Mechanisms of sympathetic restraint in human skeletal muscle during exercise: role of α-adrenergic and nonadrenergic mechanisms

Abstract

Sympathetic vasoconstriction is mediated by α-adrenergic receptors under resting conditions. During exercise, increased sympathetic nerve activity (SNA) is directed to inactive and active skeletal muscle; however, it is unclear what mechanism(s) are responsible for vasoconstriction during large muscle mass exercise in humans. The aim of this study was to determine the contribution of α-adrenergic receptors to sympathetic restraint of inactive skeletal muscle and active skeletal muscle during cycle exercise in healthy humans. In ten male participants (18–35 yr), mean arterial pressure (intra-arterial catheter) and forearm vascular resistance (FVR) and conductance (FVC) were assessed during cycle exercise (60% total peak workload) alone and during combined cycle exercise + handgrip exercise (HGE) before and after intra-arterial blockade of α- and β-adrenoreceptors via phentolamine and propranolol, respectively. Cycle exercise caused vasoconstriction in the inactive forearm that was attenuated ~80% with adrenoreceptor blockade (%ΔFVR, +81.7 ± 84.6 vs. +9.7 ± 30.7%; P = 0.05). When HGE was performed during cycle exercise, the vasodilatory response to HGE was restrained by ~40% (ΔFVC HGE, +139.3 ± 67.0 vs. cycle exercise: +81.9 ± 66.3 ml·min⁻¹·100 mmHg⁻¹; P = 0.03); however, the restraint of active skeletal muscle blood flow was not due to α-adrenergic signaling. These findings highlight that α-adrenergic receptors are the primary, but not the exclusive mechanism by which sympathetic vasoconstriction occurs in inactive and active skeletal muscle during exercise. Metabolic activity or higher sympathetic firing frequencies may alter the contribution of α-adrenergic receptors to sympathetic vasoconstriction. Finally, nonadrenergic vasoconstrictor mechanisms may be important for understanding the regulation of blood flow during exercise.

NEW & NOTEWORTHY Sympathetic restraint of vascular conductance to inactive skeletal muscle is critical to maintain blood pressure during moderate- to high-intensity whole body exercise. This investigation shows that cycle exercise-induced restraint of inactive skeletal muscle vascular conductance occurs primarily because of activation of α-adrenergic receptors. Furthermore, exercise-induced vasoconstriction restrains the subsequent vasodilatory response to hand-grip exercise; however, the restraint of active skeletal muscle vasodilation was in part due to nonadrenergic mechanisms. We conclude that α-adrenergic receptors are the primary but not exclusive mechanism by which sympathetic vasoconstriction restrains blood flow in humans during whole body exercise and that metabolic activity modulates the contribution of α-adrenergic receptors.

INTRODUCTION

Sympathetic nervous system activity is critical for proper blood pressure regulation during moderate to high-intensity exercise using a large muscle mass in humans. In addition to increases in cardiac output, sympathetic activation elicits vasoconstriction to maintain total peripheral resistance in the face of metabolic vasodilation in active skeletal muscle (35, 52). Previous investigations in humans during whole body exercise demonstrate significant restraint of blood flow to inactive (2, 44, 58) and active tissue beds (10). Restraint of blood flow to inactive tissues directs blood flow away from areas of lower metabolic activity and toward areas of higher metabolic demand (53). The vasculature within active skeletal muscle attenuates sympathetic vasoconstriction to preserve blood flow and oxygen delivery (functional sympatholysis), but it does not abolish the vasoconstrictor effects of muscle sympathetic nervous system activity (MSNA) (10, 37, 46). Therefore, sympathetic restraint of blood flow remains even in highly active skeletal muscle (6, 9, 28). The precise regulation of peripheral resistance in active and inactive tissues maintains arterial blood pressure during whole body exercise (34, 63) and enables the appropriate distribution of blood flow to and within contracting skeletal muscle (24).

Numerous investigations in humans on sympathetic restraint or sympathetic vasoconstrictor responsiveness during exercise have used smaller muscle mass exercise (e.g., knee extension exercise, handgrip exercise) to isolate local signaling mechanisms independent of changes in hemodynamics or MSNA (53). However, much less is known regarding the mechanisms of sympathetic vasoconstriction during large muscle mass exercise. In this context, it is established that sympathetic neural firing increases during moderate to high-intensity cycle exercise (52) and that the associated restraint of blood flow is neurogenic in nature (2). Under resting conditions, tonic sympathetic vasoconstrictor activity is mediated primarily (>90%) by norepinephrine binding α -adrenergic receptors (19). However, the increase in MSNA burst frequency and associated neural patterns observed during exercise are distinct from MSNA patterns observed during resting conditions (26, 29, 52). Evidence from animal models using electrical stimulation of sympathetic neurons shows that the contribution of norepinephrine, adenosine triphosphate, and neuropeptide Y varies based upon neural stimulation frequency, pattern, and arteriole order (45, 56). Therefore, the natural recruitment patterns of sympathetic neurons during exercise may alter the strength and mechanisms by which the sympathetic nervous system mediates vasoconstriction (19, 39). Additionally, alternative local vasoconstrictor mechanisms such as myogenic tone may also contribute to exercise-induced vascular restraint under conditions of increased perfusion pressure (1). While plasma norepinephrine concentration is often associated with vascular restraint during exercise in humans (44, 58, 62), the contribution of α-adrenergic receptors to vascular restraint of inactive and active tissues during large muscle mass exercise in humans is currently unclear.

Therefore, the first aim of this study was to establish the contribution of α-adrenergic receptors to vasoconstriction in inactive skeletal muscle during moderate intensity cycle exercise. The second aim was to determine the contribution of α-adrenergic receptors to the restraint of active skeletal muscle vasodilation during combined cycle and handgrip exercise. We hypothesized that 1) local intra-arterial α-adrenergic blockade will abolish vasoconstriction in inactive skeletal muscle during moderate intensity cycle exercise and 2) during moderate intensity cycle exercise, the vasodilatory response to hand-grip exercise (HGE) will be restrained due to increased α-adrenergic signaling.

