| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1School of Kinesiology, 2Department of Medical Biophysics, and 3Lawson Health Research Institute, The University of Western Ontario, London, Ontario, Canada
Submitted 14 August 2006 ; accepted in final form 6 January 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
33 mmHg] and Hyp (PETCO2 17 mmHg) conditions. Phosphorus-31 magnetic resonance and near-infrared spectroscopy were used simultaneously to monitor intramuscular acid-base status, high-energy phosphates, and muscle oxygenation. Resting intracellular hydrogen ion concentration ([H+]i) was lower (P < 0.05) in Hyp [90 nM (SD 3)] than Con [96 nM (SD 4)]; however, at end exercise, [H+]i was greater (P < 0.05) in Hyp [128 nM (SD 19)] than Con [120 nM (SD 17)]. At rest, [PCr] was not different between Con [36 mM (SD 2)] and Hyp [36 mM (SD 1)]. The time constant (
) of PCr breakdown during transition from rest to exercise was greater (P < 0.05) in Hyp [39 s (SD 22)] than Con [32 s (SD 22)], and the PCr amplitude was greater (P < 0.05) in Hyp [26% (SD 4)] than Con [22% (SD 6)]. The deoxyhemoglobin and/or deoxymyoglobin (HHb) was similar between Hyp [13 s (SD 8)] and Con [10 s (SD 3)]; however, the amplitude was increased (P < 0.05) in Hyp [40 arbitrary units (au) (SD 23)] compared with Con [26 au (SD 17)]. In conclusion, our results indicate that Hyp-induced hypocapnia enhanced substrate-level phosphorylation during moderate-intensity exercise. In addition, the increased amplitude of the HHb response suggests a reduced local muscle perfusion in Hyp compared with Con. phosphorus-31 magnetic resonance spectroscopy; near-infrared spectroscopy; muscle oxygen utilization; respiratory alkalosis; hypocapnia
) of 20–65 s (58). This rate of adaptation of oxidative phosphorylation results in a reliance on substrate-level phosphorylation [anaerobic glycolysis and phosphocreatine (PCr) breakdown via the creatine kinase reaction], with possible accumulation of metabolites that have been associated with fatigue (i.e., H+, Pi) (11). Control of oxidative phosphorylation during moderate-intensity exercise has been linked in some way to the creatine kinase reaction {e.g., cytosolic free [ADP] (5), phosphorylation potential ([ATP]/[ADP][Pi]) (59), or Gibbs free energy of ATP hydrolysis (
GATP) (45)}, and this is supported by the tight coupling of PCr breakdown and the phase II kinetics of pulmonary O2 uptake (
O2), reflecting muscle O2 consumption (44, 49). However, other factors, including O2 availability (56) and activation of enzymes [e.g., pyruvate dehydrogenase (PDH)] and provision of substrates (16), have been also implicated to control oxidative phosphorylation at the onset of exercise.
Controlled voluntary hyperventilation (Hyp) has been shown to slow (21, 41) or have no effect (57) on the on-transient response time of pulmonary O2 during moderate-intensity exercise. Furthermore, a previous study has demonstrated an increased contribution of ATP supply from substrate-level phosphorylation during constant-load cycling at 55% of maximal O2 (40). The mechanism for this slowed adaptation of O2 during Hyp is presently unclear but has been suggested to be due to a compromised O2 delivery due to a H+-induced leftward shift in the oxyhemoglobin dissociation curve (21), decreased blood flow (3, 17, 27), or a lower intracellular H+ concentration ([H+]i) causing a reduced activity of PDH (40). However, the effects of Hyp and its relationship to creatine kinase-related controllers of oxidative phosphorylation, [H+]i, and local muscle oxygenation, to our knowledge, have not been examined.
