Skeletal muscle microvascular oxygenation dynamics in heart failure: exercise training and nitric oxide-mediated function

Abstract

Chronic heart failure (CHF) impairs nitric oxide (NO)-mediated regulation of skeletal muscle O₂ delivery-utilization matching such that microvascular oxygenation falls faster (i.e., speeds PO₂mv kinetics) during increases in metabolic demand. Conversely, exercise training improves (slows) muscle PO₂mv kinetics following contractions onset in healthy young individuals via NO-dependent mechanisms. We tested the hypothesis that exercise training would improve contracting muscle microvascular oxygenation in CHF rats partly via improved NO-mediated function. CHF rats (left ventricular end-diastolic pressure = 17 ± 2 mmHg) were assigned to sedentary (n = 11) or progressive treadmill exercise training (n = 11; 5 days/wk, 6–8 wk, final workload of 60 min/day at 35 m/min; −14% grade downhill running) groups. PO₂mv was measured via phosphorescence quenching in the spinotrapezius muscle at rest and during 1-Hz twitch contractions under control (Krebs-Henseleit solution), sodium nitroprusside (SNP; NO donor; 300 μM), and N^G-nitro-l-arginine methyl ester (L-NAME, nonspecific NO synthase blockade; 1.5 mM) superfusion conditions. Exercise-trained CHF rats had greater peak oxygen uptake and spinotrapezius muscle citrate synthase activity than their sedentary counterparts (p < 0.05 for both). The overall speed of the PO₂mv fall during contractions (mean response time; MRT) was slowed markedly in trained compared with sedentary CHF rats (sedentary: 20.8 ± 1.4, trained: 32.3 ± 3.0 s; p < 0.05), and the effect was not abolished by L-NAME (sedentary: 16.8 ± 1.5, trained: 31.0 ± 3.4 s; p > 0.05). Relative to control, SNP increased MRT in both groups such that trained CHF rats had slower kinetics (sedentary: 43.0 ± 6.8, trained: 55.5 ± 7.8 s; p < 0.05). Improved NO-mediated function is not obligatory for training-induced improvements in skeletal muscle microvascular oxygenation (slowed PO₂mv kinetics) following contractions onset in rats with CHF.

chronic heart failure (CHF) induces multiple alterations in cardiovascular structure and function that culminate in reduced exercise capacity and poor prognosis (8, 51, 52, 54). Derangements in the skeletal muscle O₂ transport pathway are major features of this disease (54). Within the microcirculation, impaired hemodynamic control (35, 56) compromises the dynamic matching between O₂ delivery and utilization (Q̇O₂ and V̇o₂, respectively) and reduces muscle microvascular oxygenation (PO₂mv) during transitions in metabolic demand (3, 11, 14, 19, 45). As dictated by Fick's law of diffusion, lower PO₂mv reduces the driving pressure for blood-myocyte O₂ flux and impairs oxidative metabolism and contractile performance (29, 64).

Impaired nitric oxide (NO)-mediated function is a hallmark of CHF (16, 32, 37). Notwithstanding the plethora of factors modulating muscle blood flow during exercise (including neural, mechanical, and humoral components), deterioration of NO-mediated function plays a key role in the blunted functional hyperemic response characteristic of CHF (28, 34). Accordingly, these perturbations contribute to the pathological PO₂mv profiles (characterized by lower PO₂mv and faster PO₂mv fall; i.e., speeded kinetics) during muscle contractions in CHF (19). Endurance exercise training, on the other hand, is a nonpharmacological intervention capable of improving NO-mediated function in skeletal muscle of both healthy (21, 43) and CHF (24, 40, 66) individuals. Recent evidence indicates that exercise training improves muscle PO₂mv (resulting in higher PO₂mv and slower PO₂mv kinetics) during muscle contractions in healthy young rats partly via NO-dependent mechanisms (26). Whether similar improvements in contracting muscle PO₂mv occur, and, if so, whether they are mediated via enhanced NO signaling in CHF is unknown. Given the role of skeletal muscle dysfunction in the pathophysiology and limiting symptoms of CHF (8, 51, 52, 54) and the importance of PO₂mv in setting the driving force for blood-myocyte O₂ flux and metabolic control, a better understanding of the mechanistic basis for improving contracting muscle PO₂mv with exercise training in CHF is imperative to the design of effective therapeutic treatments for this disease.

The purpose of the present investigation was to determine the effects of endurance exercise training on contracting muscle PO₂mv and the contribution of NO to these responses in rats with matched central indexes of CHF. Based on the potential improvement in NO-mediated function following exercise training in CHF (24, 40, 66), we tested the hypotheses that 1) exercise training would elevate muscle PO₂mv and slow PO₂mv kinetics (i.e., resulting in higher microvascular oxygenation across the on-contractions transient); 2) increased NO levels [via the NO donor sodium nitroprusside (SNP)] would elevate PO₂mv and slow PO₂mv kinetics to a greater extent in sedentary compared with trained CHF rats; and 3) reduced NO levels [nonspecific NO synthase inhibition with N^G-nitro-l-arginine methyl ester (l-NAME)] would lower PO₂mv and speed PO₂mv kinetics to a greater extent in trained compared with sedentary CHF rats during the transition from rest to contractions.

MATERIALS AND METHODS

A total of 26 male Sprague-Dawley rats (body mass: 346 ± 6 g; ∼2 to 3 mo old; Charles Rivers Laboratories, Boston, MA) were used in the present investigation. Rats were maintained on a 12-h:12-h light/dark cycle with food and water provided ad libitum. All experimental procedures followed guidelines established by the National Institutes of Health and were approved by the Institutional Animal Care and Use Committee of Kansas State University.

