Independent Regulation of Alpha₁ and Alpha₂ Adrenergic Receptor–Mediated Vasoconstriction in vivo

Abstract

Background

Vascular α₁- and α₂-adrenergic receptors (ARs) mediate vasoconstriction and are major determinants of peripheral vascular tone. There is wide variability in vasoconstrictor sensitivity to α₁- and α₂AR-agonists among individuals. In previous studies this variability was not explained by identified α₁- and α₂-AR genetic variants. Thus, we hypothesized that adrenergic vasoconstrictor sensitivity is determined by shared constrictor mechanisms downstream of the individual receptors and that α₁- and α₂-AR-mediated vasoconstrictor sensitivity would therefore be correlated.

Methods

Dorsal hand vein responses to increasing doses of the α₁-AR agonist phenylephrine (12 ng/min –12,000 ng/min) and the α₂-AR agonist dexmedetomidine (0.01 ng/min – 100 ng/min) were measured in healthy subjects using a linear variable differential transformer. From individual dose-response curves we calculated the dose of phenylephrine and dexmedetomidine that produced 50% (ED₅₀) of maximum venoconstriction (E_max) for each subject. We examined the correlation between phenylephrine and dexmedetomidine ED₅₀ and E_max before and after adjustment for covariates (age, gender, ethnicity, BMI, blood pressure, heart rate, and baseline plasma norepinephrine concentrations).

Results

In 62 subjects (36 males, 34 African American, 28 Caucasians) the median ED₅₀ for dexmedetomidine was 1.32 ng/min (IQR, 0.45–5.37 ng/min), and for phenylephrine 177.8 ng/min (IQR, 40.7– 436.5 ng/min). The E_max for phenylephrine was 90.8% (82.2–99.6%) and for dexmedetomidine 80.0% (64.7–95.2%). There was no correlation between individual sensitivities (ED₅₀) to phenylephrine and dexmedetomidine, before and after adjustment for covariates (p>0.30).

Conclusions

Phenylephrine and dexmedetomidine both produce strong venoconstriction in the dorsal hand vein; however, there is no significant correlation between vascular sensitivity to an α₁-AR and α₂-AR agonist. These findings suggest independent regulation of vascular α₁- and α₂-AR-mediated responses.

INTRODUCTION

Post-synaptic α₁- and α₂-adrenergic receptors (AR) are widely distributed in the peripheral vasculature [1, 2] and both mediate vasoconstriction [1, 2]. In humans, direct infusion of the α₂-AR agonist, clonidine, into the brachial artery causes vasoconstriction [3, 4], and in the forearm vasculature of young healthy men post-synaptic α₂-ARs contribute more than α₁-ARs to basal vascular tone [5]. In the hand vein, we and others have shown that both α₁-AR and α₂-AR agonists cause pronounced vasoconstriction, with sensitivity among individuals varying several fold [6–8].

However, the factors contributing to this interindividual variability in vascular sensitivity are poorly understood. Pharmacogenetic studies have not identified α₁- and α₂-AR variants that affect response to agonists substantially. For example, variants in the gene (ADRA2B) encoding the α_2B-AR, an important mediator of vasoconstriction [9], contribute only a small amount to variability in dorsal hand vein responses to the α₂-AR agonist, dexmedetomidine [10–12]. Similarly, genetic variants of the α₁-AR studied thus far do not explain the large interindividual variability in vascular response to the α₁-AR agonist phenylephrine [8].

Since genetic variants of the respective receptors studied to date do not explain variability in α-AR-mediated vascular response, this variability may be determined by factors that are not directly dependent on the receptor, and are shared by α₁- and α₂-ARs such as factors that determine the drug concentration at the site of action, or factors that determine the response of pathways downstream of the receptor. If such factors contribute to variability in α₁- and α₂-ARs responses, then one would expect the sensitivities to vasoconstriction mediated by α₁- and α₂-ARs to be correlated. However, the possibility that α₁-AR and α₂-AR vascular sensitivity are co-regulated, has not been studied.

Therefore, to examine the hypothesis that α₁- and α₂-AR-mediated vasoconstrictor sensitivity are correlated we measured responses to the α₁-AR agonist, phenylephrine, and to the highly selective α₂-AR agonist, dexmedetomidine, in healthy subjects in vivo using the dorsal hand vein model.

