Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Feb 1.
Published in final edited form as: J Physiol. 2023 Dec 4;602(3):445–459. doi: 10.1113/JP285650

Endurance exercise training changes the limitation on muscle V̇O2max in normoxia from the capacity to utilize O2 to the capacity to transport O2

Ryan M Broxterman 1,2, Peter D Wagner 4, Russell S Richardson 1,2,3
PMCID: PMC10841684  NIHMSID: NIHMS1946156  PMID: 38048175

Abstract

Maximal oxygen (O2) uptake (V̇O2max) is an important parameter with utility in health and disease. However, the relative importance of O2 transport and utilization capacities in limiting muscle V̇O2max before and after endurance exercise training is not well understood. Therefore, this study aimed to identify the mechanisms determining muscle V̇O2max pre and post endurance exercise training in initially sedentary participants. In five initially sedentary young males, radial arterial and femoral venous PO2 (blood samples), leg blood flow (thermodilution), and myoglobin (Mb) desaturation (1H nuclear magnetic resonance spectroscopy) were measured during maximal single-leg knee-extensor exercise (KE) breathing either 12, 21, or 100 % O2 both pre and post eight weeks of KE training (1 hour, 3 times/week). Mb desaturation was converted to intracellular PO2 using an O2 half-saturation pressure of 3.2 mmHg. Pre-training muscle V̇O2max was not significantly different across inspired O2 conditions (12%: 0.47±0.10; 21%: 0.52±0.13; 100%: 0.54±.01 l/min, all q>0.174), despite significantly greater muscle mean capillary–intracellular PO2 gradients in normoxia (34±3 mmHg) and hyperoxia (40±7 mmHg) than hypoxia (29±5 mmHg, both q<0.024). Post-training muscle V̇O2max was significantly different across all inspired O2 conditions (12%: 0.59±0.11; 21%: 0.68±0.11; 100%: 0.76±0.09 mmHg, all q<0.035), as were the muscle mean capillary–intracellular PO2 gradients (12%: 32±2; 21%: 37±2; 100%: 45±7 mmHg, all q<0.029). In these initially sedentary participants, endurance exercise training changed the basis of limitation on muscle V̇O2max in normoxia from the mitochondrial capacity to utilize O2 to the capacity to transport O2 to the mitochondria.

Graphical Abstract

graphic file with name nihms-1946156-f0001.jpg

We measured muscle maximal O2 uptake (V̇O2max; using radial arterial and femoral venous PO2 (blood samples) and leg blood flow (thermodilution)) and muscle intracellular PO2 (myoglobin desaturation using 1H nuclear magnetic resonance spectroscopy) before and after eight weeks of endurance exercise training. In the two graphs, muscle convective O2 transport (sigmoid lines) is described by the Fick principle (V̇O2max = Q · [CaO2 - CvO2]) and muscle diffusional O2 conductance (linear lines) is described by Fick’s law of diffusion (V̇O2 = DMO2 · k · PvO2). Pre-training, PvO2 increased without an increase in muscle V̇O2max from normoxia to hyperoxia. Post-training, PvO2 increased with a concomitant increase in muscle V̇O2max from normoxia to hyperoxia. Thus, in these initially sedentary participants, endurance exercise training changed the basis of limitation on muscle V̇O2max in normoxia from the mitochondrial capacity to utilize O2 to the capacity to transport O2 to the mitochondria.

Introduction

Maximal oxygen uptake (V̇O2max) has become a cornerstone assessment of exercise capacity in health and disease since the work of Hill and Lupton in 1923 (Hill & Lupton, 1923). V̇O2max is a strong indicator of morbidity, mortality, and exercise performance (Saltin & Astrand, 1967; Costill et al., 1973; Weber et al., 1987). The prominence of V̇O2max, as a physiological phenomenon, stems from V̇O2max being determined by the integrated capacities of the pulmonary, cardiovascular, and muscular systems to transport O2 from air to mitochondria and, ultimately, the capacity of mitochondria to utilize this O2 (Wagner, 2011) (Figure 1). The integrated performance of the lungs, heart, blood, and muscle influences every step of the O2 cascade, which together limit muscle V̇O2max as a system (i.e. never limited by an individual system or step). These principles determine V̇O2max whether for small or large muscle mass exercises, untrained or trained muscle, and healthy or diseased muscle. The relative influence of each system or step can vary such that O2 transport or O2 utilization capacities, or both, may dominate in limiting muscle V̇O2max (Wagner, 2011). Furthermore, muscle V̇O2max is condition specific (e.g. exercise modality or O2 availability) and, therefore, the limitation on muscle V̇O2max can be revealed by manipulating experimental conditions and examining the impact on the steps of the O2 cascade.

Figure 1. The principal structures, associated functions, and mass conservation equations integrated in the transport of O2 from air to muscle mitochondria and the utilization of O2 by the mitochondria during maximal single-leg knee-extensor exercise (KE).

Figure 1.

Open in a new tab

The two graphs illustrate three conservation of mass equations: 1) the Fick Principle, V̇O2 = Q ☓ (CaO2 - CvO2); 2) Fick’s Law of Diffusion, V̇O2 = DMO2 ☓ k ☓ PvO2 and V̇O2 = DMO2 ☓ (PcapO2 - PmitoO2); and 3) the Mitochondrial Respiration Curve, V̇O2 = Vmax ☓ PmitoO2 / (PmitoO2 + P50mitoO2), where Q is muscle blood flow, CaO2 is arterial O2 content, CvO2 is venous O2 content, DMO2 is muscle diffusional O2 conductance, k is a constant for the proportionality between mean capillary and femoral venous O2 partial pressures, PvO2 is venous O2 partial pressure, Vmax is maximal mitochondrial O2 utilization, PmitoO2 is mitochondrial O2 partial pressure, and P50mitoO2 is the PmitoO2 eliciting 50% of Vmax. Together, these three equations define O2 transport such that maximal O2 uptake (V̇O2max) is the same at the intersection points (black circles) in each graph (i.e. at the system solution) in order for the mass of O2 to be conserved during transport through each step.

Three types of conservation of mass equations describe O2 transport and utilization down the O2 cascade: 1) the Fick Principle, 2) Fick’s Law of Diffusion, and 3) the Mitochondrial Respiration Curve (Figure 1, note Fick’s Law of Diffusion appears in two forms because a proportionality constant is required to graph this law with the Fick Principle for illustrative purposes). Again, the relative influence of each physiological variable in these equations can vary such that O2 transport or O2 utilization capacities, or both, may limit muscle V̇O2max (Wagner, 2011). Concomitantly quantifying each step of the O2 cascade in vivo, and thereby identifying the limitations to V̇O2max, remains difficult. However, although by no means conclusive, a series of cross-sectional studies provide a patchwork of evidence supporting the general tenet that muscle V̇O2max in normoxia is the consequence of well-matched O2 transport and utilization capacities in untrained skeletal muscle, but is limited by the capacity to transport O2 in trained skeletal muscle (Gollnick et al., 1973; Holloszy & Coyle, 1984; Roca et al., 1992; Knight et al., 1993; Richardson et al., 1995b; Cardus et al., 1998; Pedersen et al., 1999; Richardson et al., 1999; Wigmore et al., 2008; Tevald et al., 2009; Boushel et al., 2011; Gifford et al., 2016). Nevertheless, the actual longitudinal effects of endurance exercise training, in initially sedentary participants, on the limitation to muscle V̇O2max, remains to be documented.