Myocardial infarction was induced in all rats via ligation of the left main coronary artery as described in detail previously (14, 48–50). Briefly, rats were anesthetized with 5% isoflurane-O₂ mixture (Butler Animal Health Supply, Dublin, OH) and intubated for mechanical ventilation with a rodent respirator (model 680; Harvard Instruments, Holliston, MA) for the duration of the surgical procedure. The heart was accessed through a left thoracotomy in the fifth intercostal space and the left main coronary artery was ligated (6-0 silk suture) at ∼1 to 2 mm distal to the edge of the left atrium. The incision was then closed, and ampicillin (50 mg/kg im) was injected locally to reduce the opportunity for infection. The analgesic agents bupivacaine (1.5 mg/kg sc) and buprenorphine (0.01–0.05 mg/kg im) were administered subsequently and anesthesia and mechanical ventilation discontinued. All rats were monitored closely for ≥6 h for the development of arrhythmias and signs of undue stress (e.g., labored breathing). Rats were also monitored daily (e.g., appetite, weight loss/gain, gait, and posture) according to an intensive 10-day post-operative plan conducted in association with the university veterinary staff. The survival rate for the rats receiving the myocardial infarction operation was 88% (23 out of 26). At the end of the experimental protocol, one rat demonstrated no discernable damage to the left ventricle along with no changes in cardiac morphology or hemodynamic function and was therefore not included in the data analyses. Accordingly, results from the present investigation are presented below for a total of 22 animals. Each animal was allowed ≥3 wk of recovery before initiation of any experimental procedure. Previous reports support that this rat myocardial infarction model induces central and peripheral vascular and microvascular dysfunction consonant with CHF (10, 48–50).

Rats were then familiarized with downhill running on a custom-built motor-driven treadmill over the course of a 1-wk period (5–10 min/day at a speed of 20 m/min and −14% grade). After the familiarization phase, rats were assigned randomly to either sedentary (CHF; n = 11) or endurance exercise trained (CHF + EXT; n = 11) groups. Sedentary rats were confined to cage activities, whereas trained rats ran 5 days/wk over 6 to 7 wk on the declined treadmill (−14% grade). All rats from the CHF + EXT group underwent the same training protocol, in which treadmill running duration and speed were increased progressively from 10 min at 25 m/min to 60 min at 35 m/min. This final workload was maintained for the final 3 to 4 wk. Downhill treadmill running was selected for the present investigation given that both level and uphill treadmill running do not recruit the rat spinotrapezius muscle (31). Previous work from our laboratory has demonstrated that downhill treadmill running specifically trains the rat spinotrapezius muscle and constitutes an effective model for whole-body exercise training (23, 26). Of note, downhill running as a model of exercise training requires faster treadmill speeds than conventional level or uphill running regimens to elicit equivalent training adaptations in rats with a similar degree of left ventricular dysfunction and CHF (48). CHF and CHF + EXT rats underwent subsequent experimental procedures (described below) after ∼5 and 10 wk from the myocardial infarction surgery, respectively. This experimental design was used herein to match central cardiac function in an attempt to isolate peripheral vascular adaptations.

Peak oxygen uptake measurements.

Peak oxygen uptake (V̇o₂peak) was measured in CHF and CHF + EXT rats during a downhill (−14% grade) running test performed in a metabolic chamber placed on the treadmill. As described previously (9, 23, 26), the speed was set initially to 25 m/min for 2 to 3 min and then increased progressively in a ramplike fashion by ∼5–10 m/min until the rat was unable to keep pace with the treadmill belt or no further elevations in V̇o₂ were observed despite continued increases in treadmill speed. At this point of the test the V̇o₂peak was measured and recorded. Alterations in gait (e.g., lowering of the hindlimbs, dropping of the tail, and elevation of the snout) normally occurred immediately before termination of the test. Gas measurements were performed in real time via an inline O₂ analyzer (model S-3A/I; AEI Technologies; Pittsburgh, PA). The analyzer was calibrated with precision-mixed gases that spanned the expected range of gas concentrations based on previous studies. We have reported previously highly reproducible V̇o₂peak measurements using the aforementioned techniques and protocol (9).

Surgical preparation.

Rats were anesthetized initially with a 5% isoflurane-O₂ mixture and maintained on 2% to 3% isoflurane-O₂ and placed on a heating pad to maintain core temperature at ∼37° to 38°C as measured via rectal probe. The right carotid artery was cannulated, and a 2-Fr catheter-tipped pressure transducer (Millar Instruments; Houston, TX) was advanced into the left ventricle for measurement of systolic and diastolic pressures and left ventricular Δpressure/Δtime (LV dp/dt). Upon completion of these measurements, the transducer was removed and the carotid artery was recannulated with a catheter (PE-10 connected to PE-50; Intra-Medic Tubing; Clay Adams Brand, Sparks, MD) for continuous monitoring of mean arterial pressure (MAP; Digi-Med BPA model 200; Louisville, KY) and infusion of the phosphorescent probe palladium meso-tetra-(4-carboxyphenyl)porphyrin dendrimer (R2; 15 mg/kg; Oxygen Enterprises, Philadelphia, PA). The tail (caudal) artery was cannulated (PE-10 connected to PE-50) for blood sampling and infusion of anesthetic agents. Blood from the tail catheter was sampled at the end of the experimental protocol for determination of arterial gases, pH, and systemic hematocrit (Nova Stat Profile M; Waltham, MA).

