INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IT IS WIDELY RECOGNIZED that a lifestyle that incorporates moderate levels of physical activity is therapeutic for prevention and treatment of cardiovascular disease. An increase in blood flow capacity of skeletal muscle vascular beds (8, 9, 19, 20, 31, 36, 37) and altered distribution of skeletal muscle blood flow during exercise (1, 36) are among a host of adaptive changes in the cardiorespiratory system produced by exercise training. The adaptations in skeletal muscle vascular beds that produce an increase in blood flow capacity of skeletal muscle include remodeling the arterial tree (structural vascular adaptation) and altered mechanisms of control of vascular conductance within and among muscles (17). The intrinsic properties of both endothelium and smooth muscle are altered by exercise training so that vasomotor reactivity of skeletal muscle arteries and exercise hyperemia are both improved by exercise training (1, 15, 27, 44). There is a growing body of evidence that training can improve endothelium-dependent vasodilation of arteries supplying blood to the skeletal muscle and that this change in endothelial function is at least partially mediated by increased expression of endothelial nitric oxide (NO) synthase (eNOS) in endothelium of arteries providing blood flow to skeletal muscle (3-5, 14, 16, 22, 25, 34, 39, 43, 45).

Exercise training-induced adaptations in normal animals also improve skeletal muscle blood flow immediately following acute-onset peripheral arterial insufficiency caused by femoral artery occlusion (51). These adaptations appear to increase collateral conductance through structural remodeling and/or altered vasoresponsiveness of vessels that become the collateral arterial circuit upon acute-onset peripheral arterial insufficiency (51). It is reasonable to propose that the chronic, daily increases in blood flow through the iliac artery and its branches over the training period produce an increased potential for endothelium-mediated vasodilation in the circumventing circuit, which contributes to the enhanced collateral blood flow. The purpose of the study reported here was to test the hypothesis that increased release of NO by NOS (likely eNOS) contributes to the increased collateral-dependent blood flow to skeletal muscle of exercise-trained rats, following acute femoral artery occlusion. We reasoned that if the improved collateral blood flow resulted from enhanced endothelium-dependent dilation of collateral circuit arteries, blockade of NOS with N^G-nitro-L-arginine methyl ester (L-NAME) would abolish the improved blood flow. On the other hand, if collateral flow is increased by prior training simply by remodeling and enlarging the circumventing arteries (structural adaptation), then treatment with L-NAME should have similar effects in sedentary and trained rats. Results presented below confirm that prior exercise training improves collateral-dependent blood flow to skeletal muscle of rats after acute femoral artery occlusion and show that this improvement results primarily from an increased conductance of the upstream collateral circuit. Importantly, our results reveal treatment with L-NAME decreased vascular conductance, in both the upstream collateral circuit and in the skeletal muscle microcirculation. The increased vascular conductance of the collateral circuit in trained rats was abolished by treatment with L-NAME. These results imply that enhanced endothelium-mediated dilation induced by training serves to increase collateral blood flow following acute arterial occlusion.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiment design. Rats were randomly divided into sedentary (28) or exercised (n = 30) groups. Rats in trained group were subjected to daily treadmill exercise for 6 wk. After 6 wk of training, rats from each group received either L-NAME (20 mg/kg body wt, intra-arterial injection; Sigma; St. Louis, MO) or the same volume of saline (control) 30 min before blood flow measurement. Collateral circulation function was assessed by measuring collateral-dependent blood flow 4 h after bilateral femoral artery ligation and arterial catheterization. The experimental design allowed direct assessment of the main treatment effects of NOS inhibition on prior physical training-induced collateral vascular function.

Animal care. Male Sprague-Dawley rats weighing ~325 g (Taconic Farms; Germantown, NY) were housed two per cage in a temperature (21°C) and light (12 h-12 h dark/light cycle) controlled room. Rats were fed Purina Rat chow and water ad libitum at all times throughout the experiment. On arrival, all rats were accustomed to handling and daily treadmill walking for 5-10 min at 20 m/min of 15% grade for ~5 days. The treadmill protocol included turning the treadmill on and off briefly to condition the rats to run at the front of the treadmill when the treadmill belts started moving. Previous results reported from our laboratory indicated that there is no detectable sign of peripheral adaptations in the animals under this treadmill conditioning protocol (6).

