Reduced vasorelaxation to estradiol and G-1 in aged female and adult male rats is associated with GPR30 downregulation

Abstract

Previously, we reported that chronic activation of the estrogen receptor GPR30 by its selective agonist G-1 decreases blood pressure in ovariectomized hypertensive mRen2.Lewis (mRen2) rats but not intact male littermates. Furthermore, G-1 relaxes female mesenteric resistance arteries via both endothelium-dependent and -independent mechanisms. Because of the lack of a blood pressure-lowering effect by G-1 in males and the potential influence of aging on estrogen receptor expression, we hypothesized that GPR30-dependent vasodilation and receptor expression are altered in males and aged females. Thus, we assessed the response to 17β-estradiol or G-1 in mesenteric arteries obtained from 15-wk-old normotensive Lewis and hypertensive mRen2 females and males as well as 52-wk-old Lewis females. Vasodilation to 17β-estradiol (E₂) and G-1 was significantly attenuated in 15-wk-old Lewis and mRen2 males compared with age-matched females. Pretreatment of male vessels with the nitric oxide synthase inhibitor l-NAME had no significant effect on the estradiol or G-1 response. In aged females, E₂ and G-1 vasorelaxation was also significantly blunted; however, l-NAME essentially abolished the response. Associated with the reduced vascular responses, GPR30 expression in mesenteric arteries was approximately 50% lower in males and aged females compared with young females. We conclude that alterations in GPR30 expression and signaling may contribute to vascular dysfunction in aging females and a greater blood pressure in hypertensive males.

the g protein-coupled estrogen receptor [G protein-coupled receptor 30 (GPR30)] is a membrane-bound receptor that mediates the acute effects of estrogen, as opposed to the long-term transcriptional effects initiated by the classical steroid receptors estrogen receptor (ER)α and ERβ (8). One commonly studied effect of this sex hormone is its ability to promote acute vasorelaxation (19). Therefore a deficit in estrogen signaling or receptors may contribute to cardiovascular disease and vasomotor instability, two common symptoms of menopause. Several clinical studies demonstrate an increased risk of coronary heart disease in women with vasomotor symptoms, further emphasizing the importance of estrogen in vascular health (32, 36). Because GPR30 is a membrane-bound estrogen receptor that activates acute signaling pathways, this novel receptor may play an important role in mediating estrogenic effects in the vasculature.

We have utilized the mRen2.Lewis (mRen2) hypertensive rat to establish the role of estrogen and GPR30 in the regulation of blood pressure, vascular reactivity, and target organ damage. The mRen2 strain is an angiotensin II (ANG II)-dependent model of hypertension that displays marked sex differences in systolic blood pressure (27). The mRen2 female is estrogen sensitive; ovariectomy markedly increases blood pressure, and estradiol replacement or the ANG II type 1 receptor antagonist olmesartan essentially normalizes blood pressure (5). Recently, we demonstrated that chronic treatment with the selective GPR30 agonist G-1 also reduces systolic blood pressure to that of intact female controls (20). In isolated mesenteric vessels from adult mRen2 females, GPR30 activation induces acute vasodilation that is not different from estradiol and comprises both endothelium-dependent (nitric oxide) and endothelium-independent mechanisms (18). Moreover, the GPR30 antagonist G15 abolishes the vascular response to both estradiol and G-1, demonstrating the selectivity of the agonist for its receptor.

In contrast to the ovariectomized mRen2 female, G-1 does not lower systolic blood pressure in hypertensive male littermates (20). However, other studies utilizing normotensive male rodents demonstrate beneficial effects in response to GPR30 activation (7, 11). Furthermore, male GPR30-knockout mice display many phenotypic alterations, including visceral adiposity and vascular dysfunction (11, 24). These studies raise the question of whether estrogen expression and function are comparable in males and females. Aging is another important consideration in the analysis of ER function. A decline in endogenous estrogens in combination with aging may lead to alterations in ER expression or signaling and promote deleterious cardiovascular outcomes in response to hormone replacement (35). Indeed, this differential response to estrogens is present in the mRen2 model; estrogen depletion via ovariectomy is detrimental in young females but beneficial in aged females fed a high-salt diet (40). So although it is widely accepted that estrogen influences vascular tone and blood pressure in a normal setting, the ability of this hormone to promote vascular health in the face of age-related changes is yet to be determined. Therefore, the current study assessed the hypothesis that GPR30-dependent vasodilation and receptor expression are altered by sex and aging.