Therefore, the aim of this study was to measure the determinants of skeletal muscle V̇O2max in normoxia before and after endurance exercise training in initially sedentary participants. This was achieved by combining the direct measurement of active muscle V̇O2 (using blood flow and arterial-venous O2 content difference across the leg) with the measurement of muscle intracellular PO2 (using proton magnetic resonance spectroscopy, 1H-MRS) before and after eight weeks of single-leg knee-extensor exercise (KE) training. This approach facilitated the rare opportunity to assess the majority of the steps, along the O2 cascade, from air to muscle mitochondria at maximal exercise. Maximal KE is well suited for assessing the determinants of muscle V̇O2max as this exercise modality isolates a single muscle group, allowing muscle specific measurements using both the Direct Fick and 1H-MRS (Richardson et al., 1995b; Richardson et al., 1999). We tested the hypotheses that in normoxia: 1) Pre-training muscle V̇O2max would be determined by well-matched capacities to transport and utilize O2, but 2) post-training muscle V̇O2max would be limited by the capacity to transport O2.

Methods

Ethical Approval

The Institutional Review Boards of both Universities approved the protocol (990152) in compliance with the Declaration of Helsinki except for registration in a database. Participants provided written informed consent prior to the start of experimental testing.

Participants

Five healthy, initially sedentary young males without a history of cardiac or pulmonary disease completed this study. All participants were not performing, and had no history of, regular physical activity at the time of the study. The participants completed studies in two locations (University of California, San Diego for the measurement of V̇O2max and University of Pennsylvania for the measurement of quadriceps myoglobin O2 saturation by MRS, a technique not available at UCSD).

Experimental Overview

Participants performed an incremental maximal exercise test to determine their initial whole-body V̇O2max on a cycle ergometer (25 W/min increments). Prior to the first experimental visit, participants completed two to three visits for familiarization with the KE modality. The final two familiarization visits included an incremental maximal KE test and a rehearsal for the planned experimental KE tests. Within a fourteen-day timespan, participants completed the catheter-based studies to measure muscle V̇O2 in San Diego, CA, and the 1H-MRS-based studies to measure intracellular PO2 in Philadelphia, PA. This procedure was completed both pre- and post-exercise training. The exercise training, performed in San Diego, CA, consisted of eight weeks of supervised KE training. Participants trained three times per week and each visit lasted approximately 1 hour. The training consisted of a 2 week repeating cycle of varied exercise (both intervals and longer duration exercise bouts), as previously utilized (Esposito et al., 2011; Broxterman et al., 2021).

For both the catheter- and 1H-MRS-based studies, the participants performed KE protocols in hypoxia, normoxia, and hyperoxia breathing either 12 % O2, ambient air, or 100 % O2, respectively. The sequence of the inspired O2 conditions was varied to avoid ordering effects. The KE protocols began with an unweighted warm-up for approximately 5 minutes followed by a series of stepwise increments in work rate every two to three minutes up to maximal work rate. The incremental and peak work rates for the 1H-MRS-based studies were matched to the catheter-based studies (within participants and within inspired O2 conditions). Near the end of each stage during the catheter-based studies, blood samples (radial arterial and femoral venous) were collected and blood flow measured, by thermodilution, in the femoral vein. The progressive exercise protocols, to maximum, in each inspired O2 condition were completed in approximately 8–16 minutes and these protocols, with differing inspired O2 conditions, were separated by at least 2 hours of rest. All data reported are for maximal work rates in each inspired O2 condition.

Exercise Model

The custom-made KE ergometer (Andersen et al., 1985) was constructed from nonmetallic materials so the same ergometer could be used for both the catheter- and 1H-MRS-based studies. Participants lay in a semi-recumbent position (approximately 45° of hip flexion) with the KE ergometer in front of them. A fiberglass bar connected the crank of the ergometer to an ankle brace worn by the participant. Participants performed 60 contractions per minute, allowing the momentum of the flywheel to return the relaxed leg to the start position. This exercise modality isolates muscle contraction to the quadriceps femoris muscle group and contraction of this muscle group caused the lower part of the leg to extend from approximately 90 to 170° of knee flexion. This ergometer was a prototype that did not allow work rate to be measured in conventional units and required a considerable amount of internal work. Thus, KE intensity data are presented as a percentage of pre-training normoxic leg V̇O2max.

Catheter-based Studies at UCSD

Participants arrived at the laboratory in the morning when two catheters (radial artery and femoral vein) and a thermocouple (femoral vein) were placed using sterile technique, as detailed previously (Poole et al., 1992; Richardson et al., 1995b). Blood flow was measured using thermodilution and thermal balance principles (Andersen & Saltin, 1985). Iced saline was infused through the femoral venous catheter at a rate to decrease blood temperature, measured by the indwelling thermocouple, by approximately 1°C for ≈15–20 s. The saline infusion rate was measured by the weight change in the reservoir bag suspended from a force transducer and, in combination with the blood temperature measurements, was used to calculate blood flow. Arterial and venous blood samples were collected anaerobically to measure PO2, PCO2, pH, O2 saturation, and hemoglobin (Hb) using a blood gas analyzer and CO-oximeter (IL 1306 and IL 482; Instrumentation Laboratories Inc., Lexington, MA). The blood gas analyzer and CO-oximeter were calibrated before each measurement and demonstrated acceptable reproducibility (SD of repeated determinants: PO2 and PCO2 1.5 mmHg; pH 0.003). Arterial (CaO2) and venous (CvO2) blood O2 concentrations were calculated as 1.39 ml O2 per g Hb ☓ Hb g per 100 ml blood ☓ O2 saturation % + 0.003 ml O2 per 100 ml blood☓ PO2 mmHg. The O2 hemoglobin dissociation curve standard P50 was determined for each subject using the measured O2 saturation and PO2 from all of their blood samples (Simonson et al., 2014). The CaO2–CvO2 difference was determined as the difference between arterial and venous O2 concentrations. Leg V̇O2 was calculated as blood flow ☓ (CaO2–CvO2). Leg O2 delivery was calculated as blood flow ☓ CaO2. Blood lactate concentration was measured using whole blood samples treated with a cell lysing agent (1500; Yellow Springs Instrument Co., Yellow Springs, OH). Net venous lactate outflow was calculated as leg blood flow L per min ☓(venous-arterial lactate mmol per L). Heart rate was measured with an electrocardiogram.