METHODS

Subjects

The Institutional Review Board of Vanderbilt University Medical Center approved the study protocol, and subjects gave written informed consent. Male and female Caucasians and African-Americans were eligible for the study if they were 18 to 45 years of age and had no clinically significant abnormalities according to medical history, physical examination, and laboratory testing.

Sixty-two subjects were studied. Ethnicity, family history of hypertension, and exercise history were determined by self-report, and body mass index (BMI) was calculated. Subjects took no medications for at least 2 weeks, and abstained from alcohol and caffeine for at least 5 days before the study. Each subject received a diet containing 150 mmol/day of sodium, 70 mmol/day of potassium, and 600 mmol/day of calcium for the 5 days prior to each study day. Studies were performed in the morning after an overnight fast, in the same temperature-controlled room.

Measurement of Vascular Responses

Hand vein responses to the α₂-AR agonist dexmedetomidine HCl (Abbott Laboratories, Chicago, IL, USA) [6] and the α₁-AR agonist phenylephrine HCL (Elkins-Sinn, Cherry Hill, N.J., USA) [8] were measured on separate days. Venous responses were measured in a dorsal hand vein by use of a linear variable differential transformer (LVDT), (Schaevitz, model 100 MHR) as previously described [13, 14]. This instrument, when mounted on the hand, measures and records changes in the diameter of the vein.

Subjects rested on a comfortable bed and remained supine throughout the study. The subject’s arm was placed on a support sloping upward. A 23-gauge needle was inserted into a suitable dorsal hand vein, and an infusion of normal saline was administered for at least 30 minutes. Following three stable baseline measurements of hand vein diameter, drugs were infused into the vein over which the LVDT was mounted.

Phenylephrine (12 ng/min to 12,000 ng/min) or dexmedetomidine (0.01 ng/min to 100 ng/min) were administered in increasing doses with each dose infused for 7 minutes using a Harvard syringe pump and response recorded during the last 2 minutes of each infusion. The total flow rate into the vein was maintained constant by changing the rate of infusion of saline and drug. Heart rate was monitored continuously with a bedside cardiac monitor, and blood pressure was measured in the arm on the side opposite the side receiving the hand vein infusion using a semiautomated device (Dinamap MPS, Johnson and Johnson Medical, Tampa, Fla. USA). A blood sample for measurement of plasma norepinephrine concentration was obtained on the first study day at baseline before drug infusion.

Analysis of hand-vein response to dexmedetomidine and phenylephrine

Venoconstriction was expressed as the percentage reduction in vein diameter from maximal dilation, which was defined as the average of three stable baseline measurements of hand vein diameter. Measurements for the phenylephrine and dexmedetomidine hand vein responses were plotted as individual semi-logarithmic dose-response curves and analyzed using a sigmoid dose-response model (Prism 5.01 software). The dose that produced 50% (ED₅₀) of maximum venoconstriction (E_max) was determined for each subject, and these values were converted to molar concentrations to compare sensitivity (ED₅₀) to dexmedetomidine and phenylephrine for each subject.

Statistical analyses

Continuous parameters were expressed as median and interquartile range (IQR). ED₅₀ values are presented in original units (ng/min), and in log transformed molar units [log femtomole/min (log fmol/min)] for the comparison of ED₅₀ values for dexmedetomidine and phenylephrine.

ED₅₀ values for dexmedetomidine and phenylephrine in the same individuals were compared using Wilcoxon signed rank test, as were the E_max values. The Kruskal-Wallis test was used to compare dexmedetomidine ED₅₀ among quartiles of phenylephrine ED₅₀ values.

Correlations between the ED₅₀ for dexmedetomidine and phenylephrine, the E_max for dexmedetomidine and phenylephrine, and between plasma norepinephrine concentrations and the ED₅₀ for dexmedetomidine and phenylephrine were calculated using Spearman’s test.

To evaluate the effects of covariates on the relationship between dexmedetomidine and phenylephrine ED₅₀ values, we performed multiple linear regression analyses using log transformed dexmedetomidine ED₅₀ as outcome and log transformed phenylephrine ED₅₀ as the independent variable, adjusting for the covariates age, gender, ethnicity, BMI, mean arterial pressure, heart rate, and plasma norepinephrine concentrations. Similarly, we used multiple linear regression analysis to evaluate the effect of these covariates on phenylephrine log transformed ED₅₀.