METHODS AND MATERIALS

Ethical Approval

All experimental procedures were approved by the University of British Columbia Clinical Research Ethics Board (CREB ID: H17-02687) and conformed to the standards set in the latest revision of the Declaration of Helsinki (except for registration in a database). Verbal and written informed consent was obtained from all participants before testing. This investigation was part of the Global REACH expedition to Peru baseline testing that occurred during March 2018. Several participants volunteered for multiple studies conducted at the University of British Columbia (Kelowna, British Columbia; 344 m). However, the current investigation was addressed before or at least 24 h after participation in any other study that would alter hemodynamics.

Participants

Ten healthy young men (age, 27 ± 4 yr; weight, 76.1 ± 8.7 kg; height, 177.3 ± 5.8 cm) volunteered to participate. All participants were nonsmokers; free of any cardiovascular, respiratory, and cerebrovascular diseases; nondiabetic; and not taking any prescription drugs. Our investigation focused exclusively on male volunteers because female volunteers were prioritized for participation in a concurrent investigation focused exclusively on sex differences in altitude acclimatization.

Arterial Catheterization, Arterial Blood Pressure, Heart Rate, and Skin Conductance

Under local anesthesia (2% lidocaine), a 20-gauge, 7.6-cm catheter (Arrow, Markham, ON, Canada) was inserted under sterile conditions into the brachial artery of the nondominant arm for administration of pharmacological agents via a standard infusion syringe pump (PhD Ultra Syringe Pumps, Harvard Apparatus, Holliston, MA), blood sampling, and continuous measurement of mean arterial pressure (MAP). The arterial catheter was connected to an in-line, wasteless sampling port (VAMP; Edwards Lifesciences), and a pressure transducer (TruWavetransducer; Edwards Lifesciences) was placed at the level of the right atrium (22, 47). Heart rate (HR) was measured via a three-lead electrocardiogram (ADInstruments, Sydney, Australia). Cutaneous blood velocity and temperature were measured using a laser-Doppler monitor (moorVMS-LDF, Moor Instruments, Ltd., Devon, UK) throughout the study and kept cool using a fan directed toward the forearm to limit the contribution of skin blood flow to the estimation of skeletal muscle blood flow through the brachial artery (22, 47).

Forearm Blood Flow, Vascular Resistance, and Conductance

Mean blood velocity and brachial artery diameter were measured via a 12-MHz linear-array ultrasound probe (Vivid 7, General Electric, Milwaukee, WI) with mean blood velocity recorded into LabChart using a universal Doppler audio translator (uDAT, V3.0) (25). Brachial artery diameter was simultaneously screen captured (Epiphan and Camtasia) and analyzed off-line using automated wall tracking software (Brachial Analyzer software, Medical Imaging Applications, LLC, Coralville, IA). Forearm blood flow (FBF) was calculated as previously described (47). Forearm vascular resistance (FVR) was calculated as FVR = (MAP/FBF) (in mmHg·ml⁻¹·min⁻¹), and forearm vascular conductance (FVC) was calculated as FVC = (FBF/MAP) × 100 (in ml·min⁻¹·100 mmHg⁻¹) (40).

Muscle Sympathetic Nervous Activity

In 7 out of 10 participants (n = 7), acceptable recordings of multiunit MSNA were measured at rest and during moderate (60% total peak workload)-intensity dynamic leg cycling exercise as proof of concept that the exercise intensity and duration were sufficient to elicit elevated MSNA. In addition, in a subset of participants (n = 4), MSNA recordings were maintained during combined cycle and HGE. Neural activity was recorded from the radial nerve by an experienced microneurographer (GM) using standardized ultrasound-guided microneurographic techniques (14, 36). Nerve signals were amplified (100× preamplifier and variable gain isolated amplifier, Neuroamp Ex, ADInstruments, Sydney, Australia) band-pass filter (300–2,000 Hz), rectified, and integrated (time decay constant, 0.1 s) (LabChart Pro V8.3.1, ADInstruments, Sydney, Australia). Identification of multiunit burst of MSNA was quantified as burst frequency (in bursts/min) in accordance to standardized guidelines (65). Given the clear sympathetic response to exercise in control conditions, coupled with the high reproducibility of the exercise pressor response and the duration of the protocol (~4 h), MSNA recordings were discontinued after control conditions were completed to facilitate subject comfort for the remaining protocols (Fig. 1).

Fig. 1.

Schematic outline of the experimental protocol. Instrumentation infers the placement of the cardiovascular hemodynamic measurement devices, muscle sympathetic nervous activity (MSNA) electrodes along with brachial catheter of the participants. During this time, participants performed maximal voluntary contractions (MVC). Sympathetic activity measurements indicated by white arrows. Ultrasound measurements indicated by blue arrows. Peak WL, maximal cycle peak workload; HGE, handgrip exercise; Ex, exercise. Localized forearm infusion of sodium nitroprusside (SNP) indicated for flow match trial. Localized forearm infusion of phentolamine and propranolol indicated by α-β-blockade and maintained throughout the second portion of the protocol. Time is displayed in minutes. Subject sample size, n = 10.

Assessment of V̇o₂max and Isometric Maximal Voluntary Contraction

All participants performed a stepwise cardiopulmonary exercise test on an upright cycle ergometer (Velotron, Racermate; Maker of CompuTrainer and Velotron, Seattle, WA) until volitional exhaustion. After an individually selected warmup, the test began at 150 W and gradually increased by 25 W every minute. Five minutes after the initial test, a supramaximal test to volitional exhaustion was performed to confirm attainment of the maximal rate of oxygen consumption (V̇o₂). The total peak workload for recumbent cycle exercise was calculated using a previously established predictive equation, and subsequent exercise bouts were performed at 60% of the estimated recumbent peak workload (64). Additionally, 60% total peak workload was verified with the attainment of HR ≥ 60% of predicted maximal HR, and sympathetic response was confirmed when possible with direct MSNA recordings. Forearm maximal voluntary contractions (MVC) was calculated as the average of three maximal voluntary isometric contractions. Experimental HGE was performed using 15% of MVC for a total of 3 min with a duty cycle of 1-s contraction and 2-s relaxation (22) using a grip force transducer rated for use up to 800 N (MLT004/ST, ADInstruments, Sydney, Australia).