Therefore, to examine the effects of Hyp on PCr kinetics, [H+]i, and local muscle oxy- and deoxygenation kinetics, we used a combination of phosphorus magnetic resonance spectroscopy (31P-MRS) and near-infrared spectroscopy (NIRS) during the transition from rest to moderate-intensity plantar flexion exercise. Specifically, we tested the hypotheses that Hyp would result in 1) a decreased [H+]i, because [CO2] would be reduced in muscle; 2) greater PCr breakdown during exercise, consistent with slowed kinetics of muscle O2 consumption; and 3) a faster increase and a greater amplitude change in relative deoxyhemoglobin and/or deoxymyoglobin (HHb) due to a greater imbalance between O2 utilization and O2 availability, suggesting a reduced O2 delivery to the active muscle.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Experimental set-up. Single-leg plantar flexion exercise was performed using a custom-built ergometer designed to fit the bore of the MRS system. The apparatus has been described in detail previously (46). During the exercise protocols, the subjects lay supine with the legs extended in the ergometer. The ergometer used a cable-and-pulley system in which the subjects raised and lowered a suspended reservoir. Isotonic contractions were performed at the rate of 0.5 Hz (i.e., 1-s contraction, 1-s relaxation). Continuous feedback of range of movement (ROM) was provided to the subjects by a light-emitting diode, which turned on when the targeted ranges for each contraction were met (i.e., start of contraction: foot plate at 90° relative to the apparatus, which was parallel to the ground; end of contraction: 35° from start). ROM and pacing (set by a metronome) were monitored and recorded during the protocols (Spike2 software, version 4.13, Cambridge Electronic Design, Cambridge, UK).
Protocol.
Initially, each subject performed an incremental ramp plantar flexion protocol to determine the work rate that corresponded to the moderate-intensity domain (42). During the ramp protocol, the work rate was increased at the rate of 0.7 W/min until volitional fatigue. For each subject, pHi was plotted against work rate, and piecewise linear regression was used to determine the onset of acidosis, or intracellular pH threshold (TpHi). The moderate-intensity work rate used in this study was set at 75% of TpHi.
During the constant-load trials, each subject used the same loads for both Con and Hyp conditions. The trials were randomized, and the subjects performed three to five trials in each condition. We performed a number of trials so that the 95% confidence interval (C95) of the on-transient PCr time constant (PCr) was 5 s or less in each condition. A maximum of two trials was performed on a day, with a minimum of 30-min rest between trials. Each Con trial consisted of resting accommodation (20 min), moderate-intensity exercise (6 min), and recovery (10 min). The Hyp trials consisted of normal breathing during rest (3 min) and controlled voluntary hyperventilation (see below) during rest (20 min), exercise (6 min), and recovery (10 min). The load was placed on the apparatus 30 s before the start of exercise, causing the relaxed foot to be moved to a dorsiflexed position before the start of exercise.
Hyp was induced by breathing at the rate of 30 breaths/min (i.e., the same pace as the contractions during exercise). The subjects wore a noseclip, and they breathed through a mouthpiece (Vacumed 1002) that was attached by flexible autoclavable laboratory tubing (TYGON R-3603) to a gas analyzer (Puritan-Bennett, PB-253 Airway Gas Monitor). End-tidal PCO2 (PETCO2) and breathing frequency were monitored throughout the trials, and verbal feedback on PETCO2 was provided to the subjects to ensure a PETCO2 of 20 mmHg or less.
31P-MRS. 31P-MRS data were collected using a 64-cm bore, 3.0-T superconducting magnet (IMRIS, Winnipeg, MB, Canada) interfaced with a SMIS console (Surrey Medical Imaging Systems, Guilford, UK). A 6 x 9 cm 31P surface coil was placed primarily under the muscle belly of the lateral gastrocnemius on the right leg. The 1H signal was used to shim the magnet homogeneity, and the homogeneity was adjusted until the proton signal from water produced a peak with a full width at one-half maximum of 20–25 Hz and was approximately Lorentzian in shape.
Free induction decay (FID) data were collected every 3 s throughout the protocols. During analysis of the resting data, 20 subsequent spectra were averaged to produce one spectrum each minute. During exercise and recovery, two subsequent spectra were averaged, providing a time resolution of 6 s. All FID data were acquired using a 3-ms adiabatic 90°-radio frequency pulse, 12-µs delay time, 5.0-kHz receiver bandwidth, and 2,048 complex data points.