A numerical integration procedure was used to calculate mean capillary PO2 which was subsequently used to calculate muscle diffusional O2 conductance (DMO2) (Gayeski & Honig, 1988; Roca et al., 1989; Wagner, 1992; Knight et al., 1993; Richardson et al., 1995b). Briefly, mean capillary PO2 was calculated as the average of all PO2 values computed at equal time intervals along the capillary wall from the arterial to venous end. The mean capillary PO2 estimation included a Bohr Integration for the oxyhemoglobin dissociation curve using blood temperature and the participant’s measured O2 hemoglobin dissociation curve P50, arterial and venous pH, and arterial and venous PCO2. The DMO2 calculation utilized measured muscle intracellular PO2, determined by 1H-MRS, which circumvented the prior assumption of a zero intracellular PO2 during maximal exercise (Wagner, 1992). As a lumped parameter describing the whole muscle, DMO2 is considered to be constant along the length of the vasculature from artery to vein and it is assumed that the only explanation for O2 remaining in the femoral venous blood (O2 efflux from the muscle microcirculation) is diffusion limitation (Hogan et al., 1991; Knight et al., 1993; Richardson et al., 1998; Vogiatzis et al., 2015). Muscle DMO2 was calculated as leg V̇O2 / (mean capillary PO2 - intracellular PO2) and is considered to encompass all phenomena that facilitate O2 unloading at the muscle.

1H-MRS-based Studies at the University of Pennsylvania

KE exercise was performed in a 2.0 Tesla imaging magnet with a custom-built spectrometer (Oxford Instruments, Cambridge, UK) using a 7 cm diameter surface coil double-tuned to proton (84.45 MHz) and phosphorus (34.59 MHz) (Schnall et al., 1985). The coil was placed over the quadriceps approximately 20–25 cm proximal to the knee, which isolated signal detection predominantly to the rectus femoris (Ackerman et al., 1980). The theories behind O2-sensitive Mb signals have been published, in detail, previously (Bertinini & Luchinat, 1986; Richardson et al., 1995b). Inside the magnet, the coil was centered in the homogeneous portion of the magnetic field by aligning the water resonance in the center of applied x, y and z gradient fields. The main magnetic field was then shimmed until the linewidth of the water resonance at half-height decreased to <20 Hz. After this, nominal 90° hard pulses were calibrated at both frequencies. Proton spectra were obtained using a modified super-WEFT water suppression sequence. Water and fat resonances were inverted using a 12-ms hyperbolic secant pulse with an inversion bandwidth of 2 KHz. 65 ms later, the remaining spins were excited with a 0.5-ms Gaussian pulse, centered 6,650 Hz downfield from the water resonance frequency. This Gaussian pulse had previously been calibrated on the water resonance to achieve the same signal intensity as the nominal 90° proton hard pulse. Proton free-induction decays were then sampled over a spectral bandwidth of 20 kHz using 512 points and a dwell time of 50 μs. Total repetition time was 130 ms for this sequence. The position and line width of water resonance was checked periodically to confirm that there was minimal motion. The data were apodized with 100-Hz exponential weighting after data collection. The free-induction decays were then Fourier transformed and manually phased to generate the frequency spectra. Areas were obtained under the peak resonating approximately 73 parts per million down-field from the residual water signal. Before analysis, the outer 128 points on either side of the spectrum were removed. The remaining portion was baseline corrected by fitting the spectrum to a fifth-order polynomial in the frequency domain, which was then subtracted from the spectrum. To ensure the deoxy-Mb signals were not altered, the spectrum in the region of 65–80 parts per million down-field from water resonance was removed before the baseline routine. After baseline correction, spectra were manually phased and areas were obtained through integration of the peak arising 73 parts per million from the water resonance. No other peaks were visible within 10 parts per million of the deoxy-Mb resonance at any time.

The plateaued deoxy-Mb signals obtained during minutes 9 to 10 of cuff ischemia (cuff inflated to 270 mmHg) were used to estimate total Mb concentration within the muscle and to normalize signal areas to determine fractional deoxy-Mb (fdeoxy-Mb). The O2-binding curve for Mb was used to convert the signal to PO2 values as PO2 = [(1 - f) / f] ☓ P50, where 1 - f is the fraction of Mb that is oxygenated, f is the fraction of Mb that is deoxygenated, and P50 is the O2 pressure at which 50% of the Mb-binding sites are bound with O2. The temperature-dependent Mb P50 of 3.2 mmHg was used (Rossi-Fanelli & Antonini, 1958). Of note, Mb saturation is, also, reported so that PO2 can be recalculated using any P50.

Statistical Analyses

Data are presented as mean ± standard deviation. Variables were compared using two-way repeated-measures ANOVA, correcting for multiple comparisons by controlling the False Discovery Rate (q-value). Variables were considered significantly different when q<0.05. Pre- and post-training DMO2 were compared using a paired t test, and significant differences were accepted when p<0.05. All statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, CA).

Results

Participant characteristics

Participants had the following characteristics: age 27 ± 5 yr, height 176 ± 8 cm, body mass 73 ± 8 kg, and, on entry to the study, cycle V̇O2max 2.6 ± 0.3 l/min or 35.2 ± 4.9 ml/kg/min.

Effects of Inspired O2 on O2 Transport and O2 Utilization Pre-training

Figure 2, Figure 3, Figure 4A, and Table 1 present, in detail, the effects of inspired O2 on O2 transport and O2 utilization pre-training. Arterial PO2, arterial hemoglobin O2 saturation, and CaO2 were significantly different across all inspired O2 conditions pre-training (all q<0.003; Table 1, Figure 4A). Muscle O2 delivery was significantly greater in hyperoxia than hypoxia (q=0.006; Table 1). There was no significant effect of inspired O2 on muscle blood flow or CaO2–CvO2 (both q>0.151; Table 1). Mean capillary PO2 was significantly different across all inspired O2 conditions pre-training (all q<0.013; Table 1, Figures 2B and 4A). Pre-training deoxy-Mb was significantly greater (Table 1), and muscle intracellular PO2 significantly lower (both q<0.026; Table 1, Figures 2C and 4), in hypoxia than normoxia. Pre-training the muscle mean capillary–intracellular PO2 gradient was significantly greater in normoxia and hyperoxia than hypoxia (both q<0.024; Table 1, Figures 2D and 4). Pre-training muscle V̇O2max did not change significantly across O2 conditions (all q>0.174; Table 1, Figure 2A).

Figure 2. Muscle maximal O2 uptake (V̇O2max) (Panel A), mean capillary O2 partial pressure (PO2) (Panel B), muscle intracellular PO2 (Panel C), and the muscle mean capillary to intracellular PO2 gradient (Panel D) during maximal single-leg knee-extensor exercise (KE) in hypoxia (12% O2), normoxia (21% O2), and hyperoxia (100% O2) both pre-training and post-training.

Figure 2.

Open in a new tab

Individual data are presented as well as mean ± standard deviation. ASignificantly different from pre-training normoxic values (Q<0.05). BSignificantly different from pre-training hypoxic values (Q<0.05). CSignificantly different from the same pre-training oxygen condition values (Q<0.05). DSignificantly different from post-training normoxic values (Q<0.05). ESignificantly different from post-training hypoxic values (Q<0.05).

