INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

TREADMILL EXERCISE IS ACCOMPANIED by a large rise in blood flow to the muscles engaged in producing locomotion. This rise is largely attributable to a rise in the calculated vascular conductance across the muscles, although a moderate rise in arterial pressure sometimes contributes as well. The rise in calculated vascular conductance at exercise onset is attributable to the mechanical action of the skeletal muscle pump (16, 21, 22, 25) and to the production, release, and diffusion of chemical substances that initiate the relaxation of vascular smooth muscle (19). These two factors work together to raise calculated conductance (21, 25) but are restrained during locomotion by increased activity of sympathetic vasoconstrictor nerves (20, 21). The relative contribution of the muscle pump and metabolic vasodilation in raising blood flow, as well as the identity of the vasodilator substances, are poorly understood (1, 7, 10, 12). A recent addition to the long list of proposed vasodilator substances is nitric oxide (NO), and tests of the importance of this factor in producing active hyperemia have been carried out using substances that inhibit NO synthase (NOS). Most studies have found that the increase in muscle vascular conductance that accompanies treadmill locomotion is little influenced by NOS inhibition alone, but that the absolute values of conductance achieved during locomotion are reduced (8, 14). Such ambiguous results provide an unclear picture as to the importance of NO during locomotion.

In the present study, we sought to determine whether a potent contribution of NO to exercise vasodilation is masked by alterations in autonomic function following NOS inhibition. Because the rise in vascular conductance that accompanies locomotion is normally retrained by the sympathetic nervous system (20, 21), the same total rise in conductance could be achieved in the absence of NO via a reduction in the level of restraint imposed by the sympathetic nerves. A reduction in sympathetic activity is not unexpected, given 1) the rise in arterial pressure that accompanies systemic NOS inhibition, and 2) that baroreceptor reflexes remain active during dynamic exercise (2, 17). In this way, the local vascular control mechanisms that work to ensure an adequate blood supply for the muscle, and the systemic vascular control mechanisms that work to match total vascular conductance and cardiac output (to ensure an adequate blood pressure), could both still achieve their respective goals in the absence of NO production via a balanced alteration in the intensity of the competing influences they exert. To test this possibility, we compared the rise in vascular conductance in active muscle that accompanies treadmill locomotion before and after NOS was inhibited with N-nitro-L-arginine methyl ester (L-NAME) during the time that autonomic ganglionic neurotransmission was blocked by hexamethonium. In addition, we evaluated the effects of L-NAME on the time course of the rise in iliac vascular conductance that accompanies locomotion. The rationale here was that if augmented NO formation in response to locomotion does not contribute to vasodilation, then L-NAME treatment would not be expected to alter the time course of vasodilation. Experiments were carried out in chronically prepared conscious dogs.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All procedures met National Institutes of Health guidelines and were reviewed and approved by the Institutional Animal Care and Use Committee of the University of Iowa. Six mongrel dogs (18-24 kg body wt) of either sex were selected for their willingness to run on a motor-driven treadmill. The dogs were familiarized with treadmill running in a series of training sessions before the following aseptic surgical procedures were performed.

Surgical preparation. Dogs were anesthetized with thiopental sodium, intubated, ventilated, and maintained with halothane. The animals had ultrasonic transit-time blood flow transducers (Transonic, Ithaca, NY) and vascular occluder cuffs placed bilaterally on the iliac arteries through a midline abdominal incision. One dog had these devices implanted on the terminal aorta, and the measured values of blood flow in this dog were on average three times greater than in the other dogs. Blood flows measured in this dog were divided by a factor of 3 before they were averaged with the results from the remaining animals. A catheter was inserted into the aorta for measurement of systemic arterial pressure and in a femoral artery and vein for measurement of hindlimb perfusion pressure and infusion of drugs, respectively. All leads were tunneled to exit sites on the back. Skin patches delivering a total of 75 µg/h of fentanyl were placed on the dogs for 72 h after surgery to control postoperative pain, and the dogs were given cephalexin (500 mg po bid) continually after surgery throughout the time that data were collected.