Local α- and β-Adrenergic Receptor Antagonist Infusion

All drug infusions were administered through the brachial artery catheter and normalized to forearm volume (FAV) to ensure the effects of pharmacological agents remained localized to the forearm vasculature. Forearm volume was measured using the partial Frustrum model that has been validated and previously described (13). Because both α₁- and α₂-adrenergic receptors may contribute to postjunctional vasoconstrictor responses, we used a nonselective α-adrenergic blockade to inhibit both receptor subtypes. This was accomplished via infusion of phentolamine mesylate (Sandoz, Inc., Princeton, NJ) at a dose of 12 μg·dl⁻¹·FAV⁻¹·min⁻¹ for 10 min (loading dose) and maintained at 5 μg·dl⁻¹·FAV⁻¹·min⁻¹ throughout experimental trials (47). Blockade of prejunctional α₂-adrenergic receptors can facilitate endogenous norepinephrine release from sympathetic nerves, which binds β-adrenergic receptors on endothelial cells, causing a nonspecific vasodilation (59). Therefore, the nonselective β-adrenergic antagonist propranolol hydrochloride (West-ward Pharmaceutical, Corp., Eatontown, NJ) was infused at a dose of 10 μg·dl⁻¹·FAV⁻¹·min⁻¹ over 10 min (loading dose) and maintained at 5 μg·dl⁻¹·FAV⁻¹·min⁻¹ throughout experimental trials (47, 59) to limit any nonspecific contribution of β-adrenergic vasodilation. The efficacy of these doses for inhibiting adrenergic vasoconstriction are very well documented (16, 47). While a nonselective α-adrenergic agonist was not available at the time of study, blockade efficacy was confirmed (n = 4) to be greater than ~95% effective by direct challenge via intra-arterial infusion of an α₁-adrenergic agonist phenylephrine at a dose of 0.250 μg·dl⁻¹·FAV⁻¹·min⁻¹ over 3 min and are supported by the abolishment of vasoconstrictor responses during cycle exercise.

Experimental Protocol

The experimental protocol is presented in Fig. 1. All participants arrived at the laboratory having abstained from exercise and alcohol for at least 24 h and caffeine for 12 h and having consumed a light meal 2 h before testing. Upon arrival, participants were positioned on a semirecumbent cycle ergometer (Lode Angio Imaging cycle ergometer, Quinton Instruments, Groningen, The Netherlands). After catheterization and instrumentation, participants completed a minimum of 30 min of rest to allow hemodynamics to stabilize in the semirecumbent position.

Experimental protocol 1.

The aim of protocol 1 was to determine the contribution of α-adrenergic receptors to restraint of inactive skeletal muscle blood flow during cycle exercise. Participants performed cycle exercise at 60% of peak workload for 3 min. This intensity and duration were chosen 1) to induce a robust increase in MSNA (36, 52) and vasoconstriction (42, 58) within 3 min and 2) to avoid a secondary rise in skin blood flow with prolonged exercise (42). The arms were abducted at 90° laterally and supported at the level of the heart, and FBF was measured for 3 min at rest and throughout cycle exercise to assess vasoconstriction in the inactive forearm.

Experimental protocol 2.

The aim of protocol 2 was to determine the mechanisms of restraint of active skeletal muscle blood flow during cycle exercise-induced vasoconstriction. After the initial 3 min of cycle exercise (protocol 1), subjects continued to cycle at 60% peak workload while adding rhythmic HGE at 15% MVC for 3 min. The vasodilatory response to handgrip exercise was measured during the entire 3 min of combined cycle exercise and HGE. Subjects were then allowed to recover for a minimum of 30 min. Once baseline hemodynamics were reestablished, HGE was performed alone to measure the vasodilatory response to forearm exercise during resting conditions (47). All experimental procedures were then repeated after intra-arterial blockade of forearm α- and β-adrenergic receptors.

High-flow control.

Blockade of α-adrenergic receptors causes a local vasodilation that may influence subsequent vascular responses because of a rise in baseline blood flow and conductance, independent of α-adrenergic signaling, per se. Therefore, in a subset of participants (n = 5) before α- and β-adrenergic blockade trials (Fig. 1), a nitric oxide (NO) donor (sodium nitroprusside; SNP; Hospira Pharmaceutical, Lake Forest, IL) was infused intra-arterially at a dose of 4.0 µg dl⁻¹ FAV⁻¹ min⁻¹ to induce a rise in resting forearm blood flow independent of changes in α-adrenergic signaling to serve as a “high-flow control” condition. Sodium nitroprusside was chosen as the control vasodilator because it has repeatedly been shown to have no independent effect on α-adrenergic vasoconstriction in forearm skeletal muscle of humans (22, 50, 61). Importantly, the effects of SNP infusion are localized to the forearm, have rapid onset and offset kinetics, and had no impact on systemic hemodynamics or subsequent exercise pressor responses.

Data Acquisition and Analyses

All cardiovascular variables were sampled at 1 kHz via an analog-to-digital converter (Powerlab 16/30, ADInstruments, Sydney, Australia), displayed on LabChart (LabChart Pro v8.3.1, ADInstruments, Sydney, Australia) and analyzed off-line.

Experimental protocol 1.

