Interval sprint training enhances endothelial function and eNOS content in some arteries that perfuse white gastrocnemius muscle

M. H. Laughlin, C. R. Woodman, W. G. Schrage, D. Gute, E. M. Price

Abstract

The purpose of this study was to test the hypothesis that interval sprint training (IST) selectively increases endothelium-dependent dilation (EDD) and endothelial nitric oxide synthase and/or superoxide dismutase-1 protein content in arteries and/or arterioles that perfuse the white portion of rat gastrocnemius muscle (WG). Male Sprague-Dawley rats completed 10 wk of IST (n = 62) or remained sedentary (Sed) (n = 63). IST rats performed six 2.5-min exercise bouts, with 4.5 min of rest between bouts (60 m/min, 15% incline), 5 days/wk. EDD was assessed from acetylcholine (ACh)-induced increases in muscle blood flow measured in situ and by ACh-induced dilation of arteries and arterioles [first to third order (1A–3A)] that perfuse red gastrocnemius muscle (RG) and WG. Artery protein content was determined with immunoblot analysis. ACh-induced increases in blood flow were enhanced in WG of IST rats. eNOS content was increased in conduit arteries, gastrocnemius feed artery, and fourth-order arterioles from WG and fifth-order arterioles of RG but not in 2As from RG. EDD was examined in 2As and 3As from a subset of IST and Sed rats. Arterioles were canulated with micropipettes, and intraluminal pressure was set at 60 cmH2O. Results indicate that passive diameter (measured in 0 calcium PSS) of WG 2As was similar in IST and Sed, whereas diameter of WG 3As was greater in IST (96 ± 8 μm) than Sed (73 ± 4 μm). WG 2As and 3As of IST rats exhibited greater spontaneous tone, but sensitivity to stretch, phenylephrine, and sodium nitroprusside was similar to Sed arterioles. ACh-induced dilation was enhanced by IST in WG 2As but not in RG 2As or WG 3As. We conclude that IST induces vascular adaptations nonuniformly among arteries that perfuse WG muscle.

  • superoxide dismutase
  • blood flow
  • endothelial-derived factors
  • exercise
  • endothelial nitric oxide synthase

exercise training produces improved cardiorespiratory performance in health and improved quality of life in disease (6, 37). Key components in training-induced increases in cardiorespiratory performance during exercise appear to be increases in skeletal muscle blood flow capacity and altered control of total and regional muscle blood flow during exercise (23, 37, 38). Phenotypic expression of skeletal muscle fiber type also has powerful influences on vascular design and vascular control in skeletal muscle, which may influence training-induced vascular adaptations in skeletal muscle (3, 9, 21, 23, 28, 29).

Skeletal muscle tissue is composed of fibers that can be grouped into three classes on the basis of their contractile and metabolic properties (38). Slow-twitch oxidative (SO) fibers have metabolic machinery capable of synthesizing ATP aerobically and a vascular support system that delivers sufficient oxygen to support prolonged, often continuous, activity. At the other end of the skeletal muscle phenotype spectrum are fast-twitch, glycolytic (FG) fibers, characterized by a more anaerobic metabolic profile and a limited vascular support. FG fibers execute brief but forceful contractions. Fast-twitch, oxidative, glycolytic (FOG) fibers are intermediate in contractile properties and possess metabolic and vascular support systems similar to those of SO fibers. There is evidence that vascular structure and function are matched to muscle fiber characteristics, within skeletal muscle (3, 16, 21, 23, 25). There is also evidence that the relative importance of control mechanisms differ among vasculature in muscles composed of different fiber types (13, 15, 22, 23, 39) and that the relative importance of endothelium-mediated influences on vascular resistance varies with muscle fiber type (13, 15, 22, 23). Intrinsic function of vascular smooth muscle and/or endothelium of skeletal muscle arterioles is different in skeletal muscle composed of different fiber types (1, 2, 30, 33).

It seems reasonable to propose that fiber type-related differences in intrinsic function of smooth muscle and endothelium of skeletal muscle arterioles interact with muscle fiber recruitment patterns during exercise so that the distribution and type of adaptive changes induced by training vary with the type of training preformed. For example, high-intensity exercise (high speed, uphill running) has been shown to produce the greatest relative increase in contractile activity in fast-twitch, white skeletal muscle, like the white portion of the gastrocnemius muscle (WG) (11, 19). After interval sprint training (IST), WG exhibits the largest relative increase in oxidative capacity (11), capillary density (14), and blood flow capacity of the extensor muscles examined in rats (24, 41, 42). The increased blood flow capacity of WG does not appear to be mediated solely by increased capillarization (14, 42). It is possible that blood flow capacity is increased because of improved vasodilator capacity, perhaps as a result of improved endothelium-dependent vasodilation.

One potential mechanism for improved endothelium-dependent dilation (EDD) of arteries is increased release of and/or bioavailability of nitric oxide (NO) as a result of increased expression of endothelial NO synthase (eNOS) and/or superoxide dismutase-1 (SOD-1) in the endothelial cells of the arteries (8, 26). On the basis of these observations, the purpose of the present study was to test the hypothesis that a program of IST selectively increases acetylcholine (ACh)-induced EDD (responses to ACh) and increases eNOS and/or SOD-1 protein content in arteries and/or arterioles that perfuse the WG. In addition, because exercise training has been reported to alter vasomotor responsiveness of vascular smooth muscle (4, 5, 32, 35), we examined responses of isolated arterioles and feed arteries (FAs) to changes in transmural pressure (myogenic responses), phenylephrine (PE; endothelium-independent constrictor agent), and sodium nitroprusside (SNP; endothelium-independent dilator).

METHODS

Animals

Male Sprague-Dawley rats were obtained from Harlan at ∼6 wk of age in 5 groups of 25. Animals were housed in pairs in temperature (24°C)- and light (12:12-h light-dark cycle)-controlled rooms. Rat chow and water were available ad libitum. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Missouri.

