Ascorbate attenuates cycling exercise-induced neuromuscular fatigue but fails to improve exertional dyspnea and exercise tolerance in COPD

Abstract

We examined the effect of intravenous ascorbate (VitC) administration on exercise-induced redox balance, inflammation, exertional dyspnea, neuromuscular fatigue, and exercise tolerance in patients with chronic obstructive pulmonary disease (COPD). Eight COPD patients completed constant-load cycling (∼80% of peak power output, 83 ± 10 W) to task failure after intravenous VitC (2 g) or saline (placebo, PL) infusion. All participants repeated the shorter of the two exercise trials (isotime) with the other infusate. Quadriceps fatigue was determined by pre- to postexercise changes in quadriceps twitch torque (ΔQ_tw, electrical femoral nerve stimulation). Corticospinal excitability before, during, and after exercise was assessed by changes in motor evoked potentials triggered by transcranial magnetic stimulation. VitC increased superoxide dismutase (marker for endogenous antioxidant capacity) by 129% and mitigated C-reactive protein (marker for inflammation) in the plasma during exercise but failed to alter the exercise-induced increase in lipid peroxidation (malondialdehyde) and free radicals [electron paramagnetic resonance (EPR)-spectroscopy]. Although VitC did, indeed, decrease neuromuscular fatigue (ΔQ_tw: PL −29 ± 5%, VitC −23 ± 6%, P < 0.05), there was no impact on corticospinal excitability and time to task failure (∼8 min, P = 0.8). Interestingly, in terms of pulmonary limitations to exercise, VitC had no effect on perceived exertional dyspnea (∼8.5/10) and its determinants, including oxygen saturation ($\text{S}{\text{p}}_{{\text{O}}_2}$) (∼92%) and respiratory muscle work (∼650 cmH₂O·s·min⁻¹) (P > 0.3). Thus, although VitC facilitated indicators for antioxidant capacity, diminished inflammatory markers, and improved neuromuscular fatigue resistance, it failed to improve exertional dyspnea and cycling exercise tolerance in patients with COPD. As dyspnea is recognized to limit exercise tolerance in COPD, the otherwise beneficial effects of VitC may have been impacted by this unaltered sensation.

NEW & NOTEWORTHY We investigated the effect of intravenous vitamin C on redox balance, exertional dyspnea, neuromuscular fatigue, and exercise tolerance in chronic obstructive pulmonary disease (COPD) patients. Acute vitamin C administration increased superoxide dismutase (marker of antioxidant capacity) and attenuated fatigue development but failed to improve exertional dyspnea and exercise tolerance. These findings suggest that a compromised redox balance plays a critical role in the development of fatigue in COPD but also highlight the significance of exertional dyspnea as an important symptom limiting the patients’ exercise tolerance.

INTRODUCTION

Patients with chronic obstructive pulmonary disease (COPD) are characterized by increased skeletal muscle fatigue and, likely as a consequence, impaired exercise tolerance (1, 2). Fatigue, defined as a reversible reduction in force/power-generating capacity of an individual muscle or muscle group, can be peripheral and/or central in origin (3). Peripheral fatigue is the result of biochemical changes within skeletal muscle, yielding an attenuated response to neural stimulation. Central fatigue refers to a failure of the central nervous system (CNS) to maintain force/power, resulting in a reduction in descending neural drive and/or output from the spinal motoneurons (4). Interestingly, redox balance may play a role in both forms of fatigue. Specifically, reactive oxygen species (ROS) contribute to peripheral fatigue by impairing excitation-contraction coupling (5, 6) and stimulating central fatigue by augmenting the discharge rate of metabosensitive group IV afferents (7). Compared with their healthy counterparts, patients with COPD suffer from greater levels of ROS (8–11) and systemic inflammation (12, 13) even at rest, but particularly during exercise. Consequently, the contribution of free radicals to fatigue is likely greater in patients with COPD and could help to explain the faster development of fatigue during exercise in this population (14).

Recent studies, employing single-joint, small-muscle mass exercise and antioxidant interventions, identified ROS as a contributor to the impaired development of fatigue (using intravenous ascorbate, VitC) (15), and limited endurance capacity (utilizing N-acetylcysteine) (16), in patients with COPD. However, as limb and respiratory muscles are the main source of augmented ROS in COPD (8, 17, 18), exercise engaging only a small muscle mass, with low ventilatory requirements (19), likely evokes substantially less systemic oxidative stress compared with a locomotor activity, such as cycling, involving a large muscle mass and demanding significant respiratory muscle work. Of note, COPD-specific pulmonary limitations (14, 20, 21) have, in addition to altered intrinsic muscle characteristics (22–24), been identified as significant contributors to the greater exercise intolerance and compromised fatigue resistance in COPD patients. Indeed, although limb discomfort can limit exercise in most healthy individuals (25), there is considerable evidence that exertional dyspnea is key in limiting exercise in COPD (26, 27). Therefore, cycling exercise may be better suited to investigate the overall impact of exercise-induced redox balance, inflammation, and exertional dyspnea on fatigue in patients with COPD.

