Speeding of V̇o₂ kinetics with endurance training in old and young men is associated with improved matching of local O₂ delivery to muscle O₂ utilization

Abstract

The time course and mechanisms of adjustment of pulmonary oxygen uptake (V̇o₂) kinetics (time constant τV̇o_2p) were examined during step transitions from 20 W to moderate-intensity cycling in eight older men (O; 68 ± 7 yr) and eight young men (Y; 23 ± 5 yr) before training and at 3, 6, 9, and 12 wk of endurance training. V̇o_2p was measured breath by breath with a volume turbine and a mass spectrometer. Changes in deoxygenated hemoglobin concentration (Δ[HHb]) were measured by near-infrared spectroscopy. V̇o_2p and Δ[HHb] were modeled with a monoexponential model. Training was performed on a cycle ergometer three times per week for 45 min at ∼70% of peak V̇o₂. Pretraining τV̇o_2p was greater (P < 0.05) in O (43 ± 10 s) than Y (34 ± 8 s). τV̇o_2p decreased (P < 0.05) by 3 wk of training in both O (35 ± 9 s) and Y (22 ± 8 s), with no further changes thereafter. The pretraining overall adjustment of Δ[HHb] was faster than τV̇o_2p in both O and Y, resulting in Δ[HHb]/V̇o_2p displaying an “overshoot” during the transient relative to the subsequent steady-state level. After 3 wk of training the Δ[HHb]/V̇o_2p overshoot was attenuated in both O and Y. With further training, this overshoot persisted in O but was eliminated after 6 wk in Y. The training-induced speeding of V̇o_2p kinetics in O and Y at 3 wk of training was associated with an improved matching of local O₂ delivery to muscle V̇o₂ (as represented by a lower Δ[HHb]/V̇o_2p). The continued overshoot in Δ[HHb]/V̇o_2p in O may reflect a reduced vasodilatory responsiveness that may limit muscle blood flow distribution during the on-transient of exercise.

the rate of adjustment of the primary component (phase II) of pulmonary oxygen uptake (V̇o_2p) closely reflects the adjustments of oxidative metabolism at the skeletal muscle level (25, 39, 49). The phase II V̇o_2p kinetics during the on-transient of moderate-intensity exercise is slower in older compared with younger adults (2, 5, 8, 12). A slower adjustment of oxidative metabolism results in a larger O₂ deficit and greater reliance on substrate-level phosphorylation to provide ATP in sufficient amounts to sustain any given activity. As such, older adults may experience premature fatigue and reduced tolerance to exercise (15).

In some conditions (e.g., hypoxia, β-adrenergic receptor blockade, reduced arterial perfusion) and subject groups (e.g., those with chronic heart failure, peripheral vascular disease, diabetes) it has been suggested that O₂ delivery may constrain V̇o₂ kinetics (34, 47, 55). Although bulk delivery of O₂ to the exercising limb does not seem to limit the rate of adjustment of V̇o₂ kinetics (6), the local distribution of blood flow within the active muscles appears to be critical in the matching of O₂ delivery to O₂ utilization during the kinetic phase of adjustment to the increased energy demand with exercise (19). Near-infrared spectroscopy (NIRS) data in young adults have shown that the rate of deoxygenation is very rapid (14) and suggest that the rate of increase in microvascular blood flow in the active tissues is slow compared with the increase in muscle O₂ utilization (V̇o_2m) (14, 31). An age-related reduction in local (microvascular) blood flow is reflected by a greater ratio of change in deoxygenated hemoglobin to change in V̇o_2p (Δ[HHb]/ΔV̇o_2p) in older compared with younger men (13) as well as a transiently greater fall in microvascular Po₂ (Po_2mv) in older compared with young rats (4). Thus older adults rely more on O₂ extraction during the on-transient of exercise, probably because of a lower local (microvascular) blood flow-to-V̇o_2m ratio.

Endurance training results in faster V̇o_2p kinetics in both older (1, 6) and younger (7, 21, 40, 46) individuals. However, the mechanisms underlying that faster response have not been clearly elucidated, nor has the time course of adaptation of V̇o_2p been established in older adults. Information on the time course of the training-induced speeding of V̇o₂ kinetics may shed light on factors limiting the adjustment of V̇o_2m and on how and whether their contributions are influenced by aging and training status. As such, the main goal of this study was to determine the time course and mechanism of adaptation for phase II V̇o_2p in older and younger men throughout a 12-wk endurance training program. We hypothesized that there would be an improved microvascular O₂ delivery in the exercise transient in response to the endurance training that would be associated with a faster adjustment of V̇o_2p kinetics observed early in training in both older and young adults. Improvements in microvascular O₂ delivery would be indicated by a better matching between the rate of adjustment of muscle deoxygenation relative to phase II V̇o_2p (i.e., the Δ[HHb]-to-ΔV̇o_2p ratio), which represents a decreased reliance on O₂ extraction for a given V̇o_2p.