Physical training. Animals in the trained group were walked on a motor-driven rodent treadmill (Quinton model 42-15) as done previously (11). A daily progressively intense treadmill training program was applied. Rats ran at 30 m/min at 15% grade at the beginning of the exercise program; the exercise duration increased from ~20 min/day to ~55 min/day during the 6-wk training duration. When rats were able to run 60 min, then 1-min sprints at 60 m/min were introduced, beginning the 60th min and progressing to the 50th, 40th, etc. The total run time (min) was recorded for each rat.

Surgical preparation for femoral artery ligation and catheterization. With the rat under ketamine-acepromazine (100 mg/0.5 mg per kg body wt) anesthesia, both left and right femoral arteries were isolated and ligated with 3-0 surgical silk sutures about 5-6 mm distal to the inguinal ligament as done previously (53, 54). A polyethylene-50 catheter was placed in the left carotid artery and advanced to the arch of the aorta for delivering L-NAME or saline, for monitoring blood pressure and heart rate, and for infusing radioactive microspheres. A second catheter was placed in the caudal artery for monitoring arterial blood pressure and obtaining the reference blood sample during microsphere infusion. A third catheter was inserted into the left distal femoral artery with the tip of catheter just above the knee. This catheter monitored the distal blood pressure past the collateral vascular circuit before or during exercise. All catheters were exteriorized at the back of the neck. The arteries were catheterized early in the day. The fully alert animals were run on the treadmill for blood flow measurement following more than 4 h recovery, as done previously (48).

L-NAME administration. The NOS inhibitor L-NAME was dissolved in saline at 2 g/100 ml and administered intra-arterially (20 mg/kg body wt) 30 min before the blood flow measurement. The effectiveness of NOS inhibition was confirmed by increased systemic blood pressure.

Blood flow determination. Collateral blood flow was assessed under vasodilation conditions induced by treadmill running.

1

1

Resistance and conductance calculations. Collateral circuit resistance was calculated as the pressure drop across the upper hindlimb (P_caudal  P_{distal femoral artery}) divided by the blood flow to the distal limb tissues. Distal tissue resistance was calculated as the pressure drop across the distal tissue (P_{distal femoral artery}  P_venous, which was taken to be zero) divided by distal tissue blood flow. Conductances were calculated as the reciprocal of resistance.

Data analysis. All data are expressed as means ± SE. ANOVA was used to assess the main treatment effects of L-NAME and physical training and the L-NAME-physical training interactions. ANOVA with repeated measures was applied to data when applicable. P < 0.05 was recognized as a significant difference. The treatment differences across groups were determined by Newman-Keuls Multiple-Range tests (2).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effectiveness of training program. Exercise-trained rats exhibited lower body weights and lower hindlimb muscle weights than sedentary rats in both the control and L-NAME-treated groups (Table 1). Also, both groups of exercise-trained rats exhibited increased citrate synthase activity in the white quadriceps muscle relative to their respective sedentary control values (Table 1). Finally, improvement in treadmill run time (Fig. 1) observed over the first 2 wk of training was similar to that reported previously with this training program (51).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Daily run time of trained animals during the study.

Heart rate and blood pressure. Heart rates and blood pressures were similar in sedentary and exercise-trained rats preexercise and during exercise (Table 2). L-NAME treatment increased blood pressure in both sedentary and exercise-trained rats (Table 2). The increase tended to be greater in the exercise-trained rats. Of interest, pressure in the distal femoral artery was less in trained rats treated with L-NAME. There was a main treatment effect of exercise to increase heart rates at both the low and high treadmill speeds.

Muscle blood flow. As has been our experience, there was excellent agreement between left and right kidney blood flows and between muscles of the left and right limbs. These data indicate that labeled microspheres were properly mixed into cardiac output for measurements of regional blood flows. As shown in Table 3, calf muscle blood flows were similar within groups during high and low speed exercise. Because blood flows were similar for corresponding tissues across limbs and similar at both running speeds, the data have been combined for ease of presentation. Blood flow data were not obtained from all animals entered in the study due to common surgical and/or catheter problems and inadequate running performance, usually during the second higher, running speed. The number of observations of each group is noted as n in the tables and figures.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Collateral dependent blood flow to calf muscles while rats ran on the treadmill. *Greater than all other groups (P < 0.05); **greater than NG-nitro-L-arginine methyl ester (L-NAME) sedentary group (P < 0.05).

