INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SKELETAL MUSCLE HYPEREMIA at the onset of exercise reflects the transition from relatively low oxygen demands at rest to the high oxygen requirements associated with exercise. The mechanism(s) by which vasodilation occurs in active skeletal muscles is poorly understood, but a neural mechanism is particularly appealing because of the rapidity of the increase in skeletal muscle blood flow at the onset of exercise. Although -adrenergic receptors and muscarinic receptors are present in the arterial vasculature of skeletal muscle (5, 18, 23, 43) and can produce skeletal muscle vasodilation, these receptors are not essential for the rapid vasodilation that occurs in active skeletal muscle at the onset of exercise (8) or for sustained skeletal muscle hyperemia during steady-state exercise (2, 4, 7, 19, 23). However, other neurogenic vasodilator agents such as nitric oxide (10, 11), vasoactive peptides (1), ATP (24), and histamine (20, 44) may be released from postganglionic sympathetic nerve terminals and mediate the rapid increase in skeletal muscle blood flow at the onset of exercise. Indeed, nitric oxide is responsible for the hindlimb vasodilation elicited by electrical stimulation of the posterior hypothalamus (31) or lumbar sympathetic nerves (10, 11).

Although chronic sympathectomies in dogs (12) and humans (9) failed to alter the exercise-induced increases in skeletal muscle blood flow, rats exercised minutes after sympathectomy demonstrated higher blood flows during steady-state exercise than intact rats did (35). These results suggest the possibility that there may be some chronic adaptation to surgical sympathectomy that obscures the physiological role of autonomic input to the vasculature. Thus it would be advantageous to use an experimental intervention that acutely interrupts autonomic efferent mechanisms.

The purpose of the present study was to examine the role of the autonomic nervous system in controlling blood flow to active skeletal muscles at the onset of exercise. It is important to look specifically at the rapid increase in blood flow at the onset of exercise because it has been suggested that the mechanism mediating the initial skeletal muscle hyperemia at the onset of exercise may be different from that which sustains blood flow during steady-state exercise (17). These experiments tested the hypothesis that the release of neurotransmitters from the autonomic nervous system is essential for skeletal muscle vasodilation at the onset of exercise.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All experimental procedures were approved by the Institutional Animal Care and Use Committee and conducted in accordance with the American Physiological Society's "Guiding Principles in the Care and Use of Animals." Seven mongrel dogs, weighing between 18 and 22 kg, were selected for their willingness to run on a motorized treadmill and were instrumented during two separate sterile surgical procedures. For both procedures anesthesia was induced with thiopental sodium (15-30 mg/kg; Gensia Pharmaceuticals, Irvine, CA). After intubation with a cuffed endotracheal tube, a surgical level of anesthesia was maintained through mechanical ventilation with 1.5% halothane (Halocarbon Laboratories, River Edge, NJ) and 98.5% oxygen. Antibiotics (cefazolin sodium; Apothecon, Princeton, NJ) and analgesic drugs (buprenorphine hydrochloride, 0.3 mg; Reckitt and Colman, Kingston-Upon-Hull, UK) were given postoperatively. During the first surgical procedure the carotid arteries were placed in skin tubes in the neck so that they could be cannulated percutaneously to measure arterial blood pressure. In the second surgery all dogs were instrumented with flow probes (4- or 6-mm ultrasonic transit time flow probes; Transonic Systems, Ithaca, NY) around the external iliac arteries to measure hindlimb blood flow. The cables were tunneled under the skin to the back, and the dogs were given 2 wk to recover from flow probe implantation before any experimental procedures were performed.

All experiments were performed in a laboratory in which the temperature was maintained below 20°C. A 20-gauge Teflon catheter (Insyte; Becton Dickinson, Sandy, UT) was inserted retrogradely into the lumen of the carotid artery and attached to a solid-state pressure transducer (Ohmeda, Madison, WI). The flow probes were connected to a transit time flowmeter (Transonic Systems). To eliminate autonomic influence during exercise, each dog received an intravenous bolus infusion of 10 mg/kg hexamethonium (Sigma, St. Louis, MO) and 0.2 mg/kg atropine (Sigma) while sitting quietly in a sling. This combination of drugs was shown previously (37) to be an effective means of producing ganglionic blockade in dogs. Before control experiments were performed, intravenous saline infusions were given. After the drug or saline treatment was completed, resting measurements were made and the dog was moved from the sling to a treadmill. The dog sat quietly on the treadmill until blood flow stabilized. Exercise was then initiated, and the dog ran on the motorized treadmill at 3 miles/h (4.8 km/h), 0% grade. At least 24 h separated each experiment; all control experiments and autonomic blockade experiments were performed in duplicate and the data averaged for each dog.