NIRS. A continuous-wave NIRS system was used and consisted of a tungsten-halogen light source, two nonmagnetic fiber-optic cables, and a spectrometer. The spectrometer was composed of a holographic grating located in a light-tight casing and a cooled charge-coupled device (CCD) camera (Wright Instruments, Enfield, Middlesex, UK). Near-infrared light was transmitted through one optode, and a fraction of the scattered light was collected in the range of 600–980 nm by the second optode. The optodes were spaced 4 cm apart and centered within the surface coil, allowing for simultaneous 31P-MRS and NIRS measurements from the same general region of the muscle. A custom-built plastic encasement secured the optodes to the surface coil, which was fastened to the leg by velcro straps. The intensity spectrum of the scattered light was channeled on the holographic grating where it was dispersed across the CCD chip. The absorption spectrum was then fit using a method that has previously been discussed (9, 43). The output from this fitting algorithm allowed us to measure relative changes in the concentrations of HHb, oxyhemoglobin and/or oxymyoglobin (HbO2), and total hemoglobin and/or myoglobin (Hbtot). These measurements were continuously recorded every 0.2 s and were presented as a change from the preaccommodation baseline. Since the differential path length of the photons was uncertain during exercise for each subject, the NIRS-derived data are reported in arbitrary units (au).
Data analysis.
Quantification of 31P-MRS metabolite data was performed in the time domain by fitting the FID data to a sum of damped sinusoids, which could be varied in amplitude, phase, delay time, damping constant, and frequency. This method used a priori knowledge and a nonlinear least-squares algorithm to iteratively reduce the difference between the data and the experimental model (2). The relative concentrations of the phosphate metabolites [Pi, PCr, phosphomonoesters, phosphodiesters, and 
-, 
-, and 
-ATP] were determined from the amplitude of the exponential model function at time equal zero. Phosphate peaks were corrected for partial saturation (54). Before commencing the experimental protocol, baseline spectra were acquired that consisted of an average of six acquisitions having a repetition time of 30 s each. Calculation of the longitudinal relaxation (T1) correction factors was done using the amplitude of the 31P metabolites in the baseline spectrum (no T1 effects) and the spectra collected during the resting period in which breathing was at the normal rate. Absolute PCr and Pi concentrations were calculated by using PCr/ATP and Pi/ATP ratios, respectively, and by assuming a resting ATP concentration of 8.2 mM (20). ATP peak area was determined from the average of the -, -, and -ATP peaks. Free [ADP] and [AMP] were calculated by assuming a total creatine concentration of 42 mM (20) and that the creatine and adenylate kinase reactions were at equilibrium. The equilibrium constants were adjusted for [H+]i for a free [Mg2+] of 0.6 mM (14). The diprotonated Pi concentration ([H2PO4–]) was calculated using Eq. 1 (47), and GATP was calculated using Eq. 2 (36)
 | (1) |
 | (2) |
GATP is the standard Gibbs free energy of ATP hydrolysis (–32 kJ/mol), R is a gas constant (8.3145 J·K–1·mol–1), and T is absolute temperature (311 K). The pHi was determined from the chemical shift of Pi relative to PCr (54). [H+]i was calculated from pHi.
Kinetic analysis.
PCr and HHb kinetic responses were fit with a monoexponential model of the formula described by Eq. 3 using nonlinear least-squares regression techniques (Microcal Origin, version 4.10, 1996):
 | (3) |
is the time constant defined as the duration of time through which Y changes to a value equivalent to 63% of A, and TD is the time delay. For each individual trial, the PCr data were interpolated to produce 1-s time resolution, then time-aligned and ensemble-averaged into 5-s bins to produce a single data set for each subject in each condition. The PCr exercise data were fit starting at the time corresponding to the start of exercise (TD was constrained to 0 s) and proceeded to the end of the exercise bout (i.e., 360 s). The baseline value was the average of the resting PCr data and was assigned the value of 100%. The PCr recovery data were fit starting at the time corresponding to the end of exercise (TD was constrained to 0 s) and proceeded to the end of recovery (i.e., 600 s). The baseline Y value used for determining the PCr recovery kinetics was the steady-state exercise data.
The HHb data were fit similar to the method reported by DeLorey et al. (10). For each individual trial, the HHb data were interpolated to produce 1-s time resolution. The time to the onset of an increase in HHb was determined as the first point of two consecutive points greater than 1 SD above the mean of the 3 min of resting data before exercise onset. The time delay was then calculated as the average of the trials for each subject in each condition. The HHb data were not different (P > 0.05) between repeated trials, so HHb data were time-aligned and ensemble-averaged to 5-s bins to produce a single response from each subject. HHb data were then fit from the time of initial increase in HHb to 120-s with the monoexponential model. Data were fit to 120 s because 1) this was similar to the average time that PCr reached steady state during Con and therefore was within the time period of most interest; and 2) up until this time HHb was generally well fit with a monoexponential model, whereas the HHb response was more variable during the final 240 s of exercise. The HbO2 and Hbtot were interpolated to produce 1-s time resolution for each individual trial, then time-aligned and ensemble-averaged into 5-s bins to produce a single response for each subject.
