Short-term exercise training enhances functional sympatholysis through a nitric oxide-dependent mechanism

Authors


D. S. DeLorey: Faculty of Physical Education and Recreation, University of Alberta, E-435 Van Vliet Centre, Edmonton, Alberta, Canada T6G 2H9. Email: darren.delorey@ualberta.ca

Key points

  • Sympathetic nervous system activity causes tonic vasoconstriction in resting and contracting skeletal muscle. Vasoactive molecules released from the active skeletal muscle and/or endothelium have been shown to inhibit sympathetic vasoconstriction, a phenomenon defined as functional sympatholysis.

  • A definitive mechanism responsible for functional sympatholysis has yet to be identified; however, nitric oxide (NO) appears to be involved.

  • It is unknown whether exercise training alters the inhibition of sympathetic vasoconstriction and NO-mediated sympatholysis in resting and contracting skeletal muscle.

  • The present findings demonstrate that short-term exercise training augments functional sympatholysis in a training-intensity-dependent manner through a NO-dependent mechanism. These novel findings advance our understanding of the effects of exercise training on the regulation of sympathetic vasoconstriction in resting and contracting skeletal muscle.

Abstract  We tested the hypothesis that short-term mild- (M) and heavy-intensity (H) exercise training would enhance sympatholysis through a nitric oxide (NO)-dependent mechanism. Sprague–Dawley rats (n = 36) were randomly assigned to sedentary (S) or to M (20 m min−1 5% gradient) or H exercise training groups (40 m min−1 5% gradient). Rats assigned to M and H groups trained on 5 days week−1 for 4 weeks, with the volume of training being matched between groups. Rats were anaesthetized and instrumented for stimulation of the lumbar sympathetic chain and the measurement of arterial blood pressure and femoral artery blood flow. The triceps surae muscle group was stimulated to contract rhythmically at 30 and 60% of maximal contractile force (MCF). The percentage change of femoral vascular conductance (%FVC) in response to sympathetic stimulation delivered at 2 and 5 Hz was determined at rest and during contraction at 30 and 60% MCF. The vascular response to sympathetic stimulation was reduced as a function of MCF in all rats (P < 0.05). At 30% MCF, the magnitude of sympatholysis (%FVC rest – contraction; Δ%FVC) was greater in H compared with M and S groups (Δ%FVC at 2 Hz, S, 9 ± 5; M, 11 ± 8; and H, 18 ± 7; and Δ%FVC at 5 Hz, S, 6 ± 6; M, 12 ± 9; and H, 18 ± 7; P < 0.05) and was greater in H and M compared with S at 60% MCF (Δ%FVC at 2 Hz, S, 15 ± 5; M, 25 ± 3; and H, 36 ± 6; and Δ%FVC at 5 Hz, S, 22 ± 6; M, 33 ± 9; and H, 39 ± 9; P < 0.05). Blockade of NO synthase did not alter the magnitude of sympatholysis in S during contraction at 30 or 60% MCF. In contrast, NO synthase inhibition diminished sympatholysis in H at 30% MCF and in M and H at 60% MCF (P < 0.05). The present findings indicate that short-term exercise training augments sympatholysis in a training-intensity-dependent manner and through an NO-dependent mechanism.

Abbreviations 
eNOS

endothelial nitric oxide synthase

ET

exercise training

FBF

femoral artery blood flow

FVC

femoral vascular conductance

H

heavy-intensity trained group

HR

heart rate

l-NAME

N ω-nitro-l-arginine methyl ester hydrochloride

M

mild-intensity trained group

MAP

mean arterial pressure

MCF

maximal contractile force

MT

motor threshold

nNOS

neuronal nitric oxide synthase

NO

nitric oxide

NOS

nitric oxide synthase

S

sedentary time control group

Introduction

The transition from rest to exercise is characterized by an increase in arterial pressure and a redistribution of cardiac output away from inactive tissue towards exercising skeletal muscle (Rowell, 1993). Skeletal muscle blood vessels dilate to facilitate the matching of local O2 delivery to local O2 demand (Saltin et al. 1998). The robust local vasodilatation is balanced by a concomitant increase in sympathetic nerve activity, which tonically constricts blood vessels in non-active tissue and exercising skeletal muscle in order to maintain arterial blood pressure (Rowell, 1993). Indeed, several studies have demonstrated that the sympathetic nervous system tonically constricts the vasculature of exercising muscle even during intense exercise (DiCarlo et al. 1996; Buckwalter et al. 1997; O’Leary et al. 1997). Despite the presence of tonic vasoconstriction in active skeletal muscle, it is well established that a number of substances released from the active skeletal muscle (McGillivray-Anderson & Faber, 1990, 1991; Tateishi & Faber, 1995b; Thomas et al. 1998; Thomas & Victor, 1998) and/or endothelium (Tesfamariam et al. 1987; Ohyanagi et al. 1992; Häbler et al. 1997; Dinenno & Joyner, 2004) can blunt the vascular response to sympathetic nerve activity during exercise, a physiological phenomenon termed functional sympatholysis (Remensnyder et al. 1962). While a definitive mechanism for sympatholysis has not been identified, evidence is accumulating that nitric oxide (NO) may be involved, because removal of the endothelium or blockade of NO synthase (NOS) enhances the vascular response to sympathetic stimulation (Tesfamariam et al. 1987; Ohyanagi et al. 1992; Häbler et al. 1997; Thomas & Victor, 1998).