METHODS

Subjects.

Eight older (O; 68 ± 7 yr; mean ± SD) and eight young (Y; 23 ± 5 yr) adult men volunteered and gave written consent to participate in the study. Descriptive and baseline data from these subjects were given in a separate report looking at central and peripheral adaptations to endurance training in the same men (Ref. 43; the reader is referred to this paper for further information on increases in maximal V̇o_2p, cardiac output, and arteriovenous O₂ difference). All procedures were approved by the University of Western Ontario Research Ethics Board for Health Sciences Research Involving Human Subjects. All subjects were nonobese (body mass index ≤30 kg/m²) nonsmokers and were physically active, but none had been involved in any type of endurance training program for at least 12 mo before the study. Additionally, no subjects were taking medications that would affect the cardiorespiratory or hemodynamic responses to exercise. Subjects had no history of cardiovascular, respiratory, or musculoskeletal diseases, and older subjects were medically screened by a physician and underwent a maximal exercise stress test.

Protocol.

Before training began, subjects reported to the laboratory on two separate occasions. On day 1, a maximal cycle ergometer ramp test (O 15–20 W/min; Y 25 W/min) was performed (Lode Corival 400; Lode, Groningen, The Netherlands) for determination of peak V̇o₂ (V̇o_2peak) and the estimated lactate threshold (θ_L). θ_L was defined as the V̇o₂ at which CO₂ production (V̇co₂) began to increase out of proportion in relation to V̇o₂ with a systematic rise in minute ventilation-to-V̇o₂ ratio and end-tidal Po₂ whereas minute ventilation-to-V̇co₂ ratio and end-tidal Pco₂ were stable. After this test, subjects returned to the laboratory on a different day to perform step transitions in work rate (WR) from 20 W to a moderate-intensity WR that elicited a V̇o₂ corresponding to 90% θ_L. Each subject performed two sets, each including two step transitions consisting of 6 min of pedaling at 20 W, 6 min at 90% θ_L, 8 min at 20 W, and another 6 min at 90% θ_L. Each set was separated by a 30-min period of rest sitting on a chair. Identical procedures were repeated after weeks 3, 6, 9, and 12 of training. However, since the step transitions were performed at the same absolute intensity as the initial tests, the order for the ramp test and the step transitions was randomly assigned. At least 24 h was allowed between the ramp test and the step transitions.

Training.

The endurance training program consisted of three exercise sessions per week on a stationary cycle ergometer (Monark Ergomedic 874E; Monark Exercise, Varberg, Sweden) for a total duration of 12 wk. Training intensity was adjusted at 3-wk intervals to reflect changes in fitness level. During the first 10 wk, each session consisted of continuous training (CT) for 45 min at a power output that elicited ∼70% of the V̇o_2peak observed during the incremental ramp test. During the final 2 wk of training (6 exercise sessions), each individual in each group (O and Y) was assigned to one of two subgroups: 1) CT as described above or 2) high-intensity interval training (HIT), performing 10–12 exercise bouts each lasting 1 min at 90–100% of the peak power output achieved during the incremental ramp test, with 1-min rest separating bouts. Since V̇o_2peak was likely to plateau after ∼8 wk of CT (45), and this study was also designed to look at changes in V̇o_2peak with training (see Ref. 43; data not further considered in the present study), HIT was used as a strategy for progressive and continued gains in the exercise program resulting in further increases in V̇o_2peak favored by peripheral adaptations (11).

Measurements.

Gas exchange measurements were similar to those previously described (2). Briefly, inspired and expired flow rates were measured with a low-dead-space (90 ml) bidirectional turbine (Alpha Technologies VMM 110) that was calibrated before each test by using a syringe of known volume. Inspired and expired gases were sampled continuously (every 20 ms) at the mouth and analyzed for concentrations of O₂, CO₂, and N₂ by mass spectrometry (PerkinElmer MGA-1100) after calibration with precision-analyzed gas mixtures. Changes in gas concentrations were aligned with gas volumes by measuring the time delay for a square-wave bolus of gas passing the turbine to the resulting changes in fractional gas concentrations as measured by the mass spectrometer. Data collected every 20 ms were transferred to a computer, which aligned concentrations with volume information to build a profile of each breath. Breath-by-breath alveolar gas exchange was calculated by using algorithms of Beaver et al. (3).

Heart rate (HR) was continuously monitored by electrocardiogram using PowerLab (ML132/ML880; ADInstruments, Colorado Springs, CO) with a three-lead arrangement. Data were recorded with LabChart v4.2 (ADInstruments) on a separate computer.

