivity and incidence of cardiovascular disease in postmenopausal women and the beneficial effects of estrogen replacement. The results of the present study also suggest that, in addition to antithrombotic and anti-inflammatory actions, cyclooxygenase-1 inhibitors may provide a selective benefit in males with respect to desensitization towards ₁-adrenergic-mediated contractile stimuli.