All tests were two-sided, and p-values < 0.05 were considered significant. Statistical analyses were performed with the statistical software SPSS v. 17 (SPSS Inc, Chicago, Il, USA).

RESULTS

Subjects

We studied 62 subjects [36 men (56.3%), 34 African Americans (53.1%), and 28 Caucasians (43.8%)] with a median age (IQR) of 25.0 years (22.0– 32.0), and BMI 25.3 kg/m² (22.7– 28.3). Demographic and baseline characteristics are shown in Table 1.

Table 1.

Subject characteristics and baseline measurements [Number (percentage); Median (IQR)]

Characteristic	Value
Men	36 (56.3%)
Race: African Americans:     Caucasians	34 (53.1%): 28 (43.8%)
Family history of hypertension	23 (35.9%)
Regular exercise, >4 times/week	23 (35.9%)
Age (years)	25.0 ( 22.0– 32.0)
BMI (kg/m2)	25.3 (22.7– 28.3)
Systolic blood pressure (mm Hg)	109.5 (102.8– 116.3)
Diastolic blood pressure (mm Hg)	61.0 (56.0– 67.0)
Mean arterial pressure (mm Hg)	78.3 (72.6–82.3)
Heart rate (beats per minute)	60.0 (53.0– 66.0)
Plasma norepinephrine (pg/mL)	156.0 ( 127.0– 202.0)

Hand-vein responses dexmedetomidine and phenylephrine

The median ED₅₀ for dexmedetomidine was 1.32 ng/min (0.45–5.37 ng/min) [3.75 log fmol/min (3.27–4.36 log fmol/min)], and the ED₅₀ for phenylephrine was 177.8 ng/min (40.7– 436.5 ng/min) [5.94 log fmol/min (5.31–6.34 log fmol/min)]. The ED₅₀ in molar units was significantly smaller for dexmedetomidine than that for phenylephrine (p<0.001; Figure 1), suggesting greater vascular sensitivity to dexmedetomidine in the dorsal hand vein. The E_max for phenylephrine [90.8% (82.2–99.6%)] was significantly greater than that for dexmedetomidine [80.0% (64.7–95.2%)], (p<0.001; Table 2; Figure 2).

Figure 1. Hand vein sensitivity to dexmedetomidine and phenylephrine.

The horizontal line represents the median, the error bars represent the IQR

Data are presented in log fmole/min.

P<0.001 comparing the sensitivities to phenylephrine and dexmedetomidine

Table 2.

Hand vein responses to phenylephrine and dexmedetomidine

	Phenylephrine	Dexmedetomidine	P-value
ED50 (ng/min)	177.8 ng/min (40.7– 436.5)	1.3 ng/min (0.45–5.37)	<0.001*
Emax (%)	90.8 (82.2–99.6)	80.0 (64.7–95.2)	<0.001

Data are expressed as median (IQR)

E_max – Maximal venoconstriction, expressed as the percentage reduction in vein diameter from maximal dilation

ED₅₀ - The dose of phenylephrine and dexmedetomidine that produced 50% of maximum venoconstriction (E_max) for each subject

^*P-value for the comparison of ED₅₀ of phenylephrine and dexmedetomidine in fmole/min (see also Figure 1)

Figure 2. Maximal effect (E_max) for dexmedetomidine and phenylephrine.

E_max – Maximal venoconstriction, expressed as the percentage reduction in vein diameter from maximal dilation

The horizontal line represents the median and the bars represent the IQR;

P<0.001 comparing E_max for dexmedetomidine and phenylephrine

There was no correlation between individual sensitivities (ED₅₀) to phenylephrine and dexmedetomidine in all subjects [r= 0.12 (p=0.34)], or when African Americans [r= 0.06 (p=0.74)] and Caucasians [r= 0.21 (p=0.29)] were analyzed separately (Table 3; Figure 3). Dexmedetomidine ED₅₀ did not differ significantly among subjects grouped by quartiles of phenylephrine ED₅₀ (p=0.22) (Table 4).

Table 3.