IST Program

For each group of rats, all 25 rats began treadmill running 1–2 wk after delivery from the breeder. The rats ran for 10–30 min/day, 5 days/wk, at a speed of 30 m/min with 0° incline. After 2 wk of treadmill familiarization the five rats from each group that did not easily adapt to treadmill exercise were removed from the study. The remaining 20 rats were randomly assigned to sedentary control (Sed) or IST groups. Sed rats were restricted to their cages and only placed on the treadmill with no exercise for 5 days/wk, whereas the IST group continued the training program. Duration and intensity of exercise were gradually increased so that by 3–4 wk the rats were able to run six alternating bouts of 2.5 min of running, at 60 m/min, up a 15% grade, and 4.5 min of rest between bouts, 5 days/wk. IST rats continued training at this intensity for 6–8 wk. This training program has been demonstrated to produce ∼50% increases in oxidative capacity, increased blood flow capacity and increased capillarity in extensor muscles composed of FG fibers, like WG (14, 24, 41, 42).

Experimental Design

We conducted three experiments to test the hypothesis that IST increases EDD in the muscle tissue with the greatest relative increase in activity during training bouts (WG) and increases eNOS and SOD-1 protein content in arteries and/or arterioles that perfuse and/or are located in the WG. First, we measured ACh-induced increases in muscle blood flow in isolated perfused rat hindquarters and measured the regional distribution of blood flow in situ with the radiolabeled-microsphere technique. We observed increased ACh-induced dilation that was of greatest magnitude in WG of IST rats. Therefore, we determined the distribution of IST-induced increases in eNOS protein content in arteries and arterioles that perfuse gastrocnemius muscle. Finally, on the basis of these results, we examined EDD in a subset of arterioles where IST increased eNOS expression and arterioles where eNOS expression was not altered by IST to determine the functional significance of these alterations eNOS expression.

Citrate synthase assay. After euthanasia, samples of red and white portions of the vastus lateralis muscle were obtained for biochemical analysis of citrate synthase activity to determine training effectiveness. We examined red and white portions of the vastus lateralis muscle because arterioles were isolated from the triceps surae skeletal muscles. Muscle samples were frozen in liquid N2 and stored at -70°C until processed. Citrate synthase activity was measured from whole muscle homogenates by using the spectrophotometric method of Srere (43).

Endothelium-Dependent Increases in Muscle Blood Flow

We used two techniques to examine effects of EDD on muscle blood flow in IST rats. First, we used isolated perfused rat hindquarters in which blood flow was measured during isogravimetric conditions. Because we observed enhanced ACh-induced increases in blood flow in IST hindquarters, we used the radiolabeled-microsphere technique in another group of rats to measure regional blood flow and determine whether IST produced greater changes in WG than in RG muscle. Total blood flow was measured with a Transonic flowmeter and flow probes, calibrated with a graduated cylinder and stopwatch. Regional blood flow was measured with the radiolabeled-microsphere technique as described by Woodman et al. (44). The purpose of this experiment was to test the hypothesis that IST enhanced endothelium-mediated vasodilation in the muscle tissue with the greatest relative increase in blood flow during training bouts (WG). Specifically, our hypothesis was that ACh infusion into the femoral artery would produce greater increases in blood flow to the WG. ACh was selected for this study because it is known to be a potent endothelium-dependent dilator in rats and because it has minimal systemic effects due to acetylcholinesterase activity. Thus we were able to produce maximal ACh-induced vasodilation of one leg without producing decreased blood pressure and the associated reflex responses.

Perfused hindquarters experiments. The isolated, perfused rat hindquarters preparation was used as described previously (24, 27, 41). Rats were anesthetized with pentobarbital sodium (65 mg/kg ip). The skin, abdominal muscles, and viscera were then divided from the hindquarters, and the aorta and vena cava were isolated. Heparin was administered (1,000 IU iv), and the aorta was cannulated with PE-160 tubing and perfused with Tyrode solution containing 5% bovine serum albumin by using a Gilson Minipuls 2 (Middleton, WI) peristaltic pump (24, 27, 41). The caudal vena cava was canulated with PE-205 tubing, and outflow was collected for reperfusion. Arterial and venous pressures were measured with Statham P23 AC pressure transducers. Arterial pressure was measured from a cannula in the tail artery and venous pressure from a side branch of the venous catheter. The hindquarters were placed on a metal grid and suspended from a Grass Instruments model FT03C transducer (Quincy, MA) to measure hindquarters weight. After suspension of the hindquarters, perfusate flow was adjusted to keep weight constant, isogravimetric.

After the preparation had remained isogravimetric for 15 min, PE was added in 1 μM increments to the perfusate until arterial pressure reached 70 mmHg. ACh-induced dilation was examined in a dose-response fashion under constant flow conditions. ACh was infused intra-arterially at 0.001, 0.01, 0.1, and 1.0 μg/min. During each infusion condition, perfusion pressure was allowed to stabilize before the ACh dose was increased. The ACh dose-response relationship was reexamined 20 min after the addition of NG-nitro-l-arginine methyl ester (l-NAME; 300 μM) to the perfusate. The ACh dose-response relationship was subsequently examined 10 min after the addition of indomethacin to the perfusate (2.8 μM). Finally, to examine endothelium-independent dilator responses the hindquarters was exposed to infusion (ia) of SNP at 0.1, 1.0, 10, and 100 μg/min.

Regional blood flow experiments. To examine blood flow with the microsphere technique, rats were instrumented with catheters for administering microspheres and obtaining a reference withdrawal sample. Rats were also instrumented with a Transonic flow probe on the left femoral artery and an arterial catheter in the femoral artery via the epigastric artery, for infusion of drugs (44). Experiments were conducted on two groups of rats: IST and Sed. In each animal, heart rate, aortic pressure, total (femoral flow probe), and regional (microspheres) blood flow were measured under baseline conditions and during intra-arterial infusion of ACh into the femoral artery. On the basis of pilot studies, we infused ACh starting at a rate of 200 ng·kg-1·min-1 and increased dose by 200 ng·kg-1·min-1 up to doses of 1,600–2,000 ng·kg-1·min-1, where we observed maximal effects. Regional blood flows were measured under baseline conditions, during ACh infusion at rates that produced 50% maximal blood flow, and during infusion of ACh at rates that produced maximal ACh-induced increases in blood flow as measured with the flow probes.

After the last microsphere infusion during ACh infusion, soleus and gastrocnemius muscles were removed for the determination of blood flow. After the wet weight of the samples was determined, the samples were placed in counting vials and counted in a gamma counter.

