Contribution of nitric oxide to β₂-adrenoceptor mediated vasodilatation in human forearm arterial vasculature

Abstract

Aims

β₂-adrenoceptor agonists are generally considered to produce endothelium independent vasodilatation through adenylate cyclase. We determined whether nitric oxide contributes to β₂-adrenoceptor vasodilatation in human arterial vasculature.

Methods

Forearm blood flow responses to brachial intra-arterial infusions of ritodrine (2.5–50 μg min⁻¹), a selective β₂-adrenoceptor agonist, were determined in 24 healthy, normotensive subjects (mean age 22 years, 5F) on two occasions with initial and concomitant administration of l-NMMA (800 μg min⁻¹), an NO synthase inhibitor, or noradrenaline (5–30 ng min⁻¹), a control constrictor not affecting basal NO activity. Responses to the endothelium dependent vasodilator serotonin (n = 6) and an endothelium independent vasodilator GTN (n = 9) were also determined.

Results

Maximal dilatation to ritodrine during l-NMMA infusion (310±32%; mean±s.e.mean) was reduced compared to that during noradrenaline infusion (417±41%, P < 0.05), as were summary responses (1023±101 vs 1415±130; P < 0.05). Responses to GTN were unaffected by l-NMMA compared to noradrenaline; max 177±26 vs 169±20%, 95% CI for difference −33,48; P = 0.68; summary response 361±51 vs 396±37, 95% CI −142,71; P = 0.46. Dilator responses to serotonin were reduced by l-NMMA; max 64±20 vs 163±26%, P < 0.01; summary response 129±36 vs 293±60; P < 0.05) and to a greater extent than ritodrine (58±7 vs 25±14%, P < 0.05).

Conclusions

β₂-adrenoceptor mediated vasodilatation in the human forearm has an NO mediated component. The underlying mechanism for this effect is unclear, but flow mediated vasodilatation is unlikely to be responsible.

Introduction

Nitric oxide (NO), synthesised from l-arginine in the vascular endothelium, plays a major role in the control of vascular tone [1]. Basal release of NO from endothelial cells contributes to maintenance of vascular resistance in the human forearm.

β₂-adrenoceptor agonists have generally been considered to cause vasodilatation by an endothelium-independent mechanism through activation of adenylate cyclase, and the subsequent production of adenosine 3′,5′-cyclic monophosphate (cAMP) [2]. However, several recent studies suggest that the endothelium may play a role in β-adrenoceptor mediated vasodilatation. In rat resistance vessels, vasodilatation to isoprenaline, a non-selective β-adrenoceptor agonist, but not salbutamol, a selective β₂-adrenoceptor agonist, is partially attenuated by NO synthase inhibition suggesting involvement of β₁ or β₃ adrenoceptors or a non β-adrenoceptor mechanism [3]. In rat aortic rings, isoprenaline- and salbutamol-mediated relaxation is accompanied by an increase in intracellular levels of both cAMP and cGMP suggesting the involvement of guanylate cyclase [4]. Pre-treatment with an NO synthase inhibitor blunted the dilator response to isoprenaline and the levels of cGMP, without affecting cAMP levels [4]. In canine coronary resistance vessels, intracoronary administration of the NO synthase inhibitor l-NAME blunted isoprenaline mediated increases in coronary blood flow [5]. l-NAME also partially blunts the relaxation caused by salbutamol in the rat hindlimb [6]. However involvement of endothelium dependent NO mechanisms in β-adrenoceptor mediated vasodilatation has not been a universal finding. In rat femoral artery and aorta, isoprenaline vasodilatation has been reported to be unaffected by NO synthase inhibitors or removal of the endothelium [7, 8].

The aims of the present study were to determine whether NO contributes to β₂-adrenoceptor vasodilatation in human forearm arterial vasculature.

