Role of adenylyl cyclase in reduced β-adrenoceptor-mediated vasorelaxation during maturation

Beta-adrenergic receptor (βAR)-dependent blood vessel relaxation is impaired in older animals and G protein activation has been suggested as the causative mechanism. Here, we investigated the role of βAR subtypes (β1AR, β2AR, and β3AR) and cAMP in maturation-dependent vasorelaxation impairment. Aortic rings from 15 Sprague-Dawley male rats (3 or 9 weeks old) were harvested and left intact or denuded of the endothelium. Vascular relaxation in aortic rings from younger and older groups was compared in the presence of βAR subtype agonists and antagonists along with cAMP and cGMP antagonists. Isolated aortic rings were used to evaluate relaxation responses, protein expression was evaluated by western blot or real time PCR, and metabolites were measured by ELISA. Expression of βAR subtypes and adenylyl cyclase was assessed, and cAMP activity was measured in vascular tissue from both groups. Isoproterenol- and BRL744-dependent relaxation in aortic rings with and without endothelium from 9-week-old rats was impaired compared with younger rats. The β1AR antagonist CGP20712A (10-7 M) did not affect isoproterenol or BRL744-dependent relaxation in arteries from either group. The β2AR antagonist ICI-118,551 (10-7 M) inhibited isoproterenol-dependent aortic relaxation in both groups. The β3AR antagonist SR59230A (10-7 M) inhibited isoproterenol- and BRL744-dependent aortic ring relaxation in younger but not in older rats. All βAR subtypes were expressed in both groups, although β3AR expression was lower in the older group. Adenylyl cyclase (SQ 22536) or protein kinase A (H89) inhibitors prevented isoproterenol-induced relaxation in younger but not in older rats. Production of cAMP was reduced in the older group. Adenylyl cyclase III and RyR3 protein expression was higher in the younger group. In conclusion, altered expression of β3AR and adenylyl cyclase III may be responsible for reduced cAMP production in the older group.


Introduction
Blood vessel relaxation induced by the catecholamine system is mediated through b adrenergic receptors (bAR). Although vascular bAR were originally classified as beta 2 (b2AR) and beta 1 (b1AR) adrenergic receptor subtypes, later studies described atypical b adrenergic receptors that included b3AR (1). Further characterization described b3AR as a relevant mediator in catecholamine-induced vasodilation (2)(3)(4). Catecholamine-dependent blood vessel relaxation is impaired in older animals (5). In humans, loss of vascular relaxation in elderly people could contribute to clinical development of hypertension or cardiovascular diseases, as orthostatic hypotension and arterial insufficiency are present in older populations (6). Several authors have explored the mechanism involved in bARdependent impaired vascular relaxation. Initial reports suggested decreased cAMP production and phospho-diesterases as the mechanism involved in the adrenergic vasorelaxation impairment (5). Later on, decreased function of the stimulatory GTP-binding protein (Gas) was associated with b2AR-induced vasodilation impairment in aortas from aged rats (7). This idea was further explored by several authors that either supported decreased (8) or increased (9) Gas protein expression as the mechanism for age-dependent adrenergic vasorelaxation impairment. More recently, a differential activation of Ga protein has been suggested as an explanation for this controversy: activation of Gas decreased, whereas Gai increased in older animals (10).
Most of the studies on age-related bAR-mediated vascular relaxation have focused on b1AR or b2AR, but few studies have evaluated the role of b3AR (1,4). However, it has been suggested that the three bAR subtypes work in concert to render adrenergic vascular relaxation. All bAR receptors couple to Gas protein and elicit cAMP production (11). Under normal stimulation, b2ARand b1AR-dependent cAMP production is modulated by PKA phosphorylation, leading to reduced cAMP production and reduced b2ARand b1AR-mediated vasorelaxation (12). However, b3AR does not require this PKA phosphorylation mechanism, which allows a prolonged vasorelaxation effect compared with that induced by b2AR and b1AR (13). These combined results suggest that age-dependent changes in adrenergic vasorelaxation could be associated with differential expression of b-adrenergic receptor second messengers.
Understanding the causes of age-related decline in bAR-mediated vasodilation-signaling may help identify a therapeutic target for treating cardiovascular diseases. In the present study, we investigated bAR subtypes and cAMP-dependent vasorelaxation in maturation-dependent adrenergic vasorelaxation. We evaluated the mechanisms of bAR-mediated cAMP signaling by testing vasorelaxation and cAMP activity in aortic rings from 3-and 9-week-old rats in the presence of specific b1AR, b2AR, b3AR, and cAMP antagonists. We also evaluated protein content and mRNA expression of bAR in thoracic aorta.