Vascular resistance and conductance. Collateral circuit and distal tissue resistances (R_collateral and R_{distal tissue}) and their corresponding conductances were calculated as illustrated in Fig. 3. There was a main treatment effect of training to decrease R_collateral (P < 0.05) and of L-NAME to increase R_collateral (P < 0.001; cf., Fig. 4). R_{distal tissue} was decreased by training (P < 0.05) in the absence of L-NAME; all other groups are not different.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Schematic of vascular circuit used to calculate resistances (R) and conductances (C) of the collateral circuit and distal tissue. P, pressure; Q, blood flow.

View larger version (37K): [in this window] [in a new window]   Fig. 4.   Resistances of the collateral circuit and distal tissue. Collateral circuit resistance, as a percentage of total circuit resistance, is given for each group at top of each bar. Main treatment effect of training to reduce Rcollateral circuit (P < 0.05); *Rcollateral circuit less than L-NAME groups (P < 0.05); **Rdistal tissue less than all other groups (P < 0.05).

View larger version (34K): [in this window] [in a new window]   Fig. 5.   Conductances of collateral circuit and distal tissue. *Cdistal tissue greater than all other groups (P < 0.05); **Ccollateral circuit greater than all other groups (P < 0.05); Ccollateral circuit greater than L-NAME sedentary group (P < 0.05).

Blood flows to nonhindlimb tissues. Kidney blood flows were greater (P < 0.001) in the exercise groups (Table 5). After L-NAME treatment, blood flows to the kidneys were decreased in both groups, but blood flow to the nonhindlimb muscles was not significantly changed. Blood flow to nonhindlimb muscles were similar among trained and sedentary groups, with or without L-NAME treatment.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mechanisms for vascular adaptation induced by chronic exercise training in skeletal muscle can be grouped into two major categories: structural adaptations and adaptations in vascular control. Structural adaptations occur in response to exercise training in at least two forms: vascular remodeling and growth (i.e., growth of vessels such as increased length and cross-sectional area and/or diameters of the existing large and small arteries and veins) and angiogenesis (i.e., increased numbers of capillaries and other microvessels per gram of muscle). Adaptive changes in vascular control can be the result of altered neurohumoral control of the vascular bed or changes in intrinsic properties of vascular smooth muscle cells and/or endothelial cells of the arteries (16, 18, 30). Present results indicate that exercise training of normal rats produce adaptations in vascular control in the arterial circuit, which becomes the upstream component of the collateral circulation following acute femoral occlusion.

Blood flow is provided to skeletal muscle by branching networks of arteries that connect the aorta to the capillaries. The relative importance of vascular control mechanisms is not constant along the length of the arterial network; indeed, endothelium-mediated control may be of greater significance in the conduit arteries that form circumventing circuit, including the internal iliac artery. Consistent with this notion is evidence that NO is of greater importance in larger arteries than in small resistance arteries (7, 10). Also, NO is the primary vasodilator substance released by the endothelium in larger arteries (21), whereas in smaller arteries and arterioles prostaglandin production also appears to be important in determination of tone (10, 14, 32). Exercise training increases transport capacity of both the conduit/feed arteries and resistance arteries/arterioles of skeletal muscle (16). Delp and colleagues (3-5) reported that a moderate intensity, endurance training program did not alter lumen diameters or wall thickness of abdominal aorta of rats. However, Segal and colleagues (33) reported that treadmill exercise training in rats produced 12-18% increases in the medial wall thickness in the abdominal aorta, femoral, axillary, superior mesenteric, and coeliac arteries and increased total wall area in the aorta and femoral and axillary arteries. There were no differences between lumen diameter in any of these arteries (33). Huonker and colleagues (13) reported that the diameter of femoral arteries of endurance-trained humans is greater than the diameter of femoral arteries in sedentary adults, and Miyachi and colleagues (24) reported similar results for the cross-sectional area of the aorta in young healthy men. There is limited information concerning the effects of exercise training on the size and/or number of small arteries and arterioles in skeletal muscle vascular beds. The fact that precapillary vascular resistance is decreased in skeletal muscle of trained rats suggests that training induces changes in precapillary resistance vessels (19, 35-37).