Chronic endurance exercise training is known to affect the skeletal muscle vasculature in a number of ways. Following exercise training, it is generally believed that the vascular response to known vasodilators is augmented (Sun et al. 1994; Delp, 1995; Koller et al. 1995; Laughlin et al. 2004; Donato et al. 2007), skeletal muscle vascular resistance is reduced during submaximal exercise (Proctor et al. 2001) and skeletal muscle vascular conductance at peak exercise is augmented (Snell et al. 1987; Mourtzakis et al. 2004); however, these adaptations may be dependent upon the limb and vascular segment investigated (Jasperse & Laughlin, 2006). We recently reported that the vascular response to lumbar sympathetic chain stimulation in resting skeletal muscle was augmented following 4 weeks of exercise training in a training-intensity-dependent manner (Jendzjowsky & DeLorey, 2012). Whether exercise training alters the vascular response to sympathetic stimulation and the blunting of sympathetic vasoconstriction in contracting muscle has not been established.

Our understanding of how the intensity of chronic endurance exercise training affects the regulation of skeletal muscle sympathetic vasoconstriction in contracting skeletal muscle is also limited, despite some evidence that training adaptations in the cardiovascular system and in other systems appear to be sensitive to the intensity of training (Gibala, 2007; Laughlin & Roseguini, 2008; Hafstad et al. 2011).

Therefore, the purpose of the present study was to investigate the effects of 4 weeks of mild- and heavy-intensity exercise training on the magnitude of vasoconstriction in response to sympathetic stimulation in resting and contracting skeletal muscle. It was hypothesized that exercise training would augment sympatholysis in a manner dependent on the intensity of exercise training. We also hypothesized that the augmented sympatholysis would be due to an increased NO-mediated inhibition of sympathetic vasoconstriction in exercise-trained rats.

Methods

Animals and animal care

Male Sprague–Dawley rats (∼8 weeks old) were obtained from the institutional breeding colony and housed in pairs in a 12 h–12 h light–dark cycle, environmentally controlled room (22–24°C, 40–70% humidity). Water and rat chow (Lab Diet 5001; PMI Nutrition, Brentwood, MO, USA) were provided ad libitum. All experiments were conducted in accordance with the Canadian Council on Animal Care Guidelines and Policies with approval from the Animal Care and Use Committee: Health Sciences for the University of Alberta.

Chronic endurance exercise training

All rats were habituated to the laboratory and exercise by running on a treadmill (Panlab LE8710, Barcelona, Spain) for 10 min day−1 for 5 days at 10 m min−1, 0% gradient. After familiarization, rats were randomly assigned one of three groups: (i) time control sedentary (S); (ii) mild-intensity exercise training (M; 20 m min−1, 5% gradient); or (iii) heavy-intensity exercise training (H; 40 m min−1, 5% gradient). Training volume was matched between exercise groups by having animals run the same distance (600 m) during each training bout. Rats were trained on 5 days week−1 for 4 weeks. On the first training day, H rats ran 15 intervals of 1 min at 40 m min−1 5% gradient interspersed with rest periods of equal time. With each subsequent training bout, run time was increased, while rest time was maintained. This training progression allowed all rats in the H group to run continuously at the prescribed speed and gradient within 11 ± 2 days, consistent with previous studies from our laboratory (Jendzjowsky & DeLorey, 2011). Rats randomized to the M group ran at the prescribed speed, gradient and distance immediately following familiarization and maintained this for the entire exercise programme. This training protocol has previously been shown to increase heart mass, heart mass-to-body mass ratio, soleus citrate synthase activity and endothelium-dependent vasodilatation (Jendzjowsky & DeLorey, 2012). Sedentary time control animals were handled and weighed daily.

Instrumentation

Twenty-four hours after the last training session, rats were anaesthetized by inhalation of isoflurane (3.5%, balance O2). The right jugular vein was then cannulated and anaesthesia was maintained with α-chloralose (8–16 mg kg−1 h−1) and urethane (50–100 mg kg−1 h−1). The depth of anaesthesia was assessed by the stability of arterial blood pressure and heart rate (HR) and the absence of a withdrawal reflex in response to a painful stimulus (i.e. paw pinch). Core temperature was monitored by rectal probe and maintained at 36–37°C by an external heating pad (TCAT-2, Physitemp, Clifton, NJ, USA). A tracheotomy was performed to allow spontaneous breathing of room air. The left brachial artery was cannulated and connected to a solid-state pressure transducer (Abbott, North Chicago, IL, USA) for the continuous measurement of arterial blood pressure. The left femoral artery and vein were cannulated for the delivery of pharmacological agents. Blood flow was measured using a flow probe (0.7 V; Transonic Systems, Ithaca, NY, USA) placed around the right femoral artery and connected to a flowmeter (T106; Transonic Systems). Heart rate was derived from the arterial blood pressure waveform. Arterial blood samples were taken at rest and at the end of each contraction bout for the measurement of arterial inline image, arterial inline image and pH (VetStat; IDEXX Laboratories, Markham, ON, Canada).

Muscle contraction

The right sciatic nerve was exposed and instrumented with a nerve cuff electrode. The triceps surae muscle group was dissected free and attached to a force transducer (model MLT1030/D; AD Instruments, Colorado Springs, CO, USA) via the calcaneal tendon. Maximal contractile force (MCF) was determined by stimulation of the triceps surae muscle group with 25 impulses of 1 ms duration delivered at 100 Hz, 10 times motor threshold (MT). The optimal muscle length for tension development was determined by progressively lengthening the muscle and repeating the nerve stimulation until a plateau in tension (peak minus baseline) was observed. Rhythmic contractions of the triceps surae muscles were produced at 30% MCF (40 Hz, 0.1 ms pulses in 250 ms trains at a rate of 60 trains min−1 at ∼2.0 times MT) and 60% MCF (40 Hz, 0.1 ms pulses in 250 ms trains at a rate of 60 trains min−1 at ∼5.5 times MT).