Material and Methods
Fifteen male Sprague-Dawley rats (3 or 9 weeks old) were used throughout the experiments. Rats were fed with a balanced diet and had water ad libitum. All procedures conformed to the National Institutes of Health "Guide for the Care and Use of Laboratory Animals" (1996) and were approved by the Institutional Ethics Review Committee for Animal Experimentation of the Center of Research and Advanced Studies, IPN (approval No. 479-10).

Preparation of rat aortic rings
Rats were sacrificed by cervical dislocation. The thoracic aorta was carefully removed, and placed directly into ice-cold Krebs-Henseleit bicarbonate solution (117.8 mM NaCl, 6.0 mM KCl, 1.6 mM CaCl 2 , 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 24.2 mM NaHCO 3 , 11.0 mM glucose, 0.027 mM EDTA), equilibrated with 95% O 2 and 5% CO 2 , pH 7.4. Then, periadventitial fat was removed. Care was taken during the complete procedure to ensure the integrity of the vascular endothelium. For experiments with endothelium-denuded aortic rings, the endothelium was removed by gently rubbing the intimal surface of the vessels. Aortic rings (3-4 mm wide) were mounted in 5 mL water-jacketed organ baths at 37°C and equilibrated for 1.5 to 2.0 h. A maximum of two aortic rings from either 3-or 9-week-old rats were used simultaneously. Basal tone was set at 1.5 and 2.0 g, respectively. Changes in tension were measured using a pressure transducer (TSD125) connected to a computerized data acquisition system (MP150-BIOPAC, BIOPAC Systems, USA). This procedure was determined to produce optimal conditions for reproducible isometric force development.

Experimental procedure
The presence of endothelium was confirmed by assessing the effectiveness of acetylcholine (ACh, 10 -6 M) to relaxed aortic rings pretreated with phenylephrine (PE, 10 -6 M). The endothelium denudation was confirmed by the incapability of aortic rings to relax in response to ACh (10 -6 M). Concentration-response curves for ACh, isoproterenol, dobutamine, salbutamol, BRL3744, and sodium nitroprusside were obtained in aortic rings pretreated with PE by cumulative addition of the drug to the organ bath. The concentration was increased only after maximal response to the previous concentration was attained. Specific inhibitors and antagonists were used as follows: bAR antagonists, 10 -7 M CGP20712A for b1AR, 10 -7 M ICI-118,551 for b2AR, and 10 -7 M SR59230A for b3AR; adenylyl cyclase inhibitor, 10 -7 M SQ22536; PKA inhibitor, 10 -7 M H89; guanylyl cyclase inhibitor, 10 -7 M ODQ; and PGK inhibitor, 10 -7 M K75823. The isoproterenol concentration-response curve or specific agonist curve were determined, then aortic rings were incubated with the inhibitor or antagonist for 30 min, and the isoproterenol concentration-response curves were repeated. Vascular relaxation is reported as percentage, considering basal tension before PE stimulus as 100% vascular relaxation and PE-induced tension as 0% vascular relaxation.

Gene expression analysis
Aortas from 3-and 9-week-old rats were homogenized and total RNA was extracted using TRIzol (Life Technologies, USA). RNA integrity was checked in agarose gels, and 1.0 mg RNA was used for reverse transcriptase reactions. Gene expression analysis was performed using the FastStart SYBR Green Master (Rox) kit (Roche Applied Science, USA) and a 7500 Real Time Thermal Cycler (Applied Biosystems, USA). Specific primers for adenylyl cyclase subtypes and the calcium-related protein RyR3 target genes are shown in Table 1.
Relative gene expression was normalized to the constitutive expression of a-actin. Gene expression values were determined using the 2 -DDCT formula. Data in figures represent 1 of the 5 different experiments. AR/actin densitometry value ratios were calculated for each group and reported as means ± SE.

cAMP enzyme immunoassay
Aortic rings were incubated in 500 mL of Krebs solution alone or with cAMP-specific phosphodiesterase inhibitor (10 -7 M rolipram) at 37°C in a shaker bath. cAMP stimulation was initiated by adding 10 -5 M isoproterenol for 15 min. Samples were rapidly frozen after collection, and intracellular cAMP levels were determined using a cyclic AMP EIA kit (Cayman Chemical, USA) according to the manufacturer's instructions.