Methods

Study population

Studies were performed on twenty six healthy volunteers (21 male, 5 female; aged 18–35 years) in the morning between 10.00 and 13.00 h. All subjects were non-smokers, taking no medications, and were normotensive with a resting, supine blood pressure of less than 140/90 mm Hg. Informed written consent was obtained from subjects and studies were approved by the Newcastle Joint Ethics Committee. On the day of the studies, subjects refrained from caffeine and alcohol and consumed no food for at least 2 h prior to the studies.

Study drugs

All study drugs were dissolved in sterile normal saline (0.9%NaCl w/v: Fresenius Ltd, Cheshire, UK). The following drugs were administered: noradrenaline (Sanofi Winthrop, Surrey, UK), l-N^G-monomethyl-arginine (Clinalfa AG Switzerland), ritodrine hydrochloride (Solvay, Southampton, UK), serotonin hydrochloride (Clinalfa AG Switzerland), glyceryl trinitrate (David Bull laboratories, Warwick, UK), and lignocaine hydrochloride 1% w/v (Antigen Pharmaceuticals, Southport, UK).

Study protocol

Studies were performed in a quiet, temperature-controlled room (25–27° C), with subjects resting supine for 30 min prior to the studies. Subjects were studied on two occasions, at least 7 days apart. On each occasion a 27 swg cannula (Cooper’s Needle Works, UK), was inserted into the brachial artery of the non-dominant arm following administration of 1% lignocaine as local anaesthesia. Drugs or physiological saline were infused continuously at 1.0 ml min⁻¹. Forearm blood flow (FBF) was measured at 10 min intervals during 30 min saline infusion to establish baseline FBF values. On the first occasion, the NO synthase inhibitor l-NMMA (800 μg min⁻¹) was infused for 5 min. In studies examining dose-response of FBF to brachial artery infusions of l-NMMA (200–1600 μg min⁻¹) in 18 young female subjects we have established this dose produces maximal constriction, and inhibition of NOS activity (unpublished observations). Ritodrine, a selective β₂-adrenoceptor agonist was then co-infused at 5 doses (2.5, 5, 10, 20, 50 μg min⁻¹) with l-NMMA, each for 5 min. On the second occasion, noradrenaline (5–30 ng min⁻¹), a control vasoconstrictor not inhibiting NO activity was infused, in incremental doses to produce a similar degree of constriction to that produced with l-NMMA and then co-infused with ritodrine. Responses to serotonin (6–60 ng min⁻¹), an endothelium dependent vasodilator, and glyceryl trinitrate (GTN) (250–1000 ng min⁻¹), an endothelium independent vasodilator were studied, during l-NMMA and noradrenaline infusions in respectively 6 and 9 subjects (of the same 26 volunteers), to provide positive and negative study controls. Following infusion of serotonin or GTN, l-NMMA or noradrenaline infusions were continued, and FBF returned to initial constricted levels before dose response curves to ritodrine were constructed.

Forearm blood flow measurements

Forearm blood flow was measured simultaneously in both arms by venous occlusion plethysmography [9] using galidinium in silastic strain gauges, and expressed as ml 100 ml⁻¹ forearm min⁻¹. During recording, the hands were excluded from the circulation by inflation of the wrist cuffs to 200 mm Hg, the upper arm cuffs were then inflated to 50 mm Hg for 10 s in each 15 s cycle. The average of five consecutive measurements was used for analysis.

Data analysis

Forearm blood flow responses were expressed as the percentage change of FBF in the infused arm compared to the baseline FBF of the infused arm. Within subject differences of percentage changes inFBF in the control and infused arms were assessed using two separate repeated measures analysis of variance (ANOVA). Further analysis was undertaken if ANOVA suggested a statistically significant change in FBF over time. Repeated measures analysis of variance was also undertaken using absolute FBF measurements with the same results. The overall response to each drug in each subject was assessed by two measures; the maximal response and a summary response, calculated as the summation of the percentage dilator responses for the individual doses of the infused drug (arbitrary units). Data are expressed as (means±s.e.mean), and were compared using paired Student’s t-test. P < 0.05 was considered statistically significant.