As noted above, metabosensitive group IV afferents can be stimulated by free radicals and have been documented to facilitate central fatigue (28, 29); however, this may be achieved, at least in part, by impairing corticospinal (CS) excitability, which negatively impacts neural muscle activation (30–33). Indeed, CS projections constitute the major motor pathway linking the brain to the working skeletal muscle (34), and limited research in this area has previously linked COPD to a dysfunctional CS pathway and compromised cortical excitability (35, 36). However, these findings were obtained during acute exacerbations of COPD and are limited to studies of muscles of the hand at rest. Presently, there are no data documenting the impact of ROS on CS excitability during locomotor exercise in patients with COPD.

Consequently, utilizing intravenous VitC, this study aimed to investigate the effect of exercise-induced changes in redox balance, inflammation, exertional dyspnea, neuromuscular fatigue, and cycling exercise tolerance in patients with COPD. We hypothesized that VitC administration would 1) increase antioxidant capacity, attenuating both oxidative stress and inflammation as a result of exercise, and 2) attenuate both exertional dyspnea and the development of fatigue, in part by increasing CS excitability, and therefore improve exercise tolerance in patients with COPD.

METHODS

Participants

Eight patients with COPD [GOLD classification 1–3; forced expiratory flow in 1 s (FEV₁): 60 ± 4% predicted; peak O₂ consumption ($\dot{\text{V}}{\text{O}}_{2\text{p}\text{e}\text{a}\text{k}}$): 17 ± 2 mL·kg⁻¹·min⁻¹; peak work rate (W_peak): 106 ± 12 W] volunteered to participate in the study. All patients had a history of smoking (1–3 packs/day); one participant was actively smoking at the time of the experiment but refrained from the use of tobacco products for 12 h before all data collection. The subjects were not engaged in regular physical activity and did not participate in a pulmonary rehabilitation program. All participants were utilizing inhaled bronchodilators before and throughout the study but refrained from using them the morning of the testing sessions. To avoid the potential interaction with ROS (37, 38), none of the participants used supplemental oxygen. Patient characteristics, including general hematological parameters obtained from a blood draw, are summarized in Table 1. All participants refrained from consuming supplemental vitamins for 1 wk before and during their participation in the study. Written informed consent was obtained from all participants. All experimental procedures were approved by the University of Utah and the Salt Lake City Department of Veterans Affairs Medical Center Institutional Review Boards and conformed to the Declaration of Helsinki, except for registration in a database.

Table 1.

Subject characteristics

Age, yr	65 ± 3
Height, cm	180 ± 1
Weight, kg	94 ± 3
$\text{S}{\text{p}}_{{\text{O}}_2}$, %	95 ± 0.4
FVC, L	3.6 ± 0.2
FVC, % predicted	76 ± 4
SVC, L	4.1 ± 0.3
SVC, % predicted	85 ± 5
FEV1, L	2.0 ± 0.2
FEV1, % predicted	60 ± 3
FEV1/FVC	60 ± 2
FEV1/FVC, % predicted	75 ± 3
FEF25-75, L/s	1.4 ± 0.3
FEF25-75, % predicted	44 ± 6
IC, L	3.4 ± 0.3
IC, % predicted	100 ± 9
ERV, L	0.65 ± 0.10
ERV, % predicted	47 ± 6.6
Glucose, mg/dL	115.0 ± 13.2
Cholesterol, mg/dL	179.8 ± 19.0
Triglycerides, mg/dL	119.5 ± 16.0
HDL, mg/dL	45.9 ± 3.5
LDL, mg/dL	112.4 ± 14.0
WBC, K/μL	6.9 ± 1.0
RBC, M/μL	4.7 ± 0.2
Hemoglobin, g/dL	14.5 ± 0.6
Hematocrit, %	43 ± 2

Values are means ± SE. ERV, expiratory reserve volume; FEF_25-75, forced expiratory flow from 25 to 75% of forced vital capacity (FVC); FEV₁, forced expiratory flow in 1 s; HDL, high-density cholesterol; IC, inspiratory capacity; LDL, low-density cholesterol; RBC, red blood cells; $\text{S}{\text{p}}_{{\text{O}}_2}$, oxygen saturation; SVC, slow vital capacity; WBC, white blood cells. N = 8.

Experimental Protocol

Exercise.

All participants were thoroughly familiarized with the experimental procedures during two preliminary visits; all exercise sessions were separated by a minimum of 48 h. During the first visit to the laboratory, participants performed a maximal incremental cycling exercise test to determine W_peak and $\dot{\text{V}}{\text{O}}_{2\text{p}\text{e}\text{a}\text{k}}$. After a 2-min warm-up at 20 W, the workload was increased by 5–15 W/min (14) until participants could no longer continue the exercise. Then, with the goal of minimizing the potential influence of learning effects, all participants practiced the constant-load cycling exercise (79 ± 7% W_peak) to task failure (T_lim), at a self-selected constant pedal frequency (85 ± 10 rpm), several times. Task failure was defined as a >10% drop (in rpm) from the target cadence. Participants were enrolled in the experimental trials after a reproducible cycling time to task failure (coefficient of variation < 10%) was achieved during the practice sessions. This required two to four practice trials, performed on separate days. The two subsequent visits were single blinded and in random order: 1) with intravenous bolus infusion of saline (0.9% NaCl infused at 1 mL/min for 20 min, before any cycling exercise) (placebo trial, PL) and 2) with intravenous bolus infusion of ascorbate (100 mg/mL l-ascorbic acid dissolved in normal saline, infused at 1 mL/min for 20 min, before any cycling exercise) (experimental trial, VitC). A 20-gauge intravenous catheter was placed at the antecubital vein near the elbow and was used to infuse isotonic saline or ascorbate and to take blood samples before and immediately after exercise for later analysis. Finally, to allow “isotime” comparisons, all participants returned for a final visit during which the shorter of the two trials was repeated (isotime) with the other infusate (PL or VitC).