Statistical analysis
Concentration-response curves were compared using two-way analysis of variance (ANOVA) and specific differences between concentrations and rats were compared by modified Bonferroni t-test. Comparisons between two or more groups were performed by one-way ANOVA followed by Newman Keuls test to identify specific differences. Statistical analyses were performed using GraphPad Software Prism 5 for Mac OS X version 5.0 (USA). Data are reported as means ± SE and Po0.05 was considered statistically significant.

Isoproterenol-dependent vascular relaxation
PE produced similar dose response-dependent contractions in endothelium-intact aortic rings from 3-and 9-week-old rats. Maximal contractions were 1.22 ± 0.13 and 1.49 ± 0.25 g, respectively. In endothelium-denuded rings from 3-and 9-week-old rats, maximal contractions elicited by PE were 1.32 ± 0.12 and 1.67 ± 0.15 g, respectively. Isoproterenol on precontracted (PE treated) aortic rings from 3-and 9-week-old rats produced a concentration-dependent relaxation in vessels from both groups ( Figure 1 and Table 2). However, isoproterenolinduced vascular relaxation was lower in aortic rings from 9-week-old rats than in 3-week-old rats ( Figure 1A). This impairment was larger in endothelium-denuded aortic rings than in aortic rings with intact endothelium. Maximal % of isoproterenol-induced relaxation was 100±0% compared with 77.3 ± 2.1% in endothelium-intact aortic rings from 3-and 9-week-old rats, respectively. Whereas, isoproterenol-induced relaxation was 98.1±1.5% compared with 47.2±1.7% in endothelium-denuded aortic rings from 3-and 9-week-old rats, respectively ( Figure 1A). No difference in aortic ring relaxation was observed with acetylcholine or sodium nitroprusside agonists treatment in either group.

Relaxation response to adrenergic agonists in older rats
To identify the possible adrenergic receptor involved in impaired adrenergic relaxation in 9-week-old rats, we tested different b-adrenergic agonists ( Table 2). Dobutamine (b1 agonist, Figure 1B) and salbutamol (b2 agonist, Figure 1C) induced dose-dependent aortic ring relaxation in both younger and older rats. There were no differences in relaxation responses between 3-and 9-week-old rats in the presence or absence of endothelium. BRL3744 (b3 agonist) induced concentration-dependent vascular relaxation in both groups; however, BRL3744-mediated aortic ring relaxation was lower in 9-week-old rats compared with that of younger rats ( Figure 1D). BRL 3744-mediated impairment was more severe in endotheliumdenuded aortic rings than in endothelium-intact aortic rings. Maximal relaxation induced by BRL3744 was 31.4 ± 4.5 and 2.1 ± 0.6% in endothelium-denuded aortic rings of 3-and 9-week-old rats, respectively, compared with 72±2.7 and 10.2±2.5% in endothelium-intact aortic rings of 3-and 9-week-old rats, respectively ( Figure 1D).