Local muscle oxygenation profiles of the quadriceps vastus lateralis muscle were made with NIRS (Hamamatsu NIRO 300, Hamamatsu Photonics, Hamamatsu, Japan). Optodes were placed on the belly of the muscle midway between the lateral epicondyle and greater trochanter of the femur. The optodes were housed in an optically dense plastic holder, secured on the skin surface with tape, and then covered with an optically dense black vinyl sheet, thus minimizing the intrusion of extraneous light. The thigh was wrapped with an elastic bandage to minimize movement of the optodes.

The physical principles of tissue spectroscopy are described in detail by Elwell (20), and the manner in which these are applied have been explained by DeLorey et al. (14). Briefly, one fiber-optic bundle carried the NIR light produced by the laser diodes to the tissue of interest while a second fiber-optic bundle returned the transmitted light from the tissue to a photon detector (photomultiplier tube) in the spectrometer. Four different wavelength laser diodes (775, 810, 850, and 910 nm) provided the light source. The diodes were pulsed in a rapid succession, and the light was detected by the photomultiplier tube for online estimation and display of the concentration changes from the resting baseline oxyhemoglobin (HbO₂), deoxyhemoglobin (HHb), and total hemoglobin (Hb_tot). In this study, we used an interoptode spacing of 5 cm. Given the uncertainty of the optical path length in the vastus lateralis at rest and during exercise, NIRS data are presented as delta (Δ) arbitrary units (a.u.). NIRS-derived signal was zero set before the onset of exercise while subjects were quietly seated on the cycle ergometer. The raw attenuation signals (in optical density units) were transferred to a computer for later analysis. Changes in light intensities were recorded continuously at 2 Hz.

Data analysis.

V̇o₂ data were filtered by removing aberrant data points that lay outside 4 SD of the local mean. The data for each transition then were linearly interpolated to 1-s intervals and time aligned such that time zero represented the onset of exercise. Data from each transition were ensemble averaged to yield a single, averaged response for each subject. This transition was further time averaged into 10-s bins to provide a single time-averaged response for each subject. The on-transient response for V̇o₂ was fitted with a monoexponential model of the form

$$Y_{\left(t\right)}=Y_{\mathrm{B}\mathrm{s}\mathrm{l}\mathrm{n}}+\mathrm{A}\mathrm{m}\mathrm{p}\left[1-e^{-\left(t-\mathrm{T}\mathrm{D}\right)/\tau }\right]$$ (1)

where Y_(t) represents V̇o₂ at any time (t); Y_Bsln is the baseline V̇o₂ during 20-W cycling; Amp is the steady-state increase in V̇o₂ above the baseline value; τ is the time constant defined as the duration of time for V̇o₂ to increase to 63% of the steady-state increase; and TD is the time delay (such that the model is not constrained to pass through the origin). The phase I-phase II transition was constrained to a constant of 25 s. Data were modeled from the beginning of phase II to 4 min (240 s) of the step transition. The model parameters were estimated by least-squares nonlinear regression (Origin, OriginLab, Northampton, MA) in which the best fit was defined by minimization of the residual sum of squares and minimal variation of residuals around the y-axis (Y = 0). The 95% confidence interval for the estimated time constant was determined after preliminary fit of the data with Y_Bsln, Amp, and TD constrained to the best-fit values and the τ allowed to vary.

HR data were determined from the R-R interval on a second-by-second basis and edited and modeled in the same manner as the V̇o₂ data described above. The on-transient HR response was modeled from the onset of exercise to 240 s with the exponential model described in Eq. 1.

The NIRS-derived Δ[HHb] data were time aligned and ensemble averaged to 5-s bins to yield a single response for each subject. The Δ[HHb] profile has been described to consist of a time delay at the onset of exercise, followed by an increase in the signal with an “exponential-like” time course (14). The time delay for the Δ[HHb] response (TD-Δ[HHb]) was determined with second-by-second data and corresponded to the time between the onset of exercise and the first point at which the Δ[HHb] signal started to systematically increase. Determination of the TD-Δ[HHb] was made on individual trials and averaged to yield a single value for each individual. The Δ[HHb] data were modeled from the end of the TD-Δ[HHb] to 90 s of the transition with an exponential model as described in Eq. 1. The τ[HHb] described the time course for the increase in Δ[HHb], while the overall time course of Δ[HHb] from the onset of exercise was described by the effective Δ[HHb] (τ′Δ[HHb] = TD-Δ[HHb] + τΔ[HHb]).

The second-by-second Δ[HHb] and V̇o_2p data were normalized for each subject (0–100% of the response). The normalized V̇o_2p was left shifted by 20 s to account for the phase I-phase II transition so that the onset of exercise coincided with the beginning of phase II V̇o_2p, which has been previously described to coincide with muscle V̇o₂ within 10% (49). Data were further averaged into 5-s bins for statistical comparison of the rate of adjustment for Δ[HHb] and ΔV̇o_2p. Additionally, an overall Δ[HHb]-to-ΔV̇o_2p ratio for the adjustment during the exercise on-transient was derived for each individual as the average value from 20–150 s into the transition. The start point was selected to be 20 s to begin beyond the physiological TD-Δ[HHb] derived from NIRS. An end point of 150 s was selected to ensure that both the Δ[HHb] and ΔV̇o_2p signals had already reached 100% of their amplitudes.