Cycle ergometer setup.

Participants were positioned on the cycle ergometer (Velotron, Elite Model; Racer Mate, Seattle, WA) with their feet fastened securely to the pedals. The participants’ preferred seat height was recorded during the initial visit and fixed for all remaining trials. A mouthpiece attached to a one-way nonrebreathe valve (Hans Rudolph Inc., Shawnee, KS) was fixed, with a clamp, directly in front of the participant at a comfortable height to ensure consistent posture and minimal head movement. The crank angle of the cycle ergometer was monitored continuously via a calibrated linear encoder mounted on the crankshaft. All stimulations during cycling were elicited at a fixed point of the crank cycle during the peak electromyogram (EMG) burst in the cycle, measured on the right vastus lateralis (VL) (knee angle ∼100°).

Measurements

Exercise responses.

Ventilation and pulmonary gas exchange were measured at rest and during exercise with a metabolic cart (Medgraphics Ultima CFX; MGC Diagnostics, St. Paul, MN). Heart rate was determined from the R-R interval of a 12-lead electrocardiogram (Nasiff Cardiocard, Central Square, NY), and arterial oxygen saturation ($\text{S}{\text{p}}_{{\text{O}}_2}$) was estimated with a forehead pulse oximeter (Nellcor N-595, Pleasanton, CA). Ratings of exertional dyspnea were taken every minute during exercise with Borg’s modified CR10 scale (39). Esophageal pressure (Pes) was measured from a balloon placed, via the nares, in the lower third of the esophagus (40). Work of breathing (W_r) was estimated via the pressure-time products, calculated as described previously (41). The depth of the balloon within the esophagus was marked during the initial trial and kept constant through the remaining experiments.

Exercise flow volume measurements.

While participants were seated in a chair, baseline forced vital capacity (FVC) and slow vital capacity (SVC) maneuvers were performed, during each visit, before exercise testing and according to American Thoracic Society/European Respiratory Society standards (42). The largest FVC and SVC were used to determine the maximal flow-volume loop (MFVL) and vital capacity, respectively, for all flow limitation measurements before and during exercise (43). During rest and every other minute during exercise, participants were asked to perform a maximal inspiration to establish maximal inspiratory capacity (IC) needed to correctly place the exercise tidal flow-volume loop within the MFVL performed at the beginning of the trial (43). Expiratory reserve volume (ERV) was then determined from the four consecutive breaths immediately following the IC maneuver. Maximal inspiratory pressure (MIP) was defined as the lowest Pes established during the IC maneuver. After the IC assessment, ∼10 tidal flow-volume loops were recorded with commercial software (Breeze; MGC Diagnostics). The percentage of expiratory flow limitation (EFL) was determined as the percentage of the exercise tidal flow-volume loop that either met or exceeded the boundary of the MFVL (44). Dynamic hyperinflation (DH) was defined as an increase in ERV over functional residual capacity (FRC) (45). Of note, for reasons related to patient safety (i.e., dizziness/syncope following FVCs), MFVLs were obtained while participants were seated in a chair (vs. on the bike) and, to minimize the delay between the end of exercise and the assessment of postexercise neuromuscular function, only before the start of exercise. Consequently, because of postural differences between sitting in a chair and on a bike, the potential for a bronchodilation-induced widening of the MFVL during exercise, and the fact that the MFVL was obtained outside a body plethysmograph, the utilized preexercise MFVLs might not fully reflect the patients’ true MFVLs during upright cycling and could have influenced the quantification of expiratory flow limitations (46).

Torque and electromyogram recordings.

Torque and EMG recordings were measured as described previously (47). Briefly, quadriceps torque of the right leg was measured with a calibrated linear strain gauge (MLP 300; Transducer Techniques, Temecula, CA). Electromyogram recordings were recorded with surface electrodes (Ag-AgCl, 10-mm diameter) placed over the muscle belly of the vastus lateralis (VL) and biceps femoris (BF) in a bipolar configuration. Surface electrodes were placed according to SENIAM-recommended guidelines (48). EMG signals were amplified 1,000 times (Neurolog Systems; Digitimer, Welwyn Garden City, UK), band-pass filtered (20–1,000 Hz, NL-844; Digitimer), and analog to digitally converted at a sampling rate of 2,000 Hz with a 16-bit Micro 1401 mk-II and Spike 2 data collection software (Cambridge Electronic Design Ltd, Cambridge, UK).

Stimulations.

Two forms of stimulations were used during each session: 1) electrical motor nerve stimulation (MNS) and 2) transcranial magnetic stimulation (TMS).

electrical motor nerve stimulation.