Data collection. Catheters were connected to pressure transducers (P10EZ, Ohmeda, Madison, WI), and flow transducers were connected to flowmeters (T106, Transonic). Signals were displayed on a pen recorder and were digitized at 250 Hz. Average values of each signal were written to a fixed disk of a microcomputer twice per second.

Experimental protocols. The animals performed treadmill exercise at three intensities (3.2 km/h at 0% grade, 6.4 km/h at 0% grade, and 6.4 km/h at 10% grade) for 3 min in no regular order. The animals were allowed to recover for at least 3 min before exercise was repeated at a different exercise intensity. They were then given 10 mg/kg iv hexamethonium, 0.1 mg/kg iv atropine, and 1 mg/kg iv captopril to block autonomic function and the renin-angiotensin system, respectively (20). Drugs were acquired from Sigma Chemical (St. Louis, MO). The efficacy of these drugs was inferred from the exaggerated fall in arterial pressure that accompanied locomotion after autonomic blockade. The three bouts of exercise were then repeated. Some animals ran for only 1 min at some workloads, owing to reduced exercise capacity after autonomic blockade. Animals were then given 10 mg/kg iv L-NAME to inhibit NO production, and exercise was repeated as described above. The efficacy of NOS inhibition was inferred from the rise in arterial pressure elicited by this drug. On a separate day, animals performed locomotion as described above with and without L-NAME alone.

Data analysis. Iliac vascular conductance was calculated as iliac blood flow divided by arterial pressure. Arterial pressure, iliac blood flow, and iliac vascular conductance, averaged over 15-s periods immediately before the onset of locomotion and at the end of the first minute of locomotion, were taken to represent resting and exercising values, respectively. We evaluated the effects of L-NAME treatment as follows. First, we compared the absolute level of iliac vascular conductance achieved during steady-state exercise with and without L-NAME during the time that autonomic function was blocked. Second, we compared the increase in iliac vascular conductance, calculated as exercising vascular conductance minus resting vascular conductance, with and without L-NAME during the time that autonomic function was blocked. Finally, to evaluate the effects of L-NAME on the time course of the rise in iliac vascular conductance, the rise in iliac vascular conductance that was achieved over the first minute of each exercise bout was rescaled from 0 to 100%. A 3-s running average was applied to these data to smooth the responses, and the time taken to achieve 75% completion of the total rise was identified.

Statistical analysis. The effects of L-NAME treatment were tested statistically by performing multiple linear regressions with a computer spreadsheet program. Dummy variables were used as independent variables to encode treatment effects (drug administration and exercise intensity) and to account for interindividual variability among animals (24). Data are presented as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A representative example of the response to treadmill locomotion at a moderate level of intensity (6.4 km/h at 10% grade) during combined autonomic blockade and NOS inhibition is shown in Fig. 1. At the onset of locomotion, arterial pressure fell and iliac blood flow and vascular conductance rose. In Fig. 1 (bottom panel), the increase in vascular conductance that occurred over the first minute of exercise is rescaled from 0 to 100%, and the time taken to achieve 75% completion of the rise is identified by the arrow. In this example, the animal ran for 3 min. Importantly, it can be seen that the change of each variable in response to locomotion is substantially complete by 45 s of exercise.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Typical response to treadmill locomotion in a single animal that ran for 3 min following combined autonomic and nitric oxide (NO) synthase inhibition. Locomotion caused a decrease in arterial pressure and increases in iliac blood flow and vascular conductance. Arrow depicts time taken for iliac vascular conductance to achieve 75% completion of the total rise achieved in the first minute of exercise.