METHODS

Animals.

Congenic mRen2 rats were obtained from the Wake Forest Hypertension Center Transgenic colony, whereas Lewis rats were obtained from Harlan (Indianapolis, IN). Rats were housed in an Association for Assessment and Accreditation of Laboratory Animal Care-approved facility in a temperature-controlled room (22 ± 2°C) with a 12:12-h light-dark cycle, free access to food and water, and daily monitoring by veterinary staff. Systolic blood pressure was measured by tail cuff plethysmography, as described previously (20). All procedures were approved by the Animal Care and Use Committees at both Tulane University and Wake Forest University.

Vascular reactivity.

Second-order mesenteric arteries were isolated at 15 (“young”) or 52 wk of age (“aged”) and mounted on a wire myograph system (Model 620M; Danish Myo Technology, Aarhus, Denmark), as described previously (18). Vessels were initially stimulated with 0.08 M potassium chloride (KCl), 10⁻⁵ M phenylephrine (PE), and 10⁻⁶ M acetylcholine (ACh; Sigma, St. Louis, MO) to assess vessel viability and endothelial function. Vessels with >50% relaxation to ACh were considered endothelium intact. The responses to increasing concentrations (10⁻⁹ to 10^−5.5 M) of G-1 (Cayman Chemical, Ann Arbor, MI) and estradiol (E₂; Sigma) were measured in PE-constricted vessels for 5 min and compared with vehicle (1:10,000 DMSO). Some vessels were pretreated with 10⁻⁴ M N^G-nitro-l-arginine methyl ester (l-NAME) 10 min prior to PE contraction (Sigma). Responses were recorded using Chart 5 (AD Instruments, Colorado Springs, CO) and are expressed as the percentage (%) of PE contraction.

Vaginal smears.

Estrous cycle phase was determined by serial vaginal smears. Five-hundred microliters of saline was flushed gently into the entrance of the vagina two to three times. The resulting cell suspension was analyzed under a microscope for identification of cell types present during different phases of the estrous cycle (22). Persistent vaginal cornification was characterized by a predominance of cornified cells in the smear for 4 continuous days.

Serum E₂.

Trunk blood was collected, allowed to clot, and centrifuged at 1,800 g to obtain the serum. Serum samples were extracted and assayed using the Cayman Estradiol EIA kit (Cayman Chemical) according to the manufacturer's instructions.

Immunoblotting.

Mesenteric arteries were homogenized in 60 μl of NP-40 cell lysis buffer (Invitrogen, Grand Island, NY) containing protease inhibitor cocktail (Sigma) for 10 min with a Wheaton glass pestle and incubated for 30 min at 4°C. Samples were centrifuged, and the protein concentration of the supernatant was determined using the Pierce BCA Protein Assay Kit (Thermo Scientific Pierce, Rockford, IL). Eighty micrograms of protein was loaded onto a Novex 10% Bis-Tris gel (Invitrogen) and run at 150 V for 1.5 h before being transferred to a nitrocellulose membrane, using the I-Blot (Invitrogen) at 10 V for 8 min. Membranes were blocked for 1 h in 5% milk and incubated in GPR30 primary antibody (1:400, sc-134576, lot no. K2311; Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4°C and LI-COR goat anti-rabbit secondary antibody (1:10,000) for 30 min. Blots were scanned using the Odyssey infrared imager (LI-COR Biosciences, Lincoln, NE). Membranes were reprobed with anti-β-actin (monocolonal antibody no. 3700; Cell Signaling Technology, Beverly, MA) as a loading control.

Statistics.

All measurements are expressed as means ± SE. All data were analyzed using GraphPad Prism version 6 (GraphPad Software, San Diego, CA). Two-way ANOVA was used for concentration response curves, whereas one-way ANOVA was used for all other data sets. A confidence limit of 95% was considered significant, and Tukey's multiple comparison test determined differences between groups.

RESULTS

Blood pressure and baseline indices of experimental groups.