The position of the stimulating electrode on the femoral nerve (located high in the femoral triangle) that elicited the greatest compound muscle action potential (M wave) in VL and quadriceps twitch torque was determined by delivering low-intensity single-pulse stimuli (200-µs pulse width; 100–150 mA) with a cathode probe (with the anode fixed between the greater trochanter and iliac crest) connected to a 400-V stimulator (model DS7AH; Digitimer). Once established, the cathode electrode was fixed in this optimal position. Thereafter, the stimulation intensity was increased in 30-mA increments until the size of the M wave exhibited no further increase (i.e., maximal M wave, M_max) at rest and during a 50% quadriceps maximum voluntary contraction (MVC). The stimulation intensity was set at 130% of that which elicited M_max and kept constant throughout the session. Optimal position for the MNS electrode and stimulator intensity were reestablished for all trials (PL: 640 ± 82 mA, VitC: 667 ± 94 mA).

transcranial magnetic stimulation.

A double-cone coil (diameter 130 mm) connected to a magnetic stimulator (Magstim 200; The Magstim Company Ltd, Dyfed, UK) was used to elicit motor evoked potentials (MEPs) in the VL. First, during a 20% quadriceps maximum voluntary contraction (MVC), the optimal coil position (posterior to anterior direction of current flow in the motor cortex) to preferentially activate the left motor cortex was determined by finding the location that elicited the greatest MEP and twitch torque for a given stimulator output (position relative to vertex: ∼2–3 cm lateral). This location was marked directly on the scalp for accurate placement throughout the session. Active motor threshold (AMT), defined as the smallest intensity of stimulation used to elicit a MEP, was then established during a VL contraction evoking an EMG response that corresponded to 20% of maximal VL EMG output. The stimulator intensity was set to 120% of AMT (72 ± 9% of maximum stimulator output) during the 20% of maximal EMG contraction to ensure a clear MEP. This established TMS intensity was used to measure CS excitability during the pre- and postexercise assessment of quadriceps function. TMS intensities were reestablished in every session. The output intensity of the stimulator was similar between sessions (P = 0.4).

Neuromuscular assessment of quadriceps function and corticospinal excitability.

Participants were seated comfortably on a custom-built chair with full back support, such that the hip and knee were at ∼120° and ∼90° of flexion, respectively. A cuff attached to the strain gauge was also attached ∼2 cm above the lateral malleolus of the right leg. Cuff position was marked at the ankle to ensure identical positioning between testing sessions. Then, three sets of contractions, separated by 45 s, were performed. In each set, participants performed a 3-s MVC during which a superimposed twitch was elicited via MNS to calculate the voluntary quadriceps activation (VA) (49). Immediately after the MVC, the femoral nerve was stimulated again to evoke a potentiated quadriceps resting twitch (Q_tw,pot). Then, for the measurement of CS excitability, three single TMS pulses and one MNS pulse (randomized) were elicited during a sustained isometric knee extension corresponding to 20% of each participant’s maximal EMG output (obtained during MVC). Three sets were performed before infusion (Pre) and after infusion (Post-Inf). Two sets were performed immediately after exercise (Post).

Stimulations during cycling.

Three TMS and one MNS were elicited at the beginning of exercise (∼5 s after target workload was reached), every other minute during exercise, and within the final minute of the trial. Each stimulation was separated by at least five full pedal revolutions. Intensity of stimulator outputs remained constant from the established intensities for CS excitability before exercise. EMG during cycling was normalized to M_max and served as a surrogate for central motor drive.

Analysis of neuromuscular data.

The twitch amplitude from both MNS- and TMS-evoked twitches was determined by taking the peak torque minus the onset torque. The peak-to-peak amplitude and area of MEPs and M_max were measured between cursors placed to encompass all phases of evoked potentials in the VL. The area of each MEP was normalized to that of M_max to account for changes in muscle sarcolemmal excitability. The cycling EMG signal was rectified and waveform average analysis was performed over a 20-s segment, immediately after the stimulation set had been completed. The reference point for overlaying and averaging was taken as the same point on the crank angle that was used to elicit stimulations (knee angle ∼100°). EMG during each revolution was averaged over a 100-ms window (50 ms before and 50 ms after the ∼100° knee angle). Root-mean-squared electromyogram (EMG_rms) was also measured during the static contractions from a 100-ms segment before the point of stimulation.

The contractile properties of quadriceps muscle fibers were assessed from resting twitches elicited by MNS. Contraction time (CT) was represented by the time from the onset of the twitch to peak amplitude. Rate of force development (RFD) was determined as the twitch amplitude normalized to CT. Maximal rate of force development (MRFD) was calculated by measurement of the steepest rate of torque development. This was determined as the highest positive derivative of the torque for an interval of 10 ms between two cursors placed either side of the rise in torque. Maximal rate of relaxation (MRR) was calculated as the steepest rate of decline in torque, determined by the highest negative derivative of the torque for an interval of 10 ms between two cursors placed either side of the fall in torque. Half-relaxation time (RT_0.5) was taken as the interval between the peak amplitude and the point at which torque was reduced by 50%.

Antioxidant capacity, inflammation, oxidative stress, and free radicals.