Correlation between measures of phenylephrine and dexmedetomidine vascular response

	ED50 Rho (p-value)	Emax Rho (p-value)
All subjects (n=62)	0.12 (p=0.38)	0.15 (p=0.23)
African Americans (n= 34)	0.06 (p=0.74)	0.18 (p=0.32)
Caucasians (n=28)	0.21 (p=0.29)	0.17 (p=0.39)

E_max – Maximal venoconstriction, expressed as the percentage reduction in vein diameter from maximal dilation

ED₅₀ - The dose of phenylephrine and dexmedetomidine that produced 50% of maximum venoconstriction (E_max) for each subject

Figure 3. Relationship between log ED₅₀ for phenylephrine and dexmedetomidine.

ED₅₀ - The dose of phenylephrine and dexmedetomidine that produced 50% of maximum venoconstriction (E_max) for each subject

ED₅₀ is expressed in fmol/min

Spearman rho for the correlation: r= 0.12 (p=0.34)

Table 4.

Comparison of dexmedetomidine ED₅₀ in quartiles of phenylephrine ED₅₀

Phenylephrine ED50 quartile	1st quartile (lowest quartile)	2nd quartile	3rd quartile	4th quartile (highest quartile)	P- value
Dexmedetomidine ED50 (ng/min, median, IQR)	0.64 (0.30–1.89)	3.23 (0.26–34.57)	1.20 (0.43–5.08)	1.63 (0.52–9.48)	0.22

ED₅₀ - The dose of phenylephrine and dexmedetomidine that produced 50% of maximum venoconstriction (E_max) for each subject

Similarly, there was no correlation between individual E_max values for phenylephrine and dexmedetomidine in all subjects [r= 0.15 (p=0.23)], or when African Americans [r= 0.18 (p=0.32)] and Caucasians [r= 0.17 (p=0.39)] were analyzed separately (Table 3).

Determinants of hand-vein responses to dexmedetomidine and phenylephrine

In univariate analysis, plasma norepinephrine concentrations and phenylephrine ED₅₀ were significantly correlated [r=0.28 (p=0.037)], suggesting that higher plasma norepinephrine concentrations are associated with reduced phenylephrine sensitivity. In contrast, there was no significant correlation between plasma norepinephrine concentrations and dexmedetomidine ED₅₀ [r=−0.01 (p=0.96)].

In multiple linear regression analyses, phenylephrine ED₅₀ was not associated with dexmedetomidine ED₅₀ (β=0.25, p= 0.31), and neither was any of the other covariates (all p-values > 0.15). With phenylephrine ED₅₀ as the dependent variable, the association between phenylephrine ED₅₀ and plasma norepinephrine was weakened after adjustment for covariants, and of borderline statistical significance (p=0.063).

DISCUSSION

The major new finding of this study is the lack of correlation between the vascular sensitivities (ED₅₀) for dexmedetomidine and phenylephrine in vivo in humans. Therefore, our findings suggest that using the hand vein model with α₁- and α₂-AR-specific agonists provides independent information about α₂-AR and α₁-AR-mediated vasoconstrictor responses.

Interactions between the signaling pathways of α₁-, α₂- and β2-ARs were suggested by previous in vitro studies [15, 16]. Simultaneous activation of α_2B-ARs and β₂-ARs decreased the threshold concentration of epinephrine required for α_2B-AR down-regulation, and this was associated with up-regulation of GRK3 expression [15]. Co-activation of α_1a-ARs and β₂-ARs resulted in a facilitatory interaction which led to increases in calcium influx from the extracellular compartment [16]. Thus, we tested the hypothesis that variability in downstream constrictor pathways or other factors that are shared by α₁- and α₂-ARs contribute to the interindividual differences in α₁- and α₂-AR-mediated vasoconstrictor responses in the human hand vein. Our findings suggest that the large variability among individuals in both α₁- and α₂-AR-mediated hand vein vasoconstrictor responses is not explained by interindividual differences in factors that are shared by both receptors.

We found that in the dose-range of agonists used in this study (selected previously to elicit maximal response in the hand vein without systemic hemodynamic effects [6, 8]), the human dorsal hand vein is more sensitive to dexmedetomidine than to phenylephrine, since a lower molar concentration was required to produce half the maximal constriction. Also, although both agonists resulted in pronounced venoconstriction (E_max >80%), phenylephrine resulted in a larger median maximal venoconstrictor effect than dexmedetomidine.