As reported previously by our laboratory (27), systolic blood pressures were significantly higher in mRen2 males (203 ± 8 mmHg) compared with Lewis males (116 ± 5 mmHg, P < 0.001; Fig. 1A). Systolic blood pressures were also greater in mRen2 females (137 ± 5 mmHg) vs. Lewis female controls (111 ± 3 mmHg, P < 0.05). A signfiicant sex difference in blood pressure was evident in mRen2 (P < 0.001) but not Lewis rats (P > 0.05). Body weights were significantly higher in males vs. females (P < 0.001; Fig. 1B) and were also lower in Lewis females vs. other female groups (P < 0.05). Uterine weights were similar in all females (P = 0.90; Fig. 1C). Aged females displayed indices of reproductive senescence, including an extended estrous cycle or persistent vaginal cornification, as assessed by serial vaginal smears, and significantly lower serum E₂ (Fig. 1D) (38). Interestingly, the vasocontrictor response to 10⁻⁵ M PE was significantly enhanced in the mRen2 males vs. all other groups (P < 0.01; Fig. 2A); however, ACh relaxation was similar across all groups (P = 0.09; Fig. 2B).

Fig. 1.

A: systolic blood pressure was significantly greater in mRen2.Lewis (mRen2) males vs. all other groups (*P < 0.001; n = 4–6). Systolic blood pressure was also greater in mRen2 females vs. Lewis females and aged Lewis females (†P < 0.05). B: the body weights of both male groups were significantly higher than all females (*P < 0.001; n = 5). In addition, the body weights of Lewis females were significantly less than both mRen2 females and aged Lewis females (†P < 0.05). C: uterine weights were not different among female groups (P = 0.90; n = 5). D: compared with the mean serum estradiol level in young females across all phases of the estrous cycle, estradiol was decreased significantly in aged females exhibiting constant estrous (*P < 0.05; n = 5).

Fig. 2.

A: phenylephrine (PE) contraction was significantly greater in male mRen2 mesenteric vessels vs. all other groups (*P < 0.01; n = 26–33 vessels). B: acetylcholine (ACh) relaxation in endothelium-intact vessels was not different among groups (P = 0.09; n = 21–24 vessels).

Vasodilation in adult female and male rats.

To assess the effect of sex on GPR30-mediated responses in the vaculature, vessels isolated from 15-wk-old Lewis and mRen2 male rats were compared with female littermates. Dose-dependent relaxation to E₂ was significantly attenuated in male vessels (P < 0.01; Fig. 3A) and was similar between the two strains of male rats (P > 0.05). In experiments using the selective GPR30 agonist G-1, vasorelaxation was similary reduced in both male Lewis and mRen2 vessels compared with females (P < 0.05; Fig. 3B). The nitric oxide synthase inhibitor l-NAME did not alter the E₂ response (P > 0.05; Fig. 4A) or the G-1 response (P > 0.05; Fig. 4B) in male Lewis and mRen2 vessels. Moreover, the results obtained in endothelium-denuded Lewis male vessels were not different from l-NAME pretreatment (P > 0.05, n = 6–7/group; data not shown).

Fig. 3.

A: compared with mRen2 and Lewis females, mesenteric arteries from male rats showed significantly less vasodilation in response to estradiol (E₂; *P < 0.05). B: relaxation to the G protein-coupled receptor 30 (GPR30) agonist G-1 was also blunted in male vessels (*P < 0.01).

Fig. 4.

A: N^G-nitro-l-arginine methyl ester (l-NAME) had no significant effect on the response to E₂ in male arteries (P > 0.05). B: l-NAME also had no significant effect on the response to G-1 in male arteries (P > 0.05). NS, not significant.

Vasodilation in aged females.

To examine the impact of age on GPR30-dependent vasorelaxation, vessels were isolated from 52-wk-old female Lewis rats. Compared with young Lewis females, vessels from aged females relaxed significantly less in response to both E₂ and G-1 (P < 0.05; Fig. 5A). In contrast to males, l-NAME essentially abolished both E₂ and G-1 responses (P < 0.001; Fig. 5B) in aged female vessels. Results similar to l-NAME were obtained in endothelium-denuded vessels from aging females (P > 0.05, n = 5–7/group; data not shown). To determine whether estrous cycle status influenced this response, GPR30-mediated vasorelaxation in young Lewis females cycling normally was stratified according to estrous cycle phase, as determined by serial vaginal smears. As shown in Fig. 6, there were no significant differences in the vasodilatory response during different phases of the estrous cycle (P > 0.05).