Application of the microsphere technique. Radiolabeled (46Sc, 85Sr, 113Sn, or 57Co) microspheres (New England Nuclear) of 15-μm diameter were used as described in detail (24, 41, 42). For each measurement, 500,000 spheres suspended in 0.1 ml of saline containing <0.1% Tween 80 were infused into the aortic catheter (2–6 s) followed by the infusion of 1 ml of warm (37°C) saline over 20–30 s. Before infusion the sphere suspension was mixed in an ultrasonicator for 10 min and then on a Vortex for 1–2 min. The reference withdrawal sample was taken at a rate of 0.618 ml/min for 10 s before and for 60 s after sphere infusion. The volume of blood withdrawn was replaced with donor rat blood. Microsphere-blood mixing on infusion was monitored for each infusion by comparing blood flow to the right and left kidneys.

eNOS and SOD-1 Protein Content of Arteries of Sed and IST Rats

To determine the effects of IST on eNOS and SOD-1 expression in the arteries that provide blood flow to the gastrocnemius muscle, we used three groups of 25 rats as described in Experimental Design (14, 42). At the completion of the exercise training program, IST and Sed rats were anesthetized with an injection of pentobarbital sodium (50 mg/kg ip), and samples of abdominal aorta, iliac, femoral, popliteal, and renal arteries were isolated, placed in microcentrifuge tubes, and frozen at -70°C. In addition, we collected samples of FAs and arterioles isolated from the red portion of the medial head of gastrocnemius muscle (RG) and from the WG. We focused on these two muscle tissues because the WG is composed of 90% FG fibers, whereas the RG is composed primarily of high-oxidative muscle fibers (WG: 10% FOG, 90% FG; RG: 6% FG, 60% FOG, 34% SO) (7, 19).

Preparation of arterioles. The triceps surae muscle group was dissected, and the medial head of the gastrocnemius was transferred to a chamber containing cold (4°C) MOPS-buffered physiological saline solution (PSS) containing (in mM) 145.0 NaCl, 4.7 KCl, 2.0 CaCl2, 1.17 MgSO4, 1.2 NaH2PO4, 5.0 glucose, 2.0 pyruvate, and 0.02 EDTA, and 3.0 MOPS (pH 7.4). Overlying muscle tissue was dissected away, revealing the arteriolar tree of the medial head of the gastrocnemius muscle. Single FAs and first-order arterioles (1A) were isolated from the gastrocnemius muscle. Second (2A)-, third (3A)-, fourth (4A)-, and fifth (5A)-order arterioles were isolated from the RG and WG as illustrated in Fig. 1. (44). The soleus muscle was also removed and transferred to the chamber containing cold PSS, and soleus FAs were removed. FAs and arterioles were placed in microcentrifuge tubes and frozen at -70°C.

Fig. 1.

Arteriolar tree of the medial head of the rat gastrocnemius muscle. Sample vascular cast (Microfil) showing arterioles within the left medial head of the gastrocnemius muscle of a sedentary rat. FA, gastrocnemius feed artery. The method used to name branch orders of arterioles, from first order (1A) through fifth order (5A) is illustrated.

Immunoblot analysis. Relative differences in eNOS and SOD-1 protein expression in arteries and arterioles were assessed by using immunoblot analysis as described previously (26, 36). Briefly, abdominal aorta, iliac, femoral, popliteal, and renal arteries were solubilized in 20 μl Laemmli buffer (18), boiled, and sonicated for 2 min. Total protein content in individual arteries was measured by using the NanoOrange protein assay (Molecular Probes, Eugene OR). Equal amounts of total protein from each artery and arteriole sample (8 μg/lane) were loaded onto 5–12% NuPage Bis Tris gradient gels (Invitrogen) electrophoresed under reducing conditions, and transferred to polyvinylidene difluoride membrane (Hybond-enhanced chemiluminescence, Amersham). The membranes were blocked for 1 h at room temperature with 5% nonfat milk in TBS-tween (20 mmol/l Tris·HCl, 137 mmol/l NaCl and 0.1% Tween 20) and incubated overnight at room temperature with primary antibody against eNOS (1:1,600; Transduction Laboratories). Blots were subsequently incubated for 1 h with secondary antibody (1:2,500; horseradish peroxidase-conjugated anti-mouse; Sigma Chemical, St. Louis, MO). Specific eNOS protein was detected by enhanced chemiluminescence (Amersham) and evaluated by densitometry using NIH Image software (National Institutes of Health, Bethesda, MD). All blots were reblocked for 1 h at room temperature and incubated overnight with a polyclonal antibody against SOD-1 (1:1,600, Stressgen). Blots were then incubated for 1 h with secondary antibody (1:2,500; horseradish peroxidase-conjugated anti-rabbit; Sigma Chemical). Data were standardized such that the mean value of the Sed arteries was set to 1.0 and IST data were expressed as fold increase relative to the Sed arteries. Immunoblot analysis was performed on tissues from three groups of Sed and IST rats. Immunoblot analysis of gastrocnemius muscle FAs and arterioles (1A–5A) was performed by using the procedures described under Preparation of arterioles for arteries except that 8–10 arterioles of a given branch order were pooled per sample. Preliminary experiments revealed that pooling of 8–10 arterioles was needed to provide sufficient material for measurement of total protein and subsequent immunoblot analysis. The Student's t-test was used to determine statistical significance of differences in immunoblot results between Sed and IST samples.

Vasoconstrictor and Vasodilator Responses of Isolated Arterioles

Vasodilator responses of 2As from RG and WG muscle were evaluated because the immunoblot results indicated that 2As from WG exhibited increased eNOS expression, whereas RG 2As did not. In addition, 2As in RG and WG have consistent anatomy and location within the arteriolar tree. We also examined vasodilator responses of gastrocnemius FAs and WG 3As. We included WG 3As because the immunoblot results indicated their eNOS expression was not changed by IST. Thus we assessed the impact of changes in eNOS expression on vasodilator function in WG arterioles. Finally, vasodilator responses of gastrocnemius FAs were examined because the immunoblot results suggested increased eNOS expression in gastrocnemius FAs and because FAs are a primary site of regulation of vascular resistance in skeletal muscle tissue (10, 40). We used an isolated arteriolar preparation to allow examination of intrinsic vasoreactivity of FAs and arterioles from IST and Sed skeletal muscle without the confounding effects of pressure differences, neural stimulation, hormonal influences, or metabolic and mechanical influences from surrounding muscle that may be present in in vivo or whole muscle in vitro preparations.