The vasoconstrictor response to cycle exercise-induced vasoconstriction in the inactive forearm was quantified as changes in both FVC and FVR. The change in FVR was the primary variable of interest because MAP was elevated during cycle exercise compared with rest. Absolute and relative change in FVR were calculated to account for baseline differences in FVR after α-β-adrenergic blockade. Therefore, the increase in FVR during cycle exercise was calculated as follows:

$$\text{ΔFVR}=\left({\text{FVR}}_{\text{60\%\hspace{0.17em}Ex}}–{\text{FVR}}_{\text{baseline}}\right)$$

$$\text{\%ΔFVR}=\left({\text{FVR}}_{\text{60\%\hspace{0.17em}Ex}}–{\text{FVR}}_{\text{baseline}}\right)/\left({\text{FVR}}_{\text{baseline}}\right)·\text{100}\%$$

Similar calculations were made for FBF and FVC.

Experimental protocol 2.

Local forearm vasodilation is the primary variable changing during hand-grip exercise with a minor impact on systemic MAP (30); therefore, during protocol 2, forearm vasodilation is most reliably quantified as a change in conductance (6, 33, 40). The vasodilatory response to HGE was quantified as the change in FVC from rest during control conditions (i.e., low SNA), and during cycle exercise (i.e., high MSNA). The effect of sympathetic restraint on the vasodilatory response to HGE during cycle was quantified in absolute and relative terms as follows:

$$\text{ΔFVC}=\left({\text{ΔFVC}}_{\text{60\%\hspace{0.17em}Ex+15\%\hspace{0.17em}HGE}}–{\text{ΔFVC}}_{\text{15\%\hspace{0.17em}HGE}}\right)$$

$$\text{\%ΔFVC}=\left({\text{ΔFVC}}_{\text{60\%\hspace{0.17em}Ex+15\%\hspace{0.17em}HGE}}–{\text{ΔFVC}}_{\text{15\%\hspace{0.17em}HGE}}\right)/\left({\text{ΔFVC}}_{\text{15\%\hspace{0.17em}HGE}}\right)·\text{100}$$

Similar calculations were made for FBF and FVR.

All statistical analyses were planned a priori. Hemodynamic data were analyzed using a two-way repeated measure analysis of variance (ANOVA). Confidence intervals were adjusted using the Bonferroni correction method. Relative and absolute changes in FVR and FVC were analyzed using two-tailed matched-paired t-tests (control vs. α-β-blockade condition; α-β-blockade vs. SNP). Differences between MSNA burst frequencies were analyzed using two-tailed matched-paired t-tests (rest vs. cycle exercise; cycle exercise vs. cycle combine with HGE). All statistical analysis was completed using IBM SPSS (version 24, IBM statistics, Armonk, NY) and Prism GraphPad (version 8, GraphPad Software, San Diego, CA) and are reported as means ± SD unless stated otherwise. Statistical significance was defined as P < 0.05.

RESULTS

Subject Characteristics and Systemic and Forearm Hemodynamics

Subject characteristics are presented in Table 1. Systemic and forearm hemodynamics are presented in Tables 2 and 3. Baseline hemodynamics were well maintained throughout protocol 1, with small variations in HR (P = 0.04) and no difference in MAP between control, α-β-blockade, and SNP conditions (Table 2). Heart rate and MAP increased similarly from rest to 60% exercise in all conditions (all, P < 0.01 vs. rest) supporting the repeatability of the exercise intervention. As anticipated, local α-β-blockade increased inactive FBF and FVC with no impact on systemic hemodynamics (Table 2). During protocol 2, there were minor increases in HR and MAP in response to HGE that were similar throughout all trials (Table 3). Cycle exercise increased baseline HR, MAP, and FVR and decreased FVC and FBF compared with resting baseline conditions (all, P < 0.05 vs. rest baseline; Table 3). Phentolamine abolished the vasoconstrictor response to phenylephrine (ΔFVC, control, −25.9 ± 15.4; α-β-blockade, +6.5 ± 29.7 ml·min⁻¹·100·mmHg⁻¹; P = 0.23).

Table 1.

Subject characteristic

Characteristics
Subjects, n	10
Age, yr	27 ± 4
Height, cm	177.3 ± 5.8
Weight, kg	76.1 ± 8.7
V̇o2, ml·min−1·kg−1	45.9 ± 7.2
V̇o2, ml/min	3528.1 ± 623.0
Total peak workload, W	333 ± 48
60% total peak workload, W	200 ± 29
MVC, kg	50 ± 10
15% MVC, kg	8 ± 2

Values are means ± SD. V̇o₂, rate of maximal oxygen uptake; MVC, maximal voluntary contraction.

Table 2.

Hemodynamics at rest and during 60% exercise between control and α-β-blockade

	Control		α-β-Blockade		SNP
	Baseline	60% Exercise	Baseline	60% Exercise	Baseline	60% Exercise
n	10		10		5
HR, beats/min	62 ± 10	143 ± 9*	56 ± 10†	138 ± 8*	67 ± 7‡	136 ± 28*
MAP, mmHg	93 ± 8	121 ± 12*	96 ± 9	121 ± 9*	95 ± 17	121 ± 11*
MBV, cm/s	5.5 ± 1.3	4.9 ± 2.3	16.2 ± 3.6†	19.9 ± 6.3*†	22.3 ± 6.1‡	10.5 ± 5.5*‡
FBF, ml/min	44.3 ± 10.9	39.7 ± 14.6	129.1 ± 36.1†	161.0 ± 65.0†	178.1 ± 63.3	74.4 ± 37.2*‡
FVC, ml·min−1·100 mmHg−1	48.3 ± 13.3	30.3 ± 12.2*	133.0 ± 38.7†	126.2 ± 55.5†	188.8 ± 68.5	62.6 ± 33.5*‡
FVR, mmHg·ml−1·min−1	2.2 ± 0.6	4.0 ± 2.0*	0.8 ± 0.2†	0.9 ± 0.4†	0.6 ± 0.2	2.0 ± 0.9*‡
Skin conductance, AU	21.4 ± 6.9	51.8 ± 19.3*	41.2 ± 13.5†	76.7 ± 25.1*†	46.1 ± 11.5	80.2 ± 22.0*
MSNA, burst/min	13.6 ± 2.6	28.6 ± 5.1*	—	—	—	—

Values are means ± SD. HR, heart rate; MAP, mean arterial pressure; MBV, mean blood velocity; FBF, forearm blood flow; FVC, forearm vascular conductance; FVR, forearm vascular resistance; AU, arbitrary units; MSNA, muscle sympathetic nervous activity.