Isoflurane-O₂ mixture inhalation was discontinued progressively after catheter placement procedures, and rats were kept under anesthesia with pentobarbital sodium administered intra-arterially to effect. The level of anesthesia was monitored frequently via the blink and toe-pinch reflexes and supplemented as necessary. Overlying skin and fascia from the middorsal region of the rat were reflected surgically to expose the right spinotrapezius muscle. The spinotrapezius muscle exhibits a mixed fiber-type composition and oxidative capacity (13) that resemble closely the human quadriceps (38), thus representing a useful analog of human locomotor muscle. The muscle was moistened frequently throughout the surgery and experimental protocol via superfusion of Krebs-Henseleit bicarbonate-buffered solution, consisting of (in mM) 4.7 KCl, 2.0 CaCl₂, 2.4 MgSO₄, 131 NaCl, and 22 NaHCO₃, equilibrated with 5% CO₂ and 95% N₂ (pH 7.4, warmed to 37° to 38°C). Surrounding tissue was covered with Saran wrap (Dow Brands, Indianapolis, IN). Stainless steel electrodes were sutured to the rostral (cathode) and caudal (anode) regions of the spinotrapezius muscle for electrically induced twitch contractions. A previous report from our laboratory has demonstrated that these surgical procedures do not impact the microvascular integrity and responsiveness of the rat spinotrapezius muscle (1).

Experimental protocol.

Three distinct contraction bouts were performed under control (5 ml Krebs-Henseleit), SNP (NO donor; 5 ml of a 300 μM solution), and l-NAME (nonisoform-specific NO synthase inhibitor; 5 ml of a 1.5-mM solution) superfusion conditions. Drugs were purchased from Sigma-Aldrich (St. Louis, MO), and concentrations were chosen based on previous studies from our laboratory (19, 20, 26, 27). All solutions were maintained at ∼37° to 38°C. Although superfusion order was randomized between control and SNP conditions, l-NAME was the last treatment due to its relatively long half-life. Importantly, a recent report from our laboratory demonstrated that the superfusion protocol used herein (see below) does not exert residual (i.e., post-washout) SNP alteration and/or impairment of skeletal muscle blood flow, V̇o₂ or PO₂mv responses either at rest or during contractions (27). The spinotrapezius was superfused with each solution (average flow rate of ∼1.5 ml/min) for a total time of 3 min, followed by ∼3 min incubation period to allow resting muscle PO₂mv to stabilize. Subsequently, electrical stimulation (1 Hz, ∼7 V, 2-ms pulse duration) of the muscle was evoked via a stimulator (model s48; Grass Technologies, Quincy, MA) for 3 min. This stimulation protocol evokes an approximately four- to fivefold increase in blood flow together with approximately six- to sevenfold increase in metabolic rate (which corresponds to ∼30% spinotrapeziusV̇o₂peak) above resting with either minor or no alterations in blood pH that are consistent with moderate intensity exercise (4, 27). The muscle was then allowed to recover for ∼25 min before the next condition was initiated (stimulation parameters were held constant). During the recovery period following the SNP condition, the muscle was superfused at an average flow rate of ∼1.5 ml/min with Krebs-Henseleit to wash out SNP. A >30-min recovery period between consecutive bouts (i.e., 3 min off-transition, ∼25 min passive recovery, 3 min superfusion, ∼3 min incubation) was used herein to prevent any priming and drug-ordering effects that could confound the experimental interpretation of the PO₂mv responses to muscle contractions (5, 20). We have reported previously that the spinotrapezius preparation exhibits reproducible PO₂mv responses from three contraction bouts separated by ∼30 min (26). At the end of the experimental protocol, rats were euthanized with intra-arterial pentobarbital sodium overdose (>50 mg/kg). For each rat, the lungs and heart were dissected, weighed, and normalized to body mass. Myocardial infarct size was determined via planimetry as described in detail previously (19).

Spinotrapezius muscle PO₂mv measurement.

PO₂mv was measured by phosphorescence quenching using a frequency domain phosphorometer (PMOD 5000; Oxygen Enterprises, Philadelphia, PA). As described in detail previously (4), this technique applies the Stern-Volmer relationship (59), which describes quantitatively the O₂ dependence of the phosphorescent probe (R2) via the following equation:

$$P{O}_{2}mv=\left[\left(\mathit{\tau °}/\tau \right)-1\right]/\left({\text{k}}_Q\times \mathit{\tau °}\right)$$

where k_Q is the quenching constant and τ and τ° are the phosphorescence lifetimes in the absence of O₂ and the ambient O₂ concentration, respectively. The phosphor R2 (τ° = 601 μs and k_Q = 409 mmHg⁻¹/s at pH 7.4 and temperature ∼38°C) (41) was infused ∼15 min before initiation of the first muscle contraction. R2 contains Pd-porphyrin cores that bind to biological macromolecules (primarily albumin in blood plasma) (67). This facilitates its uniform distribution in the plasma and provides a signal corresponding to the volume-weighed O₂ pressure in the microvascular compartment (primarily the PO₂ within capillaries, which volumetrically constitutes the major intramuscular space) (53). Restriction of the R2 probe to the microvascular space is also promoted by its negative charge (55). The common end of the bifurcated light guide was positioned ∼2–4 mm superficial to the dorsal surface of the exposed spinotrapezius muscle. The phosphorometer modulates sinusoidal excitation frequencies between 100 Hz and 20 kHz and allows phosphorescence lifetime measurements from 10 μs to ∼2.5 ms. The excitation light (524 nm) was focused on a randomly selected area of ∼2 mm diameter of exposed muscle and has a penetration depth of ∼500 μm. PO₂mv was measured continuously and recorded at 2-s intervals throughout the duration of the experimental protocol.