Lumbar sympathetic chain stimulation

Following a laparotomy, the great vessels were temporarily retracted and the lumbar chain was exposed by dissection with a blunt glass pipette. A bipolar silver-wire stimulating electrode was attached to the lumbar sympathetic chain between L3 and L4. The electrodes were embedded and electrically isolated in a rapidly curing, non-toxic silicone elastomer (Kwiksil; WPI, Sarasota, FL, USA). The electrodes delivered constant-current stimulations through an isolated stimulator (DS3; Digitimer, Welwyn Garden City, UK).

Following a 20 min stabilization period, the following experiments were conducted in a total of 36 rats (S, n = 11; M, n = 12; and H, n = 13).

Series 1: effects of exercise training on the magnitude of vasoconstriction in response to sympathetic stimulation at rest and during muscle contraction The skeletal muscle vascular response evoked by lumbar sympathetic chain stimulation (1 min of 1 ms, 1 mA pulses delivered at 2 and 5 Hz in random order) was determined at rest and during muscle contraction at 30 and 60% of MCF. Bouts of muscle contraction were 8 min in duration, completed in random order, and separated by 60 min of recovery. During each bout, stimulations of the lumbar sympathetic chain were delivered 3 and 6 min after the onset of contraction.

Series 2: effects of exercise training on NO-mediated inhibition of sympathetic vasoconstriction at rest and during muscle contraction Following completion of series 1 and a 30 min recovery period, a bolus injection of the non-selective NOS inhibitor, Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME) was delivered (5 mg kg−1, i.v.). After ∼20 min and stabilization of haemodynamic parameters, stimulation of the lumbar sympathetic chain was repeated at rest and during contraction at 30 and 60% MCF as described in the previous subsection.

Time control studies Given the relatively long duration of the experimental protocol, an additional group of animals (n = 5) were used to investigate whether the vascular response to sympathetic stimulation in resting and contracting skeletal muscle was reproducible over time. Briefly, the skeletal muscle vascular response evoked by lumbar sympathetic chain stimulation was determined at rest and during muscle contraction at 30 and 60% of MCF (time 1). Following a recovery period (∼30 min), stimulation of the lumbar sympathetic chain was repeated at rest and during contraction at 30 and 60% MCF (time 2).

Upon completion of all experiments, animals were killed by anaesthetic overdose with alpha-chloralose and urethane (IV) and the heart was dissected free for measurement of cardiac mass.

Drugs

All drugs were purchased from Sigma-Aldrich (Oakville, ON, Canada) and dissolved in 0.9% physiological saline.

Data analysis

Data were recorded using Chart 7™ data acquisition software (AD Instruments, Colorado Springs, CO, USA). Arterial blood pressure and femoral artery blood flow (FBF) were sampled at 100 Hz and femoral vascular conductance (FVC) and HR were calculated. The magnitude of vasoconstriction in response to sympathetic stimulation was determined by calculating the mean of the response to sympathetic stimulation and expressing it as the percentage change from the preceding 1 min steady-state value for mean arterial blood pressure (MAP), FBF and FVC. The magnitude of sympatholysis was calculated as the difference between the percentage change in FVC in response to sympathetic stimulation at rest and the percentage change in FVC in response to sympathetic stimulation during muscular contraction. All data are expressed as means ± SD.

Data were analysed by three-way repeated-measures ANOVA (group × contractile condition × drug condition). The effect of exercise training and NOS inhibition on the magnitude of sympatholysis was determined by two-way repeated-measures ANOVA (group × drug condition; STATISTICA 10; Statsoft Inc., Tulsa, OK, USA). When significant F-ratios were detected, Student–Newman–Keuls post hoc analysis was performed. A P value <0.05 was considered statistically significant.

Results

All rats randomized to exercise training groups completed the assigned training protocol. Body mass was lower (P < 0.05), whereas heart mass and the heart-to-body mass ratios were increased (P < 0.05) in exercise-trained compared with S rats (Table 1). Resting HR, MAP, FBF and FVC were similar (P > 0.05) between S and exercise-trained animals (Table 2). Arterial blood gases and pH were within normal limits at rest and during each contractile bout and were not different (P > 0.05) between groups at any time point (Table 3).

Table 1.  Indices of training efficacy
GroupBody mass (g)Heart mass (g)Heart-to-body mass ratio
  1. In this and the following tables, values are means ± SD. *Significant training-intensity-dependent difference (P < 0.05). †Significant difference from sedentary control group (P < 0.05). ‡Significant difference from mild-intensity group (P < 0.05).

Sedentary time control457 ± 371.5 ± 0.10.32 ± 0.03*
Mild-intensity trained414 ± 40†1.7 ± 0.2†0.40 ± 0.04*
Heavy-intensity trained385 ± 31†‡1.7 ± 0.2†0.43 ± 0.04*
Table 2.  Baseline haemodynamic variables
GroupDrug conditionHR (beats min−1)MAP (mmHg)FBF (ml min−1)FVC (ml min−1 mmHg−1)
  1. Abbreviations: FBF, femoral blood flow; FVC, femoral vascular conductance; HR, heart rate; and MAP, mean arterial blood pressure. *Significant main effect of l-NAME (P < 0.05).

Sedentary time controlControl359 ± 3693 ± 93.1 ± 0.60.03 ± 0.01
  l-NAME321 ± 51*130 ± 14*3.3 ± 0.80.02 ± 0.01*
Mild-intensity trainedControl337 ± 5094 ± 113.0 ± 0.60.03 ± 0.01
  l-NAME312 ± 42*131 ± 11*3.1 ± 1.00.02 ± 0.01*
Heavy-intensity trainedControl335 ± 2891 ± 103.0 ± 0.80.03 ± 0.01
  l-NAME298 ± 49*126 ± 13*2.8 ± 0.80.02 ± 0.01*
Table 3.  Arterial blood gases and pH
 Series 1: controlSeries 2: l-NAME
Rest30% MCF60% MCFRest30% MCF60% MCF
 
  1. Arterial inline image, arterial inline image and pH values are presented for sedentary, mild- and heavy-intensity trained rats, during control and l-NAME conditions at rest and at 30 and 60% maximal contractile force (MCF).