Statistical analysis. Statistical analyses were performed using SigmaStat v3.1. PCr and HHb kinetic parameter estimates were analyzed using one-way ANOVA for repeated measures. In addition, all 31P-MRS- and NIRS-derived data, as well as PETCO2, were binned into 60-s intervals and analyzed for condition, time, and interaction effects during rest, exercise and recovery by repeated-measures ANOVA. When a significant interaction effect was found, a post hoc analysis was performed using Tukey's post hoc test at specific time points. Significance was set at the P < 0.05 level for all comparisons. Data presented in the text and tables are reported as means (SD), and figures show means (SE).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
31P-MRS.
Before the start of the hyperventilation maneuver, [H+]i was similar in Hyp and Con. [H+]i did not change during the resting accommodation (Con) but decreased throughout the resting hyperventilation (Hyp), such that [H+]i was lower (P < 0.05) in Hyp than in Con during the final 5 min of the preexercise rest period (Fig. 1). Exercise at 75% TpHi was associated with a small rise in [H+]i in both conditions (equivalent to a mean drop of 0.07 pHi units in Con) (Fig. 1). There were no differences in [H+]i between conditions early in exercise; however, by end exercise, the [H+]i in Hyp was greater (P < 0.05) than in Con (Fig. 1). Furthermore, the net accumulation of [H+]i (i.e., difference in [H+]i from end of rest; [H+]i) was greater in Hyp than Con from 90 s to the end of exercise (Fig. 2). [H+]i was similar in both conditions immediately after exercise, but [H+]i was lower (P < 0.05) in Hyp than in Con during the final 5 min of recovery (Fig. 1). [H+]i at end of recovery were similar to preexercise values in each condition.
|
|
(48)] was similar in Hyp [49%/min (SD 20)] and Con [55%/min (SD 28)], suggesting a similar ATP cost of contractions (12). PCr breakdown was greater (P < 0.05) in Hyp than Con after 2 min of exercise (Fig. 1). As seen in Fig. 3, the [PCr] decreased toward a steady state, with the time course being well fit with a monoexponential model. The on-transient PCr [Hyp, 39 s (SD 22); Con, 32 s (SD 22)] and amplitude change [Hyp, 26 % (SD 4); Con, 22 % (SD 6)] were both greater (P < 0.05) in Hyp than Con (Table 1). The off-transient PCr had a tendency (P = 0.15) to be longer in Hyp than Con (Table 2), while the amplitude change was greater (P < 0.05) in Hyp than Con (Table 2). The off-transient PCr time constants were not different than the on-transient PCr time constants in either Hyp or Con.
|
|
|
GATP were not different between conditions. During steady-state exercise (i.e., the average of the final 3 min of exercise), there was a greater [Pi], reduced phosphorylation potential, and lower (less negative) GATP in Hyp than Con, while there was a tendency for free ADP to be increased (P = 0.11) in Hyp vs. Con (Table 3). Glycolytic flux may be affected by Pi, monoprotonated Pi, ADP, and AMP. There were no measurable differences in these metabolites immediately before exercise or during the early stages of exercise (i.e., 30 s). However, during the steady-state phase of the exercise, Pi and monoprotonated Pi were greater (P < 0.05) in Hyp than Con, while there was a tendency for ADP (P = 0.11) and AMP (P = 0.07) to be elevated in Hyp vs. Con (Table 3). [ATP] was similar in both conditions throughout the protocol and was not different from resting levels during exercise or recovery.
|
|
, and mean response time (i.e., TD + ) were similar in Con and Hyp; however, the amplitude was greater (P < 0.05) in Hyp than Con (Table 4; Figs. 3 and 4). HbO2 and Hbtot were not different during exercise, although HbO2 tended to be lower (P = 0.16) in Hyp than Con (Fig. 4). No differences in Hbtot, HbO2, and HHb were observed during recovery (Fig. 4).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Rest.