Analysis of muscle PO₂mv kinetics.

The kinetics of PO₂mv were described by nonlinear regression analysis using the Marquardt-Levenberg algorithm (SigmaPlot 11.2; Systat software, San Jose, CA) for the onset of contractions. Transient PO₂mv responses were fit with either a one- or two-component model as follows.

One-component:

$$P{O}_{2}m{v}_{\left(\text{t}\right)}=P{O}_{2}m{v}_{\left(\text{BL}\right)}-\mathrm{\Delta }P{O}_{2}mv\left(1-e^{-\left(t-\text{TD}\right)/\tau }\right)$$

Two-component:

$$P{O}_{2}m{v}_{\left(\text{t}\right)}=P{O}_{2}m{v}_{\left(\text{BL}\right)}-{\Delta }_{1}P{O}_{2}mv\left(1-e^{-{\left(t-{\text{TD}}_{\text{1}}\right)/\tau }_1}\right)+{\Delta }_{2}P{O}_{2}mv\left(1-e^{-{\left(t-{\text{TD}}_{\text{2}}\right)/\tau }_2}\right)$$

where PO₂mv_(t) is the PO₂mv at a given time t; PO₂mv_(BL) corresponds to the precontracting resting PO₂mv; Δ₁ and Δ₂ are the amplitudes for the first and second components, respectively; TD₁ and TD₂ are the independent time delays for each component; and τ₁ and τ₂ are the time constants (i.e., time to 63% of the response) for each component. Goodness of fit was determined using three criteria: the coefficient of determination, sum of squared residuals, and visual inspection.

The mean response time (MRT) was used to describe the overall dynamics of the PO₂mv fall following the onset of muscle contractions:

$$\text{MRT}=\text{TD}+\tau$$

where TD and τ are defined above. The MRT analysis was limited to the first component of the PO₂mv response given that inclusion of an emergent second component underestimates the actual speed of the PO₂mv fall for the onset of contractions (26).

Movement of the light guide and/or animal was avoided to monitor the same sampling site in a given animal throughout all experimental conditions. However, alteration of the PO₂mv measurement plane (e.g., due to deep sighs of the animal, etc.) during muscle contractions precluded kinetic curve fitting in some instances. Therefore, PO₂mv kinetics from the present investigation are given for the following animal numbers under the specified conditions: CHF control (10/11), CHF SNP (10/11), CHF l-NAME (10/11), CHF + EXT control (11/11), CHF + EXT SNP (7/11) and CHF + EXT l-NAME (11/11).

Citrate synthase measurement.

The activity of the mitochondrial enzyme citrate synthase (a marker of oxidative capacity) from the spinotrapezius and select individual hindlimb muscles or muscle parts (soleus, red gastrocnemius, and white gastrocnemius) was measured in duplicate from muscle homogenates by a modification of the method described by Srere (62). Upon termination of the experimental protocol and euthanasia, muscles were removed, dissected free of connective tissue, and weighed. Citrate synthase activity was measured spectrophotometrically (Spectramax M5 microplate; Molecular Devices, Sunnyvale, CA) in 300 μl aliquots at 30°C.

Statistical analysis.

Data comparison was performed using unpaired Student's t-test or two-way repeated-measures ANOVA where appropriate. F-statistics were calculated using Type III (adjusted) sums of squares due to the unbalanced nature of the data. Student-Newman-Keuls post hoc tests were used to determine where the differences were located. A one-tailed test was performed when a priori directional hypotheses were tested (19, 20, 27). The level of significance was set at p < 0.05. Results are presented as means ± SE.

RESULTS

Morphological and hemodynamic variables revealed impaired ventricular function consistent with the presence of a substantial myocardial infarction and LV wall destruction in both groups (Table 1). CHF and CHF + EXT rats showed similar central indexes of moderate compensated heart failure (Table 1; p > 0.05 for all).

Table 1.

Morphological and hemodynamic characteristics of CHF and CHF + EXT rats

	CHF	CHF + EXT
LVEDP, mmHg	17 ± 2	18 ± 2
LV dp/dt, mmHg/s	6,300 ± 347	5,982 ± 282
RV, mg	350 ± 23	378 ± 26
RV/body mass, mg/g	0.74 ± 0.06	0.70 ± 0.05
Lung, mg	2,606 ± 304	2,420 ± 304
Lung/body mass, mg/g	5.55 ± 0.75	4.51 ± 0.61
Infarct size, %	38 ± 2	37 ± 2

Values are means ± SE. LVEDP, left ventricular end-diastolic pressure; LV dp/dt, left ventricular Δpressure/Δtime; RV, right ventricle; CHF, chronic heart failure; EXT, endurance exercise trained.

The greater body weight in CHF + EXT compared with CHF rats (CHF: 479 ± 11; CHF + EXT: 541 ± 10 g; p < 0.05) can be ascribed to the experimental design used herein, i.e., endpoint was ∼5 and 10 weeks post-myocardial infarction for CHF and CHF + EXT rats, respectively. CHF + EXT rats had higher V̇o₂peak than CHF rats (CHF: 63 ± 2; CHF + EXT: 68 ± 2 ml·kg⁻¹·min⁻¹; p < 0.05). Citrate synthase activity was higher in the soleus (CHF: 29.0 ± 1.0; CHF + EXT: 36.9 ± 1.7 μmol·g⁻¹·min⁻¹), red gastrocnemius (CHF: 39.4 ± 1.4; CHF + EXT: 55.1 ± 1.3 μmol·g⁻¹·min⁻¹) and spinotrapezius (CHF: 19.7 ± 0.7; CHF + EXT: 21.6 ± 0.8 μmol·g⁻¹·min⁻¹) muscles from CHF + EXT rats (p < 0.05 for all). Citrate synthase activity from the white gastrocnemius was not different between CHF and CHF + EXT (10.6 ± 0.8 and 12.2 ± 1.0 μmol·g⁻¹·min⁻¹, respectively; p > 0.05).