For all saline and VitC infusion trials, venous blood samples were obtained from the antecubital vein at rest (Pre) and immediately after exercise (Post). Plasma samples were stored at −80°C until later analysis. All samples were analyzed as described previously (15). Briefly, endogenous antioxidant activity was determined by quantifying plasma superoxide dismutase (SOD) and catalase (CAT) activity (50) (Cayman Chemical Company, Ann Arbor, MI). C-reactive protein level, a marker of inflammation, was determined by an immunoassay on plasma (R&D Systems, Minneapolis, MN). Lipid peroxidation, a marker of oxidant damage, was assessed by plasma malondialdehyde (MDA) levels (Bioxytech LPO-586, Foster City, CA). Oxygen-centered free radical levels were assessed by spin trapping and electron paramagnetic resonance (EPR). Three milliliters of venous blood was collected into a Vacutainer containing 1 mL of the spin trap α-phenyl-tert-butylnitrone (PBN) (0.0140 mol/L). After centrifugation, the PBN was extracted from the serum supernatant with toluene, and the adduct (200 µL) was pipetted into a precision-bore quartz EPR sample tube (Wilmad, Vineland, NJ) previously flushed with compressed N₂. EPR was then performed at 21°C with an EMX X-band spectrometer (Bruker, Massachusetts) with commercially available software (version 2.11, Bruker Win EPR System).

Statistical Analysis

Data are presented as means ± SE. A two-way repeated-measures ANOVA (condition × time) was used to compare the effect of VitC on physiological parameters; a Tukey post hoc test was used to correct for multiple comparisons if a main effect was identified. A one-way repeated-measures ANOVA was used to compare the effect of treatment (VitC vs. PL) on physiological parameters averaged over the final minute of exercise and indexes of fatigue; a Tukey post hoc test was used to correct for multiple comparisons if a main effect was identified. Statistical significance was set at P < 0.05.

RESULTS

Impact of VitC on Exercise-Induced Changes in Blood-Borne Biomarkers

Neither the VitC nor the PL infusion evoked any side effects, and all participants remained unaware of the experimental condition. The effects of VitC infusion on the pre-/postexercise change in antioxidant capacity, inflammation, lipid peroxidation, and free radicals are documented in Table 2. VitC infusion significantly increased end-exercise plasma ascorbate levels, whereas there was no change in the PL condition. VitC infusion increased endogenous antioxidant capacity during exercise as reflected by the significant increase in end-exercise SOD levels. Furthermore, whereas C-reactive protein was significantly increased after exercise in the PL condition in all participants, this increase was prevented with the VitC infusion (P = 0.3).

Table 2.

Quantitative assessment of antioxidants, oxidative stress, and inflammation at baseline and after exercise following an intravenous saline or ascorbate infusion

	PL	VitC
Ascorbate Pre, µg/mL	8 ± 1	7 ± 3
Ascorbate Post Ex, µg/mL	8 ± 3	91 ± 30*†
CRP Pre, ng/mL	1,674 ± 379	1,620 ± 357
CRP Post Ex, ng/mL	2,782 ± 660*	1,757 ± 384†
MDA Pre, µM	4.2 ± 0.4	3.9 ± 0.3
MDA Post Ex, µM	4.4 ± 0.4	4.3 ± 0.2
SOD Pre, U/mL	2.31 ± 0.36	1.95 ± 0.42
SOD Post Ex, U/mL	2.95 ± 0.35*	4.46 ± 0.44*†
CAT Pre, μM	113 ± 26	98 ± 18
CAT Post Ex, μM	163 ± 26*	184 ± 21*
EPR (AUC) Pre	2.7 ± 0.6	2.7 ± 0.6
EPR (AUC) Post Ex	9.5 ± 4.0*	9.8 ± 3.4*

Values are means ± SE. AUC, area under the curve; CAT, catalase; CRP, C-reactive protein; EPR, electron paramagnetic resonance; MDA, malondialdehyde; PL, placebo (saline); Post Ex, after exercise; Pre, baseline; SOD, superoxide dismutase; VitC, ascorbate. *P < 0.05 vs. Pre. †P < 0.05 vs. PL. N = 8.

Cycling Time to Task Failure and Exertional Dyspnea

There was no difference in cycling time to task failure between PL and VitC (P = 0.8; Table 3). Exertional dyspnea was not different between PL and VitC (P = 0.8).

Table 3.

Responses to final minute of constant-load exercise

	PL	VitC
Work rate, W	83 ± 10	83 ± 10
Work rate, % of peak	79 ± 3	79 ± 3
Exercise time, min	8.0 ± 1	8.0 ± 1
$\text{S}{\text{p}}_{{\text{O}}_2}$, %	92 ± 1	93 ± 1
HR, beats/min	126 ± 9	128 ± 10
Exertional dyspnea (0–10)	8.6 ± 0.8	8.4 ± 0.8
Ti, s	0.57 ± 0.06	0.57 ± 0.04
Te, s	1.07 ± 0.11	1.10 ± 0.06
Ti/Ttot	0.35 ± 0.01	0.34 ± 0.01
Vt, L	1.95 ± 0.2	1.97 ± 0.21
fR, breaths/min	39.4 ± 4.3	37.1 ± 2.1
$\dot{\text{V}}\text{E}$, L/min	74 ± 8	72 ± 8
$\dot{\text{V}}{\text{O}}_2$, L/min	1.59 ± 0.15	1.50 ± 0.18
$\dot{\text{V}}{\text{O}}_2$, %$\dot{\text{V}}{\text{O}}_{2\text{max}\hspace{0.17em}}$	99 ± 3	97 ± 4
$\dot{\text{V}}\text{C}{\text{O}}_2$, L/min	2.03 ± 0.2	1.97 ± 0.23
$\dot{\text{V}}\text{E}$/$\dot{\text{V}}{\text{O}}_2$	49 ± 4.8	52 ± 6.3
$\dot{\text{V}}\text{E}$/$\dot{\text{V}}\text{C}{\text{O}}_2$	37 ± 3	39 ± 4
ICdyn, L	2.7 ± 0.2	2.6 ± 0.2
EFL, %	72 ± 8	74 ± 9
MIP, cmH2O	−40 ± 4	−38 ± 4