Vasoconstrictor responses. After demonstration of functional endothelium and vascular smooth muscle, arterioles were allowed to achieve and maintain a stable diameter for at least 10 min before administration of a vasoactive agent. Myogenic responses to changes in intraluminal pressure were examined in arterioles and FAs that developed >20% spontaneous tone. In 2As, pressures were set to 60 cmH2O for equilibration. Pressures were then increased to 70 and 80 cmH2O and then decreased to 50 cmH2O. After each change in intraluminal pressure, the vessel was allowed 3–5 min to respond and the diameter again measured at the new intraluminal pressure. WG 3As were studied with a similar protocol except that pressure started at 53 cmH2O and was increased to 63 and 73 cmH2O and decreased to 43 cmH2O. In FAs pressure began at 90 cmH2O and was increased to 100 and 110 cmH2O, followed by a decrease to 80 cmH2O. After completion of the myogenic response experiment, pressures were returned to those used for each vessel during the equilibration period. At this time cumulative dose-response curves for PE were constructed by adding PE to the bath in whole-log increments in concentration. After each dose, FAs and arterioles were allowed 3 min before the next dose was administered.

Endothelium-dependent responses to ACh. After arterioles maintained ∼50% constriction for 10 min, an ACh dose-response curve was constructed by adding ACh (10-9 to 10-4 M) to the bath in half-log increments in concentration. After each dose of ACh, the arteriole achieved a stable diameter before the next dose was administered. Typically, arterioles achieved stable diameter in <3 min.

In some arterioles, we used blockers of select endothelial cell signaling pathways to examine the effects of IST on the relative importance of each pathway in ACh-induced vasodilation. Arterioles were treated with NW-nitro-l-arginine (l-NNA; 300 μM) to inhibit NOS activity, indomethacin (50 μM, to inhibit cyclooxygenase activity), and l-NNA + indomethacin to inhibit both pathways. Responses to repeat exposure to the EC50 and maximal dose of ACh were compared with the untreated responses.

SNP experiments. Dilation of FAs and arterioles in response to SNP was examined to evaluate endothelium-independent dilator responses. After arterioles maintained ∼50% constriction for 10 min a SNP dose-response curve was constructed by adding SNP (10-9 to 10-4 M) to the bath in whole-log increments in concentration. After each dose of SNP, the arteriole was allowed to achieve a stable diameter before the next dose was administered. After the final dose-response curve, each arteriole was exposed to PSS containing zero calcium for 1 h, and passive diameter was measured as described previously (1, 2). Internal and external diameters were measured with the video micrometer as described previously (1, 2). Wall thickness was calculated as external diameter - internal diameter divided by two.

Drugs and Solutions

Warm PSS-albumin containing 10 g/l bovine serum albumin (fraction V, >98% pure, United States Biochemical) was utilized within the arteriolar lumen and was pH 7.4 at 37°C. Warm PSS without albumin was utilized as superfusate for the isolated arteriole experiments and was pH 7.4 at 37°C. All PSS was prepared in advance, filtered through a 0.2-μm filter, and frozen for use on the day of experiment. All drug stocks were prepared in distilled, deionized water and frozen for later use. NaCl, KCl, and CaCl2 were obtained from Fisher Scientific (Pittsburgh, PA). All other chemicals were obtained from Sigma Chemical.

Data Analysis

Arteriolar diameter data were expressed as 1) absolute diameter (in μm), 2) percentage of the possible response between the baseline diameter before the first dose of the agent being utilized and maximal calcium-free diameter, and 3) percentage of the maximal response from baseline diameter. A one-way ANOVA was utilized (Super-ANOVA statistical software) when comparing maximal diameters, beginning tone, maximal responses, and the EC50 data (Prism statistical software) between arteries or arterioles from IST and Sed rats.

In blood flow experiments, an ANOVA was used to compare various parameters within muscles or tissues across conditions and to compare a parameter (blood flow) across muscles (or tissues). Duncan's new multiple-range test was used to determine differences among treatment means (19, 20).

Dose-response relationships were compared between groups (e.g., Sed vs. IST) by utilizing two-way repeated-measures ANOVA with repeated measures on one factor (dose) (SuperANOVA statistical software). Differences between groups that were identified by ANOVA were located utilizing Tukey's multiple-comparisons post hoc test. Significance for all analyses performed was set at P < 0.05 for all statistical comparisons.

RESULTS

IST rats were smaller (body weight = 402 ± 8 g) than Sed rats (427 ± 8 g). Heart weights of IST (1.40 ± 0.04 g) and Sed (1.43 ± 0.04 g) rats were similar as were heart weight-to-body weight ratios (3.49 ± 0.09 g/kg for IST and 3.34 ± 0.08 g/kg for Sed). Oxidative capacity of the white portion of the vastus lateralis muscle was significantly greater in the IST (13.1 ± 1.6 mmol·min-1·g-1) rats compared with Sed (7.3 ± 0.5 mmol·min-1·g-1) values. These results indicate that the IST rats underwent the expected adaptations to this training program (14, 24, 41, 42).

Endothelium-Dependent Increases in Muscle Blood Flow

Total hindquarters flows under isogravimetric conditions were 9.6 ± 1.9 and 3.1 ± 1 ml·min-1·100 g-1, respectively, in IST and Sed rats. The greater isogravimetric blood flow and decreased pre- and postcapillary resistances in hindquarters from IST rats confirm previous results that IST increases isogravimetric flow in perfused hindquarters of rats (41).