Statistics.

Data are presented as means ± SD. Paired and unpaired t-tests and repeated-measures analysis of variance (ANOVA) were used to determine statistical significance for the dependent variables. The ANOVA model was described as S₁₆ × T₅ × A₂ such that subjects (S; no. of subjects) are crossed with testing time (T; 5 testing times: pretraining, week 3, week 6, week 9, and posttraining) and age (A; older and young adults). A Tukey post hoc analysis was used when significant differences were found for the main effects of each dependent variable. Pearson product moment correlation coefficients were used to determine the degree of association between key variables. The ANOVA and correlation coefficients were analyzed by SPSS version 15.0 (SPSS, Chicago, IL). Statistical significance was declared when P < 0.05.

RESULTS

Subject characteristics and pretraining peak exercise values are listed in Table 1. Compliance with the training program was 94 ± 1% (28/30 training sessions) and 95 ± 1% (29/30 training sessions) in O and Y, respectively. V̇o_2peak was significantly increased by 3 wk in both O (by 10 ± 9%) and Y (by 12 ± 6%). Further training resulted in a total improvement from pre- to posttraining that represented 31 ± 10% in O and 18 ± 10% in Y. As noted in methods, groups were split after the 10th week of training; however, since training type (i.e., continuous vs. interval) did not affect any of the kinetic parameters, data are not presented separately.

Table 1.

Subject characteristics and peak exercise responses

	n	Age, yr	Height, cm	Body Mass, kg	Peak WR, W	Peak HR, beats/min	V̇o2peak, l/min	V̇o2peak, ml·kg−1·min−1
Older	8	68 ± 7	177 ± 9	82 ± 8	188 ± 44	144 ± 22	2.3 ± 0.5	28 ± 7
Young	8	23 ± 5*	178 ± 5	80 ± 8	314 ± 41*	189 ± 7*	3.8 ± 0.5*	48 ± 6*

Values are means ± SD for n subjects. WR, work rate; HR, heart rate; V̇o_2peak, peak oxygen uptake.

^*P < 0.05 compared with older.

V̇o₂ kinetics.

O had a greater phase II V̇o₂ time constant (τV̇o_2p) compared with Y. Pretraining τV̇o_2p was 43 ± 11 s in O and 34 ± 8 s in Y. τV̇o_2p decreased significantly by 3 wk of training in both O (35 ± 9 s) and Y (22 ± 8 s), with no further changes seen with continued training (Fig. 1, Table 2). After 3 wk of training, τV̇o_2p in O was similar to that observed in Y before training. No testing time-by-age interactions were detected, reflecting a similar rate of adaptation of V̇o_2p kinetics in both O and Y and a maintained difference between age groups across time.

Fig. 1.

Changes in the phase II pulmonary oxygen uptake (V̇o₂) time constant (τV̇o_2p) over the course of the endurance training program in older and young adults. Pre, pretraining; W3, W6, W9, weeks 3, 6, and 9 of training; Post, posttraining. *P < 0.05 compared with pretraining; NS, not significant.

Table 2.

Kinetics parameters for V̇o₂, HR, and Δ[HHb] in O and Y from pretraining through posttraining

	Pretraining	Week 3	Week 6	Week 9	Posttraining
Phase II τV̇o2p, s
O§	43 ± 10	35 ± 9*	34 ± 8*	33 ± 8*	32 ± 7*
Y	34 ± 8	22 ± 8*	19 ± 6*	20 ± 7*	19 ± 7*
τHR, s
O§	49 ± 15	35 ± 10*	38 ± 10*	31 ± 10*	31 ± 11*†
Y	45 ± 14	28 ± 10*	27 ± 7*‡	29 ± 13*	26 ± 7*†
τΔ[HHb], s
O	13 ± 11	15 ± 7	11 ± 4	13 ± 4	12 ± 4
Y	11 ± 4	12 ± 2	10 ± 2	11 ± 2	10 ± 1
TD-Δ[HHb], s
O§	7 ± 3	9 ± 2	8 ± 2	8 ± 1	9 ± 1
Y	6 ± 1	6 ± 1	6 ± 1	6 ± 2	7 ± 1
τ′Δ[HHb], s
O§	20 ± 12‡	24 ± 8‡	19 ± 4‡	21 ± 4‡	19 ± 4‡
Y	17 ± 4‡	18 ± 2	16 ± 2	17 ± 3	16 ± 1

Values are means ± SD. τ, Time constant of response; V̇o_2p, pulmonary V̇o₂; Δ[HHb], change in deoxygenated hemoglobin concentration; TD, time delay; τ′Δ[HHb], sum of effective τΔ[HHb] and TD-Δ[HHb];

^*Significantly different from pretraining values (P < 0.05);

^†significantly different from week 6 (P < 0.05),

^‡significantly different from phase II τV̇o_2p at the same testing time (P < 0.05);

^§significantly different from Y (P < 0.05).