Values are means ± SE. EFL, expiratory flow limitation; f_R, respiratory frequency; HR, heart rate; IC_dyn, dynamic inspiratory capacity; MIP, maximal inspiratory pressure; PL, placebo; $\text{S}{\text{p}}_{{\text{O}}_2}$, oxygen saturation; Te, expiratory time; Ti, inspiratory time; Ttot, respiratory cycle duration; $\dot{\text{V}}\text{C}{\text{O}}_2$, CO₂ production; $\dot{\text{V}}\text{E}$, minute ventilation; VitC, ascorbate; $\dot{\text{V}}{\text{O}}_2$, oxygen consumption; $\dot{\text{V}}{\text{O}}_{2\text{max}\hspace{0.17em}}$, maximal $\dot{\text{V}}{\text{O}}_2$; Vt, tidal volume. N = 8.

Neuromuscular Function and Corticospinal Excitability

The indexes of neuromuscular function (MVC, Q_tw,pot, VA, M_max, MEP) were not affected by the infusion of VitC (P > 0.4) or saline (P > 0.5) and were similar between days (P > 0.4). Exercise-induced changes in neuromuscular function are presented in Table 4. Preexercise M_max area (∼0.03 mVs) and peak-to-peak amplitude (∼4.4 mV) were similar between conditions before exercise (P > 0.3) and did not change after PL or VitC infusion (P > 0.3) or after exercise (P > 0.3). Exercise in the PL condition induced a significant degree of peripheral fatigue, as indicated by the reduction in both Q_tw,pot and MVC (Fig. 1). VitC significantly reduced the pre-/postexercise fall in Q_tw,pot (by ∼22%) and MVC (by ∼29%) but had no effect on VA (Fig. 1).

Table 4.

Exercise-induced changes in neuromuscular function

	PL	VitC
RFD, Nm/s	0.27 ± 0.04*†	0.35 ± 0.06*
RFD, %c	−30 ± 6†	−17 ± 9
CT, ms	138 ± 18*†	116 ± 13*
CT, % vs. Pre	12 ± 3	5 ± 6
MRFD, Nm/s	594 ± 106*†	706 ± 120*
MRFD, % vs. Pre	−33 ± 7†	−17 ± 10
MRR, Nm/s	305 ± 71*	314 ± 41*
MRR, % vs. Pre	−16 ± 12	−7 ± 10
RT0.5, ms	148 ± 19*†	121 ± 13*
RT0.5, % vs. Pre	11 ± 4	4 ± 5
VL SP, ms	106 ± 14	102 ± 3
VL SP, % vs. Pre	−6 ± 6	4 ± 3

Values are means ± SE. CT, contraction time; MRFD, maximal rate of force development; MRR, maximal rate of relaxation; PL, placebo; Pre, baseline; RFD, rate of force development; RT_0.5, half-relaxation time; VitC, ascorbate; VL SP, silent period of the vastus lateralis. *P < 0.05 vs. Pre. †P < 0.05 vs. PL. N = 8.

Figure 1.

Neuromuscular fatigue and corticospinal excitability represented as % change from pre- to postexercise. A: maximal voluntary contraction (MVC). B: potentiated quadriceps twitch torque (Q_tw,pot). C: voluntary activation (VA). D: changes in motor evoked potentials (MEPs) were evoked during a quadriceps contraction corresponding to 20% of the electromyogram (EMG) obtained during the preexercise MVC. Data are presented as means ± SE. PL, placebo; VitC, ascorbate. *P < 0.05 vs. PL. N = 8.

Various EMG data obtained during cycling exercise are presented as percent change from the first stimulation set, occurring ∼5 s after the onset of exercise (Fig. 2A). VitC attenuated the rise in EMG during exercise (P < 0.05). M_max area, determined during cycling, remained unaltered in both conditions (PL 28 ± 3 µV·s; VitC 28 ± 4 µV·s; P > 0.4). MEPs were unchanged before, during, and after exercise in both conditions (Fig. 2B and Fig. 1D).

Figure 2.

Changes in electromyography (EMG, A) and motor evoked potentials (MEPs, B) during cycling exercise. Patients were stimulated during the first minute of cycling and every other minute thereafter until exhaustion. The 50% time point corresponds with the stimulation set closest to 50% of time to exhaustion. All MEPs were normalized to maximal M wave (M_max) from the same stimulation set. Data are presented as means ± SE. PL, placebo; VitC, ascorbate; VL, vastus lateralis. *P < 0.05 vs. VitC. N = 8.