Results for ACh-induced vasodilation in isolated perfused hindquarters are summarized in Fig. 2A. These results indicate that ACh produced more vasodilation in IST hindquarters and suggest that ACh-induced release of endothelium-derived relaxing factors is increased by IST. To examine the role of the l-arginine NO synthase pathway, the hindquarters were treated with l-NAME to block NO production via this pathway. l-NAME produced dramatic increases in perfusion pressure (constant-flow conditions) (Sed arterial pressure: pretreatment = 111 ± 34 mmHg, l-NAME = 218 ± 12 mmHg; IST arterial: pretreatment = 70 ± 8 mmHg, l-NAME = 180 ± 15 mmHg). l-NAME also blunted ACh-induced vasodilation in IST preparations and blocked it in Sed hindquarters. When the hindquarters were treated with l-NAME and indomethacin, there was no significant change in perfusion pressures and there was no longer significant dilation in response to ACh in either group.

Fig. 2.

A: acetylcholine (ACh)-induced changes in vascular resistance of isolated perfused hindquarters. Values are means ± SE. IST, interval sprint-trained rats (n = 7); Sed, sedentary control rats (n = 3). *Vascular resistance was decreased significantly more in the IST rats than in Sed (P < 0.05) at ACh infusion rates of 1 and 10 μg/min. B and C: microsphere measurements of ACh-induced increases in blood flow (B) and vascular conductance [the change (Δ) in conductance produced by ACh infusion; C] in the white portion of gastrocnemius muscle (WG) of IST (n = 5) and Sed rats (n = 6). Values are means ± SE. Pre, blood flow before ACh infusion; ACh, blood flow during ACh infusion. ANOVA indicates that ACh-induced increases in blood flow and conductance of IST muscle but not in WG of Sed rats. Blood flow was greater in WG of IST than in Sed during ACh infusion.

Regional blood flow measurements revealed that ACh produced modest (5–10 ml·min-1·100 g-1) increases in blood flow to the soleus and RG in both Sed and IST rats and that there was no difference between IST and Sed flow responses to ACh in these muscle samples. In contrast, ACh did not alter blood flow in the mixed portion of the gastrocnemius muscle in Sed rats and produced a small amount of increased blood flow in IST rats. Importantly, ACh infusion produced increases in blood flow and vascular conductance in WG muscle of IST, whereas ACh infusion did not increase WG blood flow or conductance in Sed rats (Fig. 2B).

Immunoblot Results

eNOS protein content of arteries of Sed and IST rats. Figure 3A presents a sample immunoblot for gastrocnemius muscle arterioles, illustrating the manner the samples were loaded on the gels, which allows comparisons between IST and Sed samples. Figure 3B presents average eNOS protein content for arteries throughout the arteriolar tree of the medial head of the gastrocnemius muscle. These results indicate that IST increases eNOS protein content of gastrocnemius FAs, WG 2As, WG 4As, WG 5As, RG 3As, and RG 5As. Statistical analysis indicates that FAs, WG 4As, and RG 5As of IST rats had significantly greater eNOS content compared with Sed (P < 0.05) and that the RG 4As of IST rats had significantly less eNOS compared with Sed (P < 0.05). eNOS protein content did not appear to be changed in the other arterioles of RG.

Fig. 3.

Effects of IST on endothelial nitric oxide synthase (eNOS) protein content of gastrocnemius muscle arteries. A: sample immunoblot showing eNOS protein content of human endothelial cell lysate (HEL; an internal control sample); fifth (RG 5A)-, fourth (RG 4A)-, third (RG 3A)-, second (RG 2A)-, and first (G 1A)-order arterioles of red gastrocnemius muscle; and gastrocnemius feed arteries (GFA) of sedentary (S) and IST rats (I). Note that, for each branch order, the S and I samples are paired. Rainbow molecular mass markers are illustrated on the far left from the rainbow marker sample (markers are shown because the rainbow molecular mass markers do not react with eNOS antibodies). B: average eNOS protein content in arteries and arterioles of IST rats expressed relative to the eNOS content of paired Sed samples. If relative eNOS content is >1, then it indicates an increase in eNOS content in the IST artery. eNOS protein content was quantified by scanning densitometry with NIH Image software. GFA are data from 2 sets of IST and Sed animals. Gastrocnemius 1A–5A data are from 4 different groups of Sed and IST rats so means and SE are presented. Each group of Sed and IST animals consisted of 5–10 rats, so there are data from 10 to 30 different rats included in these averages. Statistical analysis was conducted with groups of rats as the n (i.e., 2 for GFA data and 4 for arterioles). Values are means ± SE. *IST value is different from Sed, P < 0.05 (by 1-way Student's t-test).

We expected the results from Sed rats would reveal increasing eNOS content, relative to total arteriolar protein content, with increasing branch order (1A–5A). This expectation was based on the decrease in the number of smooth muscle cells and endothelial cells that occurs with increasing branch order. To evaluate this possibility in Sed animals we expressed the eNOS content of 1A, 2A, and 3A relative to the eNOS content of the 1A on each blot (importantly, comparisons among branch arterioles were only made if all branch orders were present on the same blot). The results of this analysis are shown in Fig. 4 and indicate that eNOS content is greater in 1As than in WG 2As and WG 3As, whereas RG 2As have greater eNOS content than WG 2As and WG 3As.

Fig. 4.

Comparison of average eNOS protein content in arterioles of sedentary rats. All eNOS protein content is expressed relative to the eNOS protein content of G 1A. eNOS protein content was quantified by scanning densitometry with NIH Image software. As for Fig. 3, results are from 4 different groups of Sed rats. Values are means ± SE.

Figure 5A presents a sample eNOS immunoblot of paired samples for the aorta and average results for the conduit arteries that perfuse the gastrocnemius muscle. IST produced the greatest increase in eNOS protein content in distal aorta, iliac artery, and femoral artery and modest increases in the popliteal artery. Interestingly, IST did not appear to increase eNOS protein content in the renal or mesenteric arteries.

Fig. 5.

A: Average eNOS protein content data (left) and sample immunoblot (right) showing eNOS protein content of HEL and paired samples of aorta from S and I rats. Values are means ± SE. Note that the S and I samples are paired from left to right in the sample blot. Rainbow molecular mass markers are illustrated on the far right. B: average superoxide dismutase-1 (SOD-1) protein content data (left) and sample immunoblot (right) showing SOD-1 protein content of HEL and paired samples of aorta from S and I rats. Values are means ± SE. Note that the S and I samples are paired from left to right in the sample blot. Rainbow molecular mass markers are illustrated on the far right. Average eNOS and SOD-1 protein contents are presented for distal aorta (D Aor; n = IST 12 and n = Sed 17), iliac artery (iliac; n = IST 16 and n = Sed 16), femoral artery (Fem; n = IST 18 and n = Sed 15), popliteal artery (Pop; n = IST 14 and n = Sed 17), renal artery (Renal; n = IST 16 and n = Sed 15), and mesentaric artery (Mes; n = IST 5 and n = Sed 5) with IST protein content expressed relative to the eNOS or SOD-1 content of paired Sed samples. Protein content was quantified by scanning densitometry with NIH Image software. *IST value is different from Sed, P < 0.05.