^*P < 0.05, rest vs. 60% exercise.

^†P < 0.05, control vs. +α-β-blockade.

^‡P < 0.05, α-β-blockade vs. sodium nitroprusside (SNP).

Table 3.

Hemodynamics during active muscle perfusion between control and α-β-blockade

	Rest + 15% HGE			60% Cycle Exercise + 15% HGE
	Rest baseline	HGE	∆	Cycle baseline	HGE	∆
HR, beats/min
Control	61 ± 10	60 ± 11	−1 ± 5	143 ± 9‡	148 ± 10*‡	6 ± 3‡
α-β-Blockade	54 ± 9†	57 ± 9*	3 ± 3†	138 ± 8‡	144 ± 6*‡	6 ± 3
MAP, mmHg
Control	90 ± 6	91 ± 6	1 ± 2	121 ± 12‡	123 ± 12*‡	1 ± 4
α-β-Blockade	92 ± 10	95 ± 8†	3 ± 5	121 ± 9‡	119 ± 8*‡	−1 ± 7
MBV, cm/s
Control	5.9 ± 1.1	22.0 ± 6.1*	16.2 ± 6.1	4.9 ± 2.3	19.1 ± 10.4*	13.6 ± 10.1
α-β-Blockade	16.6 ± 3.4†	35.1 ± 6.9*†	18.5 ± 6.7	19.9 ± 6.3†	32.9 ± 10.3*†	16.7 ± 10.7
FBF, ml/min
Control	45.5 ± 10.7	173.6 ± 64.9*	128.1 ± 62.8	36.7 ± 14.6	134.7 ± 79.5*	98.0 ± 75.4
α-β-Blockade	124.4 ± 26.6†	283.5 ± 90.7*†	159.1 ± 77.0†	161.0 ± 65.0†	288.2 ± 100.9*†	127.2 ± 72.1
FVC, ml·min−1·100 mmHg−1
Control	50.2 ± 9.9	189.4 ± 68.2*	139.3 ± 67.0	30.3 ± 12.2‡	112.2 ± 70.3*‡	81.9 ± 66.3‡
α-β-Blockade	136.5 ± 35.0†	300.2 ± 96.6*†	163.7 ± 82.3	126.2 ± 55.5†	232.4 ± 85.7*†	106.2 ± 55.9
FVR, mmHg·ml−1·min−1
Control	2.1 ± 0.4	0.6 ± 0.2*	−1.5 ± 0.4	4.0 ± 2.0‡	1.2 ± 0.7*‡	−1.0 ± 0.9
α-β-Blockade	0.8 ± 0.2†	0.4 ± 0.1*†	−0.4 ± 0.1†	0.9 ± 0.4†	0.5 ± 0.2*†	−0.3 ± 0.2†
Skin Conductance, AU
Control	21.4 ± 6.9	53.6 ± 27.1*	32.1 ± 27.7	51.8 ± 19.3‡	98.0 ± 33.4*‡	46.2 ± 23.2‡
α-β-Blockade	39.6 ± 13.6†	77.3 ± 22.6*†	37.7 ± 17.3	76.7 ± 25.1‡	107.4 ± 26.9*	30.7 ± 16.9†

Values are means ± SD. HR, heart rate; MAP, mean arterial pressure; MBV, mean blood velocity; FBF, forearm blood flow; FVC, forearm vascular conductance; FVR, forearm vascular resistance; AU, arbitrary units.

^*P < 0.05, rest and cycle baseline vs. handgrip exercise (HGE).

^†P < 0.05, control vs. α-β-blockade.

^‡P < 0.05, 15% HGE vs. 60% exercise + 15% HGE.

Protocol 1: Mechanisms of Exercise-Induced Vasoconstriction in Inactive Skeletal Muscle

Systemic and forearm hemodynamics for each time point are presented in Table 2. Representative blood velocity tracings from each trial are presented in Fig. 2, A–C, absolute changes in FBF, FVR, and FVC are presented in Fig. 2, D–F, and relative change in FBF, FVR, and FVC are presented in Fig. 2, G–I. Three minutes of cycle exercise significantly increased MSNA burst frequency, confirming elevated sympathetic vasoconstriction (baseline, 13.6 ± 2.6 vs. 60% exercise; 28.6 ± 5.1 bursts/min; P < 0.01; Table 2). The rise in MSNA during cycle exercise was associated with a maintenance of FBF (baseline, 44.2 ± 10.9 vs. 60% exercise; 36.7 ± 14.6 ml/min; P = 0.08; Fig. 2, A and D) despite a significant rise in MAP (ΔMAP: +29 ± 7 mmHg; P < 0.01; Table 2). The observed maintenance of FBF in the face of a large change in perfusion pressure was due to an increase in local forearm vascular tone (ΔFVR: +81.7 ± 84.6%; P = 0.02; Fig. 2, H and I). As anticipated, forearm α-β-adrenergic blockade increased inactive MBV (P < 0.01; Fig. 2B), FBF (P < 0.01) and FVC (P < 0.01) but had no impact on the HR response (ΔHR, control: +81 ± 10; α-β-blockade, +82 ± 11 beats/min; P = 0.83) or MAP response (ΔMAP, control, +29 ± 7; α-β-blockade, +25 ± 6 mmHg; P = 0.54) to cycle exercise (Table 2). Local α-β-adrenergic blockade attenuated the rise in forearm vascular tone during cycle exercise by ~80% (ΔFVR, control, +81.7 ± 84.7% vs. α-β-blockade, +9.7 ± 30.7%; P = 0.05; Fig. 2, H and I). Lastly, cycle exercise increased skin conductance similarly in all conditions (Δskin conductance, control, +30.4 ± 17.2 arbitrary units (AU); α-β-blockade, +35.5 ± 15.7 AU, SNP: +32.9 ± 14.5 AU; P = 0.77; Table 2).

