| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biomedical Sciences and Technologies and M.A.T.I. Centre of Excellence, University of Udine, Udine, Italy
Submitted 3 January 2008 ; accepted in final form 23 April 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PCr). High-energy phosphate concentration (by 31P-NMR spectroscopy) in the calf muscles were measured during three repetitions of the rest-to-work transition of moderate aerobic square-wave exercise on nine healthy volunteers, while resting [PCr] was estimated from the appropriate spectroscopy data. PCr concentration decreased significantly (22 ± 6%) from rest to steady-state exercise, without differences among the three repetitions. Absolute resting [PCr] and PCr were consistent with literature values, amounting to 27.5 ± 2.2 mM and 23.9 ± 2.9 s, respectively. No significant relationships were detected between individual PCr and mechanical power, fraction or absolute amount of PCr hydrolyzed, or change in ADP concentration. On the contrary, individual PCr (s) was linearly related to absolute resting [PCr] (mM), the relationship being described by: PCr = 0.656 + 0.841·[PCr] (n = 9, R = 0.708, P < 0.05). These data support the view that in humans PCr concentration sets the time course of the oxidative metabolism in skeletal muscle at the start of exercise, being one of the main controllers of oxidative phosphorylation. mitochondrial respiration; 31P-NMR spectroscopy; work onset
O2) increases at a slower rate than the (stepwise) increase in ATP utilization. However, ATP concentration in working muscles is maintained at a constant level (15, 26) since the hydrolysis of phosphocreatine (PCr) fills the gap between energy requirement and energy yield by oxidative pathways. Thus, if no other energy sources are involved, e.g., in the absence of any detectable lactate production (8), at any time during the exercise the sum of the energy per unit of time derived from net oxygen consumption (O2) and from the PCr breakdown (PCr) provides the overall required power (W). Under these conditions, both O2 and PCr approach a new steady state in a complementary fashion since PCr represents a readily available source of energy that the muscle can use to cover the increased energy demand until oxidative metabolism meets the energy requirement. As a first approximation, after the onset of a constant-load exercise the time course of O2 and of PCr concentration ([PCr]) follow a first-order monoexponential kinetics, characterized by the same time constant (37, 38, 56 among many others).
The factors affecting the rate of adjustment of these two energy sources during the rest-to-work transients (or during the work-to-rest transient) have been studied in some detail (21, 26, 32). However, they are still not fully understood, and, to our knowledge, the relationships between high-energy phosphate concentration and/or their changes during transients and the time constant of the O2 and [PCr] kinetics have never been studied directly in humans. Nevertheless, on the basis of the analysis of steady-state exercise data, PCr concentration at the start of a sudden increase in energy demand has been suggested to be one of the main factors determining the [PCr] time constant (14). To the authors' knowledge, the only alternative model that predicts the relationship between the PCr kinetics on a sudden increase in energy demand and the concentration of the main phosphagens (or change thereof) is that of Korzeniewski and Zoladz (34). Applying this model, it is concluded that the [PCr] kinetics is determined by the amount of PCr that has to be transformed into creatine during the transition. Since workload determines the amount of PCr hydrolyzed, a consequence of this prediction is that the PCr kinetics ought to be related also to work intensity; this last relationship, however, is in contrast to the conclusions of Binzoni et al. (4), who stated that the half-time of the muscle PCr kinetics was independent of workload.
The aim of this study was to verify experimentally in humans, under physiological conditions, the hypothesis that the time constant of the [PCr] kinetics during rest to moderate plantar flexion exercise transition is set by the phosphocreatine concentration at rest, but it is independent of the amount of hydrolyzed phosphocreatine.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Exercise protocol and 31P-NMR spectroscopy. 31P-NMR spectra were collected from the right calf of the subjects both at rest and during dynamic plantar flexion exercise using a standard whole body Magnetom SP 4000, 1.5-T scanner (Siemens, Erlangen, D) located at the Radiology Unit of the School of Medicine of the University of Udine (Italy).
Lying supine on a specially designed nonferromagnetic calf ergometer (17), each subject performed consecutively three exercise trials, separated by at least 10 min rest. Each trial consisted of 2.5 min rest followed by 6 min of rhythmic plantar flexion. The exercise, of moderate intensity, consisted of synchronous plantar flexions of 
30° with both feet against the same force, provided by a given number and type of rubber band and at the same frequency of 0.83 Hz (imposed by a metronome), with the knee extended. The parameters determining actual mechanical power (i.e., number and type of rubber bands, frequency and amplitude of the flexions) were the same for all the volunteers and were chosen on the basis of our previous experience (17, 18) to obtain an exercise intensity below the intracellular lactate threshold (38). Subjects were asked to carefully control the frequency and amplitude of the flexions; this last could be monitored on a light-emitting diode (LED) bar fixed at eye level. Adjustable straps and belts maintained the subjects' feet and body in the appropriate position on the pedals and on the main frame, thus minimizing unwanted movements and muscle contractions. Actual mechanical power was calculated from the outputs of the force and displacement transducers of the ergometer (see 17 for details).