SOD-1 protein content of arteries of Sed and IST rats. Figure 6A presents a sample SOD-1 immunoblot for gastrocnemius arterioles of Sed and IST rats, and Fig. 6B presents the average results showing that WG 4As of IST rats appear to have increased SOD-1 protein content and that RG 4As of IST rats appear to have decreased SOD-1 content relative to Sed. Statistical analysis revealed that RG 4As have significantly less SOD-1 in IST than in Sed (P < 0.05; Fig. 6B).

Fig. 6.

Effects of IST on SOD-1 protein content of gastrocnemius muscle arteries. A: sample immunoblot showing SOD-1 protein content of HEL; RG 5A, RG 4A, RG 3A, RG 2A, and G1 A; and GFA of S and I rats. Note that for each branch order the S and I samples are paired. Rainbow molecular mass markers are illustrated on the far left. B: average SOD-1 protein content in arteries and arterioles of IST rats expressed relative to the content of paired Sed samples. If relative SOD-1 content is >1, it indicates an increase in SOD-1 content in the IST artery. SOD-1 protein content was quantified by scanning densitometry with NIH Image software. GFA are data from 2 sets of IST and Sed animals. Gastrocnemius 1A–5A data are from 4 different groups of Sed and IST rats, so means and SE are presented. Each group of Sed and IST animals consisted of 5–10 rats so there are data from 10 to 30 different rats included in these averages. Statistical analysis was conducted with groups of rats as the n (i.e., 2 for GFA data and 4 for arterioles). *IST value is different from Sed, with P < 0.05 (by 1-way Student's t-test).

The results presented in Fig. 5B indicate that SOD-1 protein content in popliteal arteries of IST rats was greater than in Sed. IST did not alter SOD-1 protein content in the other conduit arteries.

Vasoconstriction and Vasodilation Responses of Isolated Arterioles

Characteristics of arterioles. The structural characteristics of the arterioles utilized in vasomotor function experiments are presented in Fig. 7. The results in Fig. 7A indicate that WG 3As from the IST rats had significantly greater internal diameter than those of Sed rats, whereas IST did not alter maximal diameter (maximal calcium-free diameters) of the RG 2As, WG 2As, or FAs. There were no differences in wall thickness of the Sed and IST arteries (Fig. 7B). FAs, WG 2As, and WG 3As from IST rats all had greater spontaneous tone than did the comparable Sed vessels (Fig. 7C). In contrast, the spontaneous tone of RG 2As of Sed and IST rats was not significantly different. It is also of interest that the FAs generally had less spontaneous tone than did the arterioles. Results indicate that, although IST arterioles tended to have greater spontaneous tone, IST arterioles exhibited myogenic reactivity similar to Sed arterioles (data not shown), maintaining diameter relatively constant across the range of intraluminal pressures examined.

Fig. 7.

Characteristics of arteries used in vasomotor reactivity experiments. A: diameters of arteries and arterioles. B: wall thickness of arteries and arterioles. C: spontaneous tone of arteries and arterioles. Values are means ± SE for 8 Sed and 6 IST rats. *IST and Sed value significantly different, P < 0.05.

Responses of arterioles to PE. All four types of arteries constricted in a dose-dependent manner in response PE treatment (Figs. 8 and 9). The maximal constriction of all three types of Sed and IST arterioles was similar in that PE caused complete closure of the arterioles. In contrast, the gastrocnemius muscle FAs from Sed rats exhibited greater constriction than did IST gastrocnemius muscle FAs. Because the arterioles from IST rats began with greater tone we evaluated sensitivity of the arteries to PE by expressing the data as percentage of maximal constrictor response (Figs. 8 and 9, C and D). This analysis indicates that the gastrocnemius muscle FAs of Sed rats were also more sensitive to PE than GFAs from IST rats (Fig. 9D), whereas there were no significant differences between PE sensitivity of the three types of arteriole from IST and Sed rats.

Fig. 8.

Effects of IST on vasoconstrictor responses of arterioles to increasing doses of phenylephrine (PE). Values are means ± SE for absolute diameters (A and B) and percentage of maximal response (C and D) for Sed and IST rats. RG 2A: IST n = 4 and Sed n = 4. WG 2A: IST n = 6 and Sed n = 5. B, baseline.

Fig. 9.

Effects of IST on vasoconstrictor responses of arterioles to increasing doses of PE. Values are means ± SE expressed as absolute diameter (A and B) and as percentage of maximal response (C and D). WG 3A: IST n = 6 and Sed n = 4. GFA: IST n = 5 and Sed n = 8. Note that Sed GFAs exhibited complete constriction, whereas IST GFAs did not completely close. *IST and Sed value significantly different, P < 0.05.

Vasodilation responses. All arterioles and gastrocnemius muscle FAs exhibited dose-dependent vasodilation in response to ACh (Fig. 10). There were no statistically significant differences between the responses of RG 2As or WG 3As of IST and Sed rats. In contrast, the WG 2As of IST rats exhibited greater vasodilation to ACh (Fig. 10B) than Sed WG 2As. Surprisingly, gastrocnemius muscle FAs of IST rats showed less maximal dilation in response to ACh than gastrocnemius muscle FAs of Sed rats. Results of the SNP experiments are presented in Fig. 11 and demonstrate that SNP-induced dilation was similar in arterioles and gastrocnemius muscle FAs of Sed and IST rats.

Fig. 10.

ACh-induced vasodilation of arterioles from RG and WG. Values are means ± SE expressed as percent possible dilation. for Sed and IST rats. A: RG 2A (IST n = 6 and Sed n = 7). B: WG 2A (IST n = 6 and Sed n = 8). C: WG 3A (IST n = 5; Sed n = 5). D: GFA (IST n = 3 and Sed n = 4). *IST and Sed value significantly different, P < 0.05.