The effect of autonomic blockade on the response of blood pressure to graded exercise is presented in Table 1. Locomotion elicited marked falls in arterial pressure after autonomic blockade. Group mean data from five dogs depicting the response of arterial pressure are shown in Fig. 2. The data in Fig. 2, A-C, were collected in a single day (n = 5, dogs 1-5), and the data in Fig. 2, D-F, were collected on a separate single day (n = 5, dogs 2-6). Mean arterial pressure values during rest and exercise under the different conditions are plotted versus the values of iliac blood flow measured under control conditions for each day (autonomic function intact, Tables 1 and 2), which were selected as an index that scales linearly with exercise intensity. Figure 2, A-C, depicts the effects of NOS inhibition (dashed lines) during the time that autonomic function was blocked. It can be seen that NOS inhibition elicited substantial increases in arterial pressure. Figure 2, D-F, depicts the effects of NOS inhibition alone (dashed lines), which also raised arterial pressure.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Response of mean arterial pressure to graded treadmill locomotion plotted vs. iliac blood flow, which was selected as an index of relative severity of exercise. A-C: N-nitro-L-arginine methyl ester (L-NAME) treatment following autonomic blockade with hexamethonium (Hex) raised arterial pressure above pressure observed during autonomic blockade alone. D-F: L-NAME treatment alone when autonomic function is intact raised arterial pressure compared with control. See text for further explanation.

The effect of autonomic blockade on the response of iliac blood flow to graded exercise is presented in Table 1. Autonomic blockade elicited moderate reductions in iliac blood flow. Group mean data depicting the rise in iliac blood flow in response to graded treadmill locomotion are shown in Fig. 3. Data are plotted as described for Fig. 2. Figure 3, A-C, depicts the effects of NOS inhibition (dashed lines) during the time that autonomic function was blocked. The statistical tests confirmed what is visually apparent in Fig. 3, A-C: L-NAME did not alter the absolute blood flow achieved during locomotion (P = 0.55) nor the increase in blood flow that accompanied locomotion (i.e., no change in the slope; P = 0.65; n = 5; dogs 1-5). Figure 3, D-F, displays data collected on a separate day and control data (autonomic function intact) are plotted as solid lines. L-NAME treatment alone reduced the absolute level of blood flow achieved during exercise (P = 0.01) but did not alter the increase in blood flow that accompanied locomotion (i.e., no change in slope, P = 0.44, n = 5, dogs 2-6).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Increases in iliac blood flow in response to graded treadmill locomotion plotted vs. iliac blood flow, which was selected as an index of relative severity of exercise. A-C: L-NAME treatment following autonomic blockade with Hex had no effect on iliac blood flow beyond that elicited by autonomic blockade alone. D-F: L-NAME treatment alone when autonomic function was intact reduced iliac blood flow compared with control. See text for further explanation.

The effect of autonomic blockade on the response of iliac vascular conductance to graded exercise is presented in Table 1. Locomotion was accompanied by an exaggerated rise in vascular conductance after autonomic blockade. Group mean data depicting the rise in iliac vascular conductance in response to graded treadmill locomotion are shown in Fig. 4. Data are plotted as described for Fig. 3 except that iliac vascular conductance was selected as an index that scales linearly with exercise intensity. Figure 4, A-C, depicts the effects of NOS inhibition (dashed lines) during the time that autonomic function was blocked. Again the statistical tests confirmed what is visually apparent: L-NAME significantly reduced the absolute level of vascular conductance achieved during locomotion (P = 0.001) and significantly reduced the rise in vascular conductance that accompanied locomotion (i.e., reduced the slope, P = 0.001, n = 5, dogs 1-5). The effects of L-NAME on the absolute values of conductance achieved during graded treadmill locomotion and the change in conductance from rest to exercise are directly compared in Fig. 5. Similar to its effects on blood flow, L-NAME treatment alone (depicted in Fig. 4, D-F) reduced the absolute level of vascular conductance achieved during exercise (P = 0.00001) but did not alter the increase in vascular conductance (i.e., no change in slope, P = 0.47, n = 5, dogs 2-6).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Increases in iliac vascular conductance in response to graded treadmill locomotion plotted vs. iliac vascular conductance, which was selected as an index of relative severity of exercise. A-C: L-NAME treatment following autonomic blockade with Hex caused reductions in both rise in conductance in response to locomotion and absolute value of conductance achieved during exercise compared with autonomic blockade alone. D-F: L-NAME treatment alone when autonomic function was intact reduced the absolute value of conductance achieved during locomotion but did not change the rise in conductance that accompanied locomotion compared with control. See text for further explanation.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Effect of NO synthase inhibition by L-NAME treatment following autonomic blockade with Hex. L-NAME reduced both the absolute value of iliac vascular conductance achieved during graded treadmill locomotion (A: n = 5, dogs 1-5; P < 0.001 for main effect of L-NAME on conductance) and the change in vascular conductance from rest to exercise (B: n = 5, dogs 1-5; P < 0.001 for main effect of L-NAME on change in conductance).