Arterial blood pressure (mmHg) and right and left external iliac blood flow (ml/min) were simultaneously written to paper on a polygraph recorder (model 7; Grass, Warwick, RI) and stored on both a videocassette data recorder (model D, Vetter, Rebersburg, PA) and a computer (Apple 8500 Power PC) using a MacLab data acquisition system at 100 Hz (ADInstruments, Castle Hill, Australia). Data were analyzed off-line using the MacLab software to calculate mean arterial pressure, heart rate, and iliac blood flow. Right and left iliac blood flow were summed to yield a value for hindlimb blood flow. Similarly, hindlimb vascular conductance (blood flow/mean arterial pressure) was calculated as the summation of right and left iliac conductance. While each dog sat in the sling, 10-s averages of resting blood flow and conductance were recorded before and after ganglionic blockade for statistical analysis. Hindlimb blood flow and conductance were also averaged over 1-s intervals for the first 30 s of exercise.

To analyze the time course of vasodilation at the onset of exercise, a two-way (drug × time) repeated-measures analysis of variance was performed on the hindlimb conductance values at 5-s intervals over the first 30 s of exercise. In addition, a separate two-way repeated-measures analysis of variance was performed for heart rate, mean arterial pressure, hindlimb blood flow, and hindlimb conductance values at rest and during steady-state exercise (at 1 min of exercise). An  level of 0.01 was used to establish statistical significance. Where significant F ratios were found, a Tukey's post hoc test was performed. All descriptive statistics are presented as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Table 1 provides hemodynamic data at rest in the sling and at 1 min of exercise with and without autonomic blockade. Autonomic blockade at rest produced increases in heart rate, hindlimb blood flow, and hindlimb conductance and decreases in mean arterial pressure (P < 0.01). Exercise during ganglionic blockade caused further decreases in mean arterial pressure but no significant change in heart rate (P < 0.01). In contrast, exercise under control conditions produced increases in mean arterial pressure and heart rate compared with the same values at rest (P < 0.01).

Figure 1 is an original tracing from an individual dog at the beginning of exercise on the treadmill at 3 miles/h, 0% grade, after control treatment with saline. There were rapid increases in iliac blood flow and conductance in both hindlimbs that exceeded the eventual steady-state values. This "overshoot" in blood flow and conductance is readily apparent in this original record and is in sharp contrast to the response seen with autonomic blockade shown in Fig. 2.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Original tracings from 1 dog at onset of exercise at 3 miles/h, 0% grade, under control (saline) conditions. Zero on time scale represents onset of exercise. There were rapid increases in iliac blood flow and conductance at initiation of treadmill exercise. Increases in blood flow and conductance at onset of exercise substantially exceed levels at 30 s of exercise.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Original tracings from 1 dog at onset of exercise at 3 miles/h, 0% grade, after ganglionic blockade. Zero on time scale represents onset of exercise. There were rapid increases in iliac blood flow and conductance at initiation of treadmill exercise. However, initial increases in blood flow and conductance do not exceed levels at 30 s of exercise. "Overshoot" in blood flow and conductance is absent with autonomic blockade. Note marked sustained drop in blood pressure with initiation of treadmill exercise.