Results

Ritodrine infusions

Data for ritodrine responses on two male subjects were excluded from the data analysis, as forearm blood flow did not return to baseline values after initial infusion of serotonin. Baseline FBF did not significantly differ between the 2 study days (3.54±0.26 and 3.60±0.26 ml 100 ml⁻¹ min⁻¹, P = 0.81). Forearm blood flow in the control arm did not change during the course of the studies. Constriction (%±s.e.mean) to l-NMMA and noradrenaline was similar; 24±2 vs 26±3%, P = 0.60; 95% CI for difference (−5,8). Maximal dilatation to ritodrine (%±s.e.mean) was blunted during l-NMMA compared with noradrenaline co-infusion; 310±32 vs 417±41%, P < 0.05 (Figure 1). Summary measure responses (mean±s.e.mean) were similarly reduced during l-NMMA co-infusion; 1023±101 vs 1415±130; P < 0.05.

Figure 1.

Vasodilator responses to brachial artery infusions of ritodrine in 24 healthy young subjects during co-infusion of either l-NMMA (▴) or noradrenaline (▪) on two separate occasions. Changes in forearm blood flow in the infused arm are compared with values at rest following 30 min saline infusion. Maximal forearm blood flow responses to ritodrine were blunted during l-NMMA infusion compared with noradrenaline (P < 0.05). Data are mean±s.e.mean.

Serotonin and GTN infusions

Baseline FBF did not significantly differ between the 2 study days during either the serotonin (3.33±0.31 and 3.14±0.28 ml 100 ml⁻¹ min⁻¹, P = 0.66) or GTN infusions (3.77±0.59 and 3.93±0.63 ml 100 ml⁻¹ min⁻¹, P = 0.62). Pre-constriction to l-NMMA and noradrenaline was similar prior to serotonin infusions; 21±3 vs 20±7%, P = 0.86; 95% CI (−14,12). Maximal dilatation was blunted during l-NMMA co-infusion compared to noradrenaline; 64±20 vs 163±26%, P < 0.01 (Figure 2). Serotonin summary measure results were also blunted; 129±177 vs 293±60; P < 0.05. l-NMMA blunted serotonin induced vasodilatation to a greater extent than ritodrine; 58±7 vs 25±14%, P < 0.05. Comparison of the blunting of vasodilatation at a dose of ritodrine (2.5 μg min⁻¹) that elicited a similar degree of dilatation to serotonin (150%vs 163%) gave similar results; 58±7 vs 22±10%, P = 0.09. Constriction to l-NMMA and noradrenaline was similar during the GTN infusions; 24±4 vs 29±4%, P = 0.57; 95% CI (−13,22). Maximal dilatation to GTN was unaffected by l-NMMA; 177±26 vs 169±20%, P = 0.68; 95% CI (−33,48), (Figure 3). Summary measure responses were similarly unchanged; 361±51 vs 396±37; P = 0.46; 95% CI (−142,71).

Figure 2.

Vasodilator responses to brachial artery infusion of the endothelium dependent vasodilator serotonin in six healthy young subjects, during co-infusion of either l-NMMA (▴) or noradrenaline (▪). Data are mean±s.e.mean, and are changes compared with baseline FBF following 30 min saline infusion. Maximal forearm blood flow responses to serotonin were blunted during l-NMMA compared with noradrenaline co-infusion (P < 0.05).

Figure 3.

Vasodilator responses to brachial artery infusions of the endothelium independent vasodilator GTN in seven healthy subjects, during co-infusion of either l-NMMA (▴) or noradrenaline (▪). Data are mean±s.e.mean, and are changes compared with baseline FBF following 30 min saline infusion. Maximal forearm blood flow responses were unchanged during l-NMMA compared with noradrenaline co-infusion (P = 0.68).