The amplitude of the increase in V̇o_2p across all testing times was lower in O (0.55 ± 0.29 l/min) compared with Y (1.23 ± 0.20 l/min), reflecting the lower WR in O (i.e., O 68 ± 15 W, Y 128 ± 28 W). No changes in the functional V̇o_2p gain were observed with training (with a mean overall ΔV̇o_2p/ΔWR in O of 11.4 ± 1.3 ml·min⁻¹·W⁻¹ and in Y of 11.5 ± 0.9 ml·min⁻¹·W⁻¹).

HR kinetics.

The τHR was greater in O compared with Y. After 3 wk of training τHR decreased in O and in Y, and it remained decreased compared with pretraining for the remainder of the study (Table 2). In O, τHR was not different from τV̇o_2p at any testing time. In Y, the only significant difference was a greater τHR compared with τV̇o_2p at week 6.

Δ[HHb] kinetics.

The amplitudes of the increase in Δ[HHb] (O 10 ± 5 a.u., Y 13 ± 6 a.u.) as well as τΔ[HHb] (O 13 ± 6 s, Y 11 ± 2 s) were similar in both groups. The overall time course of Δ[HHb], as reflected by τ′Δ[HHb], was longer (P < 0.05) in O (21 ± 7 s) compared with Y (17 ± 3 s). No changes in response to training were observed for TD-Δ[HHb], τΔ[HHb], or τ′Δ[HHb] in either O or Y (Table 2).

Before training, τ′Δ[HHb] adjustment was shorter than τV̇o_2p (P < 0.05) in both O and Y (Table 2, Fig. 2A1 and Fig. 3A1), which resulted in the calculated Δ[HHb]-to-ΔV̇o_2p ratio displaying a transient “overshoot” during the exercise on-transient relative to the subsequent steady-state level (Fig. 2A2 and Fig. 3A2). After 3 wk of training in Y, τ′Δ[HHb] and τV̇o_2p were similar (Table 2), and the Δ[HHb]/ΔV̇o_2p overshoot was attenuated (Fig. 3, B1 and B2). With further training, the overshoot was eliminated in Y (Fig. 3, C1–E1 and C2–E2). In O, after 3 wk of training the Δ[HHb]/ΔV̇o_2p overshoot (Fig. 2B2) also was attenuated; however, in this group, the τ′Δ[HHb] remained shorter (P < 0.05) than the τV̇o_2p (Table 2). No further attenuations in the Δ[HHb]/ΔV̇o_2p overshoot were observed with continued training in O (Fig. 2, C2–E2). The reductions in τV̇o_2p in both O and Y with training were closely associated with a lowered Δ[HHb]-to-ΔV̇o_2p ratio during the exercise on-transient (r: O = 0.93, P < 0.05; Y = 0.98, P < 0.05; Fig. 4).

Fig. 2.

1: Group mean profiles for the adjustment of change in deoxygenated hemoglobin concentration (Δ[HHb]) and V̇o_2p (left shifted such that data from phase I V̇o_2p were not included) during the initial 180 s of a step transition in work rate in older adults pretraining (A), at week 3 (B), week 6 (C), and week 9 (D) of training, and posttraining (E). ●, Time points at which the relative increase of Δ[HHb] is greater than the relative increase of V̇o_2p (P < 0.05). 2: Group mean profiles for the adjustment of Δ[HHb]/ΔV̇o_2p during the initial 180 s of a step transition in work rate in older adults pretraining (A), at week 3 (B), week 6 (C), and week 9 (D) of training, and posttraining (E). a.u., Arbitrary units. *Δ[HHb]/ΔV̇o_2p significantly different from 1.0 (P < 0.05).

Fig. 3.

1: Group mean profiles for the adjustment of Δ[HHb] and V̇o_2p (left shifted such that data from phase I V̇o_2p were not included) during the initial 180 s of a step transition in work rate in young adults pretraining (A), at week 3 (B), week 6 (C), and week 9 (D) of training, and posttraining (E). ●, Time points at which the relative increase of Δ[HHb] is greater than the relative increase of V̇o_2p (P < 0.05). 2: Group mean profiles for the adjustment of Δ[HHb]/ΔV̇o_2p during the initial 180 s of a step transition in work rate in young adults pretraining (A), at week 3 (B), week 6 (C), and week 9 (D) of training, and posttraining (E). *Δ[HHb]/ΔV̇o_2p significantly different from 1.0 (P < 0.05).