Fig. 2.

Sympathetic restraint during exercise-induced vasoconstriction. A–C: beat-by-beat forearm antegrade and retrograde velocity at rest and during 60% exercise (Ex) during control (A), α-β-blockade (B), and flow match trials (C). D–I: individual and mean absolute change within control (orange bars), α-β-blockade (blue bars), and sodium nitroprusside (SNP; gray bars) in forearm blood flow (FBF; D), forearm vascular resistance (FVR; E), and forearm vascular conductance (FVC; F). Individual and mean percent change within control, α-β-blockade, and SNP in FBF (G), FVR (H), and FVC (I). Subject sample size, n = 10.

High-Flow Control

Because α-β-adrenergic blockade causes vasodilation and an attendant increase in resting FBF, the attenuation of exercise-induced vasoconstriction by α-β-adrenergic blockade could be due to nonspecific effects of baseline vasodilation. Therefore, in a subset of five subjects, we repeated protocol 1 using an infusion of sodium nitroprusside (nitric oxide donor) to cause a nonspecific vasodilation without altering α-β-adrenergic signaling. Infusion of SNP into a resting forearm significantly increased forearm MBV (Fig. 2C), FBF (P < 0.01), and FVC (P = 0.01). In contrast to α-β-adrenergic blockade, the forearm vasoconstrictor response during cycle exercise was strikingly preserved, causing a large decrease in FBF (FBF, baseline, 178.1 ± 63.3 vs. 60% exercise, 74.4 ± 37.3 ml min⁻¹; P = 0.03; Table 1) due to an increase in local forearm vascular tone (ΔFVR, +1.4 ± 0.9 mmHg·ml⁻¹·min⁻¹; ΔFVC, –126.2 ± 58.9 ml·min⁻¹·100 mmHg⁻¹; both, P < 0.05; Fig. 2, E and F). During SNP infusion, the relative change in FVR and FVC during cycle exercise was similar or greater than control conditions (%ΔFVR, P = 0.02; %ΔFVC, P = 0.02) and was significantly greater than α-β-adrenergic blockade conditions (%ΔFVR, SNP, +248.7 ± 152.9; α-β-blockade, +9.7 ± 30.7%; P < 0.01; and ΔFVC, SNP, –65.8 ± 16.5; α-β-blockade, −6.1 ± 30.6%; P < 0.01; Fig. 2, H and I). These results confirm that nonspecific increases in baseline vascular tone do not attenuate sympathetic vasoconstriction and that α-adrenergic signaling is the primary mediator of exercise-induced sympathetic vasoconstriction in inactive skeletal muscle.

Experimental Protocol 2: Mechanism of Exercise-Induced Restraint of Active Skeletal Muscle Vasodilation.

Systemic and forearm hemodynamics for each time point are presented in Table 3. MSNA remained elevated throughout the combined cycle and HGE exercise condition, confirming sustained sympathetic vasoconstriction (MSNA, baseline, 13.6 ± 2.6 burst/min; 60% exercise, 28.6 ± 5.1 burst/min; 60% exercise + 15% HGE, 38.5 ± 8.1 burst/min; P = 0.03). HGE alone (control) increased FBF (ΔFBF, +128.1 ± 62.8 ml/min; P < 0.01) and FVC (ΔFVC, +139.3 ± 67.0 ml·min⁻¹·100 mmHg⁻¹; P < 0.01; Table 3, Fig. 3A). When HGE was performed during cycle-induced vasoconstriction, the vasodilatory response to HGE was reduced by ~40% (Fig. 3F), demonstrating significant restraint of metabolic vasodilation (ΔFVC, rest + HGE, +139.3.4 ± 67.0 ml·min⁻¹·100 mmHg⁻¹; 60% Ex + HGE, +81.9 ± 66.3 ml·min⁻¹·100 mmHg⁻¹; P = 0.01; Fig. 3, B, D, and F). Despite blockade of α-adrenergic vasoconstrictor signaling, cycle exercise-induced restraint of the vasodilatory response to HGE was unaltered (%ΔFVC, control, −50.6 ± 44.4%; α-β-blockade, −27.8 ± 53.2%; P = 0.15; Fig. 3F). There was a modest increase in skin conductance during HGE that was similar in all conditions (Δskin conductance, control, HGE, +32.1 ± 27.7 AU; cycle exercise + HGE, +46.2 ± 23.2 AU; α-β-blockade, HGE, +37.7 ± 17.3 AU; cycle exercise + HGE, +30.7 ± 16.9 AU; P = 0.14).

Fig. 3.

Active muscle perfusion during exercise-induced vasoconstriction. A and B: individual and mean absolute change comparing 15% handgrip exercise (HGE) alone vs. 60% exercise (Ex) + 15% HGE during control (orange bars) and α-β-blockade (blue bars) in forearm blood flow (FBF; A) and forearm vascular conductance (FVC; B). C and D: individual and mean absolute change from 15% HGE alone to 60% + 15% HGE comparing control vs. α-β-blockade in FBF (C) and FVC (D). E and F: individual and mean percent change from 15% HGE alone to 60% + 15% HGE comparing control vs. α-β-blockade in FBF (E) and FVC (F). Subject sample size, n = 10.

DISCUSSION

This study explored the contribution of α-adrenergic signaling to cycle exercise-induced restraint of inactive and active skeletal muscle in healthy adult males. The primary novel findings of this study are 1) that restraint of inactive forearm skeletal muscle during moderate intensity cycle exercise is mediated primarily (~80%) by α-adrenergic receptors, and 2) that vasoconstriction induced by cycle exercise restrains the vasodilatory response to active forearm skeletal muscle, however, 3) the restraint of active skeletal muscle vasodilation was not fully explained by α-adrenergic signaling. Together, these findings show that α-adrenergic receptors are the primary, but not exclusive mechanisms of sympathetic restraint in response to physiological increases in MSNA. Furthermore, skeletal muscle metabolic activity influences the contribution of α-adrenergic receptors to restraint of muscle blood flow and vascular conductance in humans.