Figure 3. The relationships between muscle maximal O2 uptake (V̇O2max) and femoral venous O2 partial pressure (PvO2) (Panel A), mean capillary PO2 (Panel B), and the muscle mean capillary to intracellular PO2 gradient (Panel C), and muscle O2 delivery (Panel D) during maximal single-leg knee-extensor exercise (KE) in hypoxia (12% O2), normoxia (21% O2), and hyperoxia (100% O2) both pre-training and post-training.

Figure 3.

Open in a new tab

Data are presented as mean ± standard deviation. Pre-training muscle V̇O2max did not change across O2 conditions. The hypoxic and normoxic data exhibited a linear relationship extending from the origin of the graphs, but the hyperoxic data fell to the right of this relationship. These data indicate that pre-training muscle V̇O2max in normoxia was not limited by the capacity for O2 transport. Post-training muscle V̇O2max increased from hypoxia to normoxia to hyperoxia. The data for all O2 conditions exhibited a linear relationship extending from the origin of each graph. These data indicate that post-training muscle V̇O2max in normoxia was limited by the capacity for O2 transport and not the capacity for O2 utilization.

Figure 4.

Figure 4.

Open in a new tab

The O2 cascade from artery to muscle tissue during maximal single-leg knee-extensor exercise (KE) in hypoxia (12% O2), normoxia (21% O2), and hyperoxia (100% O2) both pre-training (Panel A) and post-training (Panels B).

Table 1.

Primary O2 transport and utilization variables at maximal knee-extensor exercise in hypoxia (12% O2), normoxia (21% O2), and hyperoxia (100% O2)

Pre-training Post-training
Hypoxia Normoxia Hyperoxia Hypoxia Normoxia Hyperoxia
Percentage of inspired O2 12 21 100 12 21 100
Percentage of pre-training normoxic leg V̇O2max 93 ± 18 100 106 ± 23 117 ± 24 134 ± 21 150 ± 30
Muscle V̇O2max (l/min) 0.47 ± 0.10 0.52 ± 0.13 0.54 ± 0.07 0.59 ± 0.11CD 0.68 ± 0.11C 0.76 ± 0.09CDE
Muscle blood flow (l/min) 4.6 ± 1.2 4.4 ± 1.4 4.4 ± 1.2 4.3 ± 1.0 4.6 ± 1.1 5.0 ± 1.1
Muscle O2 delivery (l/min) 0.8 ± 0.3 0.9 ± 0.3 1.0 ± 0.3B 0.8 ± 0.2D 1.0 ± 0.2 1.1 ± 0.2DE
CaO2 (ml/100 ml) 17.1 ± 1.5A 20.6 ± 1.2 22.6 ± 1.0AB 18.6 ± 1.4CD 21.0 ± 0.7 22.6 ± 0.6DE
CaO2 - CvO2 (ml/100 ml) 10.5 ± 1.2 12.4 ± 2.9 12.9 ± 3.5 13.9 ± 1.5C 15.3 ± 2.0C 15.5 ± 1.7
PaO2 (mmHg) 48 ± 8A 125 ± 7 623 ± 15AB 52 ± 7CD 122 ± 9C 650 ± 22CDE
PaCO2 (mmHg) 33 ± 3 34 ± 4 33 ± 6 31 ± 3 35 ± 5 35 ± 2
PvO2 (mmHg) 21 ± 4A 26 ± 5 29 ± 3AB 20 ± 4D 23 ± 2 26 ± 4DE
PvCO2 (mmHg) 57 ± 9 68 ± 11 83 ± 6 72 ± 15 75 ± 16 72 ± 11
O2-hemoglobin P50 (mmHg) 26.4 ± 0.4 26.4 ± 0.4 26.4 ± 0.4 26.9 ± 0.7C 26.9 ± 0.7C 26.9 ± 0.7C
SaO2 (%) 83 ± 6A 98 ± 0 99 ± 0AB 87 ± 5CD 98 ± 1 98 ± 1CE
SvO2 (%) 32 ± 9A 40 ± 13 46 ± 16A 22 ± 9CD 27 ± 8C 34 ± 10CDE
Arterial pH 7.41 ± 0.03 7.41 ± 0.05 7.41 ± 0.08 7.41 ± 0.04 7.40 ± 0.04 7.39 ± 0.02
Venous pH 7.24 ± 0.06 7.19 ± 0.07 7.20 ± 0.11 7.18 ± 0.08 7.18 ± 0.08 7.17 ± 0.07
Deoxy-Mb (% max cuff signal) 46 ± 4A 31 ± 4 29 ± 14 79 ± 20C 53 ± 11C 53 ± 8
Muscle intracellular PO2 (mmHg) 3.8 ± 0.6A 7.2 ± 1.4 10.0 ± 6.6 1.1 ± 1.2CD 3.1 ± 1.1C 3.0 ± 1.1
Mean capillary PO2 (mmHg) 33 ± 5A 41 ± 4 50 ± 9AB 33 ± 3D 40 ± 3 48 ± 6DE
Muscle diffusional O2 conductance (ml/min/mmHg) 17.3 ± 2.8 19.0 ± 3.8
Heart rate (beats/min) 134 ± 10 119 ± 20 124 ± 23 135 ± 24 128 ± 26 134 ± 36
Net venous lactate outflow (mmol/min) 7.3 ± 2.6 8.3 ± 2.9 5.9 ± 4.2 10.2 ± 3.2 9.1 ± 3.0 8.24 ± 1.9
Open in a new tab

Mean ± SD. Muscle V̇O2max, quadriceps maximal oxygen uptake; CaO2, arterial oxygen content; CaO2 - CvO2, arterial to venous oxygen content difference; PaO2, arterial partial pressure of oxygen; PaCO2, arterial partial pressure of carbon dioxide; PvO2, venous partial pressure of oxygen; PvCO2, venous partial pressure of carbon dioxide; O2-hemoglobin P50, oxygen partial pressure at which 50% of hemoglobin binding sites are bound with oxygen; SaO2, arterial hemoglobin oxygen saturation; SvO2, venous hemoglobin oxygen saturation; Deoxy-Mb, quadriceps myoglobin deoxygenation; Mb-associated PO2, quadriceps intracellular PO2, quadriceps intracellular partial pressure of oxygen.

A

Significantly different from pre-training normoxic values (Q<0.05).

B

Significantly different from pre-training hypoxic values (Q<0.05).

C

Significantly different from the same pre-training oxygen condition values (Q<0.05).

D

Significantly different from post-training normoxic values (Q<0.05).

E

Significantly different from post-training hypoxic values (Q<0.05).