Sedentary time control
 Arterial inline image (mmHg)91 ± 390 ± 289 ± 288 ± 387 ± 491 ± 3
 Arterial inline image (mmHg)38 ± 139 ± 138 ± 141 ± 240 ± 238 ± 2
 pH7.42 ± 0.017.43 ± 0.017.42 ± 0.017.41 ± 0.017.42 ± 0.017.42 ± 0.01
Mild-intensity trained
 Arterial inline image (mmHg)90 ± 489 ± 487 ± 290 ± 489 ± 489 ± 3
 Arterial inline image (mmHg)41 ± 241 ± 340 ± 339 ± 341 ± 239 ± 1
 pH7.42 ± 0.017.43 ± 0.017.41 ± 0.027.41 ± 0.027.41 ± 0.027.40 ± 0.03
Heavy-intensity trained
 Arterial inline image (mmHg)91 ± 389 ± 390 ± 490 ± 490 ± 390 ± 4
 Arterial inline image (mmHg)40 ± 240 ± 239 ± 240 ± 240 ± 240 ± 1
 pH7.41 ± 0.027.42 ± 0.027.40 ± 0.027.40 ± 0.027.41 ± 0.027.42 ± 0.02

Series 1: magnitude of vasoconstriction in response to sympathetic stimulation in resting and contracting skeletal muscle

The response to lumbar sympathetic stimulation delivered at 2 and 5 Hz in resting skeletal muscle in a representative animal is shown in Fig. 1A. The magnitude of vasoconstriction in response to sympathetic stimulation delivered at 2 Hz was increased (P < 0.05) by exercise training in a training-intensity-dependent manner (Fig. 2). In response to 5 Hz stimulation, a greater (P < 0.01) constriction was seen in M and H compared with S rats (Fig. 2).

Figure 1.

Original data from a representative animal illustrating the response of mean arterial blood pressure (MAP), femoral blood flow (FBF), femoral vascular conductance (FVC) and contractile force to lumbar sympathetic stimulations delivered at 2 and 5 Hz in resting skeletal muscle (A) and during skeletal muscle contraction at 60% of maximal contractile force (MCF; B) 
Arrow denotes the onset of contraction.

Figure 2.

The percentage change of FVC in response to 2 Hz (left panel) and 5 Hz sympathetic stimulation (right panel) at rest and during contraction at 30 and 60% of maximal contractile force (MCF) in sedentary (open), mild-intensity (light grey) and heavy-intensity trained rats (dark grey) during control conditions 
Values are means ± SD. *Significant difference between all groups at specified contractile state (P < 0.05). †Significant difference from the sedentary control group at specified contractile state (P < 0.05).

The response to lumbar sympathetic chain stimulation delivered at 2 and 5 Hz during skeletal muscle contraction in a representative animal is shown in Fig. 1B. Muscle contraction at 30% MCF produced a similar (P > 0.05) increase in skeletal muscle blood flow and vascular conductance in S, M and H groups (Table 4). Muscle force production was not different (P > 0.05) between groups at 30% MCF (S, 763 ± 122 g; M, 774 ± 107 g; and H, 745 ± 89 g). The magnitude of vasoconstriction in response to lumbar sympathetic stimulation at 2 Hz was greater (P < 0.05) in M and H compared with S rats at 30% MCF. In contrast, the response to sympathetic stimulation delivered at 5 Hz was similar (P > 0.05) between trained rats and the S group at 30% MCF (Fig. 2).

Table 4.  Muscle contraction-induced hyperaemia during two acute intensities of sustained rhythmic contractions
Contractile force (% of maximum)GroupDrug conditionFBF (mL min-1)FVC (mL min-1 mmHg-1)
  1. The increase of femoral artery blood flow (FBF) and femoral vascular conductance (FVC) in response to muscle contraction. *Significant main effect of muscle contractile force (P < 0.05). †Significant main effect of l-NAME (P < 0.05).

30%Sedentary time controlControl3.04 ± 0.750.032 ± 0.007
   l-NAME3.56 ± 1.21†0.030 ± 0.010†
 Mild-intensity trainedControl3.72 ± 1.30.039 ± 0.011
   l-NAME4.12 ± 1.32†0.033 ± 0.011†
 Heavy-intensity trainedControl3.72 ± 1.170.041 ± 0.010
   l-NAME3.79 ± 1.02†0.031 ± 0.010†
60%Sedentary time controlControl5.30 ± 1.160.050 ± 0.010*
   l-NAME4.90 ± 1.57†0.042 ± 0.012*†
 Mild-intensity trainedControl5.88 ± 1.880.058 ± 0.015*
   l-NAME6.04 ± 1.34†0.051 ± 0.011*†
 Heavy-intensity trainedControl5.77 ± 1.760.060 ± 0.011*
   l-NAME5.26 ± 1.70†0.044 ± 0.015*†

Compared with rest, the magnitude of vasoconstriction in response to sympathetic stimulation was diminished (P < 0.05) during contraction (i.e. sympatholysis) at 30% MCF in S, M and H rats (Fig. 2). However, the magnitude of sympatholysis (%FVC rest minus contraction; Δ%FVC) was greater (P < 0.05) in H compared with M and S groups during 2 Hz sympathetic stimulation and greater (P < 0.01) in H compared with S rats during 5 Hz sympathetic stimulation (Fig. 3).