The present results demonstrate that, in the presence of bilateral femoral artery occlusion, calf muscle blood flow is 90% greater (P < 0.001) during exercise in rats previously trained than in rats kept sedentary (Fig. 2). This study confirms and extends our previous observation (51) in showing that this improvement in collateral blood flow is dependent on NOS activity. Because the microvascular arteries and arterioles of the calf muscles are expected to be near maximally dilated during exercise at these intensities in rats with bilateral femoral occlusion, we proposed that the training-induced adaptations are focused in the vessels of the iliac arterial tree and not the microvasculature of the calf muscles (51). In the present study we measured distal femoral pressures to more rigorously evaluate this hypothesis. Figure 3 illustrates a hemodynamic schematic that allows analysis of the relative contributions of the distal tissue vascular resistance and that of the upstream collateral circuit. By measuring distal femoral pressure, as shown, we can divide resistance into collateral (R_c) and distal resistance (R_distal). With the use of blood flow data for the distal hindlimb, arterial pressure (P_a) and distal femoral pressure (P_distal), both R_c and R_distal can be calculated as presented in Fig. 4. There are several noteworthy aspects. First, a comparison of scales in Fig. 4 demonstrates that R_c is the major site of resistance determining blood flow downstream, amounting to 75-85% of the total resistance. Thus a reduction in R_c would have the greatest impact to increase blood flow to the calf muscle. Second, the results presented in Fig. 4 indicate that training significantly decreased R_c (P < 0.05). This implies that meaningful adaptations were induced in the conduit vessels of the thigh by prior exercise training. Third, L-NAME inhibition of NOS increased R_c (P < 0.001) in both sedentary and trained animals. This implies that NO-sensitive dilation of upstream collateral conduits was important in determining downstream collateral blood flow during treadmill running in both sedentary and trained animals. This finding for normal sedentary rats confirms the work of Unthank and colleagues (41, 42). Interestingly, whereas NOS inhibition increased vascular resistance, it did not appear to alter the distribution of resistance in the entire circuit (Fig. 4). Fourth, prior training significantly decreased R_distal (P < 0.005). Whereas this decrease in resistance would contribute to increased collateral blood flow, it accounted for only about one-third of the flow increase to the distal muscles of trained rats, because R_distal is not the major resistance in the circuit. Thus, relative to increasing blood flow to calf skeletal muscle, the most important site for vascular adaptation is the upstream collateral circuit.

Applying the same hemodynamic schematic (Fig. 3), we calculated collateral circuit conductance (C_c) and distal tissue conductance (C_distal) using blood flow data for the distal leg, arterial pressure (P_a), and distal femoral pressure (P_distal) as presented in Fig. 5. The directional differences between groups revealed in these conductance data (Fig. 5) are generally consistent with the analysis of the resistance data (Fig. 4). Thus, in the absence of L-NAME, C_c was greater in trained rats than sedentary rats. L-NAME treatment reduced C_c to similar values in both trained and sedentary groups. This result emphasizes the importance of increased NOS-generated NO as a determinant of the training-induced increase in C_c during exercise. It is also interesting that L-NAME produced a significant decrease in C_distal of trained rats but not sedentary rats. This indicates that prior exercise training also resulted in an increase in the relative importance of NOS-generated NO as a determinant of C_distal during exercise. However, C_distal was similar in sedentary and trained rats during L-NAME treatment.

The results of this study support the interpretation that the exercise training-induced improvement of blood flow during exercise in femoral occluded rats is primarily the result of decreased resistance of the collateral circuit. It is also apparent that the large decreases in conductance, produced by L-NAME treatment, were centered in the collateral arteries upstream from the distal tissue, in exercise-trained rats (Fig. 5). We believe that this interpretation of our results is not confounded by nonspecific systemic effects of L-NAME treatment (e.g., blood pressure, resting muscle blood flow, and oxygen consumption, and/or sympathetic outflow), because of internal comparisons within our data set showing that L-NAME treatment had minimal effects on blood flow to normal, nonischemic tissues. For example, blood flows to the normal nonischemic muscles (diaphragm, abdominal, and psoas) of the sedentary animals were similar to that measured in the sedentary L-NAME animals (cf., Table 5). Similarly, comparisons between these groups illustrate that L-NAME did not alter renal blood flows (cf., Table 5). In contrast, the L-NAME effect on collateral-dependent tissue (e.g., calf muscle) was substantial. It appears more likely that these effects of L-NAME in the collateral-dependent tissues resulted from inhibition of NO synthase. Thus we believe that there is little compelling evidence for a generalized alteration in blood flow control to active muscle or a problematic nature of untoward effects of L-NAME that undermine the basic conclusions of this study.