Averaged across the three intensities of locomotion, L-NAME treatment alone (autonomic function intact) significantly lengthened the time to 75% completion of the total rise in vascular conductance achieved in the first minute of locomotion from 5.2 ± 3.2 to 11.0 ± 4.0 s (P = 0.0001, n = 5, dogs 2-6).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We found that absent the potentially confounding influence of the sympathetic system on muscle vascular conductance, L-NAME treatment reduced both the increase in muscle vascular conductance from rest to exercise and the absolute value of vascular conductance achieved during graded treadmill locomotion. Importantly, the absolute levels of blood flow were unaffected by L-NAME treatment under the condition of autonomic blockade, suggesting strongly that the concentration of most other vasoactive substances was unaltered as well. Furthermore, L-NAME slowed the rate of rise of vascular conductance, a finding consistent with the idea that augmented NO formation normally contributes to the vasodilation that accompanies locomotion. To our knowledge, this is the first unambiguous (see the following text) demonstration of an important role for NO in mediating vasodilation of active muscle during locomotion.

A number of previous studies have tested for a contribution by NO in eliciting the vasodilation that accompanies a broad range of modes of exercise. A common finding following NOS inhibition is that the absolute values of muscle vascular conductance are reduced at rest and during exercise, but that the magnitude of the increase in conductance seen in response to exercise is similar to the rise seen when NO function is intact. In agreement, this is what we found, as is shown in Fig. 4, D-F. Interpretation of such findings is problematic. On the one hand, it has been argued that the similar rise in conductance in response to exercise indicates that NO is relatively unimportant in bringing about this response (15); that is, the exercising conductance is simply biased downward in proportion to the reduction in resting conductance. On the other hand, it has been argued that the failure to achieve the same absolute value of conductance (or resistance) during exercise is evidence that NO is an important determinant of muscle vasodilation during exercise (3, 4). The situation is often further complicated by two additional factors. First, the blood flow through the active muscles is sometimes (4, 6, 15) but not always (15, 18, 26) reduced by NOS inhibition, meaning that the concentrations of vasoactive substances that accumulate in a manner dependent on blood flow could sometimes be altered as well. Second, the extent of vasodilation that accompanies many forms of exercise is restrained by the sympathetic nervous system (see Table 1) (9, 21), meaning that potential deficits in the local regulation of vascular conductance after NOS inhibition could be masked by alterations in sympathetic outflow. For example, the tendency for NOS inhibition to raise arterial pressure likely leads to a reduction in sympathetic outflow during exercise, as has been shown when arterial pressure is raised by other pharmacological agents during exercise (17).

The results of our experiment overcome these limitations in a manner that reveals a far less ambiguous picture of the extent to which NO contributes to the vasodilation of active muscle during locomotion. Inasmuch as autonomic function was blocked, alterations in sympathetic outflow could not confound our results. Furthermore, NOS inhibition during the time that autonomic function was blocked did not alter the absolute values of iliac blood flow at rest or during exercise (Fig. 3, A-C). This means that the concentration of most other vasoactive substances was likely to be similar (except perhaps for other substances released in response to increases in the shear stress imposed on vascular endothelium; see below). Thus it is most likely that the differences we observed in iliac vascular conductance were due exclusively to the presence or absence of NO. Finally, within this favorable setting, we found that the absolute value of vascular conductance achieved during locomotion and the rise in vascular conductance that accompanied locomotion were both reduced by NOS inhibition.