α-Adrenergic Restraint of Inactive Skeletal Muscle

In humans, the sympathetic nervous system exerts a tonic vasoconstrictor tone on peripheral vascular beds that is predominantly mediated by α-adrenergic receptors (17, 18) to maintain total peripheral resistance, and MAP. Recently, Fairfax and colleagues (19) demonstrated that α-adrenergic signaling accounts for >90% of the beat-by-beat vasoconstrictor response to bursts of MSNA during resting conditions in humans. However, the contribution of α-adrenergic receptors to sympathetic restraint of inactive skeletal muscle during whole body exercise in humans is not clearly defined. In humans, there is an increase in MSNA within 3 min of initiating moderate to high-intensity cycle exercise (52) that is associated with vasoconstriction in inactive skeletal muscle and viscera (2, 44, 58) and is critical for the proper maintenance of MAP in the face of metabolic vasodilation (11, 34). Blair and colleagues found that bretylium tosylate, an antagonist of sympathetic neural activity, abolished the forearm vasoconstrictor response to lower body exercise. Bretylium tosylate is a nonspecific inhibitor of sympathetic neurotransmitter release via alteration of sympathetic neural action potentials and refractory period (3) and does not distinguish between postjunctional adrenergic and nonadrenergic mechanisms. However, this initial investigation clearly demonstrates the neurogenic nature of exercise-induced vasoconstriction (2).

In this investigation, participants exercised at 60% of their peak workload, which was confirmed to elicit a robust increase in MSNA (Table 2). The rise in MSNA increased vascular resistance, an effect that was attenuated by ~80% after local inhibition of α-adrenergic receptors (Fig. 2, H and I). Importantly, the vasoconstrictor response to exercise was unaltered by intra-arterial infusion of sodium nitroprusside (nitric oxide donor), which served as a high-flow control condition to match the hyperemia observed after local blockade of α-adrenergic receptors. Despite the significant increase in sodium nitroprusside-mediated baseline blood flow, vasoconstriction during exercise was remarkably preserved and nearly identical to control conditions (Fig. 2, H and I). These findings are in line with previous investigations showing no impact of exogenous nitric oxide signaling, or the associated increase in shear stress, on sympathetic or adrenergic vasoconstriction in humans (22, 49, 61) as opposed to other endothelium dependent vasodilators such as adenosine triphosphate (23, 30, 51). Our results confirm that postjunctional α-adrenergic receptor signaling is the primary mechanism of sympathetic vasoconstriction in inactive skeletal muscle during physiologically induced increase in MSNA in humans.

α-Adrenergic Restraint of Active Skeletal Muscle

During exercise, MSNA is directed toward inactive and active contracting skeletal muscles (54, 58). Seminal investigations from Remensnyder and colleagues (46) demonstrate that sympathetic vasoconstrictor activity is significantly attenuated during muscle contractions but is not abolished. Residual sympathetic restraint of contracting skeletal muscle remains even during high-intensity exercise and is critical for the proper maintenance of blood pressure (11, 34). Investigations in dogs during dynamic treadmill exercise show that intra-arterial infusion of an α-adrenergic antagonist further increases blood flow, indicative of residual α-adrenergic restraint of contracting skeletal muscle even during high-intensity exercise (5, 9). While these investigations identify α-adrenergic restraint of contracting skeletal muscle, it is unclear if other sympathetic vasoconstrictor mechanisms independent of α-adrenergic signaling restrain active skeletal muscle blood flow during whole body exercise in humans. Investigations superimposing arm exercise onto leg exercise (55) and vice versa (12, 57, 62) have shown significant reductions in blood flow and vascular conductance to active skeletal muscle, presumed to be due to sympathetic vasoconstrictor activity, though this has not been previously confirmed experimentally in humans. In the current investigation, blockade of tonic α-adrenergic vasoconstrictor signaling increased the vasodilatory response to HGE even during resting conditions (P < 0.01, Table 3). These findings support previous investigations indicating that tonic α-adrenergic vasoconstrictor signaling at rest restrains metabolic vasodilation in healthy adults independent of changes in MSNA (47). When HGE was performed during cycle exercise-induced sympathoexcitation, the vasodilatory response was restrained by ~40% compared with HGE performed during resting conditions (Fig. 3B). Surprisingly, inhibition of α-adrenergic vasoconstrictor signaling had little impact on the cycle exercise-induced restraint of metabolic vasodilation (Fig. 3D). The lack of consistent effect of α-adrenergic blockade in contracting skeletal muscle suggests that alternative signaling mechanisms can restrain active skeletal muscle blood flow during conditions of heightened MSNA.

The apparent lack of α-adrenergic restraint of exercising FVC may be related to conditions unique to this experimental approach. First, 15% MVC handgrip exercise has been previously shown to attenuate α-adrenergic vasoconstrictor signaling (22, 31, 61). As such, the contribution of α-adrenergic receptors to sympathetic restraint may be minimized by the HGE intensity. Investigations using graded intensities of HGE could shed light on this issue. Furthermore, the increase in MSNA may have been different between control and α-β-blockade conditions; however, this is unlikely as the HR and pressor response to all conditions were highly repeatable (Table 2 and 3). Beyond these considerations, we still hypothesized that some of the exercise-induced restraint of metabolic vasodilation would be attributable to α-adrenergic receptors based upon investigations in animal models demonstrating intact α-adrenergic signaling during high-intensity dynamic exercise (9). A primary difference between the present investigation and previous investigations of Buckwalter et al. (9) is the timing of α-adrenergic blockade. Buckwalter and colleagues initiate adrenergic blockade after exercise-induced vasoconstriction was established, facilitating assessment of the combined effect of tonic (baseline) and exercise-induced α-adrenergic signaling on skeletal muscle blood flow during dynamic exercise. In the current investigation, baseline α-adrenergic signaling was removed before initiation of exercise to assess the contribution of α-adrenergic receptors to exercise-induced vasoconstriction. While this approach makes it possible to isolate the exercise-induced vasoconstriction, alternative pathways may compensate for the loss α-adrenergic signaling. From both investigations it is clear that 1) α-adrenergic signaling can restrain skeletal muscle blood flow at rest and during exercise, 2) the contribution of α-adrenergic signaling to vascular restraint is modulated by metabolic vasodilatory signaling (functional sympatholysis), and 3) alternative vasoconstrictor mechanisms may contribute importantly to reflex regulation of skeletal muscle blood flow in humans.