No differences in relative spinotrapezius muscle mass were found between CHF and CHF + EXT (0.89 ± 0.03 and 0.94 ± 0.02 mg/g, respectively; p > 0.05).

There were no differences in arterial O₂ saturation (CHF: 90.9 ± 1.3; CHF + EXT: 93.8 ± 0.6%), pH (CHF: 7.42 ± 0.01; CHF + EXT: 7.43 ± 0.01) or systemic hematocrit (CHF: 34.3 ± 1.3; CHF + EXT: 36.0 ± 0.6%) between groups (p > 0.05 for all).

Effects of exercise training on muscle PO₂mv in CHF.

MAP was not different in CHF compared with CHF + EXT rats either before or after Krebs-Henseleit superfusion (Table 2; p > 0.05 for all). The stimulation protocol used herein did not evoke significant changes in MAP in CHF or CHF + EXT rats in any of the distinct superfusion conditions (p > 0.05 for all). The mean difference in MAP between the onset and offset of contractions were as follows: CHF control: 2 ± 1; CHF SNP: 8 ± 2; CHF l-NAME: 2 ± 1; CHF + EXT control: 0 ± 3; CHF + EXT SNP: 6 ± 2; CHF + EXT l-NAME: −1 ± 1 mmHg.

Table 2.

Mean arterial pressure (expressed in mmHg) pre- and postsuperfusion of Krebs-Henseleit (control), SNP, and l-NAME in CHF and CHF + EXT rats

	Control		SNP		l-NAME
	Pre	Post	Pre	Post	Pre	Post
CHF	105 ± 3	104 ± 3	110 ± 3	101 ± 3*	112 ± 5	113 ± 5
CHF + EXT	115 ± 7	115 ± 7	114 ± 9	106 ± 7	121 ± 6	123 ± 7

Values are means ± SE. SNP, sodium nitroprusside. Significantly different from:

^*CHF post- N^G-nitro-l-arginine methyl ester (l-NAME) superfusion.

Although resting and contracting steady-state PO₂mv [PO₂mv_(BL) and PO₂mv_(SS), respectively] were not different between groups (Fig. 1 and Table 3; p > 0.05 for both), exercise training induced significant differences in the time course of PO₂mv following the onset of contractions under the control condition (Figs. 1–4 and Table 3). Specifically, the overall speed of PO₂mv fall during contractions (MRT and T₆₃) was markedly slowed in CHF + EXT compared with CHF rats (Fig. 4 and Table 3; p < 0.05 for all).

Fig. 1.

Top: Spinotrapezius muscle PO₂mv profiles (circles) and their model fits (dashed lines) from representative chronic heart failure (CHF) and CHF + endurance exercise trained (EXT) rats in the control condition. Note that exercise training slowed the overall speed of the PO₂mv fall during contractions. Middle: PO₂mv residuals demonstrate excellent model fits. Bottom: absolute PO₂mv difference between CHF + EXT and CHF responses shown at top. Note that the greatest effect of exercise training on PO₂mv kinetics occurred during the rest-contraction transient (i.e., 0–60 s). Time zero denotes the onset of muscle contractions.

Table 3.

Spinotrapezius muscle PO₂mv kinetics following the onset of contractions under control, SNP, and l-NAME conditions in CHF and CHF + EXT rats

	Control		SNP		l-NAME
	CHF	CHF + EXT	CHF	CHF + EXT	CHF	CHF + EXT
PO2mv(BL), mmHg	24.3 ± 2.3	22.6 ± 1.9	33.7 ± 2.3*	35.3 ± 2.9*	23.5 ± 2.4†	21.3 ± 1.0†
Δ1PO2mv, mmHg	10.3 ± 1.0	9.5 ± 1.1	8.8 ± 0.8	10.4 ± 2.1	11.9 ± 1.6	11.6 ± 1.0
Δ2PO2mv, mmHg	1.9 ± 0.3	2.2 ± 0.4			2.4 ± 0.4	2.3 ± 0.4
ΔTotalPO2mv, mmHg	8.8 ± 1.0	8.7 ± 1.0	8.8 ± 0.8	10.4 ± 2.1	9.7 ± 1.3	10.6 ± 0.8
PO2mv(SS), mmHg	15.5 ± 1.8	13.9 ± 1.2	24.9 ± 2.2*	24.9 ± 2.2*	13.7 ± 1.2†	10.7 ± 0.6*†
TD1, s	8.9 ± 1.5	12.6 ± 1.3#	4.0 ± 1.7*	5.7 ± 2.1*#	6.7 ± 0.7†	11.9 ± 2.1†‡
TD2, s	45.7 ± 5.7	83.3 ± 10.5			35.4 ± 4.7	60.5 ± 15.5
τ1, s	11.9 ± 1.3	19.7 ± 3.0#	39.0 ± 6.5*	49.8 ± 8.3*‡	10.1 ± 1.4†	19.1 ± 2.9†#
τ2, s	41.2 ± 12.6	57.0 ± 9.9			59.3 ± 9.1	38.0 ± 13.5
T63, s	20.4 ± 1.2	31.2 ± 2.3‡	42.2 ± 7.7*	55.0 ± 7.1*‡	17.4 ± 1.6†	31.3 ± 3.4†‡