Effects of inspired O2 on O2 transport and O2 utilization Post-training

Figure 2, Figure 3, Figure 4B, and Table 1 present, in detail, the effects of inspired O2 on O2 transport and O2 utilization KE post-training. Arterial PO2 and CaO2 were significantly different across all inspired O2 conditions post-training (all q<0.022; Table 1, Figure 4B). Arterial hemoglobin O2 saturation was significantly greater in normoxia and hyperoxia than hypoxia (both q=0.003; Table 1). There was no significant effect of inspired O2 on muscle blood flow or CaO2–CvO2 post-training (all q>0.066; Table 1). Muscle O2 delivery was significantly different across all inspired O2 conditions (all q<0.002; Table 1). Post-training mean capillary PO2 was significantly different across all inspired O2 conditions (all q<0.008; Table 1, Figures 2B and 4B). Post-training there were no differences in deoxy-Mb across inspired O2 conditions (all q>0.838), but muscle intracellular PO2 was significantly greater in normoxia than hypoxia (q=0.029). Post-training the muscle mean capillary–intracellular PO2 gradient was significantly different across all inspired O2 conditions (all q<0.029; Table 1, Figures 2D and 4B). Post-training muscle V̇O2max was significantly different across all inspired O2 conditions (all q<0.035; Table 1, Figure 2A).

Effects of KE training on muscle O2 transport and O2 utilization

Figure 2, Figure 3, Figure 4, and Table 1 present, in detail, the effects of KE training on O2 transport and O2 utilization. KE training significantly increased arterial hemoglobin O2 saturation in hypoxia and hyperoxia (all q<0.039), but not in normoxia (q=0.101; Table 1, Figure 4). CaO2 in normoxia and hyperoxia were not significantly different between pre- and post-training (all q>0.149, but CaO2 in hypoxia was significantly greater post-training (q=0.028; Table 1). KE training did not significantly affect muscle blood flow (all q>0.435; Table 1). KE training did not significantly affect muscle O2 delivery (all q>0.082; Table 1). KE training significantly increased CaO2–CvO2 in hypoxia and normoxia with an average increase of approximately 25% (all q<0.041; Table 1). KE training did not affect mean capillary PO2 (all q>0.194; Table 1, Figures 2B and 4). KE training did not significantly increase DMO2 (p=0.178; Table 1). KE training significantly increased deoxy-Mb, approximately 26% (statistical differences only in hypoxia and normoxia q<0.041; Table 1, Figures 2C and 4). Muscle intracellular PO2 was, on average, approximately 66% lower post- compared to pre-training (statistical differences only in hypoxia and normoxia q<0.026; Table 1, Figures 2C and 4). KE training did not significantly affect the muscle mean capillary–intracellular PO2 gradient in any O2 condition (all q>0.052; Table 1, Figures 2D and 4). KE training significantly increased muscle V̇O2max in all inspired O2 conditions, with an average increase of approximately 33% (all q<0.009; Table 1, Figure 2A).

Discussion

The aim of this study was to measure the determinants of skeletal muscle V̇O2max in normoxia before and after endurance exercise training in initially sedentary participants. This was achieved by combining the direct measurement of active muscle V̇O2 (using blood flow and arterial-venous O2 content difference across the leg) with the measurement of muscle intracellular PO2 (using 1H-MRS) before and after eight weeks of KE training. This approach facilitated the rare opportunity to assess both O2 transport, at the major steps of the O2 cascade, and muscle O2 utilization. The muscle mean capillary–intracellular PO2 gradient increased from hypoxia to both normoxia and hyperoxia pre-training, but muscle V̇O2max was unaltered, indicating that, contrary to our first hypothesis, the capacity to utilize O2 limited muscle V̇O2max in normoxia. Post-training, both the muscle mean capillary–intracellular PO2 gradient and muscle V̇O2max increased from hypoxia to normoxia to hyperoxia, indicating that, consistent with our second hypothesis, the capacity to transport O2 now limited muscle V̇O2max in normoxia. An additional interesting and novel finding of the current study was that muscle intracellular PO2 during maximal KE was markedly greater pre-training (4–10 mmHg) than post-training (1–3 mmHg). Thus, in these initially sedentary participants, endurance exercise training changed the basis of limitation on muscle V̇O2max in normoxia from the mitochondrial capacity to utilize O2 to the capacity to transport O2 to the mitochondria.

Pre-training limitations on muscle V̇O2max

Plotting the Fick Principle, Fick’s Law of Diffusion, and the Mitochondrial Respiration Curve illustrates how the assessed physiological variables integrate to determine muscle V̇O2max (Figure 5). Pre-training convective O2 delivery increased approximately 27% and the muscle mean capillary–intracellular PO2 gradient increased approximately 36% from hypoxia to both normoxia and hyperoxia (Table 1, Figures 2D and 4). The lack of increase in muscle V̇O2max despite the increased potential for O2 flux indicates that the capacity to utilize O2 (and not O2 transport) limited muscle V̇O2max in normoxia pre-training (Table 1, Figure 2A).

Figure 5. Muscle O2 uptake (V̇O2max) as a function of muscle venous PO2 (Panels A and C) and as a function of muscle intracellular PO2 (Panels B and D) during maximal single-leg knee-extensor exercise (KE) in hypoxia (12% O2), normoxia (21% O2), and hyperoxia (100% O2) both pre-training (Panels A and B) and post-training (Panels C and D).

Figure 5.

Open in a new tab

The lines in each graph illustrate two simultaneous O2 mass conservation equations and the circles indicate maximal muscle O2 uptake (V̇O2max) in hypoxia, normoxia, and hyperoxia, which must lie at the intersection of the two equations. In Panels A and C, muscle convective O2 transport (sigmoid lines) is described by the Fick principle (V̇O2max = Q · (CaO2 - CvO2)), where Q is muscle blood flow, CaO2 is arterial O2 content, and CvO2 is femoral venous O2 content) and muscle diffusive O2 transport is described by Fick’s law of diffusion (V̇O2 = DMO2 · k · PvO2, where DMO2 is muscle diffusional O2 conductance, k is a constant for the proportionality between mean capillary and femoral venous partial pressure of O2 (PO2), and PvO2 is femoral venous PO2). In Panels B and D, the Mitochondrial Respiration Curve is described by a hyperbolic function (V̇O2 = Vmax · PmitoO2 / (PmitoO2 + P50mitoO2), where Vmax is maximal mitochondrial O2 uptake, PmitoO2 is mitochondria PO2, P50mitoO2 is the PmitoO2 eliciting 50% of Vmax) and muscle diffusive O2 transport is dictated by Fick’s law of diffusion (VO2 = DMO2 · (PcapO2 – PmitoO2), where PcapO2 is mean capillary PO2). Muscle intracellular PO2 was used as a surrogate for PmitoO2 in these equations, as PmitoO2 cannot, currently, be measured currently and the intracellular environment is the nearest location for which PO2 can be measured. The dotted linear line in Panel B depicts the “actual” DMO2 (based on the other two inspired O2 conditions). Pre-training, the increase in mean capillary, femoral venous, and muscle intracellular PO2, without an increase in muscle V̇O2max from normoxia to hyperoxia, infers that the capacity to utilize O2 limited muscle V̇O2max in normoxia. The post-training increase in mean capillary and femoral venous PO2 with the concomitant increase in muscle V̇O2max from normoxia to hyperoxia infers that the capacity to transport O2 now limited muscle V̇O2max in normoxia