Fig. 4.

A: correlation between changes in Δ[HHb]/ΔV̇o_2p and τV̇o_2p in response to training for older men (O) and young men (Y). B: time course of changes in Δ[HHb]/ΔV̇o_2p and τV̇o_2p in Y. C: time course of changes in Δ[HHb]/ΔV̇o_2p and τV̇o_2p in O. *P < 0.05 compared with pretraining.

DISCUSSION

This study examined the effects of a 12-wk endurance training program on the time course of adaptation of V̇o_2p during transitions to moderate-intensity exercise in older and young healthy adults. The main findings were as follows: 1) The decrease in τV̇o_2p in both O and Y occurred within the first 3 wk of training, with no significant changes seen thereafter. 2) After 3 wk of training, an attenuation in the Δ[HHb]/ΔV̇o_2p overshoot during the on-transient, relative to the subsequent steady-state level, suggests a better matching of microvascular blood flow and O₂ distribution and muscle O₂ utilization in the exercise transient in both O and Y. 3) Continued training beyond 3 wk in both O and Y was not associated with further apparent improvement in the matching of blood flow to O₂ utilization or with reductions in τV̇o_2p. 4) In Y, by week 6 O₂ utilization and phase II V̇o_2p were closely matched (Δ[HHb]/ΔV̇o_2p ∼1.0), suggesting that the possibility of a O₂ delivery constraint to muscle V̇o₂ kinetics was ameliorated, whereas in O further improvements in the matching of Δ[HHb]/ΔV̇o_2p were not achieved in response to the training program and thus no additional changes in τV̇o_2p were observed, indicating that O₂ delivery remained a constraint.

Previous studies have demonstrated that endurance training results in a faster V̇o_2p kinetics in both older (1, 6) and young (7, 21, 40, 46) adults. However, the time course of this adaptation has only been explored in young (40, 46) and middle-aged (21), but not older, adults. In younger groups a faster V̇o_2p occurred in as little as 2–4 days of endurance training (40, 46), with improvements continuing up to 30 days after the start of the training program (46) and no further changes observed after 30, 60, and 90 days of training (21). The present data demonstrated faster V̇o_2p kinetics with endurance training in older adults (as well as young) within the first 3 wk, with no further significant changes thereafter (6, 9, 12 wk). Taken together, these data show that the rate of adaptation of oxidative phosphorylation at exercise onset after the initiation of an endurance training program occurs within 3 wk of training regardless of age and may continue up to 30 days (4 wk), with no further change observed with up to 12 wk. Further studies are warranted to clarify the short-term (e.g., <3 wk of training) time course of adaptation of τV̇o_2p in older adults.

What mechanisms or regulatory factors might constrain the V̇o₂ kinetics in both older and younger subjects and explain the faster V̇o₂ kinetics with exercise training? A limitation in O₂ delivery to the active tissues has been proposed as one of the likely mechanisms regulating the rate of adaptation of oxidative phosphorylation (34, 47, 55). Phillips et al. (46) hypothesized that a faster femoral artery blood velocity in the absence of increases in muscle oxidative enzyme activity (27) was the mechanism responsible for the reduction in τV̇o_2p observed early in training in young subjects. From the present study, using the HR as an estimate of “central” blood flow, it was notable that τHR was similar to τV̇o_2p in O and Y at any testing time (with the only exception being Y at week 6), suggesting that the time course of increase of central blood flow and thus central O₂ delivery was matched to muscle O₂ utilization. Training resulted in a decreased τHR in both O and Y. Importantly, τHR is only an indirect estimation of O₂ delivery. Measures of muscle conduit artery blood flow kinetics during the transition to exercise show that the rate of adjustment is similar to or faster than that of V̇o_2p (19, 31, 37). Indeed, in a training study of older adults that resulted in a faster V̇o₂ kinetics, the kinetics of femoral artery mean blood velocity was unchanged (6). Thus bulk delivery of O₂ to the exercising limb does not seem to be limiting τV̇o_2p or the adaptation to training that speeds V̇o₂ kinetics.