Pulmonary Responses

Ventilatory responses, pulmonary gas exchange, heart rate, and $\text{S}{\text{p}}_{{\text{O}}_2}$ during the final minute of exercise are reported in Table 3. During the final minute of exercise, IC was, compared with rest, decreased in both conditions (PL 2.7 ± 0.2 L, VitC 2.6 ± 0.2 L; P < 0.001). Compared with the start of exercise, inspiratory reserve volume (IRV) progressively decreased whereas ERV increased during exercise (both P < 0.05; Fig. 3C). There was no effect of VitC on IRV and ERV (P > 0.5). W_r is illustrated in Fig. 3. MIPs were not affected by VitC (P > 0.6) (Table 3). EFL significantly increased throughout the exercise. VitC had no effect on EFL (Table 3).

Figure 3.

Pulmonary mechanics and flow-volume loops. All data points were recorded every other minute while there were no femoral or transcranial magnetic stimulations. End represents the final 30 s of exercise (uninterrupted by stimulations). A: respiratory muscle work (W_R). B: inspiratory (IRV) and expiratory (ERV) reserve volume. C: representative example illustrating the maximum flow-volume loop (MFVL) and exercise-induced changes in tidal flow-volume loops. Data are presented as means ± SE. PL, placebo; $\dot{V}E$, minute ventilation; VitC, ascorbate; W_R-exp, expiratory W_R; W_R-insp, inspiratory W_R N = 8.

DISCUSSION

This study sought to investigate the effect of VitC on exercise-induced redox balance, markers for inflammation, exertional dyspnea, neuromuscular fatigue, and cycling exercise tolerance in patients with COPD. Furthermore, the impact of ROS on CS excitability, which when attenuated hinders neural muscle activation and can contribute to fatigue during locomotor exercise, was assessed in this patient population. Acute VitC administration increased the patients’ antioxidant capacity, as suggested by the 129% increase in SOD, and attenuated the development of fatigue, a critical determinant of exercise tolerance, without affecting CS excitability during cycling exercise. However, VitC failed to improve exertional dyspnea, a key factor limiting cycle exercise performance in patients with COPD. Together, these findings suggest that a compromised redox balance plays a critical role in the exaggerated development of neuromuscular fatigue in patients with COPD but also highlight the significance of exertional dyspnea as an important symptom limiting endurance exercise tolerance in this population.

Impact of VitC on Selected Blood-Borne Biomarkers

Intravenous ascorbate improved the patients’ antioxidant defense system including CAT and SOD but did not affect the exercise-induced increases in oxygen-centered free radicals and lipid peroxidation assessed in the plasma (Table 2). As VitC is, generally, thought to attenuate oxidative stress by tipping the balance between pro- and antioxidant forces toward the latter (51), these observations were unexpected but not entirely surprising. Specifically, the current indexes of VitC-induced alterations in oxidative stress were limited to plasma MDA levels and mainly (52–54), but not exclusively, oxygen-centered free radicals (i.e., EPR) and do not exclude the potential positive effects on other free radical species and indexes of oxidative stress. Furthermore, the markers of oxidative stress employed are based on venous plasma samples and therefore may not exactly reflect the effect of the VitC on the cellular redox balance within the exercising muscle. Finally, high doses of exogenous antioxidants, such as vitamin C, have been suggested to display prooxidative effects (55, 56). In this case, the improved antioxidative capacity may have now faced elevated prooxidative activity, with the net effect of a similar exercise-induced increase in oxygen-centered free radicals and lipid peroxidation in the PL and VitC trials.

Interestingly, VitC prevented the exercise-induced increase in C-reactive protein observed during PL exercise (Table 2). This confirms earlier findings (57) and suggests that intravenous ascorbate may have attenuated the heightened inflammatory response characterizing patients with COPD (58). Importantly, since CRP is only one of many indexes of inflammation, the present observations do not allow for a definitive conclusion related to the effect of VitC on the inflammatory response to exercise. Moreover, our findings cannot determine whether the invariant level of C-reactive protein following exercise with VitC is secondary to the attenuated neuromuscular fatigue and, therefore, ameliorated muscle damage, or whether the attenuated inflammatory response actually contributed to the decrease in fatigue. This conundrum, although difficult to unravel, deserves additional evaluation.

Exercise-Induced Neuromuscular Fatigue and Corticospinal Excitability

Cycling exercise resulted in substantial peripheral quadriceps fatigue that was attenuated by ∼25% when the identical bout was repeated after the VitC infusion (Fig. 1). Interestingly, as limb and respiratory muscles are recognized to be the major source of augmented ROS in COPD (8, 17, 18, 59), VitC during cycling exercise (characterized by a large active muscle mass and high respiratory muscle work) attenuated peripheral fatigue to an extent (∼25%; Fig. 1) similar to that previously documented during single-leg knee extension exercise (characterized by small active muscle mass and low respiratory muscle work) (15, 60).