Whereas prior exercise training exerts meaningful vascular adaptations to improve collateral blood flow immediately following femoral artery occlusion, this is different from the response of animals that are exercise trained after femoral artery occlusion. The training of rats with peripheral arterial insufficiency induces marked increases in collateral blood flow to the calf muscle providing a much greater absolute blood flow than the effects observed in rats trained without peripheral ischemia, as in this study (52, 54). Dramatic structural enlargement of collateral vessels appears to be a major characteristic accounting for increased absolute blood flow in claudicant rats that are trained after occlusion (52, 54).

It is interesting that distal femoral artery pressures were not increased in the exercise-trained rats relative to that of sedentary rats (Table 2). Given similar central arterial pressures and the decrease in R_c, one would expect higher distal femoral artery pressures in the exercise-trained rats. However, this prediction ignores the contribution of kinetic energy to the hemodynamic environment within the limb vasculature. It is appropriate to ignore kinetic energy, as defined by Bernoulli's principle, under normal hemodynamic conditions. However, when blood flow occurs through small circumventing channels (i.e., arising from the internal iliac) into the larger diameter distal saphenous and popliteal arteries, significant conversions can occur between potential (measured distal artery pressures) and kinetic energy. We have observed, from radiographs of arterial casts of the collateral vascular tree of femoral artery-occluded rats (47), that numerous small arteries arising from the internal iliac artery converge into two large vessels in the distal hindlimb, the popliteal and saphenous arteries. Thus under these conditions it is important to consider the effects of Bernoulli's principle where kinetic energy in the vascular circuit is primarily a function of the vessel cross-sectional area and blood flow. Because we do not know the actual diameters of the popliteal and/or saphenous arteries in the animals during exercise nor the actual distribution of blood flow delivered through the collateral vessels, we cannot precisely calculate kinetic energy. However, as predicted from the increases in blood flow to the calf muscles, the trained animals are expected to have a much greater kinetic energy during exercise. Thus the total energy (i.e., kinetic energy + P_distal) in the artery of the trained rats is expected to be greater than that in the sedentary rats, for any given diameter of the popliteal and/or saphenous arteries. Future work will be necessary to test this hypothesis.

Whereas we cannot be certain of the precise isoform of NOS that is involved, the present findings are consistent with a growing body of evidence that exercise training produces increased endothelium-mediated dilation and increase expression of eNOS in dog aortas (34, 43), rat aortas (3-5), porcine coronary resistance arteries (25, 45), and rat gracilis muscle resistance arteries (14, 39). Many believe that exercise training produces this adaptation as a result of increases in muscle blood flow during exercise bouts that generate a "shear stress" signal for vascular adaptation to exercise training (23, 28, 34, 43). Increased transcription of the eNOS gene and increased eNOS protein expression have been reported in response to increases in shear stress in cultured endothelial cell monolayers (29, 30) and isolated coronary arterioles (46). Also, chronic increases in blood flow produced by arteriovenous fistulas in dogs (23) and rats (28) have been shown to result in increased transcription of the eNOS gene and increased eNOS protein expression. Present results are consistent with the hypothesis that exercise training increases endothelium derived NO in the internal iliac arterial tree increasing vascular conductance in this portion of the collateral circuit.

If as these results suggest, exercise training increases the NO production ability of the iliac arterial tree, there could be other influences beyond the short-term control of vascular conductance as illustrated in Fig. 5. For example, NO production via NOS activity contributes importantly to angiogenesis in response to tissue ischemia (26) and chronic muscle stimulation (12). More germane to the context of the present study, NO production appears to be essential to the remodeling of conduit arteries in response to increased flow (40) and administration of angiogenic growth factors (49, 50). As a result, this apparent increased capacity for NO production by the iliac arterial tree may also translate into a greater and possibly a more rapid remodeling of the collateral circuit in response to arterial insufficiency. This contingency is supported by the additive effect of exercise training and basic fibroblast growth factor administration (52).