The present study is the first to demonstrate a reduction in in vivo muscle [H+]i as a consequence of voluntary hyperventilation-induced hypocapnia in humans, although this effect was shown previously in animal models using mechanical ventilation (23). The lower [H+]i is consistent with the decrease in intracellular CO2 stores as seen by the lowering of PETCO2 from 33 mmHg in Con to 17 mmHg in Hyp. In the present study, despite an intracellular alkalosis, and presumably a small shift in the creatine kinase equilibrium, there was no measurable effect on resting [PCr] (Fig. 2) or free [ADP] (Table 3). The lack of an effect on these metabolites, in combination with no differences between conditions in relative concentrations of Hbtot, HHb, and HbO2, suggests that there was no effect of Hyp on muscle O2 consumption during rest. For example, if there was an increase in muscle O2 consumption in Hyp, an increase in O2 extraction (evident by an elevated HHb and decreased HbO2 signal) and likely a change in [PCr] would have been expected but was not observed. The finding of hyperventilation having no measurable effect on muscle O2 consumption at rest is similar to studies that have demonstrated no changes in pulmonary O2 during hyperventilation at rest in human subjects (21, 40), but it is in apparent contrast to hypocapnia-induced alkalosis causing an increased O2 in anesthetized dogs (4) and in isolated perfused canine hindlimb muscle (19).
Exercise.
At the onset of exercise, the initial rate of PCr breakdown (48) was similar in Hyp and Con, suggesting a similar ATP cost of contractions during exercise. However, the greater amplitude of the on-transient PCr kinetic response (Table 1 and Figs. 1 and 3) indicates a greater ATP supply from PCr breakdown during the transition from rest to steady-state exercise. In addition, the elevated [H+]i during the later stages of exercise in Hyp vs. Con in the present study is consistent with previous observations that hyperventilation causes a greater muscle lactate accumulation (3, 40). An increase in muscle lactate represents a mismatch between pyruvate production and oxidation, such that any excess of pyruvate production is converted to lactate. Therefore, the greater accumulation of [H+]i (and possibly lactate) in Hyp observed in this study (Figs. 1 and 2) may be explained by a slower activation of the mitochondrial PDH complex and/or an enhanced glycolytic flux. A slowed PDH activation and an enhanced rate of glycogenolysis have both been observed during the early stages of cycling exercise during a similar hyperventilation protocol (40). An elevated glycolytic flux with a delayed PDH activation would be expected to result in a greater muscle lactate accumulation, which would contribute to an increase in [H+]i through physiochemical changes (22). Therefore, the greater [PCr] breakdown and possibly enhanced lactate accumulation indicate increased ATP production from substrate-level phosphorylation and reduced ATP supply from oxidative phosphorylation during the early stages of exercise. This agrees with the slower adaptation of pulmonary O2 observed during leg cycling exercise during similar hyperventilation protocols (21, 41).
There are at least two possible explanations of how hyperventilation could cause slowed kinetics of muscle O2 consumption. First, hyperventilation may have resulted in an increased contribution from glycogenolysis and anaerobic glycolysis during the early stages of exercise. An enhanced rate of anaerobic glycolysis has been demonstrated to directly slow the time course of adaptation of muscle O2 consumption at the onset of exercise in in silico studies (33, 34). These modeling studies attributed the slowed kinetics of O2 consumption caused by the greater transient ATP supply from anaerobic glycolysis to lower free [ADP] (and subsequently a reduced drive for oxidative phosphorylation) and a reduction in the requirements for ATP supply (33, 34). There are several possible mechanisms for hyperventilation causing an enhanced glycolytic flux at the onset of exercise. One possibility is that an intracellular alkalosis would be expected to increase the monoprotonated form of Pi, which has been implicated to convert glycogen phosphorylase b to the active a form (28). However, in the present study, monoprotonated Pi was not measurably different between conditions before exercise onset and during the early stages of exercise (Table 3). Furthermore, other metabolites typically associated with the activation of glycogenolysis and glycolysis such as AMP and ADP (8, 38) were also calculated to be similar in both conditions before exercise onset and during the early stages of exercise (Table 3). Although it was not measured in this study, it is possible that the rate of glycogenolysis was increased by elevated plasma epinephrine. Twenty minutes of voluntary hyperventilation in human subjects has been previously demonstrated to elevate plasma epinephrine by approximately threefold (37, 53), and an elevated plasma epinephrine has been associated with an enhanced rate of glycogenolysis in some (7, 24) but not all studies (39). Plasma epinephrine may enhance glycogenolysis by increasing [cAMP], which mediates a series of reactions that transforms inactive muscle phosphorylase b to the active a form (7).