31P-NMR spectra were acquired using a 5-cm-diameter surface coil positioned on the middle part of the medial belly of the gastrocnemius muscle of the right leg. The magnetic field homogeneity was adjusted by manual global shimming on the proton signal of tissue water until the peak was approximately Lorentzian in shape; the resulting mean peak width at one-half maximum was 0.27 ± 0.03 (mean ± SD) parts per million (ppm). After switching to 31P, a complete data set (composed of 120 consecutively acquired spectra) was collected throughout each trial. Each spectrum was obtained from the sum of three free-induction decays, each acquired in 2,048 complex data points with a dwell time of 250 µs and a repetition time of 0.601 s. Because of the characteristics of the spectroscopy sequence (9), a final time resolution of 3.65 s was obtained. At the start of the collection of the 46th spectrum, using the intercom of the NMR equipment, the same investigator gave verbally the signal for the start of exercise to all subjects.
Analysis of spectra. Spectroscopy data were analyzed off-line, with the MRUI software (43). Positions and areas (signal intensities) of spectral peaks were calculated by means of the time-domain VARPRO/minpack fitting program, using the appropriate starting values and without any prior knowledge. The first three spectra of each dataset were discarded, and no other correction for partial saturation was made, assuming that the T1 relaxation time of PCr remained constant throughout the exercise.
Each collected spectrum was first analyzed separately, and the area of the PCr peak was determined. Overall, 27 (9 volunteers x 3 trials each) high-resolution datasets were obtained, each composed of 117 (= 120 – 3) data points separated by 3.65-s intervals (i.e., 1 spectrum), and used for the kinetic analysis.
Subsequently, to increase the signal-to-noise ratio, the spectra of the same data series were averaged every 3 consecutive spectra, thus obtaining 39 new average spectra with a time resolution of 10.95 s. The chemical shift (
; ppm) between Pi and PCr peaks was evaluated on these new spectra, which allowed us to calculate intracellular pH applying the following equation (1):
 |
Kinetic analysis.
Custom software, based on the algorithms reported by Numerical Recipes in C (45), was used. A nonlinear iterative chi-square fitting procedure was applied to calculate the best fit for the [PCr] kinetics. Chi-square was calculated using for each data point the standard deviation obtained from the VARPRO spectral analysis program. For each set of 120 (minus 3) spectra, the areas of the PCr peak [signal intensity, S(t)] were interpolated as a function of time (t) using the following monoexponential function:
 |
is the time constant, t0 is the start time of signal change (corresponding to the start of exercise), Sr is the signal in resting conditions, and Ss is the signal intensity at the asymptote (i.e., at steady-state exercise). All four parameters (, t0, Sr, Ss), and the corresponding errors, were determined by the iterative fitting procedure, the starting values of which were as follows: = 21.9 s, t0 = time of the 46th spectrum, Sr = average value obtained over the 2.5 min rest (42 spectra), and Ss = average value of the last 25 spectra collected during steady-state exercise (in turn considered to be reached 3 min after exercise onset).
The fitting procedure was run first on the 27 individual datasets and the corresponding fractions of PCr hydrolyzed (
PCrs/PCrr) were calculated as (Sr – Ss)/Sr. Subsequently, all the datasets were oversampled in 0.36-s time bins; they were time-aligned, and thus the three profiles of each subject were ensemble averaged, yielding finally nine mean profiles. The fitting procedure was run a second time on these nine average datasets, and the obtained time constants, and corresponding errors, were assumed to be the individual values of the volunteers.
Assessment of absolute resting concentrations.
To assess the individual PCr and Pi concentrations at rest ([PCrr] and [Pir], respectively), a grand average spectrum was calculated for each subject averaging all the spectra acquired at rest (42 spectra x 3 repetitions). All the phosphagen signal intensities (Pi, PCr, 
ATP, 
ATP, and βATP) and corresponding errors were obtained, fitting the respective peaks in the time-domain with the prior knowledge and the classical constraints used for a 1.5-T spectrometer (51). The signal intensities were then brought back to the completely relaxed condition applying the appropriate saturation factors calculated on the basis of the apparent spin-lattice relaxation times (T1) (19). These are reported by the literature (9) for the different phosphagens as obtained applying the same acquisition parameters as in the present experimentation. Finally, [PCrr] and [Pir] were estimated by comparing the area under the PCr and Pi peaks, respectively, with that of the βATP, assuming an average ATP concentration of 5.72 mmol/kg wet wt for the resting muscle (2). This value corresponds to a concentration of 8.54 mmol/l of intracellular water at rest, assuming a content of 0.67 ml of intracellular water/g muscle wet wt (31). Throughout this study all concentrations are expressed in millimoles per liter intracellular water (mM).