Fig. 11.

Sodium nitroprusside-induced vasodilation of arterioles from RG and WG of the gastrocnemius muscle. Values are means ± SE expressed as percent possible dilation for Sed and IST rats. A: RG 2A (IST n = 6 and Sed n = 7). B: WG 2A (IST n = 6 and Sed n = 8). C: WG 3A (IST n = 5 and Sed n = 5). D: GFA (IST n = 3 and Sed n = 4).

DISCUSSION

The purpose of this study was to test the hypothesis that IST increases EDD and eNOS and/or SOD-1 protein content selectively in FAs and/or arterioles that perfuse and/or are located in the WG. This hypothesis was based on previous observations that IST produces increased blood flow capacity and capillarization in muscle composed of a high percentage of fast glycolytic fibers, like WG, with little or no change in capillarization of SO or FOG muscle (14, 24). Accordingly, we proposed that the changes in blood flow capacity of WG are due in part to training-induced increases in endothelium-dependent vasodilator capacity of the arteries providing blood flow to this muscle tissue. Results support the hypothesis in that some arteries providing blood flow to the WG exhibited increased EDD and/or increased expression of eNOS. However, some arteries that provide blood flow to WG did not exhibit these adaptations. The primary findings of this study are as follows. 1) ACh-induced decreases in total hindquarters vascular resistance were greater in IST rats and regional blood flow results indicate the increased dilation was greatest in WG of IST rats. 2) In both IST and Sed rats, arterioles in RG exhibit greater eNOS content than arterioles in WG. 3) IST produced an increase in eNOS content of aorta, iliac, femoral, and gastrocnemius muscle FAs, WG 2As, WG 4As, RG 3As, and RG 5As but not in other arteries and arterioles. 4) SOD-1 content was greater in IST popliteal arteries than in Sed. 5) Maximal diameter of WG 3As from IST rats was greater than that of Sed. 6) Gastrocnemius muscle FAs, WG 2As, and WG 3As isolated from IST rats exhibited greater spontaneous tone than those of Sed. 7) PE-induced vasoconstrictor responses of gastrocnemius muscle FAs from IST rats were significantly blunted compared with those of Sed gastrocnemius muscle FAs. 8) In the IST rats, an enhanced vasodilator response to ACh of gastrocnemius FAs and WG 2As appeared to be associated with increased eNOS content. IST did not alter ACh-induced dilation in RG 2As or WG 3As. On the basis of these observations, we conclude that the increase in blood flow capacity of the WG produced by IST is associated with increased eNOS protein content of some arteries that perfuse this muscle and with enhanced EDD of some arterioles located in WG muscle tissue.

Effects of Muscle Fiber-Type Composition on Arteriolar Vasoreactivity

The relative importance of mechanisms controlling vascular resistance and blood flow appear to vary in skeletal muscle tissue composed of different fiber types (13, 15, 22, 23, 39). For example, there is evidence of differences in vascular smooth muscle and in endothelium-mediated influences on vascular resistance in muscle tissue with different muscle fiber type composition (13, 15, 22, 23). In the present study, the observation that RG 2As and WG 2As exhibited similar vasoconstrictor responses to PE confirms previous results that arterioles isolated from RG and WG exhibit similar constrictor responses to catecholamines (1, 34). Of interest, whereas 2As completely closed with the highest doses of PE, WG 3As did not show complete constriction. RG and WG arterioles used in this study exhibited similar responses to SNP. In contrast, RG 2As from both Sed and IST rats exhibited greater dilation in response to ACh than did WG 2As, also confirming previous observations (1, 2). Importantly, present results demonstrate that the enhanced endothelium-dependent dilation in RG 2As vs. WG 2As of Sed rats was associated with greater content of eNOS protein in RG 2As. Indeed, RG 2As appear to contain more than twice as much eNOS protein than 1As and more than three times as much as WG 2As and 3As (Fig. 4). These results add to the growing body of literature indicating that the phenotype of the endothelium of skeletal muscle arterioles is different in skeletal muscle composed of different fiber types (1, 2, 15, 30, 33).

Effects of Training on Arterial and Arteriolar Endothelium

ACh-induced decreases in vascular resistance of rat hindquarters were greater in IST rats (Fig. 2A). Regional blood flow measurements indicate that increases in blood flow were directed to the white skeletal muscle tissue (Fig. 2B). These results are consistent with our hypothesis that increases in EDD responses would be present in the arteries perfusing WG skeletal muscle. Immunoblot analysis indicates that eNOS protein content was increased in some arteries and arterioles providing blood flow to WG muscle but not in others. Furthermore, the vasomotor reactivity results obtained with isolated arterioles and arteries are consistent with the notion that enhanced EDD is present in some arteries but not others. Importantly, WG 2As from IST rats exhibited enhanced vasodilator responses to ACh and appeared to have increased eNOS protein content, whereas the two arteries from IST rats that did not exhibit increased eNOS or SOD-1 content (RG 2As and WG 3As) did not exhibit enhanced vasodilator responses to ACh. These results are consistent with the notion that changes in eNOS protein content influence EDD. In a subset of RG 2As and WG 2As, we also examined the effects of blockade of NOS with l-NNA. In RG 2As from Sed and IST rats, l-NNA decreased sensitivity of the RG 2As to ACh so that the EC50 dose produced only 20% dilation, whereas maximal ACh-induced dilation was not altered by l-NNA. In contrast, l-NNA treatment decreased the dilation at the EC50 dose to only 20% in WG 2As from IST rats but decreased dilation to 40% in WG 2As from Sed rats (data not shown). Thus blockade of NOS activity appeared to have greater effect in the arterioles from IST rats that exhibited increased eNOS content (WG 2As) than in those showing no change in eNOS content (RG 2As) with IST. These preliminary results provide further evidence of a link between training-induced changes in eNOS expression and endothelium-dependent vasodilation.