Figure 3 presents summary data for mean arterial pressure, hindlimb blood flow, and hindlimb conductance for all seven dogs over the first 30 s of exercise under control and ganglionic blockade conditions. Hindlimb blood flow increased at the onset of exercise with ganglionic blockade despite the concomitant decrease in mean arterial pressure. Both mean arterial pressure and hindlimb blood flow were significantly lower during ganglionic blockade compared with control conditions (P < 0.01). Because the decrease in mean arterial pressure reduced the driving pressure for blood flow, it is imperative to examine the vascular conductance data under these conditions. Statistical analyses of the hindlimb conductance data at 5-s intervals over the first 30 s of exercise revealed a statistically significant difference between the control and ganglionic blockade conditions at 20, 25, and 30 s (P < 0.01) but not at 5, 10, and 15 s of exercise. It can be readily observed from Fig. 3 that there was no overshoot in hindlimb conductance after treatment with hexamethonium and atropine.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Mean arterial pressure, hindlimb blood flow, and hindlimb conductance at onset of exercise with control (saline) and experimental intervention (ganglionic blockade). Data were averaged (means ± SE) over 1-s intervals (except for time 0, which represents a 10-s average of values obtained while dogs were resting on treadmill). Hindlimb conductance SE bars have been excluded for clarity of presentation, but SE measurements over the 30 s of exercise under these 2 conditions ranged from 0.25 to 1.48 ml · min1 · mmHg1. There were no statistically significant differences in conductance responses between the 2 treatments at 5, 10, and 15 s of exercise. However, conductance values were significantly elevated with ganglionic blockade at 20, 25, and 30 s of exercise (P < 0.01).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major new finding from this study is that ganglionic blockade did not alter the rate or magnitude of the increase in hindlimb conductance over the first 15 s of exercise. The lack of an effect indicates that the autonomic nervous system is not required for the rapid increase in conductance in active skeletal muscle at the onset of mild exercise in the dog. In addition, the results show that, compared with the unblocked condition, hindlimb conductance was elevated from 20 s through 1 min of exercise with ganglionic blockade. We interpret this to indicate that sympathetically mediated vasoconstriction limits the rise in blood flow to active skeletal muscle later during exercise under normal conditions. However, because skeletal muscle blood flow during exercise was lower under the autonomic blockade conditions, the gradual increase in conductance after 20 s of exercise may reflect some contribution of metabolic vasodilation of the underperfused muscle.

The rapidity of the increase in skeletal muscle blood flow at the onset of exercise and the immediate decrease in blood flow at the cessation of exercise suggest a neural component. A mild intensity of exercise was used because we reasoned that this would minimize the metabolic contribution to vasodilation and thus provide the optimal conditions to unmask a neural component. In fact, the time course of the increase in blood flow shows an overshoot at this workload (Fig. 3). Substantial overshoot has been shown in previous studies at mild exercise intensities (26, 35) and has been attributed to neurally mediated vasodilation (3). In a recent investigation (8), our group tested involvement of -adrenergic and muscarinic receptors in skeletal muscle hyperemia using intra-arterial infusions of receptor antagonists. We concluded that neither -adrenergic nor muscarinic receptors were essential for the rapid increase in blood flow at the onset of exercise. However, a potential limitation in those experiments was the accessibility of smooth muscle -adrenergic and muscarinic receptors with intra-arterial infusion of propranolol and atropine (21, 28, 29). Nevertheless, the data from the present study support the conclusion that these receptors are not essential for the skeletal muscle vasodilation at the onset of exercise. Furthermore, the present results rule out a contribution to exercise hyperemia from nonadrenergic and noncholinergic neurotransmitters released from the sympathetic nerve terminal, such as nitric oxide, vasoactive peptides, ATP, and histamine.

Although the present study provides no evidence to support neurally mediated skeletal muscle vasodilation at the onset of exercise, the results do indicate that there is neurally mediated restraint of skeletal muscle vasodilation during exercise. This is in agreement with previous investigations that provide evidence for sympathetic restraint of blood flow to active skeletal muscle (6, 22, 34, 35, 42). We postulate that the increase in blood flow to active skeletal muscle at the onset of exercise reflects an initial vasodilation unhindered by vasoconstriction and that an overshoot becomes apparent as sympathetic vasoconstriction limits skeletal muscle vasodilation. This postulate is supported by the demonstration that conductance remained elevated in the absence of vasoconstrictor tone under ganglionic blockade. Sympathetic vasoconstriction limited blood flow to active skeletal muscle as evidenced by the reduction in hindlimb conductance after 20 s of exercise under unblocked conditions but not with ganglionic blockade. An absence of sympathetic restraint of blood flow to active skeletal muscle during the initial phase of exercise was also reported by Peterson et al. (35) in sympathectomized rats. Therefore, we conclude that sympathetic vasoconstriction restrains blood flow to active skeletal muscle during steady-state exercise.

In agreement with our results examining hindlimb vascular conductance, Sheriff et al. (37) reported similar elevations in total vascular conductance with exercise in control and autonomic blocked dogs. However, a particular strength of the present study is the measurement of blood flow to skeletal muscle rather than total cardiac output. The effects of systemic autonomic blockade on total vascular conductance reflects the withdrawal of sympathetic influences to inactive tissues, including visceral organs, as well as skeletal muscle, whereas the measurement of hindlimb vascular conductance allows more discrete examination of skeletal muscle vasodilation at the onset of exercise. Although iliac blood flow also subserves bone and skin, the increase in blood flow is directed only to muscle because there is no change in blood flow to bone and skin during exercise in dogs (30). It must be noted that Sheriff et al. (37) also showed that hindlimb conductance increased in three ganglionic blocked dogs at the onset of exercise but made no quantitative comparisons to control data in the absence of ganglionic blockade. Thus the unique contribution of the present study is the comparison of skeletal muscle vascular conductance during dynamic exercise with and without autonomic blockade.