An important aspect of the current study was the ability to calculate DMO2 using the muscle mean capillary–intracellular PO2 gradient. The slope of the linear relationship (through the origin of the graph) between V̇O2max and the muscle mean capillary–intracellular PO2 gradient defines DMO2 (Figure 3C). The computation of DMO2 is only valid under conditions where V̇O2max is measurably O2 transport limited. Therefore, pre-training DMO2 was only estimated for the hypoxia condition because muscle V̇O2max did not statistically increase from normoxia to hyperoxia (Table 1, Figure 2A). It is recognized that DMO2 necessitates the assumption that muscle V̇O2max in hypoxia was O2 transport limited, which would require a more severe hypoxia condition to establish. In support of this assumption, both the normoxia and hyperoxia data lie below the expectation from the linear relationship for the hypoxia data between muscle V̇O2max and the muscle mean capillary–intracellular PO2 gradient (Figure 3C). The data patterns and linear relationships between muscle V̇O2max and venous PO2, mean capillary PO2, and muscle O2 delivery are consistent with the relationships between V̇O2max and the muscle mean capillary–intracellular PO2 gradient (Figure 3). As a result, the linear fits in Figure 3 are based only on the hypoxia data pre-training. The deviations below the linear relationships support that muscle V̇O2max in normoxia and hyperoxia were both less than expected based on the diffusive O2 transport capacity. Together, these results support that mitochondrial oxidative capacity was being approached at the higher inspired O2 conditions and are consistent with muscle V̇O2max in normoxia being limited by the capacity to utilize O2 pre-training.

The finding that O2 utilization capacity limited muscle V̇O2max must be reconciled with the concept that O2 transport always limits V̇O2max below mitochondrial oxidative capacity, based on a hyperbolic mitochondrial respiration curve (Figure 1C) (Cano et al., 2013). While correct as a theoretical concept, the hyperbolic shape of the mitochondrial respiration curve precludes a sudden switch from O2 transport capacity limitation to O2 utilization capacity limitation. Rather, the role for O2 transport limitation becomes gradually less as muscle V̇O2max nears mitochondrial oxidative capacity and the physiological relevance of this limitation diminishes. The lack of difference in muscle V̇O2max between hypoxia and normoxia supports the conclusion that in normoxia these sedentary participants must have been operating on the practically flat part of the mitochondrial respiration curve, near mitochondrial oxidative capacity (Figure 5B). Collectively, the evidence from this study supports the conclusion that, in these sedentary participants, pre-training muscle V̇O2max during maximal KE in normoxia was limited by the capacity to utilize O2 (i.e. mitochondrial oxidative capacity).

Post-training limitations on muscle V̇O2max

In contrast to pre-training findings, muscle V̇O2max increased from hypoxia to normoxia to hyperoxia post-training (Table 1, Figure 2A). These increases in muscle V̇O2max were facilitated by increases in the muscle mean capillary–intracellular PO2 gradient (Table 1, Figures 2D and 4). The increase in muscle V̇O2max with increased O2 availability indicates that the capacity to transport O2 (and not O2 utilization) now limited muscle V̇O2max post-training (Table 1, Figure 2A). In contrast to pre-training, the increase in muscle V̇O2max from normoxia to hyperoxia post-training supports the concept that in normoxia these trained participants were now operating on the steep part of the mitochondrial respiration curve, not near mitochondrial oxidative capacity (Figure 5D). As with pre-training, all systems and steps along the O2 cascade contribute to the limitation of muscle V̇O2max. Post-training, however, the capacity to utilize O2 within the muscle had increased more than the capacity to transport O2 to the muscle. Thus, post-training muscle V̇O2max during maximal KE in normoxia was now limited by the capacity to transport O2 (i.e. convective and diffusive transport).

Using similar logic to pre-training, post-training DMO2 was estimated as the average value for the hypoxia and normoxia conditions because muscle V̇O2max increased from hypoxia to normoxia and from normoxia to hyperoxia (Table 1, Figure 2A). The hyperoxia condition was not used to estimate DMO2 because, with the current experimental design, it cannot be determined that V̇O2max was still O2 transport capacity-limited. It would require an intervention such as hyperbaric hyperoxia to establish if the muscle V̇O2max in hyperoxia was O2 transport limited. However, just as for the pre-training data, post-training muscle V̇O2max in hyperoxia did not reach the level expected from diffusive O2 transport capacity, suggesting mitochondrial oxidative capacity was being approached. As a result, the linear fits in Figure 3 are based only the hypoxia and normoxia data post-training. The data patterns and linear relationships between muscle V̇O2max and venous PO2, mean capillary PO2, and muscle O2 delivery are consistent with the relationships between V̇O2max and the muscle mean capillary–intracellular PO2 gradient (Figure 3). Together, these results are consistent with muscle V̇O2max in normoxia being limited by the capacity to transport O2 post-training.

Muscle intracellular PO2 Pre- and Post-training

Pre-training, average muscle intracellular PO2 ranged from approximately 4–10 mmHg (Table 1, Figures 2C, 4A, and 5B). As muscle V̇O2max did not change across inspired O2 conditions pre-training, even the lowest muscle intracellular PO2 values (4 mmHg in hypoxia) were apparently sufficient to achieve near-maximal skeletal muscle mitochondrial O2 utilization (Table 1, Figures 2C and 5B). This further supports the conclusion that in normoxia these sedentary participants were likely operating on the practically flat part of the mitochondrial respiration curve, near mitochondrial oxidative capacity (Figure 5B). Post-training muscle intracellular O2 was markedly less than pre-training, with a PO2 that ranged from approximately 1–3 mmHg (Table 1, Figures 2C, 4, and 5). Interestingly, in general muscle intracellular O2 in hypoxia pre-training was greater than all of the O2 availability conditions post-training. Taken together, the evidence from this study supports the conclusion that after training, in initially sedentary participants, muscle V̇O2max during maximal KE in normoxia was limited by the capacity to transport O2 (i.e. convective and diffusive transport).

Effects of KE training on the determinants of muscle V̇O2max in normoxia

Together, the Fick Principle, Fick’s Law of Diffusion, and the Mitochondrial Respiration Curve convey the effects of KE on the determinants of muscle V̇O2max in normoxia (Figure 5). KE training increased muscle V̇O2max in normoxia approximately 30% (Table 1, Figure 2A). Both pre- and post-training, muscle blood flow at maximal exercise was not affected by hypoxia or hyperoxia compared to normoxia (Table 1). Muscle blood flow during maximal KE has previously been reported both to increase with hypoxia (Rowell et al., 1986) or to be unaltered by hypoxia (Richardson et al., 1995a; Calbet et al., 2009). Rowell et al. (Rowell et al., 1986) suggested that the active muscle may be slightly “over-perfused” relative to oxygen uptake during knee-extensor exercise compared to cycle exercise. This is based on the relatively greater PvO2 values at maximal knee-extensor exercise (approximately 18–25 mmHg) compared to maximal cycle exercise (approximately <10 mmHg). Such an over-perfusion may dampen the impact of altering CaO2 on muscle blood flow during KE compared to cycle exercise. Based on the Fick Principle, this increase in muscle V̇O2max was primarily the result of a 20% increase in the CaO2–CvO2 difference, as KE training did not affect muscle blood flow (Table 1). The muscle mean capillary–intracellular PO2 gradient was also unaffected by KE training. Extending this to Fick’s Law of Diffusion, the 10% increase in muscle diffusional conductance with KE training, although not statistically significant, is likely the primary reason for the increase in muscle V̇O2max, in the face of an apparently constant mean capillary–intracellular PO2 gradient. An increase in muscle diffusional conductance combined with the increase in O2-hemoglobin P50 could have contributed to the 30% lower SvO2 and, therefore, the increase in the CaO2–CvO2 difference. The Mitochondrial Respiration Curve indicates that the increase in muscle diffusional conductance must have been accompanied by an increase in mitochondrial oxidative capacity for muscle V̇O2max to increase with KE training. The PO2 values across the O2 cascade also indicate that the primary effects of KE training on O2 transport and O2 utilization were at the level of the muscle (Figure 4). Overall, KE training increased muscle V̇O2max in normoxia primarily through increased muscle diffusional conductance and mitochondrial oxidative capacity.