As ROS are recognized to impair cellular components that result in peripheral fatigue and lower myofibrillar Ca²⁺ sensitivity and Ca²⁺ reuptake by the sarcoplasmic reticulum (61, 62), the observed improvement in peripheral fatigue during VitC may be the consequence of a positive effect of VitC on the intramuscular redox balance (15). Indeed, CT, RFD, MRFD, RT_0.5, and MRR, all variables directly influenced by myofibrillar Ca²⁺ sensitivity and SR Ca²⁺ cycling (63), were attenuated to a lesser extent during VitC compared with PL (Table 4). These findings highlight a potential role of ROS in determining the patients’ compromised peripheral fatigue resistance and suggest that VitC may attenuate peripheral fatigue in COPD, at least in part, by improved intramuscular Ca²⁺ handling. In addition to these intramuscular effects, VitC may have mediated the decrease in peripheral fatigue through its beneficial effects on limb blood flow. Indeed, endogenous antioxidants can improve exercising leg blood flow in COPD by alleviating the adverse impact of oxidative stress on various vascular control mechanisms (64). It is therefore reasonable to speculate that, given the sensitivity of peripheral fatigue to muscle O₂ delivery (65), an increase in leg blood flow, secondary to the VitC infusion, could partially account for the lower level of end-exercise neuromuscular fatigue during VitC.

Maintaining or enhancing the functional integrity of the CS pathway during physical activity is key for the CNS to optimally “drive” muscle contraction required for a given task (29). While increases in EMG (i.e., surrogate for central motor drive) enhance CS excitability during cycling exercise, peripheral fatigue inhibits the motor pathway, with the net effect of no change in CS excitability in healthy individuals (31, 32, 47, 66). Furthermore, group III/IV muscle afferents, stimulated by ROS (7) and metabolites associated with peripheral fatigue (67), can alter CS excitability (68, 69). However, despite the positive effects of VitC on redox balance and peripheral fatigue in the patients with COPD, overall CS excitability and exercise-induced changes in VA were unaltered by this intervention. Although these findings suggest that oxidative stress and peripheral fatigue do not compromise the net excitability of the CS pathway during exercise in patients with COPD, the present data cannot address their specific effect on motor cortical and spinal motoneuronal excitability, the two compartments determining the net excitability of the CS motor pathway.

Pulmonary System Responses and Endurance Exercise Tolerance

All patients exhibited typical COPD-specific pulmonary responses to exercise, including, compared with healthy individuals (14, 70), an exaggerated W_r, exercise-induced arterial desaturation, evidence of expiratory flow limitation (see methods for critical consideration), and dynamic hyperinflation (Fig. 3, Table 3). As both exaggerated W_r and arterial desaturation limit peripheral hemodynamics and limb O₂ delivery (71), these factors likely contributed, substantially, to the development of neuromuscular fatigue during exercise (14). Regardless, VitC did not alter W_r and arterial oxygenation during exercise, suggesting that the beneficial effect of intravenous ascorbate on the development of fatigue was predominantly the consequence of the aforementioned intracellular and/or vascular processes.

Exertional dyspnea has previously been identified as a key factor limiting exercise in COPD (26, 27). It may therefore be speculated that, despite a significant positive effect on neuromuscular fatigue, the lack of improvement in exercise tolerance with VitC was primarily due to the failure of this intervention to ameliorate the dyspnea exhibited by the patients. Specifically, as reflected in the patients’ similar derangements in pulmonary mechanics, high end-expiratory lung volumes, and exaggerated W_r during VitC and PL (Table 3, Fig. 3), VitC did not alter key determinants of the multifaceted mechanistic basis of exertional dyspnea (72). Regardless, the exact impact of antioxidant supplementation on endurance exercise in patients with COPD remains equivocal, with some studies reporting a positive effect (16), whereas others (60), including the present investigation, report no effect. Interestingly, in this context, even studies on healthy individuals are controversial, with some documenting an improvement in endurance performance following antioxidant supplementation (73) and others reporting no change (74).

It is important to note that the results of this study, although robust and similar in all participants, are based on a relatively small sample size. The present findings should therefore be considered a first step in investigating the utility of antioxidant supplementation for improving exercise capacity in COPD. Additional studies using larger sample sizes and perhaps different antioxidants are needed before the clinical relevance of this approach can be determined with certainty.

Conclusions

Although VitC raised markers for antioxidant capacity, diminished markers for inflammation, and improved neuromuscular fatigue resistance, VitC failed to improve exertional dyspnea and cycling exercise tolerance in patients with COPD. As dyspnea is recognized to limit exercise tolerance in COPD, the otherwise beneficial effects of VitC may have been impacted by this unaltered sensation. The outcome of this study therefore highlights the inability of VitC to improve exercise capacity in COPD and the significance of exertional dyspnea in limiting endurance exercise tolerance in this population.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Grants HL-116579 and HL-103786 as well as Ruth L. Kirschstein National Research Service Award T32 HL-139451 and by the Department of Veterans Affairs Office of Rehabilitation Research and Development (E6910-R, E1697-R, E1572-P, and E3207-R).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

T.J.H. and M.A. conceived and designed research; T.J.H., J.C.W., S.K.S., T.S.T., V.R.R., J.Z., A.D.N., and N.M.B. performed experiments; T.J.H., V.R.R. and J.Z. analyzed data; T.J.H., J.C.W., S.K.S., T.S.T., R.S.R., and M.A. interpreted results of experiments; T.J.H. prepared figures; T.J.H. drafted manuscript; T.J.H., J.C.W., S.K.S., T.S.T., R.S.R., and M.A. edited and revised manuscript; T.J.H., J.C.W., S.K.S., T.S.T., V.R.R., J.Z., A.D.N., N.M.B., R.S.R., and M.A. approved final version of manuscript.