Although the autonomic nervous system is not essential for the rapid increase in conductance to active skeletal muscle, it is clearly important in the regulation of arterial pressure during exercise. Autonomic blockade compromised the ability to maintain blood pressure during exercise by limiting changes in cardiac output and preventing systemic vasoconstriction. In the absence of compensatory changes in cardiac output and sympathetic vasoconstriction, skeletal muscle vasodilation resulted in a substantial decrease in mean arterial pressure at the onset of exercise.

The physiological mechanism(s) responsible for skeletal muscle exercise hyperemia has been a topic of investigation for over 100 years, since Gaskell's proposal that vasodilation occurred in contracting muscle due to the release of metabolites by the muscle fibers (15). Although skeletal muscle blood flow and metabolic activity increase in an exercise intensity-dependent manner and appear to be linked, a mechanistic connection or series of connections between these two events has not been established. If indeed there is a link, potential vasodilators would have to act immediately at the onset of exercise, because blood flow increases rapidly at the onset of exercise. In addition, manipulation of this vasodilator would have to substantially affect skeletal muscle hyperemia during exercise. Isolation of a single vasodilator agent may be problematic if redundant mechanisms are responsible for skeletal muscle hyperemia during exercise. A number of vasodilators such as adenosine, potassium, hypoxia, osmolarity, and ATP have received attention over the years without definitive support for any as being essential for skeletal muscle hyperemia (14, 17, 27). Nitric oxide release through shear stress (36) or directly from hemoglobin (40) has also been implicated in the regulation of skeletal muscle vasomotor tone. However, experimental results show that nitric oxide synthase blockade has, at most, a modest effect on exercising blood flow in dogs (33) and humans (13, 16, 38, 45).

The findings in this study are consistent with the suggestion that skeletal muscle hyperemia during exercise is a local phenomenon (25). One possible local control mechanism could be the muscle pump. The term "muscle pump" (also known as the venous pump) is used to describe propulsion of blood from the vasculature of skeletal muscle during contraction of the muscle. This hypothesis is attractive because activation of the muscle pump occurs with initiation of exercise, as does the increase in skeletal muscle blood flow. During dynamic exercise, muscular contraction compresses the veins and propels the blood back toward the heart. Subsequently, muscular relaxation pulls open the empty, collapsed veins, causing a drop in venous pressure, an increase in the pressure gradient, and an increase in blood flow. The muscle pump hypothesis is supported by the observation that peak conductance values for skeletal muscle are greater during dynamic exercise than are achievable with pharmacological vasodilation at rest (25). Sheriff et al. (37) showed in three dogs that hindlimb conductance was altered by increases in contraction frequency associated with increasing treadmill speed. In these dogs, autonomic blockade was employed to abolish neural influences and it was argued that metabolic effects were too slow, so the authors concluded that the muscle pump must account for the rapid initial response. More recently, Tschakovsky and colleagues (41) demonstrated that duplication of the mechanics of the muscle pump with rhythmic inflation and deflation of a forearm cuff increased forearm blood flow. However, other studies have questioned the importance of the muscle pump in skeletal muscle hyperemia (32, 39). These studies found that increased venous filling did not change blood flow during muscle contractions (32, 39). This contradicts the muscle pump hypothesis, which predicts that when resting skeletal muscle is maximally vasodilated, subsequent muscle contraction will increase skeletal muscle blood flow more than under control conditions. Although the investigation of the contribution of the muscle pump to skeletal muscle hyperemia was not the focus of the present study, our findings are compatible with this hypothesis.

After more than 100 years of research, clear demonstration of the physiological mechanism(s) responsible for skeletal muscle vasodilation during exercise remains elusive. On the basis of the results of the present study, there is little need for further investigation of a neural mechanism for the initial increase in blood flow at the onset of exercise. We conclude that autonomic innervation is not essential for the rapid skeletal muscle vasodilation at the onset of exercise. However, sympathetic vasoconstriction to active skeletal muscle is evident later in exercise.