Figure 3.

The magnitude of sympatholysis calculated as the difference between the percentage change in FVC in response to sympathetic stimulation at rest and during contraction at 30% maximal contractile force (MCF) in response to sympathetic stimulation at 2 Hz (left panel) and 5 Hz (right panel) in sedentary (open), mild-intensity (light grey) and heavy-intensity trained rats (dark grey) prior to and following NOS blockade with l-NAME (hatched bars; 5 mg kg−1i.v.) 
Values are means ± SD. †Significant difference from sedentary control group (P < 0.05). ‡Significant difference from mild-intensity trained group (P < 0.05). **Significant difference between control and l-NAME conditions within the same training group (P < 0.05).

Muscle contraction at 60% MCF produced a similar (P > 0.05) increase in skeletal muscle blood flow and vascular conductance in S, M and H animals (Table 4). Muscle force production was not different (P > 0.05) between S, M and H groups at 60% MCF (S, 1179 ± 131 g; M, 1241 ± 163 g; and H, 1199 ± 116 g). During contraction at 60% MCF, the magnitude of vasoconstriction in response to sympathetic stimulation at 2 and 5 Hz was similar (P > 0.05) in S, M and H rats (Fig. 2).

However, the magnitude of vasoconstriction in response to sympathetic stimulation was reduced (P < 0.05) compared with rest in all groups (Fig. 2). In response to sympathetic stimulation delivered at 2 Hz, the magnitude of sympatholysis was increased (P < 0.05) in a training-intensity-dependent manner during contraction at 60% MCF (Fig. 4). During sympathetic stimulation at 5 Hz, sympatholysis was greater (P < 0.05) in M and H compared with S rats during contraction at 60% MCF (Fig. 4).

Figure 4.

The magnitude of sympatholysis calculated as the difference between the percentage change in FVC in response to sympathetic stimulation at rest and during contraction at 60% maximal contractile force (MCF) in response to sympathetic stimulation at 2 Hz (left panel) and 5 Hz (right panel) in sedentary (open), mild-intensity (light grey) and heavy-intensity trained rats (dark grey) prior to and following NOS blockade with l-NAME (hatched bars; 5 mg kg−1i.v.) 
Values are means ± SD. *Significant difference between all groups (P < 0.05). †Significant difference from sedentary control group (P < 0.05). **Significant difference between control and l-NAME conditions within the same training group (P < 0.05).

Series 2: NO-mediated inhibition of sympathetic vasoconstriction

Nitric oxide synthase blockade produced a similar decrease in resting HR, FBF and FVC and increase in MAP in S, M and H rats (Table 2; P < 0.05). The vascular response to sympathetic stimulation at both 2 and 5 Hz was augmented (P < 0.05) in resting and contracting skeletal muscle in S, M and H rats in the presence of l-NAME (Fig. 5).

Figure 5.

The percentage change of FVC in response to 2 Hz (left panel) and 5 Hz sympathetic stimulation (right panel) at rest and during contraction at 30 and 60% of maximal contractile force (MCF) in sedentary (open), mild-intensity (light grey) and heavy-intensity trained rats (dark grey) during nitric oxide synthase (NOS) blockade with l-NAME (5 mg kg−1i.v.) 
Values are means ± SD. *Significant difference between all groups at specified contractile state (P < 0.05). †Significant difference from the sedentary control group at specified contractile state (P < 0.05). ‡Significant difference from mild-intensity trained group at specified contractile state (P < 0.05).

Treatment with l-NAME did not alter (P > 0.05) muscle force production. The increase in FBF during contraction at 30 or 60% MCF was also not altered (P > 0.05) by l-NAME (Table 4), but the increase in FVC during contraction at 30 or 60% was reduced (P < 0.01) by l-NAME in S, M and H rats (Table 4).

Nitric oxide synthase blockade augmented (P < 0.05) sympathetic vasoconstriction during muscular contraction in all rats (Fig. 5). In the presence of l-NAME, at 30% MCF, H rats had a greater (P < 0.05) vasoconstrictor response to sympathetic stimulation at 2 Hz compared with S and M rats, whereas 5 Hz stimulation produced a greater (P < 0.01) vasoconstriction in both M and H compared with S rats (Fig. 5). At 30% MCF, the magnitude of sympatholysis was not affected (P > 0.05) by l-NAME in S rats and M rats during 2 or 5 Hz sympathetic stimulation (Fig. 3); however, l-NAME significantly reduced (P < 0.05) the magnitude of sympatholysis during contractions at 30% MCF in H rats during sympathetic stimulation at both 2 and 5 Hz (Fig. 3).

During NOS blockade at 60% MCF, H rats had an increased (P < 0.05) vascular response to 2 and 5 Hz sympathetic stimulation compared with the S group (Fig. 5). Treatment with l-NAME did not alter (P > 0.05) the magnitude of sympatholysis in S rats during 2 or 5 Hz sympathetic stimulation (Fig. 4); however, l-NAME reduced (P < 0.05) the magnitude of sympatholysis during contractions at 60% MCF in M and H rats during both 2 and 5 Hz sympathetic stimulation (Fig. 4).

Time control studies

There was no difference (P > 0.05) in the constrictor response to sympathetic stimulation between time 1 and time 2 at rest (difference between percentage decrease in FVC at time 1 minus time 2: 2 Hz, −0.46 ± 0.53; and 5 Hz, 4.02 ± 0.55), 30% MCF (2 Hz, 1.02 ± 0.68; and 5 Hz, −2.78 ± 0.94) or 60% MCF (2 Hz, −1.82 ± 0.77; and 5 Hz, −2.20 ± 1.67).