We found a negative correlation between resting plasma norepinephrine concentrations and vascular sensitivity to phenylephrine. This finding is consistent with down-regulation of vascular α₁-ARs by norepinephrine. The modulating effect of endogenous non-selective adrenergic agonists such as norepinephrine, which activates both α₁-ARs and α₂-ARs in vivo, on hand vein responses has not been defined. We have previously shown that hand vein response to α₁-AR -mediated venoconstriction is stable over time (and is thus used to produce background vasoconstriction in studies of hand vein vasodilation) [17]. These two observations - the previously reported lack of desensitization with short-term exposure to an exogenous agonist such as phenylephrine, and our current finding of a negative correlation between resting plasma norepinephrine concentrations and vascular α₁-AR sensitivity - would be most consistent with near maximal α₁-AR desensitization at baseline conditions in vivo.

However, plasma norepinephrine concentrations are affected by several factors, including BMI, gender and ethnicity [18], and after adjustment for these and other covariates, the association between resting plasma norepinephrine concentrations and vascular sensitivity to phenylephrine was weakned and of borderline significance. Thus, it is unclear if there is a direct relationship between phenylephrine sensitivity and norepinephrine concentrations, or if it is due to factors that affect norepinephrine concentrations. To determine the specific relationship between endogenous sympathetic tone and α₁-AR-mediated local vascular response would require a study designed specifically for that purpose.

There are several methodological considerations regarding the measurement of vascular α₁-AR and α₂-AR responses to agonist in vivo. Systemic administration of either α₁-AR or α₂-AR agonists increases and decreases blood pressure, respectively, and thus results in the activation of homeostatic cardiovascular reflexes that would confound the measures of local vascular response. In particular, activation of central α₂-ARs results in a decrease in sympathetic tone, a factor that would mask the direct vasoconstriction mediated by peripheral α₂-ARs [4, 19]. Thus, although vascular α₂-ARs are functional, it has been difficult to establish their importance in the vasculature relative to α₁-ARs.

Accordingly, in order to define α₁-AR and α₂-AR vascular sensitivity it is necessary to minimize the effects on blood pressure and sympathetic activity that occur after systemic administration of agonist. Thus, we used the dorsal hand vein model [13, 14], that allows the direct infusion into the vessel studied of low doses of drugs that act on α₁- and α₂-ARs locally minimizing systemic effects [6, 13] and thus avoiding reflex cardiovascular responses that occur after systemic administration.

We defined α₁-and α₂-AR sensitivity using highly selective agonists, but cannot rule out the possibility of some non-selective α-AR activation. However, this is unlikely. First, phenylephrine and dexmedetomidine are both highly selective for their respective α-ARs, as compared to other α-ARs agonists for in vivo use [20, 21]. Second, non-selective α-AR activation by either agonist would be expected to increase the correlation between responses to dexmedetomidine and phenylephrine, whereas our study showed lack of correlation between responses.

The clinical implications of the current study are speculative. Since our findings suggest independent regulation of α₁- and α₂-AR- mediated vasoconstriction, dual blockade of both receptors peripherally could theoretically be used clinically to achieve an antihypertensive or vasodilating effect. However, while α₁-AR antagonists are extensively used in the treatment of hypertension, α₂-AR antagonists such as yohimbine cause an increase in blood pressure,[22] presumably because of the blockade of central pre-synaptic α₂-ARs, resulting in increased sympathetic outflow. Currently, no selective, peripherally acting post-synaptic α₂-AR antagonist is available for use in humans, precluding the use of α₂-AR antagonists as potential antihypertensive agent [23].

In summary, we found that phenylephrine and dexmedetomidine both produce powerful venoconstriction in the dorsal hand vein in vivo, but that there was no correlation between α₁- and α₂-AR vascular sensitivity. The large variability among individuals in α₁- and α₂-AR vascular sensitivity is not explained by differences in shared pre- or post-receptor factors.

Abbreviations and symbols

E_max

Maximal venoconstriction, expressed as the percentage reduction in vein diameter from maximal dilation

ED₅₀

The dose of agonist that produced 50% of maximum venoconstriction (E_max) for each subject

AR

adrenergic receptors

LVDT

linear variable differential transformer

IQR

interquartile range

fmol/min

femtomole/minute

BMI

body mass index