Relaxation response to adrenergic antagonists in older rats
To further explore the b-adrenergic subtypes involved in vascular relaxation impairment in older rats, we tested the effects of bAR agonists in the presence of bAR antagonists. The results are presented as pD2 values in Table 3. The presence of b1AR antagonist CGP20712A inhibited dobutamine-induced relaxation, but did not affect salbutamol-or BRL3744-induced responses. The addition of b2AR antagonist ICI-118,551 inhibited isoproterenoland salbutamol-induced relaxation, but did not affect the dobutamine-induced response. The presence of b3AR antagonist SR59230A inhibited BRL3744-induced relaxation, but did not affect dobutamine-and salbutamolinduced responses.
These antagonists were tested on isoproterenoland BRL3744-induced relaxation in aortic rings with and without endothelium. Isoproterenol-induced relaxation was tested in presence of specific bAR antagonists (Figure 2). The presence of b1AR antagonist CGP20712A did not affect isoproterenol-dependent aortic ring relaxation in younger and older rats. Addition of the b2AR antagonist ICI-118,551 inhibited isoproterenol-dependent aortic ring relaxation in both groups (Figure 2A-D). The b3AR antagonist SR59230A inhibited isoproterenol-dependent aortic ring relaxation in 3-week-old rats but not in 9-weekold rats. BRL3744-dependent vascular relaxation was not affected by the b1AR antagonist CGP20712A or the b2AR antagonist ICI-118,551 in either group. However, the b3AR antagonist SR59230A inhibited BRL3744-dependent vascular relaxation in aortic rings of 3-week-old rats, but not in those of 9-week-old rats ( Figure 3A-D).
Differential protein expression of b1AR, b2AR, and b3AR Higher levels of b1AR and b2AR protein expression were detected in vascular tissue of 9-week-old rats compared with those of 3-week-old rats). The protein Figure 1. Concentration response curves in precontracted (treated with phenylephrine) aortic rings. The relaxant response to isoproterenol (A), dobutamine (B), salbutamol (C), and BRL3744 (D) was evaluated in endothelium-intact aortic rings (+) and endothelium-denuded aortic rings ( À ) of 3-and 9-week-old rats. Data are reported as the mean±SE of 5 different animals. *Po0.05 vs 3-week-old (ANOVA followed by modified Newman Keuls t-test). Table 2. pD2 values for b-adrenergic agonist-induced relaxation in aortic rings with (+) and without (-) endothelium.    expression levels of b3AR were lower in vascular tissue of 9-week-old rats compared with that in 3-week-old rats ( Figure 4A and B). No differences were observed in b2AR receptor phosphorylation under basal conditions or after isoproterenol stimulation in vascular tissue of younger and older rats ( Figure 5A and B).

Differential response to adenylyl cyclase pathway inhibitors
The adenylyl cyclase inhibitor SQ22536 and the protein kinase A inhibitor H89 significantly inhibited isoproterenol-induced vascular relaxation in aortic rings of 3-week-old rats but not in those of 9-week-old rats  ( Figure 6A and B). The guanylyl cyclase inhibitor ODQ and the PGK inhibitor KT75823 did not affect isoproterenolinduced relaxation in aortic rings of younger and older rats ( Figure 6C and D).
cAMP production by aortic rings was measured after stimulation with isoproterenol. The results indicated that cAMP production decreased in aortic rings of 9-week-old rats (3.3 ± 1 pmol/mL) compared with that in 3-week-old rats (20±2 pmol/mL). Similar results were observed in the presence of the phosphodiesterase inhibitor rolipram (8 ± 1 and 26 ± 3 pmol/mL for 9-and 3-week-old rats). However, dose response curves of forskolin-induced  vascular relaxation showed no differences between aortic rings from younger and older rats. Thus, pD2 values were 6.9 ± 0.13 and 6.6 ± 0.2, and cAMP production was 20 ± 2 and 25 ± 5 pmol/mL for 3-and 9-week-old rats, respectively.
The expression of genes encoding adenylyl cyclase III, V, and VI showed increased expression of type V and VI and decreased expression of type III in vascular tissue of 9-week-old rats compared with that in 3-week-old rats ( Figure 7). The calcium-related protein RyR3 mRNA transcript levels increased 2.5 ± 0.4 fold in vascular tissue of 9-week-old rats compared with that in 3-week-old rats.