Discussion

The purpose of the present study was to investigate the effects of 4 weeks of mild- and heavy-intensity exercise training on the magnitude of vasoconstriction in response to sympathetic stimulation in resting and contracting skeletal muscle. Consistent with previous findings from our laboratory (Jendzjowsky & DeLorey, 2012), the present study demonstrated that short-term exercise training augmented the constrictor response to sympathetic stimulation at rest. The important novel finding in the present study was that exercise training augmented sympatholysis in contracting muscle. The training-induced improvements in sympatholytic capacity appeared to be sensitive to the intensity of the training stimulus, because rats that trained at a heavy intensity had an augmented ability to inhibit sympathetic vasoconstriction during moderate- and heavy-intensity contractions. In contrast, rats that trained at a mild- intensity exhibited augmented sympatholysis only during heavy-intensity contractions. The mechanism responsible for the augmented sympatholysis in trained animals was an augmented NO-mediated inhibition of sympathetic vasoconstriction, because NOS blockade decreased the magnitude of sympatholysis in exercise-trained rats.

To our knowledge, this is the first study to demonstrate that a prospective short-term exercise training regimen augments sympatholysis. These results provide direct evidence of a training-intensity-sensitive upregulation of sympatholysis, mediated through an NO-dependent mechanism.

The magnitude of vasoconstriction in response to sympathetic stimulation

Previous studies that have investigated the effects of exercise training on sympathetic vascular responsiveness have done so in resting skeletal muscle or in isolated vascular preparations and have produced equivocal results (Wiegman et al. 1981; Smith et al. 1988; Delp et al. 1993; Sun et al. 1994; McAllister & Laughlin, 1997; Lash, 1998; Jasperse & Laughlin, 1999; Spier et al. 1999; Laughlin et al. 2004; Donato et al. 2007). Indeed, chronic endurance exercise training has been shown to increase (Lash, 1998), decrease (Wiegman et al. 1981; Delp et al. 1993; Spier et al. 1999; Donato et al. 2007) or not alter postsynaptic α-adrenergic receptor responsiveness (Sun et al. 1994; Jasperse & Laughlin, 1999). However, studies that have investigated sympathetic vascular responsiveness following heavy-intensity aerobic exercise (McAllister & Laughlin, 1997) and sprint interval exercise training (Laughlin et al. 2004) have demonstrated an augmented response to exogenous administration of noradrenaline and phenylephrine in isolated arterioles (McAllister & Laughlin, 1997; Laughlin et al. 2004). We also recently reported augmented sympathetic vascular responsiveness in resting skeletal muscle following heavy-intensity exercise training (Jendzjowsky & DeLorey, 2012). Consistent with our previous investigation (Jendzjowsky & DeLorey, 2012), the vascular response to sympathetic stimulation delivered at 2 Hz was augmented in resting skeletal muscle in a manner dependent on the intensity of exercise training in the present study. We extend our previous findings by demonstrating that the response to sympathetic stimulation delivered at a higher frequency (5 Hz), reflective of a higher level of sympathetic outflow, was also increased in exercise-trained compared with S rats.

Compared with rest, muscular contraction diminished the magnitude of vasoconstriction in response to sympathetic stimulation in all rats. Indeed, the magnitude of vasoconstriction declined as a function of muscle contractile force in all rats. During muscular contraction at 30% MCF, exercise-trained rats showed a greater vasoconstriction in response to sympathetic stimulation delivered at 2 Hz compared with sedentary rats, whereas the response to 5 Hz sympathetic stimulation was not different between groups. This increase in vascular responsiveness to sympathetic stimulation at 2 Hz suggests that short-term exercise training resulted in vascular smooth muscle becoming more sensitive to low levels/frequencies of sympathetic outflow during muscle contraction at moderate intensities. However, it is possible that exercise training may also have altered the amount or composition of neurotransmitters released in response to stimulation of the lumbar chain or that exercise training altered local muscle metabolite production. During muscular contraction at 60% MCF, the vascular response to sympathetic stimulation was not different between groups at both 2 and 5 Hz stimulation frequencies. These findings suggest that exercise training does not alter vascular responsiveness to sympathetic stimulation during heavy-intensity contractile activity.

Proctor et al. (2001) reported that 9–12 weeks of heavy-intensity exercise training decreased leg blood flow during submaximal cycle exercise at 70 and 140 W. The lower post-training leg blood flows were achieved at similar or reduced levels of leg noradrenaline spillover, suggesting that exercise training may have augmented sympathetic vascular responsiveness in contracting muscle, consistent with the present findings.

Magnitude of sympatholysis

Previous studies completed in dynamically exercising humans and animals and in situ muscle preparations have demonstrated that sympatholysis is dependent on the intensity of exercise and/or muscle contractile force (Anderson & Faber, 1991; Ohyanagi et al. 1991; Buckwalter et al. 1997, 1998, 2001; Buckwalter & Clifford, 1999; Ruble et al. 2000, 2002; Tschakovsky et al. 2002; Rosenmeier et al. 2003; VanTeeffelen & Segal, 2003; Wray et al. 2004). In agreement with prior investigations, the ability to inhibit sympathetic vasoconstriction was related to muscle force production in all rats in the present study.