Determinants of muscle V̇O2max during small muscle mass exercise versus whole-body exercise

For both small muscle mass and whole-body exercises, the steps of the O2 cascade integrate to determine muscle V̇O2max (Wagner, 2011). But can the fundamental basis of limitation on muscle V̇O2max (i.e. O2 transport versus O2 utilization capacities) be the same for small muscle mass and whole-body exercises? In sedentary participants, muscle V̇O2max did not increase from normoxia to hyperoxia during either maximal KE or cycle exercise (Cardus et al., 1998; Pedersen et al., 1999). The capacity to utilize O2 exceeded the capacity to transport O2 during both maximal KE and cycle exercise in trained participants (Knight et al., 1993; Richardson et al., 1995b; Richardson et al., 1999; Boushel et al., 2011; Gifford et al., 2016). This is congruent with evidence that endurance exercise training increases mitochondrial oxidative capacity to a greater degree than muscle V̇O2max (Gollnick et al., 1973; Holloszy & Coyle, 1984). The pre- and post-training results of the current study are consistent with these previous findings. Gifford et al. (Gifford et al., 2016) demonstrated that V̇O2max in normoxia appeared to be limited by the capacity to utilize O2 in untrained participants and by the capacity to transport O2 in trained participants for both KE and cycle exercise. In patients with heart failure, muscle V̇O2max in normoxia was limited by the capacity to utilize O2 for both KE and cycle exercise (Esposito et al., 2011). Overall, there is evidence that muscle V̇O2max in normoxia can be limited by the capacity to utilize O2 or the capacity to transport O2 for both small muscle mass and whole-body exercises. There also is evidence that the basis of limitation on muscle V̇O2max can be the same for both small muscle mass and whole-body exercises within individuals. Whether this is always the case remains to be determined.

Key Points.

  • Maximal O2 uptake is an important parameter with utility in health and disease.

  • The relative importance of O2 transport and utilization capacities in limiting muscle maximal O2 uptake before and after endurance exercise training is not well understood.

  • We combined the direct measurement of active muscle maximal O2 uptake with the measurement of muscle intracellular PO2 before and after eight weeks of endurance exercise training.

  • We show that increasing O2 availability did not increase muscle maximal O2 uptake before training, while increasing O2 availability did increase muscle maximal O2 uptake after training.

  • The results support that, in these initially sedentary participants, endurance exercise training changed the basis of limitation on muscle maximal O2 uptake in normoxia from the mitochondrial capacity to utilize O2 to the capacity to transport O2 to the mitochondria.

Acknowledgments

The authors wish to thank all of the subjects of this study for their committed participation and sacrifice to be a part of this involved investigation.

Funding

This work was supported by National Heart, Lung, and Blood Institute grant number HL-091830 and a Ruth L. Kirschtein National Research Service Award grant number 1T32HL139451, as well as the Veterans Administration (VA) Rehabilitation Research and Development Service grant numbers E6910-R, E1697-R, E3207-R, E9275-L, and E1572-P and Clinical Science Research and Development Career Development Award (IK2CX002114).

Biography

graphic file with name nihms-1946156-b0002.gif

Ryan M. Broxterman is currently an Assistant Professor in the Division of Geriatrics at the University of Utah and the Geriatric Research, Education, and Clinical Center (GRECC) at the Salt Lake City VA Medical Center. Dr. Broxterman’s research focuses on the integration of vascular, metabolic, and neuromuscular function as determinants of exercise capacity and tolerance in health and disease.

Footnotes

Competing interests

The authors declare that they have no competing interests.

Data Availability Statement

All data supporting the results are presented in the manuscript.