Efforts to evaluate the effects of NOS inhibition on the time course of hemodynamic adjustments to exercise are less common. Shoemaker et al. (23) found that the combination of atropine and L-NAME did not alter the time course of the rise in blood flow accompanying exercise. We found that NOS inhibition slowed the time course of the rise in iliac vascular conductance when autonomic function was intact, i.e., L-NAME nearly doubled the time to 75% completion of the rise in conductance. This is further evidence that NO plays a significant role in bringing about the rise in conductance that accompanies locomotion; if augmented NO formation in response to locomotion did not induce vasodilation, then L-NAME treatment would not be expected to alter the time course of the rise in conductance. Thus it appears that attenuated release of NO following L-NAME treatment constitutes a primary deficit in the regulation of muscle vascular conductance in response to locomotion. Our data collected during autonomic blockade are consistent with the idea that less intense sympathetic vasoconstriction ensues secondary to the slowed vasodilation, thereby permitting the same rise in conductance to persist after L-NAME administration when autonomic function is intact.

Taken together, the foregoing arguments indicate that NO makes a potent contribution to the vasodilation in active muscle that accompanies locomotion. On average we found that L-NAME reduced the absolute level of iliac conductance during locomotion by 30% and reduced the rise in conductance by 32%.

In evaluating the role of NO in bringing about the local circulatory adjustments to locomotion within muscle, we have focused our analysis on muscle vascular conductance during autonomic blockade. When autonomic function is blocked, presumably only control mechanisms residing within the active muscles influence vascular tone. Thus our analysis likely provides a direct indication of the effects of NOS inhibition on the vasodilator function in the active muscle. Others have evaluated the effects of NOS inhibition on muscle blood flow during exercise (e.g., 8, 14) as well as on vascular conductance or resistance. However, during locomotion at least, any difference in the magnitude of blood flow achieved after systemic NOS inhibition will depend primarily on the flow generated by the heart and depend secondarily on how NOS inhibition affects the distribution of cardiac output among organs competing for flow. For example, muscle blood flow during exercise could be increased by systemic NOS inhibition despite reduced vasodilation in muscle. This could occur if NOS inhibition elicited a relatively greater constriction within inactive organs and cardiac output was unaltered (or increased). After autonomic blockade, we found that muscle blood flow was unaltered by L-NAME despite an attenuated rise in muscle vascular conductance during locomotion. A likely explanation for this lack of change in iliac blood flow after L-NAME administration includes 1) the relative insensitivity of stroke volume to increases in afterload (5), and 2) possible beneficial effects of systemic NOS inhibition on cardiac filling pressure (13).

The magnitude of the contribution of NO during locomotion may be somewhat underestimated in our experiment. As noted above, L-NAME did not alter the levels of iliac blood flow achieved during locomotion beyond that caused by autonomic blockade alone. The constancy of flow, coupled with the attenuated rise in muscle vascular conductance induced by NOS inhibition, is expected to increase the shear stress on the endothelial cells during locomotion compared with autonomic blockade alone. This in turn is expected to lead to an augmented release of other shear-stress-sensitive vasodilator substances such as prostaglandins (11). An augmented release of these substances would be expected to offset the deficits in the regulation of vascular conductance arising from the absence of NO. Our results demonstrate that other shear-stress-sensitive vasodilator substances cannot completely offset the absence of NO; thus these other substances are not redundant but are likely to be acting in a compensatory fashion.

In conclusion, augmented formation of NO in response to locomotory exercise normally provides an important drive to the relaxation of vascular smooth muscle within active skeletal muscle. When NOS alone is blocked, potential deficits in the regulation of muscle vascular conductance stemming from NOS inhibition are masked by withdrawal of sympathetic vasoconstrictor nerve activity such that conductance increases normally. When this confounding influence of sympathetic function is removed, the substantial (30%) reductions in both the absolute values of muscle vascular conductance during locomotion and the rise in conductance accompanying locomotion reveal the true importance of the contribution of NO to exercise hyperemia.