Fig. 5.

A: compared with 15-wk-old Lewis females, arteries from aged Lewis females dilated significantly less in response to E₂ or the GPR30 agonist G-1 (*P < 0.05). B: in aged Lewis female arteries, l-NAME pretreatment abolished the response to both E₂ and G-1 (*P < 0.001).

Fig. 6.

In young Lewis females, vasodilation in response to the GPR30 agonist G-1 did not vary across phases of the estrous cycle (P > 0.05).

Vascular GPR30 protein expression.

We assessed the relative expression of GPR30 protein in mesenteric arteries obtained from young female, aged female, and young male Lewis rats. As reported previously in both aortic and mesenteric vessels from young mRen2 females, we found a major band of 50 kDa for GPR30 in mesenteric lysates from Lewis rats (Fig. 7A). Moreover, quantification of the immunoblots revealed significantly lower GPR30 expression in both the aged female and young male Lewis groups compared with young female Lewis rats (P < 0.05; Fig. 7B).

Fig. 7.

A: a representative immunoblot comparing GPR30 protein levels in mesenteric arteries from young female Lewis, aged female, and young male Lewis rats. B: GPR30/β-actin protein expression normalized to Lewis females was reduced significantly in aged female and young male Lewis arteries (*P < 0.05; n = 4/group).

DISCUSSION

The ability of male arteries to relax in response to exogenous E₂ is evident in many vascular beds. E₂ improves coronary blood flow in both male and female spontaneously hypertensive rat, although the response is significantly greater in females (13). In contrast, E₂-induced vasorelaxation is greater in mesenteric arteries from male guinea pigs (25). In denuded rat aortic rings, there is no sex difference in E₂-induced vasodilation (9), and E₂-induced vasorelaxation is independent of sex in rat tail arteries (10). Effects of E₂ on vascular reactivity are also evident in human males, as clinical studies show an improvement in blood flow following infusion of conjugated equine estrogens or oral E₂ (2, 17). Although males used in the current study were gonad intact, castration has no effect on the vasodilatory response to E₂ in rat aortic strips and tail arteries (6, 10).

The decrease in E₂-induced vasorelaxation in males is most likely due to GPR30 downregulation in the mesenteric vasculature. GPR30 expression is also significantly reduced in male rat inferior vena cava (28) and mouse and human pancreatic islets (1, 16) but shows no sex difference in rat brain (3, 23), heart (7), or carotid and middle cerebral arteries (4). Despite differences in GPR30 levels, the presence of a detrimental phenotype in male GPR30-knockout mice suggests that this ER plays an important physiological role in males as well as females (11, 24). Although circulating estrogen levels are significantly lower in males, local production of sex hormones in the arterial wall may reach physiologically significant concentrations and provide beneficial cardiovascular effects (34).

Previous studies in young mRen2 females revealed no effect of ovariectomy on the vascular response to E₂ or G-1 despite a significant increase in blood pressure (18). However, the current study demonstrates impaired vasorelaxation and reduced GPR30 expression in mesenteric arteries from intact aged females. E₂-induced relaxation is also attenuated in the aorta of aging female spontaneously hypertensive rat, and whereas there were no differences in ERα or ERβ density, GPR30 expression was not assessed (39). The irregular cycling of aged female Lewis rats was characterized by persistent vaginal cornification, which along with the maintained uterine weight suggests a period of constant estrous (21, 37). Aged females in constant estrous had significantly lower serum E₂ levels compared with E₂ levels in young females averaged across all phases of the estrous cycle; however, the E₂ level in aged females was still within the normal range for young animals in the estrous phase and higher than what is achieved with ovariectomy (21, 26, 40). This period of constant estrous characterized by maintained E₂ levels is unique to rodents and precedes a period of persistent anestrous with very low E₂ levels, which is more akin to human menopause. The differing results in young ovariectomized vs. aged intact females raised the question of whether the low E₂ levels in aged females contributed to an attenuated response. However, no differences in GPR30-induced vasorelaxation were found across phases of the estrous cycle in young Lewis females cycling normally. The absence of an attenuated response in the normal cycling female during periods of low E₂, along with our previous results showing a maintained response in young ovariectomized females, suggests that reduced GPR30 expression and function are independent of hormonal status. Downregulation of vascular GPR30 may compromise the beneficial effects of E₂ replacement later in life.