Values are means ± SE. PO₂mv_(BL), resting PO₂mv; Δ₁PO₂mv, amplitude of the first component; Δ₂PO₂mv, amplitude of the second component; Δ_TotalPO₂mv, overall amplitude regardless of one- or two-component model fit; PO₂mv_(SS), contracting steady-state PO₂mv; TD₁, time delay for the first component; TD₂, time delay for the second component; τ₁, time constant for the first component; τ₂, time constant for the second component; T₆₃, time to reach 63% of the primary amplitude as determined independent of modeling procedures. The one-component exponential model was used to analyze the PO₂mv kinetics in the following conditions: CHF control (2/10), CHF SNP (10/10), CHF l-NAME (1/10), CHF + EXT control (7/10), CHF + EXT SNP (7/7), CHF + EXT l-NAME (5/11). Significantly different from:

^*control within group;

^†SNP within group;

^‡CHF within superfusion condition;

^#P = 0.05–0.1 vs. CHF within condition.

Fig. 4.

Spinotrapezius muscle PO₂mv kinetics (mean response time, which describes the overall dynamics of PO₂mv fall) following the onset of contractions in the CHF and CHF + EXT groups under control, SNP, and l-NAME superfusion conditions. Significantly different from: *control within group; †SNP within group; ‡CHF within superfusion condition.

Effects of altered NO on muscle PO₂mv in sedentary and trained CHF rats.

SNP superfusion did not change MAP in either CHF or CHF + EXT rats (Table 2; p > 0.05 for both). Relative to the control condition, SNP elevated PO₂mv_(BL) and PO₂mv_(SS) in both groups to similar values (Figs. 1 and 2 and Table 3; p < 0.05 for all). No differences in the amplitude of the PO₂mv response following the onset of contractions (Δ₁PO₂mv and Δ_TotalPO₂mv) were observed between CHF and CHF + EXT rats with SNP (Table 3; p > 0.05 for all). SNP slowed MRT, T₆₃, and τ₁ in both groups such that CHF + EXT rats continued to display slower PO₂mv kinetics when compared with their sedentary counterparts (Fig. 4 and Table 3; p < 0.05 for all). It is important to note that the absolute and relative changes promoted by SNP in MRT, T₆₃, and τ₁ were not different when comparing CHF and CHF + EXT rats (data not shown; p > 0.05).

Fig. 2.

Spinotrapezius muscle PO₂mv responses from representative CHF and CHF + EXT rats under sodium nitroprusside (SNP) and N^G-nitro-l-arginine methyl ester (l-NAME) superfusion conditions. Time zero denotes the onset of muscle contractions. Note that SNP slowed PO₂mv kinetics to a greater extent in CHF + EXT compared with CHF rats. l-NAME did not alter PO₂mv kinetics in either group but reduced the contraction steady-state PO₂mv only in CHF + EXT rats when compared with control. Therefore, CHF + EXT rats continued to exhibit slower PO₂mv kinetics than their sedentary counterparts during the l-NAME condition. PO₂mv kinetics parameters are shown in Table 3 and Fig. 4.

l-NAME superfusion did not change MAP in either CHF or CHF + EXT rats (Table 2; p > 0.05 for both). Relative to the control condition, l-NAME had no effects on PO₂mv_(BL) in either group but decreased PO₂mv_(SS) only in CHF + EXT rats (Figs. 1 and 2 and Table 3; p < 0.05). No differences in the amplitude of the PO₂mv response following the onset of contractions (Δ₁PO₂mv, Δ₂PO₂mv, and Δ_TotalPO₂mv) were observed between CHF and CHF + EXT rats with l-NAME (Table 3; p > 0.05 for all). Notably, l-NAME did not alter the time course of PO₂mv fall at the onset of contractions (i.e., TD₁, τ₁, MRT, and T₆₃) in CHF or CHF + EXT rats when compared with the control condition (Figs. 3 and 4 and Table 3; p > 0.05 for all). Thus CHF + EXT rats exhibited slower PO₂mv on-kinetics (i.e., TD₁, MRT, and T₆₃) than their sedentary counterparts during the l-NAME condition (Fig. 4 and Table 3; p<0.05 for all).

Fig. 3.

Mean relative spinotrapezius muscle PO₂mv responses in CHF and CHF + EXT rats under control, SNP, and l-NAME superfusion conditions. Time zero denotes the onset of muscle contractions. SE bars were omitted for clarity. Note slower PO₂mv kinetics in CHF + EXT compared with CHF rats during the control condition. SNP slowed PO₂mv kinetics in CHF + EXT compared with CHF rats, whereas l-NAME had no effects on PO₂mv kinetics in either group relative to the control condition. PO₂mv kinetics parameters are shown in Table 3 and Fig. 4.

DISCUSSION

The present investigation demonstrates that endurance exercise training fundamentally changes the microvascular oxygenation profile (i.e., slows PO₂mv kinetics) following the onset of contractions in the spinotrapezius muscle of CHF rats. Relative to the control condition, increased NO with SNP slowed PO₂mv kinetics in both groups (such that CHF + EXT rats had greater MRT of the PO₂mv fall), whereas decreasing NO bioavailability with l-NAME had no effects on PO₂mv kinetics in either group. Contrary to our hypothesis, enhanced NO-mediated function may not be obligatory for training-induced improvements in contracting muscle microvascular oxygenation in CHF.