Alternatively, hyperventilation may have contributed directly to slowed kinetics of O2 consumption, and this, in turn, may have caused the increased reliance on PCr breakdown to meet ATP requirements. A greater PCr breakdown would be expected to be associated with an increase in metabolites that have been implicated to regulate glycolysis and glycogenolysis, including Pi, ADP, and AMP (8, 38). In the present study, the elevated [Pi], free [ADP] (P = 0.11), and [AMP] (P = 0.07) during steady-state exercise in Hyp compared with Con may have had a positive allosteric effect on phosphofructokinase and enhanced glycogen phosphorylase activity (6), resulting in an enhanced glycolytic flux. Therefore, the greater PCr breakdown and possibly enhanced glycogenolysis in Hyp may have been consequent to an impaired drive for oxidative phosphorylation.
Possible effects of hyperventilation on control of oxidative phosphorylation.
In addition to providing a means of supplying ATP, PCr breakdown via the creatine kinase reaction has been associated with stimuli for driving oxidative phosphorylation (e.g., ADP, phosphorylation potential, GATP). Previous studies have demonstrated that the pulmonary O2 kinetic response amplitude and steady state-level achieved following transitions to moderate-intensity cycling exercise were not altered by hyperventilation (21, 40), suggesting that the steady-state muscle O2 consumption is unaffected by the hyperventilation condition. Therefore, in the present study, the greater PCr breakdown in Hyp suggests that there was an enhanced PCr-related stimulus (Table 3) to maintain the same rate of muscle oxidative phosphorylation.
In some metabolic control models, the rate of oxidative phosphorylation is controlled by a combination of factors (e.g., 1, 13, 32, 59). For example, in Wilson's proposed model (59), the rate of oxidative phosphorylation is controlled by a combination of the redox potential ([NAD+]/[NADH]), phosphorylation potential ([ATP]/[ADP][Pi]), and O2 availability. Furthermore, during submaximal exercise, a reduced drive from one of these mechanisms must be compensated by an increase in another to maintain the same rate of oxidative phosphorylation. In this study, during steady-state exercise, the phosphorylation potential {calculated as ln ([ATP]/[ADP][Pi])}, was lower in Hyp than Con (Table 3), and associated with this difference, the cytoplasmic free energy of ATP hydrolysis (GATP) was lower (i.e., less negative) in Hyp than Con (Table 3). Both of these factors indicate an enhanced stimulus to drive oxidative phosphorylation. Therefore, in the present study, the greater PCr breakdown in Hyp, and associated enhanced drive from the phosphorylation potential, may be in response to a reduction in mitochondrial NADH (and FADH2) and/or O2 availability.
NADH (and FADH2) provision for oxidative phosphorylation is derived mainly from the oxidation of acetyl CoA within the tricarboxylic acid (TCA) cycle. The mitochondrial PDH complex is responsible for the oxidative decarboxylation of pyruvate to form acetyl CoA, and thus a limitation in the activation of PDH at the onset of exercise could impair the delivery of carbohydrate-derived acetyl CoA and NADH units to the TCA cycle and electron transport chain. Hyperventilation-induced hypocapnic alkalosis was shown to slow PDH activation at the onset of cycling exercise at 55% of maximal O2 (40). In that study, the slowed rate of PDH activation was attributed to an intracellular alkalosis (40), which resulted in an activation of PDH kinase [pH optimum 7.0–7.2 (16)] and an inhibition of PDH phosphatase [pH optimum 6.7–7.1 (16)] (covalent regulators of the PDH complex) and combined for an overall inhibition of the PDH complex early in exercise. In the present study, the muscle [H+]i was lower in Hyp than Con immediately before the start of exercise, although the [H+]i decrease was only 6 nM lower [i.e., pHi increase from 7.00 (Con) to 7.04 (Hyp)]. Differences in [H+]i were not observed early in the transition to exercise, and thus the inhibition of PDH may have been attenuated with time in exercise as shown by LeBlanc et al. (40). However, it should be noted that previous studies have demonstrated that increasing the activity of PDH by dichloroacetate administration does not speed the on-transient kinetics of O2 consumption in in situ canine (15) and in vivo human (25, 31, 50) models, suggesting that PDH activity does not normally limit on-transient O2 kinetics.