Discussion
In the present study, we demonstrated that vascular relaxation impairment is associated with maturation, and we suggest that changes in expression of genes encoding b3AR and adenylyl cyclases are responsible for the altered vascular function. Our observation that vasorelaxation impairment induced by ACh or sodium nitroprusside did not change in 9-week-old rats compared with 3-week-old rats supports a specific role for bAR in maturation-dependent vasorelaxation impairment, as described in previous studies (7). Reduced bAR-induced vasorelaxation associated with aging has been reported in several studies (6,8,(16)(17)(18), similarly to our results. However, the age of animals in those reports ranged from 6-to 24-month-old compared with 9-week-old rats used in the present study, suggesting that this effect is related to maturation rather than with aged animals.
The presence of several bAR subtypes in rat aortic rings was demonstrated by the use of specific receptor agonists and antagonists and further supported by western blot. This suggests that all 3 adrenergic receptor subtypes are involved in the isoproterenol induced vasorelaxation, as reported previously (19). Beta adrenergic-dependent vasorelaxation has been described as endothelium-dependent and endothelium-independent in rat aorta (20,21). Our data demonstrated that endothelium removal decreased salbutamol and BRL37344induced vasorelaxation but did not affect dobutamineinduced vasorelaxation (Table 2). Thus, we suggest that adrenergic-mediated vasorelaxation of the endothelium is related to the receptor involved.
Identification of the bAR involved in maturationdependent vasorelaxation impairment was suggested by the decreased vasorelaxation in 9-week-old rats induced by specific b2 and b3 agonists, and supported by the absence of inhibitory effect of specific antagonists. More over, decreased expression of the b3 receptor further supports the role of this receptor in the impaired vasorelaxation.
Relaxation impairment is more evident if endothelium has been removed (53±3% in endothelium-denuded arteries) compared with endothelium-intact rings (33 ± 4%). This result suggests an important change of the adrenergic signaling pathway in smooth muscle cells. This is in agreement with a report showing reduction in isoproterenol-induced cAMP accumulation in cultured smooth muscle cells obtained from older rats (22). Because no differences in b2 adrenergic receptor phosphorylation were observed in younger and older rats, we suggest that vasorelaxation impairment in 9-week-old rats was not associated with changes in receptor activation.
Adenylyl cyclase stimulation with cAMP production is the classical AR signaling pathway in vascular smooth muscle (23). Therefore, we explored cAMP pathway as being responsible for the age-maturation impairment. In older rats, there was a reduction of cAMP-dependent b3 vasorelaxation, which was evidenced by the fact that neither adenylyl cyclase inhibitors nor PKA inhibitors affected isoproterenol-induced relaxation in aortic rings from 9-week-old rats. cAMP production was reduced in aortic rings from 9-week-old rats, which further supports the hypothesis that the isoproterenol effect was no longer cAMP-dependent in older rats. Reduced cAMP production can be explained by reduced initial stimulus, reduced adenylyl cyclase activation, or elevated phosphodiesterase activation. Forskolin-induced adenylyl cyclase activity did not differ in aortic ring relaxation in younger and older rats suggesting that cAMP activation is not the mechanism involved in the age-maturation cAMP reduction. However, adenylyl cyclase III expression was reduced in vascular tissue of 9-week-old rats, suggesting that reduced cAMP levels may result from a change in adenylyl cyclase subtype that interacts with b-receptors. The calcium receptor RyR3 expression levels increased in Figure 7. Comparative analysis of transcript abundance of adenylyl cyclase (AC) isoforms. qPCR was performed on cDNA from 3-and 9-week-old rats. The relative expression of AC III, V, or VI was calculated by the 2 -DDCT method and normalized with respect to the mean value of 3-week-old controls. All values were normalized. Data are reported as the mean±SE of 5 different animals. *Po0.05 vs 3-week-old rats (one-way ANOVA followed by Newman Keuls test). 9-week-old rats. Adenylyl cyclase calcium-dependent regulation has been associated with adenylyl cyclase inhibition rather than phosphodiesterase activation (24). Cytoplasmic calcium levels are reported to regulate adenylyl cyclase activity in vascular smooth muscle cells (25,26). Therefore, our data suggest that reduced cAMP production in 9-week-old rats was associated with increased calcium levels and to increased RyR3 receptor, which inhibits adenylyl cyclase III isoform in aortic rings.
The present results indicate that impaired bAR relaxation may be associated with age maturation. Our results showed evidence of this process at the onset of sexual maturation. Therefore, we suggest that increased male sexual hormones promote changes in the AR signaling pathway as part of physiological blood vessel maturation. Impaired bAR relaxation present in adult or older rats may be associated with blood vessel damage. Thus, the role of testosterone related to changes in vascular response requires further investigation.