Additionally, exercise training augmented sympatholysis in a training-intensity-dependent manner. Specifically, H trained rats had a greater degree of sympatholysis during contraction at both 30 and 60% of MCF compared with M trained rats, in which a greater magnitude of sympatholysis occurred only during contraction at 60% MCF. The commonly accepted method to assess sympatholysis is to compare the percentage decrease in FVC during contraction to the response at rest, because the percentage change in FVC has been shown to correspond to a similar percentage reduction in blood vessel radius despite differing levels of vascular conductance between rest and exercise (Buckwalter & Clifford, 2001). In the present study, we calculated the magnitude of sympatholysis as the difference between the percentage decrease in FVC at rest and during contraction. It could be argued that the increased constrictor response to sympathetic stimulation at rest equipped the exercise-trained rats with a greater capacity for sympatholysis. However, it is difficult to reconcile how an increase in constrictor responsiveness at rest would also confer a greater ability to inhibit sympathetic vasoconstriction during contraction upon the trained animals. Indeed, the greater constrictor responsiveness at rest could in fact make it more difficult for the trained animals to inhibit constriction during contraction. Thus, while the trained animals may have a greater range over which to decrease constrictor responsiveness, we believe the important consideration is that the trained animals were able markedly to inhibit the response to sympathetic stimulation during contraction, despite an increase in constrictor responsiveness at rest.

The present findings are in contrast to recent human studies. Wimer & Baldi (2012) have shown that 6 weeks of intermittent hand-grip training performed at 30% maximal voluntary contraction for 8 min, three times per week did not alter sympatholysis during hand-grip exercise. Likewise, Mortensen et al. (2012) recently reported no difference in sympatholysis between control and exercise-trained limbs following 5 weeks of single-leg knee-extension exercise training. Recent cross-sectional studies have also reported a similar magnitude of sympatholysis in trained compared with untrained humans (Wray et al. 2007; Wimer & Baldi, 2012). The reason(s) for these contrasting results are not readily apparent; however, they may be related to the intensity and volume of exercise training prescribed. As demonstrated in the present study, the intensity of training appears to influence sympathetic vascular control. Another potential explanation for the contrasting findings may be related to differences between prospective training studies and cross-sectional investigations. Functional vascular adaptations appear to occur early during training (≤6 weeks), with structural adaptations arising in response to prolonged durations of exercise training (Delp, 1998; Green et al. 2004). A major strength of the present study design was the ability to isolate the effect of training-intensity and match the total volume of exercise training completed between groups. The training history of subjects in a cross-sectional study can obviously not be controlled in this manner, and effects of training on the vasculature are subject to a variety of factors, such as training history and volume. Further studies will be required to investigate the relationships between the duration and volume of exercise training and functional sympatholysis and to determine whether improved sympatholysis is a vascular adaption that persists with more prolonged or larger volume training regimens.

A decline in postsynaptic receptor responsiveness has been mechanistically linked to sympatholysis. Although not a universal finding, tonic α1-adrenergic receptor-mediated vasoconstriction and α1-adrenergic receptor responsiveness have generally been shown to decline only during heavy-intensity exercise (Anderson & Faber, 1991; Buckwalter et al. 2001; Wray et al. 2004), whereas tonic α2-adrenergic receptor-mediated constriction and receptor responsiveness appears to be diminished during moderate- and heavy-intensity exercise (Faber, 1988a; McGillivray-Anderson & Faber, 1990, 1991; Anderson & Faber, 1991; Thomas et al. 1994; Tateishi & Faber, 1995b; Buckwalter et al. 2001; Rosenmeier et al. 2003; Wray et al. 2004). It has been argued that this intensity-dependent modulation of receptor responsiveness may be related to the distribution of receptors within the vascular tree, because receptors positioned on distal branches of the vascular tree would be in closer proximity to the interstitial environment, exposing them to larger concentrations of vasoactive molecules that oppose sympathetic vasoconstriction. Data from the rat cremaster muscle indicate that α2-adrenergic receptors are localized to small, distal secondary and tertiary arterioles, whereas α1-adrenergic receptors are primarily located on larger, proximal arterioles (Faber, 1988b; McGillivray-Anderson & Faber, 1990, 1991; Anderson & Faber, 1991). However, in the mouse gluteus maximus muscle the functional distribution of α2-adrenergic receptors was greater in proximal (1A) arterioles, while the functional distribution of α1-adrenergic receptors was greater in distal (3A) arterioles, suggesting that the distribution of receptors varies between vessel branch orders and skeletal muscle type/function (Moore et al. 2010). We did not investigate exercise-training-induced changes in the expression and/or distribution or responsiveness of individual postsynaptic receptors in the present study. Given the short-term training stimulus and relatively low volume of training used in this study, we do not believe that receptor expression and/or distribution would be altered in the present study. However, exercise training has been shown to upregulate the expression of other vasoregulatory molecules; therefore, further studies will be required to establish the effect of exercise training on postsynaptic receptor expression and distribution.

Nitric oxide-dependent inhibition of sympathetic vasoconstriction

A definitive mechanism responsible for sympatholysis has not been established. However, NO has been shown to inhibit sympathetic vasoconstriction at rest (Tesfamariam et al. 1987; Häbler et al. 1997; Nase & Boegehold, 1997) and during muscle contraction (Ohyanagi et al. 1992; Hansen et al. 1994; Thomas et al. 1998; Thomas & Victor, 1998; Sander et al. 2000). NO may mediate sympatholysis through activation of ATP-sensitive K+ channels and blunting of postsynaptic α-adrenergic receptor responsiveness (Tateishi & Faber, 1995a; Thomas et al. 1997). Consistent with previous studies, the present data demonstrate that NO inhibits sympathetic vasoconstriction in resting (Tesfamariam et al. 1987; Ohyanagi et al. 1992; Häbler et al. 1997; Nase & Boegehold, 1997) and contracting skeletal muscle (Hansen et al. 1994; Thomas et al. 1998; Thomas & Victor, 1998; Sander et al. 2000), regardless of training status. An important novel finding of the present study is that NO-mediated sympatholysis is augmented by exercise training in a training-intensity-dependent manner (Figs 4 and 5).