The slowed kinetics of O2 consumption in Hyp, and associated greater PCr breakdown, could be related, in part, to a reduced O2 availability. The hyperventilation-induced hypocapnia and associated decrease in plasma [H+] would be expected to attenuate peripheral vasodilation (18) and cause a leftward-shift in the oxyhemoglobin dissociation curve, thereby impairing both convective and diffusive O2 delivery. The NIRS-derived HHb signal represents the balance between local microvascular perfusion and muscle O2 consumption. In the present study, assuming a similar steady-state muscle O2 consumption in Hyp and Con, the greater amplitude of the HHb response in Hyp (Figs. 3 and 4) and similar HHb mean response time between conditions [despite possibly slower kinetics of O2 consumption in Hyp (21, 41)] suggest that muscle perfusion may have been reduced in Hyp, requiring a greater reliance on muscle O2 extraction.
It has been previously observed that mechanical hyperventilation reduced dog skeletal muscle blood flow (3), pig arterial blood flow (27), and rabbit microcirculatory blood flow (17), and voluntary hyperventilation lowered human forearm blood flow (30). In the present study, although Hbtot was not different between conditions, the Hbtot signal reflects the change in total [Hb] within the region of NIRS interrogation and may not be a sensitive indicator of blood flow per se (35). Rather, the Hbtot signal reflects a combination of factors that include local muscle perfusion and vasodilation and constriction.
Although we did not measure plasma [H+] in this study, the lowering of PETCO2 from 33 to 17 mmHg in Hyp during rest (and similarly during exercise) would correspond to a decrease in arterial PCO2 (PaCO2) from 32 to 18 mmHg, respectively (26), and these PaCO2 values correspond to a plasma [H+] of 35 nM (pH 7.46) and 22 nM (pH 7.65), respectively (52). A reduced plasma [H+] in Hyp results in a greater hemoglobin affinity for O2 in the blood and a leftward shift in the oxyhemoglobin dissociation curve, and therefore a reduced O2 off-loading. Assuming a temperature of 37°C, these values would result in a lower O2 half-saturation pressure of Hb (P50) in Hyp (20 mmHg) compared with Con (25 mmHg) (29). The lower plasma PO2 in Hyp would presumably reduce the driving pressure for O2 from the microvasculature into the muscle, thereby reducing diffusive O2 delivery to the active muscle. This is known to occur in an animal model; Gustafsson et al. (17) observed a 9% reduction in rabbit skeletal muscle PO2 during hypocapnia. A leftward shift in the oxyhemoglobin dissociation curve would be expected to reduce the amplitude of the HHb kinetic response during exercise. In contrast, the increased amplitude of HHb in Hyp (compared with Con) observed in the present study indicates that the reduced blood flow had a relatively greater effect on the HHb signal than the leftward shift in the oxyhemoglobin dissociation curve. This suggests that blood flow may be the dominant factor in reducing O2 delivery in Hyp.
In summary, Hyp-induced hypocapnia resulted in a reduced [H+]i at rest, no measurable differences in [H+]i during the initial minutes of exercise (i.e., during the time period that PCr adapted to a steady-state), and an elevated [H+]i during the later stages of exercise compared with Con. Despite no measurable differences in [H+]i during the early stages of exercise, we observed a greater PCr breakdown and a slower time-course for PCr breakdown in Hyp, consistent with a slower adaptation of muscle O2 consumption. In addition, the greater amplitude of HHb in Hyp but no change in the HHb mean response time between conditions may indicate that O2 delivery was impaired in the Hyp condition.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
O2 kinetics following the onset of heavy intensity cycle exercise. Pflügers Arch 447: 867–874, 2004.[CrossRef][ISI][Medline]O2 kinetics during moderate-intensity cycle exercise. Med Sci Sports Exerc 37: 1159–1164, 2004.O2) on-kinetics in skeletal muscles. Biochem J 379: 703–710, 2004.[CrossRef][ISI][Medline]O2 and muscle deoxygenation at the onset of moderate-intensity exercise with respiratory alkalosis (Abstract). Med Sci Sports Exerc 36: S61, 2004.O2 and intramuscular 31P metabolite kinetics during high-intensity exercise in humans. J Appl Physiol 95: 1105–1115, 2003.This article has been cited by other articles:
|
|
 |
|
  L. M. K. Chin, R. J. Leigh, G. J. F. Heigenhauser, H. B. Rossiter, D. H. Paterson, and J. M. Kowalchuk Hyperventilation-induced hypocapnic alkalosis slows the adaptation of pulmonary O2 uptake during the transition to moderate-intensity exercise J. Physiol., August 15, 2007; 583(1): 351 - 364. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