Despite evidence showing that the kinetics of bulk delivery of O₂ in the limbs is appropriate to meet the metabolic requirements of the active tissues, blood flow responses to exercise are mediated not only by changes in cardiac output or conduit artery flow but also by the effects of the muscle pump and various vasoactive metabolites and hormones regulating the level of constriction and dilation within the microvascular resistance vessels (17, 18, 41, 48). Recent advances with NIRS have allowed continuous assessment of muscle deoxygenation as an index of the matching of microvascular O₂ delivery to muscle O₂ utilization. DeLorey et al. (13, 14), applying this measure in conjunction with measurements of V̇o₂ kinetics, showed that the rate of adjustment of the NIRS-derived Δ[HHb] signal was faster than the adjustment of phase II V̇o_2p and that this response was exacerbated in older adults, indicating a greater fractional O₂ extraction and thus poorer blood flow distribution. Similarly, Harper et al. (31) reported that in young adults performing moderate-intensity knee-extension exercise femoral artery blood flow adjusted faster than the estimated capillary blood flow. Thus, although the rate of adjustment of blood flow in the conduit artery matches that observed for V̇o₂, the kinetics of microvascular blood flow may have a slower time course and therefore limit the rate of adjustment of V̇o₂ kinetics. With training of young subjects, based on a faster adjustment of phase II V̇o_2p with unchanged adjustment in Δ[HHb], McKay et al. (40) speculated that training-induced decreases in the τV̇o_2p were explained, at least in part, by better matching of muscle O₂ delivery to O₂ utilization. The present data showed that before training, τ′Δ[HHb] was shorter than τV̇o_2p (P < 0.05) in both O and Y (Table 2, Fig. 2A1, and Fig. 3A1), which resulted in the Δ[HHb]-to-ΔV̇o_2p ratio displaying a transient overshoot relative to the subsequent steady-state level (Fig. 2A2 and Fig. 3A2). This transient overshoot in the Δ[HHb]-to-ΔV̇o_2p ratio (values > 1.0) is consistent with a greater microvascular fractional O₂ extraction per unit V̇o_2p compared with the exercise steady state (values = 1.0) and reflects a lower O₂ delivery relative to muscle O₂ utilization in the area of the NIRS probe (slower adjustment of microvascular blood flow).

The young subjects in this study displayed a “slower” V̇o_2p kinetics than normally observed in relatively fit, young adults (τV̇o_2p ∼20 s). The adjustment of V̇o₂ in this group may be constrained by a mismatch between local muscle perfusion and metabolism that requires O₂ extraction to increase rapidly during the exercise on-transient because of a slow increase in local muscle O₂ delivery. In this regard, untrained, sedentary young (as well as older) adults, as used in the present study, may exhibit a reduced endothelium-dependent vasodilation compared with those who regularly perform aerobic exercise (16), which could contribute to a poorer microvascular blood flow and the transient overshoot in the Δ[HHb]-to-ΔV̇o_2p ratio reported in this study. Animal studies have shown that endothelium-dependent vasodilation [using acetylcholine (ACh)] and flow (shear stress)-induced vasodilation were reduced in feed arteries and in 1A arterioles of soleus muscles (oxidative) of old but not of young rats (42), which could contribute to an impaired blood flow distribution in the older adults in this study. Interestingly, at rest and during steady-state submaximal exercise total hindlimb blood flow was similar in old and young rats, but blood flow distribution to oxidative muscles was reduced and distribution to glycolytic muscles was increased in older animals (44), suggesting that the matching of blood flow and O₂ delivery to O₂ utilization in active fibers of the old rats may be compromised. Δ[HHb]/ΔV̇o_2p data presented in the present study support the idea of older individuals having a maldistribution of blood flow within the active muscles at the start of exercise.

After 3 wk of training, the τ′Δ[HHb] in Y was unchanged compared with pretraining values but now was similar to τV̇o_2p (Table 2). Also, during the transition to exercise, the overshoot in the Δ[HHb]/ΔV̇o_2p profile was attenuated in this group (Fig. 3, B1 and B2) and was not evident after further training (Fig. 3, C1–E1 and C2–E2), suggesting that local blood flow and O₂ delivery were better matched to the O₂ requirement of the active muscle. Older adults also showed an attenuated Δ[HHb]/ΔV̇o_2p overshoot after 3 wk of training (Fig. 2B2); however, τ′Δ[HHb] adaptation remained faster than τV̇o_2p (Table 2), and, unlike the response in Y, the overshoot in the Δ[HHb]/ΔV̇o_2p profile was not eliminated with continued training in O (Fig. 2, C2–E2). What mechanism might explain the improvements in the matching of local O₂ delivery to muscle O₂ utilization within 3 wk in both older and younger individuals? Changes to ACh-mediated and flow-induced vasodilation represent an important factor that may have influenced the time course of adaptation of V̇o₂ kinetics to training in older and younger groups. Older and middle-aged men who regularly perform endurance exercise (16, 54) or who have completed 3 mo of aerobic exercise (16) demonstrate a greater ACh-mediated vasodilatory response compared with their sedentary counterparts. Interestingly, exercise training was shown to restore both endothelium (52)- and flow (53)-dependent vasodilation in soleus muscle arterioles from old and young rats. However, training in young rats resulted in improvements beyond those observed in old trained animals (52, 53), supporting the finding of the present study that training adaptations did not continue beyond 3 wk in O, and with 12 wk of training, or even longer-term training (1), the τV̇o₂ does not achieve values reached in the trained young. Rapid improvements in both endothelium- and flow-mediated vasodilation were reported 12–24 h after a single bout of exercise and were sustained for 1–2 days; however, unlike the acute response, chronic exercise training induces adaptations that are twofold higher and more long-lasting, remaining for up to 1 wk after exercise (26, 30). Therefore, enhancement of endothelium- and flow-mediated vasodilation in both O and Y adults may be mainly responsible for the faster blood flow adjustment at exercise onset, and thereby lead to an attenuation or elimination of the overshoot in the Δ[HHb]/ΔV̇o_2p profile that is seen in the present study with exercise training. The high correlation and similar time course of changes in Δ[HHb]/ΔV̇o_2p and τV̇o_2p (Fig. 4) in both O and Y further support the notion that an improved O₂ distribution within the microvasculature plays a major role in the changes observed in τV̇o_2p.