Discussion

This study demonstrates that β₂-adrenergic vasodilatation in the human forearm vasculature has an NO component, accounting for some 25% of the total response. This is less than that observed with the endothelium dependent dilator serotonin, suggesting that adenylate cyclase activation remains responsible for the majority of the vasodilator action of ritodrine on human forearm arterial vasculature. Forearm vasodilatation to GTN was unaffected by NO synthase inhibition, indicating that NO activity associated with β₂-adrenoceptor vasodilatation is not due to a non-specific effect of increased flow. Noradrenaline has weak β₂-adrenergic effects, which could have opposed α-adrenergic constrictor effects, and the NO component of ritodrine’s vasodilator response may have been underestimated due to a higher level of β₂-adrenoceptor activity for any given dose of ritodrine with noradrenaline compared to l-NMMA co-infusion.

Other recently published studies agree with these findings. Two studies have reported l-NMMA blunting vasodilatation to isoprenaline, a non-selective β-adrenoceptor agonist, [10, 11] but did not exclude a β₁-adrenoceptor mediated effect. Dawes et al. [12] reported a blunted response to intra-arterial infusions of salbutamol in the human forearm following l-NMMA. In this study, baseline flow was not standardised with a control vasoconstrictor, but responses were compared with a control vasodilator, with and without l-NMMA co-infusion.

The mechanism responsible for this endothelium dependent component is unclear, but three likely possibilities exist; flow-mediated vasodilatation, endothelial cell β₂ response, or a facilitative interaction between NO system and β₂-mediated vasodilatation. Acute increases in blood flow and volume have been associated with a secondary increase in vascular NO activity in a number of settings; the phenomenon of flow-mediated vasodilatation [13, 14]. In this study, we minimised the initial contribution made by flow mediated vasodilatation by producing similar degrees of pre-constriction with l-NMMA and noradrenaline. By performing the study in this way, the possibility of a treatment order effect does exist but is very unlikely given the time interval between studies. Also the lack of attenuation of the forearm vasodilator response to GTN during l-NMMA co-infusion, observed in this and in a previous study, suggest flow-mediated vasodilatation is not responsible [15].

A more likely mechanism is endothelial cell β₂-adrenoceptors activating NO synthase. β-adrenoceptors have been characterised in both animal and human endothelial cells, with a predominance of the β₂-adrenoceptor subtype [16]. Finally a number of studies indicate functional interactions between the intracellular second messengers cAMP and cGMP. Studies in the isolated rat kidney suggest that NO can activate adenylate cyclase [17]. In pig pial arteries, isoprenaline and salbutamol infusion have been observed to increase both cAMP and cGMP levels, and the cGMP levels were subsequently attenuated by an NO synthase inhibitor [18]. Conversely other evidence suggests that cAMP increases responsiveness to basal and agonist induced NO release. In pig aortic endothelial cells, the synthesis and release of NO by bradykinin and ATP was enhanced by pre-treatment with forskolin, adenosine, and isoprenaline, all of which activate adenylate cyclase [19]. In a neuronal cell-line, the amplitude and duration of the cGMP response to bradykinin, serotonin, and endothelin was enhanced by pre-treatment with forskolin, compared to controls [20]. Enhancement of NO release was associated with an increase in internal Ca²⁺ stores. Beta-adrenoceptor induced increases in endothelial Ca²⁺ by cAMP could result in enhanced release of NO. Alternatively increases in smooth muscle cell cAMP could amplify the response to any given level of basal NO activity.

In conclusion the present studies demonstrate that β₂-adrenoceptor mediated vasodilatation is partially NO mediated in the human forearm arterial vasculature. This effect is likely to be due to a specific β₂-adrenoceptor mediated effect on NO synthase and not a non-specific flow mediated effect related to β₂-adrenoceptor mediated increases in blood flow.