The contribution of nonadrenergic mechanisms to sympathetic vasoconstriction is likely due to vasoconstrictor actions of alternative neurotransmitters including adenosine triphosphate and neuropeptide Y. Furthermore, adenosine triphosphate and neuropeptide Y are both potent vasoconstrictors that are coreleased with norepinephrine from sympathetic nerve terminals and contribute to sympathetic vasoconstrictor responses (45, 56). Release of neuropeptide Y from sympathetic nerve terminals is facilitated by increased neuronal firing frequency (45), and investigations in canine models indicate that neuropeptide Y-mediated vasoconstriction remains intact in contracting skeletal muscle independent of α-adrenergic receptor activity (7, 8). However, to date, the overall contribution of adenosine triphosphate and neuropeptide Y to whole body exercise pressor responses in humans remains unclear. Finally, other nonsympathetic vasoconstrictor mechanisms may be responsible for the observed restraint of contracting skeletal muscle. Along these lines, the increase in baseline blood pressure during cycle exercise may activate myogenic vasoconstrictor mechanisms (1, 41), oscillatory shear patterns may induce local vasoconstrictor mechanisms such as endothelin-1 (60), and local angiotensin signaling may all contribute (4).

Considerations

The current investigation assessed vasoconstrictor responses in the forearm as opposed to the leg. This approach was chosen to avoid potential systemic influences of pharmacological agents when administered to larger muscle beds. While previous research has noted differences in vasodilatory and vasoconstrictor signaling between the forearm and leg (27, 38, 43), these discrepancies are often modest and differ primarily in magnitude, while the underlying signaling mechanisms are generally well conserved between vascular beds (30, 51). We acknowledge that the arm likely plays a lesser role in blood pressure regulation due to its size; nevertheless, restraint of upper limb blood flow is still an important part of maintaining appropriate blood pressure during whole body exercise (11).

Additionally, the findings of this investigation are limited to healthy men. Future investigations will be needed to determine the contribution of α-adrenergic receptors to exercise-induced vasoconstriction in women. In women, the role of β-adrenergic receptors should be considered independently due to significant interactions between estrogen and β-adrenergic receptor signaling (20, 32, 48). Greater β-adrenergic vasodilation has been shown to buffer sympathetic vasoconstrictor signaling in women, and previous investigations have shown reduced postjunctional α-adrenergic vasoconstrictor responsiveness in women compared with men (21, 32). As such, local β-adrenergic signaling may be an important independent regulator of sympathetic transduction during exercise in women.

Finally, while every effort was made to limit cutaneous blood flow, there was still a modest but consistent increase in cutaneous vascular conductance observed during cycle exercise (protocol 1) and HGE (protocol 2). In each case, caution is warranted when interpreting cutaneous flux during conditions that alter blood pressure and presumably vascular tone, the latter of which cannot be ascertained by laser Doppler. However, it is important to acknowledge that any increase in cutaneous blood flow will overestimate FBF measured at the brachial artery (42) and will underestimate the magnitude of vasoconstriction observed. Importantly, skin conductance increased similarly throughout all conditions and is not likely to impact the conclusions of this investigation as they relate to the mechanisms of vasoconstriction during exercise. Lastly, caution is warranted when extrapolating these findings to longer duration exercise where a significant rise in core temperature and skin blood flow will alter hemodynamics considerably (42). Further investigation is warranted to investigate how the mechanisms to vascular restraint change during longer duration exercise (15).

Conclusion

Recent investigations indicate that in resting humans, sympathetic vasoconstriction is predominantly mediated by α-adrenergic mechanisms (19). This investigation has shown that during large muscle dynamic exercise (cycle exercise), α-adrenergic receptors are the primary mechanism by which sympathetic vasoconstriction occurs in inactive skeletal muscle. Furthermore, metabolic activity of skeletal muscle modulates the contribution of α-adrenergic receptors to vascular restraint during large muscle mass exercise, such that α-adrenergic signaling restrains active skeletal muscle blood flow, but the contribution is lessened by increased metabolic activity. As such, the relative contribution of alternative vasoconstrictor mechanisms may be important for understanding the exercise pressor reflex.

GRANTS

This work was supported by the Department of Sport Science, University of Innsbruck (to A.B.H., F.H., S.L.R., and J.S.L.), The Wilderness Medical Society research in-training grant (to G.M.), and by National Heart, Lung, and Blood Institute Grant 1F32-HL1-37285-0 (to C.M.H.). Some of the work contained in this study was supported by a Canada Research Chair (P.N.A.) and the Natural Sciences and Engineering Research Council of Canada (to M.M.T. and P.N.A.).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

G.M., S.A.R., J.S.L., and C.M.H.J. conceived and designed research; A.B.H., G.M., S.A.R., C.G., M.M.T., P.N.A., F.H., S.L.R., J.S.L., and C.M.H.J. performed experiments; A.B.H. and C.M.H.J. analyzed data; A.B.H., J.S.L., and C.M.H.J. interpreted results of experiments; A.B.H. and C.M.H.J. prepared figures; A.B.H. and C.M.H.J. drafted manuscript; A.B.H., G.M., S.A.R., C.G., M.M.T., P.N.A., F.H., S.L.R., J.S.L., and C.M.H.J. edited and revised manuscript; A.B.H., G.M., S.A.R., C.G., M.M.T., P.N.A., F.H., S.L.R., J.S.L., and C.M.H.J. approved final version of manuscript.