Exercise training and muscle microvascular oxygenation in CHF.

Skeletal muscle PO₂mv kinetics reflect the dynamic matching between Q̇O₂ and V̇o₂ within the microvascular space during transitions in metabolic demand (4). Therefore, slowed PO₂mv kinetics following the onset of contractions in CHF + EXT rats (Figs. 1–4 and Table 3) indicates an improved Q̇O₂/V̇o₂ matching secondary to faster adjustment in muscle Q̇O₂ compared with CHF rats (14). Importantly, enhanced microvascular Q̇O₂ responses across the rest-contraction transient after exercise training oppose the predations of CHF on muscle conductive and diffusive microvascular Q̇O₂ [i.e., mainly reduced red blood cell flux (f_RBC) and proportion of red blood cell-flowing capillaries] (35, 56) and reduce fractional O₂ extraction (thus increasing effectively the pressure gradient driving O₂ from the blood into the myocyte). Moreover, improvements in contracting muscle microvascular Q̇O₂ are crucial to support a faster rate of oxidative phosphorylation (i.e., faster V̇o₂ kinetics; as suggested by higher spinotrapezius citrate synthase activity seen herein and reported previously) (46, 58) following completion of an exercise training program in CHF individuals. At any given submaximal contractile activity, these training-induced adaptations in microvascular O₂ transport would be expected to attenuate perturbations of the intracellular milieu (e.g., changes in ADP, PCr, and Cr concentrations), reduce the rate of anaerobic glycolysis and reliance on finite energy sources, and contribute to increased exercise tolerance as documented previously (29, 30, 64, 65), thus providing a mechanistic basis for these effects.

The improvement (slowing) in muscle PO₂mv kinetics seen herein after training derives principally from adaptations in peripheral microvascular control given that no differences in central indexes of heart failure were observed between CHF and CHF + EXT groups (Table 1). This finding is consistent with our experimental design and with the notion that training-induced adaptations to left ventricular pump function (i.e., improved systolic volume and cardiac output) are a function of training frequency, intensity, and duration, such that central cardiac adaptations in CHF rats might only be observed following high-intensity sprint training regimens (47).

The peripheral adaptations observed herein are also consistent with the fact that CHF patients exhibit substantial plasticity within their skeletal muscle O₂ transport pathway (conductive and diffusive components) in response to exercise training (17) and the major role of skeletal muscle dysfunction in the pathophysiology of CHF (8, 51, 52, 54). As discussed previously (26, 54), there is an interdependence between muscle conductive and diffusive O₂ transport components setting fractional O₂ extraction (57): O₂ extraction = 1 − e^{−Do₂/βQm}, where DO₂ is muscle O₂ diffusing capacity (dictated largely by capillary hematocrit and the volume density of red blood cell-flowing capillaries) (18, 22), β is the slope of the O₂ dissociation curve in the physiologically relevant range, and Q̇_m is muscle blood flow. Because β is unlikely to be affected appreciably by exercise training, alterations in O₂ extraction (and thus PO₂mv) will depend primarily on the DO₂/Q̇_m ratio (57). It is interesting to note that although CHF induces peripheral microvascular dysfunction via impairments in both muscle DO₂ (reduced proportion of red blood cell-flowing capillaries) and Q̇_m (reduced f_RBC) (35, 56), exercise training can promote beneficial adaptations in both DO₂ (2, 44, 57) and Q̇_m (60). In this sense, slower PO₂mv kinetics following the onset of contractions in CHF + EXT rats (Figs. 1–4 and Table 3) is evidence for a lower DO₂/Q̇_m ratio (i.e., reduced fractional O₂ extraction) and suggests that adaptations in conductive (Q̇_m; mainly faster f_RBC kinetics) rather than diffusive O₂ transport are of relatively greater importance in improving contracting muscle microvascular oxygenation in this disease.

Effects of altered NO on PO₂mv kinetics in CHF.

NO and its derivatives are important modulators of skeletal muscle vascular control (63). Impairments in NO-mediated function contribute to diminished functional hyperemia (28, 34) and temporal Q̇O₂/V̇o₂ mismatch (19) during muscle contractions in CHF. Conversely, exercise training potentially ameliorates NO-induced vasodilation in CHF (24, 40, 66) and improves (slows) contracting muscle PO₂mv kinetics in healthy individuals (26). We therefore hypothesized initially that endurance exercise training would enhance muscle microvascular oxygenation following the onset of contractions (i.e., slow PO₂mv kinetics) in CHF partly via improved NO-mediated function. Surprisingly, SNP slowed the overall dynamics of the PO₂mv fall (mean response time; MRT) in CHF and CHF + EXT rats, whereas l-NAME did not alter MRT in either group (Fig. 4). Despite the small but significant reduction in the contracting steady-state PO₂mv [PO₂mv_(SS)] in CHF + EXT rats with l-NAME (Table 3), our results suggest that improved NO-mediated function is not obligatory for training-induced enhancement of muscle PO₂mv kinetics following contractions onset in rats with CHF.

These responses contrast markedly with those seen previously in healthy rats, in which SNP slowed PO₂mv kinetics to a greater extent in sedentary animals and l-NAME abolished the differences in PO₂mv kinetics between sedentary and trained animals that were evident during the control condition (26). It is therefore crucial to appreciate that in trained CHF skeletal muscle, enhanced NO-mediated function may not be obligatory to slow PO₂mv kinetics (and consequently to increase the pressure head for O₂ diffusion) at a time when V̇o₂ is rising at its fastest rate and improve muscle O₂ supply and oxidative function following contractions onset (4, 14, 29, 54, 64).