Another mechanism that may affect the vasodilatory responses to exercise is increased accumulation of reactive oxygen species (ROS). Increased ROS have been proposed to affect the nitric oxide (NO)-mediated signaling and bioavailability in older subjects because of their binding affinity to NO (54). Taddei et al. (54) showed that treating older sedentary individuals with antioxidants restored the vasodilatory capacity of NO inhibited by N^G-monomethyl-l-arginine; this suggests that the age-related endothelial dysfunction is at least in part caused by oxidative stress-induced reduction in NO bioavailability. Indeed, administration of antioxidants eliminated the Po_2mv undershoot observed in older rats’ spinotrapezius muscle during a transition from rest to moderate-intensity exercise (33).

Age-related changes in capillary structure have been proposed to affect gas exchange between the capillary and the muscle fiber (9). However, recent data showed that the capillary structure is not compromised with age (38). In fact, the ratio of capillary-to-fiber surface contact to oxidative capacity has been shown to be substantially higher in older rats (32). As such, an O₂ diffusion limitation in the old would not be explained by a reduced structural capacity for O₂ transfer per se but rather depend on the flux and distribution of red blood cells (RBC) within the capillary bed. In this regard, older rats have a decreased lineal density of RBC-perfused capillaries lying adjacent to a fiber (which determines the potential for blood-myocyte O₂ flux) and compensate for this, at least at rest, by increasing individual capillary RBC velocity and flux such that O₂ delivery (as quantified by RBC·mm⁻¹·s⁻¹) is similar in both old and young (51). However, during electrically induced contractions older rats do not show the increased capillary RBC velocity and flux observed in young rats (10). These alterations in capillary hemodynamics in the older rats are likely to reduce the convective and diffusive transport of O₂ to the myocyte.

Several studies have suggested that the locus of control for oxidative phosphorylation resides intracellularly (22, 23). At the onset of exercise, the rate of increase in V̇o₂ is determined by the phosphocreatine shuttle attenuating the increase in ADP accumulation in the mitochondria (56). Additionally, NO production competing with O₂ for the binding site of cytochrome-c oxidase has been proposed to play a role in regulating the rate at which V̇o₂ adapts (35, 36). Substrate supply, in particular related to pyruvate dehydrogenase activity, has also been thought to be one of the mechanisms related to the rate of V̇o₂ increase (24, 28, 29, 50). Although these intracellular factors (as cited above) may be the main ones controlling oxidative phosphorylation and eliciting a V̇o₂ kinetics of less than ∼20 s, for those with slower V̇o₂ kinetics the present data suggest that the matching of microvascular O₂ delivery to the metabolic demand is a key factor determining or constraining the rate at which V̇o_2p adjusts (i.e., τV̇o₂). Nevertheless, we cannot eliminate the possibility that changes in mitochondrial oxidative capacity (with training) may alter phosphorylation and/or redox potential and thereby influence the driving of oxidative phosphorylation.

In conclusion, this study demonstrates that 3 wk of endurance training resulted in a significant decrease in τV̇o_2p in both older and young adults, with no significant reductions in τV̇o_2p seen during the subsequent 9 wk of training. Additionally, τV̇o_2p measured after 3 wk of training in O was similar to that observed in Y before the start of training. Finally, an improved Δ[HHb]-to-ΔV̇o_2p ratio (reflecting a better matching of O₂ distribution within the tissues) was associated with the reduction in τV̇o_2p observed in both O and Y. Although these data do not preclude that the basic control to V̇o_2p kinetics resides within intracellular factors that were not measured in this study, it suggests that with “slower” V̇o₂ kinetics the rate of adaptation of V̇o₂ may be constrained by O₂ availability related to the matching of microvascular O₂ delivery to muscle V̇o₂.

GRANTS

This study was supported by Natural Sciences and Engineering Research Council of Canada (NSERC) research and equipment grants. Additional support was provided by Standard Life Assurance Company of Canada. J. M. Murias was supported by a doctoral research scholarship from the Canadian Institutes of Health Research (CIHR).

DISCLOSURES

No conflicts of interest are declared by the author(s).