Our hypothesis was based on the view that exercise-induced vascular adaptations occur, spatially, in the area of muscle tissue with the greatest relative increase in fiber activity during exercise training bouts. It is generally believed that exercise training increases eNOS expression in arteries perfusing tissues with increased blood flow during exercise training bouts, perhaps because of the increased wall shear stress in the arteries produced by the increased amount of blood flowing through these arteries (12). This conception of these adaptative responses suggests that in the present study, our results would find eNOS expression to be increased throughout the arteriolar tree of the gastrocnemius muscle because blood flow is increased throughout this muscle when rats run at 60 m/min, up a 15% incline (20). Results reveal that eNOS protein content was increased in aorta, iliac, femoral, and gastrocnemius FAs of IST rats and in WG 2As and WG 4As, consistent with this line of reasoning. We proposed that the greatest increases in eNOS expression would occur in the arteries perfusing the WG because this muscle has the greatest relative increase in fiber activity and blood flow during exercise at this intensity. The pattern of arteries and arterioles exhibiting increases in eNOS and/or SOD-1 protein shown in Figs. 3 and 6 indicate that IST does not uniformly increase expression of these proteins throughout the arterial tree perfusing the WG. Indeed, increases in eNOS content were also seen in the RG 5As of IST rats. Whatever the exercise-induced signal for increased expression of eNOS (i.e., increased shear stress), these results suggest that this signal is not uniformly increased throughout the arterial tree perfusing the WG. For example, if the signal is shear stress, it is known that shear stress is determined by the interactions of several factors. During exercise, it is possible that in many of the arterioles, diameter increased to a greater extent than did flow through these arterioles. If so, shear stress would either not increase or would not increase sufficiently to alter eNOS expression. We do not have data describing blood flow velocity and diameter in these arteries during exercise to rigorously evaluate this postulate.

The aorta and conduit arteries of IST rats exhibited increased eNOS content (Fig. 5). Our laboratroy previously reported increased eNOS protein content in aortas of endurance-trained rats that ran for 60 min/day, 5 days/wk (8, 9). It is striking that the IST rats used in this study also exhibit increased eNOS protein content in their aortas because these rats only ran a total of 15 min/day. However, it is important to emphasize that although the rats only ran for 15 min, aortic blood flow was likely increased during the 4.5-min rest periods as well. Because cardiac output was not measured in the rats in the present study, we do not know the magnitude or duration of increased flow during or after each 2.5-min sprint. On the basis of the behavior of the rats, we expect that aortic flow is increased throughout the 42 min of the training sessions.

We included the renal and mesenteric arteries in our analysis as controls, i.e., arteries in which we were quite sure blood flow decreased during exercise training bouts (24). Neither eNOS nor SOD-1 content were altered in these arteries from IST rats. These results are in contrast to reports in the literature that indicate that exercise training can produce increases in endothelial function and eNOS content, even in arteries not perfusing muscle tissue active during training bouts (8, 9, 13, 15, 22, 23).

An interesting and unexpected finding in our results is the increase in eNOS content of RG 3As and RG 5As and the decreased eNOS and SOD-1 content of RG 4As of IST rats. In retrospect, the increases in eNOS content are consistent with results in the literature in that whereas the largest IST-induced increases in blood flow capacity were observed in WG muscle, IST also increases blood flow capacity of RG and mixed gastrocnemius muscle (24). Only the soleus muscle showed no increase in blood flow capacity after IST. In future studies we will be able to examine EDD in RG 3As, but we cannot study endothelium-dependent responses in RG 4As or RG 5As with our present techniques because these arterioles are too small and have too many branches.

Finally, it is important to emphasize that it is tenuous to draw broad conclusions concerning the effects of muscle fiber type composition on vascular responsiveness of arterioles and/or on interactions between fiber-type composition and training-induced vascular adaptations only on the basis of the results of this study. There are two primary reasons for this caution. First, the evaluation of the potential importance of well established differences among vasomotor properties of different branch orders within an arteriolar tree requires more study (17, 31). In the present study, vasomotor responses were only examined in three branch orders of arterioles, and we were able to examine eNOS and SOD-1 protein content only in 1A–5A. Thus, although our results provide some insight into differences in eNOS and SOD-1 protein content throughout the arteriolar tree of the rat gastrocnemius muscle, we are not able to evaluate the majority of branch orders in the arteriolar tree. On the basis of preliminary observations, there are generally 12–16 arteriolar branch orders within the RG. Therefore, the present study has only examined these relationships in a small portion of the arteriolar tree. Clearly, different results may be obtained in other branch orders within the arteriolar networks of this muscle. Second, previous studies that revealed the importance of relationships among muscle fiber type composition, muscle fiber recruitment patterns and vascular control mechanisms to blood flow distribution within and among muscles were only able to do so because they examined a large number of muscles with widely varying biochemical characteristics (15, 19, 2123). Before general patterns of training-induced changes in vascular responsiveness among muscle tissues with widely varying biochemical characteristics can be established, vascular responses and endothelial phenotypes need to be examined in arterioles from more muscles with differing fiber-type composition.

Conclusions

IST produced increased ACh-induced vasodilation in the skeletal muscle tissue that has the greatest relative increase in activity during high-intensity (uphill), sprint exercise (WG). Because some arteries providing blood flow to WG muscle exhibited increased eNOS protein content, these results suggest that increased EDD contributes to the increase in blood flow capacity of WG muscle. However, in contrast to our hypothesis, increases in eNOS and/or SOD-1 protein content were more apparent in the conduit arteries (aorta, iliac, femoral, popliteal, and gastrocnemius muscle FAs) than in arterioles that perfuse WG. The observation of increased ACh-induced dilator responses of WG 2As but not RG 2As indicates that changes in eNOS protein content are associated with increased endothelium-dependent dilation. Results also indicate that IST produced a 30% increase in the maximal diameter of 3As in WG muscle. This observation suggests that structural vascular adaptation of the arteriolar tree may play an important role in increased blood flow capacity of WG muscle. For example, IST may increase blood flow capacity primarily by stimulation of increases in diameter and/or number of smaller arterioles in WG with only minor changes in vasoreactivity to vasoconstrictor and vasodilator signals. Additional research is required to assess the relative importance of structural and functional adaptation in the improvement of flow capacity in WG.

Acknowledgments

The authors thank Pam Thorne, Tammy Strawn, Denise Holiman, and Ann Melloh for excellent technical contributions to this work.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-36088 and HL-26490.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

References

View Abstract