Mechanical and chemical stimuli elicit the release of NO from NOS localized in the endothelium (eNOS; Rubanyi et al. 1986) and skeletal muscle sarcolemma (neuronal NOS; nNOS; Nakane et al. 1993; Kobzik et al. 1994) in response to muscular contraction. Both eNOS (Ohyanagi et al. 1992) and nNOS (Thomas et al. 1998, 2003; Sander et al. 2000) have been shown to inhibit sympathetic vasoconstriction in human and animal preparations (Ohyanagi et al. 1992; Hansen et al. 1994; Thomas et al. 1998; Thomas & Victor, 1998; Sander et al. 2000). In the presence of NOS blockade or removal of the endothelium, vascular responsiveness to noradrenaline and phenylephrine in the abdominal aorta (Delp et al. 1993; Spier et al. 1999) and gastrocnemius and soleus first order arterioles (Donato et al. 2007) was augmented following moderate-intensity exercise training, suggesting that eNOS-mediated inhibition of sympathetic vasoconstriction was enhanced following exercise training. Indeed, exercise training has been shown to enhance eNOS protein expression in response to moderate- (Oltman et al. 1995; Spier et al. 2004; McAllister et al. 2005) and heavy-intensity training (Laughlin et al. 2004). Expression of the nNOS isoform has also been shown to increase in response to exercise training. Ten days of exhaustive cycle exercise training increased nNOS protein expression in human skeletal muscle (McConell et al. 2007), and 4 weeks of daily swim training increased nNOS protein expression in the rat hindlimb (Tatchum-Talom et al. 2000). In contrast, run training three times per week for 6 weeks did not change nNOS protein expression in human skeletal muscle (Frandsen et al. 2000). Thus, the available evidence, although not conclusive, suggests that the expression of both eNOS and nNOS may change in response to exercise training and may be involved in the training-mediated upregulation of sympatholysis. To our knowledge, the effect of exercise training on specific NOS isoform-mediated sympatholysis has not been established.

Another potential mechanism involved in the exercise-training-mediated augmentation of NO-mediated sympatholysis may be associated with the recruitment of skeletal muscle fibres. Increasing intensities of exercise are associated with a progressive recruitment of skeletal muscle and, in particular, the recruitment of additional glycolytic muscle fibres (Laughlin & Armstrong, 1982, 1985; Laughlin et al. 1984). Although the present study did not assess the amount or pattern of muscle fibre recruitment during exercise training, it is conceivable that a larger proportion of glycolytic muscle fibres was recruited during heavy- compared with mild-intensity training (Laughlin & Armstrong, 1982, 1985; Laughlin et al. 1984). It also seems likely that exercise training resulted in a training-intensity-dependent vascular recruitment, such that a larger volume of blood vessels and vessels from different skeletal muscles may have been ‘conditioned’ in exercise-trained compared with sedentary rats. Some evidence suggests that a greater magnitude of NO-mediated sympatholysis occurs in glycolytic compared with oxidative skeletal muscle (Thomas & Victor, 1998). Therefore, if a larger number of blood vessels from glycolytic muscles were affected as a function of training intensity, it is conceivable that this may have contributed to the training-intensity-dependent upregulation of sympatholysis in the present study.

Perspectives and significance

An exercise-training-induced increase in vascular reactivity to sympathetic stimulation in resting skeletal muscle and an enhanced inhibition of sympathetic vasoconstriction during contraction appears counterintuitive. We suggest that exercise training augments the responsiveness of vascular smooth muscle to vasoactive molecules and that the skeletal muscle contractile state impacts the effectiveness and/or availability of individual vasoactive signalling molecules and influences regulation of the overall level of vascular tone. In resting skeletal muscle, when sympathetic neurotransmitters are largely unopposed by locally released vasodilators, vasoconstriction is augmented. During muscle contraction, exercise-trained rats may release a larger quantity of vasoactive molecules in response to contraction or may become more responsive to vasoactive molecules that oppose sympathetic vasoconstriction (i.e. NO), resulting in an augmented sympatholysis.

The present findings demonstrate that short-term exercise training augmented functional sympatholysis in a training-intensity-dependent manner through a NO-dependent mechanism. These novel findings advance our understanding of the effects of exercise training on the regulation of sympathetic vasoconstriction in resting and contracting skeletal muscle.

Several previous studies have shown that the ability to inhibit sympathetic vasoconstriction may become impaired with ageing, oxidative stress and disease (Thomas et al. 2001; Dinenno et al. 2005; Zhao et al. 2006; Parker et al. 2007; Kirby et al. 2011; Vongpatanasin et al. 2011). The present data demonstrate that there is considerable plasticity in the regulation of sympathetic vasoconstriction and that the skeletal muscle vasculature is remarkably responsive to a relatively modest volume of exercise training. These data suggest that exercise training may be an important component in the treatment of pathophysiological conditions characterized by elevated sympathetic outflow, increased vascular resistance and a loss of sympatholytic capacity.

Appendix

Author contributions

N.G.J. contributed to the design of experiments, data collection, analysis and interpretation, and manuscript preparation and revision. D.S.D. contributed to the conception of the project, design of experiments, data analysis and interpretation, and manuscript preparation and revision. Both authors approved this version of the manuscript.

Acknowledgements

This project was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC), Canadian Foundation for Innovation and Alberta Advanced Education and Technology. N.G.J. was supported by an NSERC Canada Graduate Doctoral Scholarship, University of Alberta Presidents’ Scholarship and Izaak Walton Killam Memorial Scholarship. The authors wish to thank Drs Janice Marshall, Andrew Coney and Kelvin Jones and Mr Neil Tyreman for their assistance with the development of the experimental preparation.

Ancillary