It is important to note that systolic blood pressures in the aged Lewis females were not different from young Lewis females or males; the use of telemetry instead of tail cuff may have detected small differences in blood pressure in normotensive rats (33). As reported previously, systolic blood pressures in mRen2 males and females were significantly higher than their Lewis counterparts (27). In vessels from mRen2 males, which display the highest systolic blood pressures, phenylephrine constriction was significantly greater compared with all other groups. In contrast, there were no differences in ACh relaxation. Whereas other hypertensive animal models exhibit a decrease in endothelium-dependent vasodilation (30, 31), mesenteric vessels from transgenic (mRen2)27 rats, the predecessor of the congenic mRen2 rats, do not show any evidence of endothelial dysfunction (29). Likewise, we show for the first time that mesenteric artery endothelial function is preserved in both male and female congenic mRen2.Lewis rats. The increased phenylephrine response in the male mRen2 suggests greater sympathetic tone, perhaps due to ANG II-induced upregulation of adrenergic receptors, rather than reduced compensatory vasodilator mechanisms (12). Moreover, the lack of a difference in systolic blood pressure between female and male Lewis rats despite reduced GPR30 expression suggests that alterations in ER signaling within the male mRen2 vasculature are not the primary defect contributing to a sustained increase in blood pressure. However, reduced GPR30 or its associated signaling pathways may potentially contribute to lack of buffering against either the progression or maintenance of hypertension in the mRen2 males.

We demonstrated that GPR30-dependent vasodilation in young female rats comprises a significant endothelial component revealed by attenuation of the response after denuding or l-NAME treatment (18). In contrast, the male response in the current study was not altered by l-NAME or denuding, suggesting a deficit in the endothelial component of GPR30 signaling. This deficit could be due to underlying endothelial dysfunction in males, as other studies demonstrate greater endothelial-dependent relaxation in females (14, 15). Although responses at submaximal concentrations of ACh were not assessed, we found no overt differences in the response to 10⁻⁶ M ACh in the current study, suggesting little or no sex differences in global endothelial function. The data from male vessels instead suggest a nitric oxide-independent mechanism, as reported previously in porcine coronary arteries (41). Thus, the reduced response in males could reflect selective downregulation of endothelial GPR30 expression. In contrast, the response in aged females was completely attenuated by l-NAME or denuding, suggesting a deficit in the vascular smooth cell response to GPR30 activation. Previously, we showed that GPR30 localized to both the intimal and medial layers of mesenteric arteries and aorta (18). In combination with the functional data shown here, we suggest a deficit in endothelial GPR30 expression in males, whereas smooth muscle cell GPR30 expression is preferentially downregulated in aging females. The mechanism for GPR30's nitric oxide-independent effects in vascular smooth muscle cells is currently under investigation.

In conclusion, mesenteric artery relaxation to E₂ and the selective GPR30 agonist G-1 was blunted in young males and aged females to a similar extent and was associated with a decrease in GPR30 protein expression. The male vascular response was not altered in the presence of l-NAME, indicating a lack of nitric oxide release in response to GPR30 activation. In contrast, vasodilation in aged females was completely dependent on the production of nitric oxide. Reduced GPR30 signaling in males may contribute to sex differences in blood pressure evident in the mRen2 and other hypertensive strains. In aging females, reduced GPR30 expression in conjunction with altered ER signaling may contribute to the loss of cardiovascular protection in response to hormone replacement therapy and instead promote detrimental estrogenic effects.

GRANTS

This work was supported by the National Heart, Lung, and Blood Institute (HL-56973, HL-51952, and HL-103974) and the American Heart Association (25515E and 3080005).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.H.L. and M.C.C. contributed to the conception and design of the research; S.H.L., A.S.d.S., and M.S.S. performed the experiments; S.H.L., A.S.d.S., and M.S.S. analyzed the data; S.H.L. and M.C.C. interpreted the results of the experiments; S.H.L. prepared the figures; S.H.L. drafted the manuscript; S.H.L. and M.C.C. edited and revised the manuscript; S.H.L., A.S.d.S., M.S.S., and M.C.C. approved the final version of the manuscript.