References

  1. Ackerman JJ, Grove TH, Wong GG, Gadian DG & Radda GK. (1980). Mapping of metabolites in whole animals by 31P NMR using surface coils. Nature 283, 167–170. [DOI] [PubMed] [Google Scholar]
  2. Andersen P, Adams RP, Sjogaard G, Thorboe A & Saltin B. (1985). Dynamic knee extension as model for study of isolated exercising muscle in humans. J Appl Physiol 59, 1647–1653. [DOI] [PubMed] [Google Scholar]
  3. Andersen P & Saltin B. (1985). Maximal perfusion of skeletal muscle in man. J Physiol 366, 233–249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bertinini I & Luchinat C. (1986). NMR Paramagnetic Molecules in Biological Systems. Benjamin/Cummings Publishing, Melo Park, CA. [Google Scholar]
  5. Boushel R, Gnaiger E, Calbet JA, Gonzalez-Alonso J, Wright-Paradis C, Sondergaard H, Ara I, Helge JW & Saltin B. (2011). Muscle mitochondrial capacity exceeds maximal oxygen delivery in humans. Mitochondrion 11, 303–307. [DOI] [PubMed] [Google Scholar]
  6. Broxterman RM, Wagner PD & Richardson RS. (2021). Exercise Training in Chronic Obstructive Pulmonary Disease: Muscle O2 Transport Plasticity. Eur Respir J. [DOI] [PubMed] [Google Scholar]
  7. Calbet JA, Radegran G, Boushel R & Saltin B. (2009). On the mechanisms that limit oxygen uptake during exercise in acute and chronic hypoxia: role of muscle mass. The Journal of physiology 587, 477–490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cano I, Mickael M, Gomez-Cabrero D, Tegner J, Roca J & Wagner PD. (2013). Importance of mitochondrial P(O2) in maximal O2 transport and utilization: a theoretical analysis. Respir Physiol Neurobiol 189, 477–483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cardus J, Marrades RM, Roca J, Barbera JA, Diaz O, Masclans JR, Rodriguez-Roisin R & Wagner PD. (1998). Effects of F(I)O2 on leg VO2 during cycle ergometry in sedentary subjects. Med Sci Sports Exerc 30, 697–703. [DOI] [PubMed] [Google Scholar]
  10. Costill DL, Thomason H & Roberts E. (1973). Fractional utilization of the aerobic capacity during distance running. Medicine and science in sports 5, 248–252. [PubMed] [Google Scholar]
  11. Esposito F, Reese V, Shabetai R, Wagner PD & Richardson RS. (2011). Isolated quadriceps training increases maximal exercise capacity in chronic heart failure: the role of skeletal muscle convective and diffusive oxygen transport. J Am Coll Cardiol 58, 1353–1362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gayeski TE & Honig CR. (1988). Intracellular PO2 in long axis of individual fibers in working dog gracilis muscle. Am J Physiol 254, H1179–1186. [DOI] [PubMed] [Google Scholar]
  13. Gifford JR, Garten RS, Nelson AD, Trinity JD, Layec G, Witman MA, Weavil JC, Mangum T, Hart C, Etheredge C, Jessop J, Bledsoe A, Morgan DE, Wray DW, Rossman MJ & Richardson RS. (2016). Symmorphosis and skeletal muscle VO2 max : in vivo and in vitro measures reveal differing constraints in the exercise-trained and untrained human. J Physiol 594, 1741–1751. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gollnick PD, Armstrong RB, Saltin B, Saubert CWt, Sembrowich WL & Shepherd RE. (1973). Effect of training on enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol 34, 107–111. [DOI] [PubMed] [Google Scholar]
  15. Hill AV & Lupton H. (1923). Muscular Exercise, Lactic Acid, and the Supply and Utilization of Oxygen. QJM 16, 135–171. [Google Scholar]
  16. Hogan MC, Bebout DE & Wagner PD. (1991). Effect of increased Hb-O2 affinity on VO2max at constant O2 delivery in dog muscle in situ. J Appl Physiol 70, 2656–2662. [DOI] [PubMed] [Google Scholar]
  17. Holloszy JO & Coyle EF. (1984). Adaptations of skeletal muscle to endurance exercise and their metabolic consequences. J Appl Physiol 56, 831–838. [DOI] [PubMed] [Google Scholar]
  18. Knight DR, Schaffartzik W, Poole DC, Hogan MC, Bebout DE & Wagner PD. (1993). Effects of hyperoxia on maximal leg O2 supply and utilization in men. J Appl Physiol 75, 2586–2594. [DOI] [PubMed] [Google Scholar]
  19. Pedersen PK, Kiens B & Saltin B. (1999). Hyperoxia does not increase peak muscle oxygen uptake in small muscle group exercise. Acta Physiol Scand 166, 309–318. [DOI] [PubMed] [Google Scholar]
  20. Poole DC, Gaesser GA, Hogan MC, Knight DR & Wagner PD. (1992). Pulmonary and leg VO2 during submaximal exercise: implications for muscular efficiency. J Appl Physiol 72, 805–810. [DOI] [PubMed] [Google Scholar]
  21. Richardson RS, Knight DR, Poole DC, Kurdak SS, Hogan MC, Grassi B & Wagner PD. (1995a). Determinants of maximal exercise VO2 during single leg knee-extensor exercise in humans. Am J Physiol 268, H1453–1461. [DOI] [PubMed] [Google Scholar]
  22. Richardson RS, Leigh JS, Wagner PD & Noyszewski EA. (1999). Cellular PO2 as a determinant of maximal mitochondrial O(2) consumption in trained human skeletal muscle. J Appl Physiol 87, 325–331. [DOI] [PubMed] [Google Scholar]
  23. Richardson RS, Noyszewski EA, Kendrick KF, Leigh JS & Wagner PD. (1995b). Myoglobin O2 desaturation during exercise. Evidence of limited O2 transport. J Clin Invest 96, 1916–1926. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Richardson RS, Tagore K, Haseler LJ, Jordan M & Wagner PD. (1998). Increased VO2 max with right-shifted Hb-O2 dissociation curve at a constant O2 delivery in dog muscle in situ. J Appl Physiol 84, 995–1002. [DOI] [PubMed] [Google Scholar]
  25. Roca J, Agusti AG, Alonso A, Poole DC, Viegas C, Barbera JA, Rodriguez-Roisin R, Ferrer A & Wagner PD. (1992). Effects of training on muscle O2 transport at VO2max. Journal of applied physiology 73, 1067–1076. [DOI] [PubMed] [Google Scholar]
  26. Roca J, Hogan MC, Story D, Bebout DE, Haab P, Gonzalez R, Ueno O & Wagner PD. (1989). Evidence for tissue diffusion limitation of VO2max in normal humans. J Appl Physiol 67, 291–299. [DOI] [PubMed] [Google Scholar]
  27. Rossi-Fanelli A & Antonini E. (1958). Studies on the oxygen and carbon monoxide equilibria of human myoglobin. Arch Biochem Biophys 77, 478–492. [DOI] [PubMed] [Google Scholar]
  28. Rowell LB, Saltin B, Kiens B & Christensen NJ. (1986). Is peak quadriceps blood flow in humans even higher during exercise with hypoxemia? Am J Physiol 251, H1038–1044. [DOI] [PubMed] [Google Scholar]
  29. Saltin B & Astrand PO. (1967). Maximal oxygen uptake in athletes. J Appl Physiol 23, 353–358. [DOI] [PubMed] [Google Scholar]
  30. Schnall MD, Subramanian VH, Leigh JS & Chance B. (1985). A new double-tuned probe for concurrent 1H and 31P NMR. J Magn Reson 65, 122–129. [Google Scholar]
  31. Simonson TS, Wei G, Wagner HE, Wuren T, Bui A, Fine JM, Qin G, Beltrami FG, Yan M, Wagner PD & Ge RL. (2014). Increased blood-oxygen binding affinity in Tibetan and Han Chinese residents at 4200 m. Exp Physiol 99, 1624–1635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Tevald MA, Lanza IR, Befroy DE & Kent-Braun JA. (2009). Intramyocellular oxygenation during ischemic muscle contractions in vivo. Eur J Appl Physiol 106, 333–343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Vogiatzis I, Habazettl H, Louvaris Z, Andrianopoulos V, Wagner H, Zakynthinos S & Wagner PD. (2015). A method for assessing heterogeneity of blood flow and metabolism in exercising normal human muscle by near-infrared spectroscopy. J Appl Physiol 118, 783–793. [DOI] [PubMed] [Google Scholar]
  34. Wagner PD. (1992). Gas exchange and peripheral diffusion limitation. Med Sci Sports Exerc 24, 54–58. [PubMed] [Google Scholar]
  35. Wagner PD. (2011). Modeling O(2) transport as an integrated system limiting (.)V(O(2)MAX). Computer methods and programs in biomedicine 101, 109–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Weber KT, Janicki JS & McElroy PA. (1987). Determination of aerobic capacity and the severity of chronic cardiac and circulatory failure. Circulation 76, VI40–45. [PubMed] [Google Scholar]
  37. Wigmore DM, Befroy DE, Lanza IR & Kent-Braun JA. (2008). Contraction frequency modulates muscle fatigue and the rate of myoglobin desaturation during incremental contractions in humans. Appl Physiol Nutr Metab 33, 915–921. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

All data supporting the results are presented in the manuscript.

RESOURCES