Potential mechanisms underlying slowed PO₂mv kinetics with training in CHF.

Although endurance exercise training improves muscle PO₂mv kinetics (potentially via a speeding of the f_RBC response as discussed above) following the onset of contractions in both healthy (26) and CHF (present investigation) rats, enhanced NO signaling does not appear to be intrinsic to training-induced adaptations in muscle microvascular O₂ exchange in CHF + EXT (Figs. 1–4 and Table 3). Despite evidence supporting an improvement in NO-induced vasodilation in CHF after exercise training (24, 40, 66), it is interesting to note that Lindsay et al. (39) reported no effects of training on acetylcholine-induced relaxation of aortic rings in CHF rats and Yi et al. (68) found preserved prostaglandin-induced relaxation of coronary arteries from trained CHF dogs. Together with the fact that impairments in NO-mediated function with CHF might be compensated by increased contribution of other vasodilators such as endothelium-derived hyperpolarizing factors (EDHF) (33, 42), these data highlight the redundancy and synergism of mechanisms regulating muscle O₂ delivery (25) and suggest that exercise training may alter the relative contribution of distinct mechanisms governing microvascular oxygenation in CHF.

Among the myriad of training-induced adaptations that might enhance contracting muscle PO₂mv kinetics in CHF + EXT are improved neurohumoral (e.g., ↓ catecholamines, arginine vasopressin, angiotensin II, endothelin-1), autonomic (↓ sympathetic tone and ↑ vagal tone), and inflammatory (e.g., ↓ tumor necrosis factor-α, interleukin-6) responses; reduced oxidative stress (e.g., ↑ superoxide dismutase, catalase, glutathione peroxidase); and structural alterations (↑ capillarity) (7, 12, 15, 51, 52, 54). Training might also produce alterations in the regional distribution of Q̇_m (49) and attenuate spatial heterogeneities in contracting muscle microvascular oxygenation (36). These and other potential adaptations may shift the balance between vasodilators and vasoconstrictors in favor of the former without obligatory improvements in NO-mediated function (e.g., ↑ EDHF and ↓ endothelium-dependent contracting factors, respectively) and thus improve muscle Q̇O₂/V̇o₂ matching (slow PO₂mv kinetics) during transitions in metabolic demand in trained CHF individuals. Future studies designed to address the relative contribution of the abovementioned factors in improving skeletal muscle PO₂mv kinetics following exercise training in CHF will be valuable.

Clinical implications.

Resolution of the mechanisms underlying the improvements in muscle microvascular oxygenation following exercise training in CHF is crucial to the development of effective therapeutic strategies targeting this disease. The current results carry important clinical implications for CHF populations by demonstrating that enhanced NO-mediated function may not be obligatory for slowed PO₂mv kinetics across the rest-contractions transient after training. Moreover, endurance exercise training might constitute a more powerful intervention than some contemporary pharmacological therapies (e.g., angiotensin-converting enzyme inhibitors and β-adrenergic blockers), which have been reported not to abrogate microvascular oxygenation deficits during muscle contractions in CHF patients (61).

Experimental considerations.

As mentioned above, downhill treadmill running was used herein as a model of endurance exercise training given that this paradigm effectively trains the rat spinotrapezius muscle (as evidenced by slowed PO₂mv kinetics, greater citrate synthase activity, and resistance to fatigue) and induces training adaptations such as increased whole-body V̇o₂peak and hindlimb muscle citrate synthase activity (present results; refs. 23, 26). In fact, the effects of downhill training on whole-body V̇o₂peak and muscle oxidative enzyme capacity observed in the current investigation are similar to those reported previously by Musch et al. (48) in male CHF rats with similar left ventricular infarct size and LVEDP subjected to uphill training (10% grade).

It could be argued that reduced driving pressure (MAP) with SNP compared with l-NAME in CHF rats (Table 2) potentially constrains blood flow dynamics and influences PO₂mv kinetics during metabolic transitions. However, previous reports from our laboratory (6) indicate that this effect is negligible or nonexistent when MAP is greater than ∼70 mmHg as herein.

Further experimental considerations involving the utilization of downhill treadmill running as a model of endurance exercise training, differences in vascular adaptations to training based on muscle fiber type composition, and potential muscle damage following eccentric exercise (downhill running) have been discussed in detail previously (26).

Conclusions

The current novel findings indicate that endurance exercise training is a powerful nonpharmacological treatment capable of improving muscle microvascular oxygenation (resulting in slowed PO₂mv kinetics) during metabolic transitions in CHF. Contrary to what was found previously in healthy skeletal muscle (26), enhanced NO-mediated function may not be obligatory for these training-induced microvascular adaptations in CHF.

GRANTS

This project was supported in part by a Fellowship from the Brazilian Ministry of Education/CAPES-Fulbright and a Doctoral Student Research Grant from the American College of Sports Medicine Foundation (to D. M. Hirai); American Heart Association Heartland Affiliate Grant 0750090Z (to T. I. Musch); Kansas State University SMILE grant; and National Heart, Lung, and Blood Institute Grant HL-108328 (to D. C. Poole).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. conception and design of research; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. performed experiments; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. analyzed data; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. interpreted results of experiments; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. prepared figures; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. drafted manuscript; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. edited and revised manuscript; D.M.H., S.W.C., C.T.H., S.K.F., D.J.M., B.J.B., T.I.M., and D.C.P. approved final version of manuscript.