Calculation of ADP concentration.
Individual resting and steady-state free cytosolic ADP concentrations were calculated from the equilibrium equation of the creatine kinase reaction:
 |
Statistical analysis. Statistical analyses were performed using the Systat package, and average values are expressed as means ± SD. The Shapiro-Wilk test was applied to verify that all the samples came from a normally distributed population. Since all variables satisfied this condition, parametric statistical tests were subsequently used. For multiple comparisons, analysis of variance for repeated measures (MANOVA) was applied, followed by specific contrasts when appropriate. Correlation between variables was assessed by means of the Pearson correlation coefficient, assuming a P < 0.05 as statistically significant. Linear fits were calculated according to the Deming regression, which takes into account errors in both coordinates. The algorithms reported by Numerical Recipes in C (45) were used, which return the chi-square probability as Q (a small value indicating a poor fit).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
PCrs/PCrr in the three repeated trials (trial effect: F = 0.103, P = NS). Intracellular pH was not significantly different in the three repetitions (trial effect, F = 0.97, P = NS); it increased significantly during the first minute of exercise (condition effect, F = 5.75, P < 0.02; difference contrast, F = 17.73, P < 0.01), returning to resting values during steady-state exercise (difference contrast, F = 0.141, P = NS). ADP concentration increased during steady-state exercise (exercise effect, F = 30.65, P < 0.001) and was not significantly different in the three repetitions (trial effect, F = 1.11, P = NS). Table 1 summarizes the average muscular PCr concentration at rest, together with the average mechanical power, the corresponding fraction of PCr hydrolyzed, the change in ADP concentration, and the pH at rest and during steady-state exercise.
|
PCrs/PCrr; average of the 3 trials) was linearly related to the corresponding actual mechanical power (in W) (n = 9, R = 0.841, P < 0.005). The increase in ADP concentration from rest to steady-state exercise ([ADPs]; µM) was linearly related to both the mechanical power (n = 9, R = 0.856, P < 0.005) and the change in PCr concentration (mM) from rest to steady-state exercise (n = 9, R = 0.940, P < 0.005). Figure 1 illustrates the PCr signal intensity (arbitrary units) as a function of time, as obtained by averaging all trials on all volunteers; each of the 27 (9 x 3) individual high-resolution data series was realigned to the appropriate starting time.
|
PCr; s) and mechanical work (n = 9, R = –0.024, P = NS) or the fraction of PCr hydrolyzed (n = 9, R = 0.137, P = NS). No significant relationship could be detected between the individual PCr and the corresponding absolute amount of PCr hydrolyzed (mM; n = 9, R = 0.256, P = NS; Fig. 2) or the increase in ADP concentration from rest to steady-state exercise (µM; n = 9, R = 0.053, P = NS; Fig. 3). In contrast, PCr increased linearly with the absolute muscular PCr concentration at rest ([PCrr]; mM) (Fig. 4; n = 9, R = 0.708, P < 0.05), the corresponding linear regression being described by:
 |
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
PCr and the resting PCr concentration is consistent with the view that, if PCr is indeed an energy reserve, the greater the store (i.e. the PCr concentration), the slower the time course by which the oxidative metabolism attains the new steady state. The same conclusion was drawn by the theoretical analysis carried out previously by our group on the energetics of muscular exercise at steady state (14). The relationship observed in the present work is consistent also with the linear model of respiratory control described about 20 years ago by Meyer (40). In this model, which was also verified on rats fed with a creatine analog (41), the creatine kinase reaction is treated as a chemical capacitor, the capacitance being proportional to total creatine content.
On the contrary, the present experimental results are in contrast with the conclusions drawn by Korzeniewski and Zoladz (34). These authors developed a dynamic model of oxidative phosphorylation in skeletal muscle by assembling quantitative knowledge about numerous enzymes, processes, metabolic blocks, and kinetic equations (33). Using this model, which can calculate over time the intermediate metabolite concentrations, an important role of the absolute amount of hydrolyzed PCr in determining PCr is predicted. However, the data used to build the model were obtained from unspecified muscle fiber types, disregarding possible differences among slow- and fast-twitch fibers. A recent experimental study seems to confirm the prediction of Korzeniewski and Zoladz (34), since it illustrates a relationship between the PCr kinetics and the amount of hydrolyzed PCr (22). However, this nonlinear relationship was obtained by pooling together both the two investigated conditions, without and with reduced blood flow (cuff stenosis), including the highest exercise intensities, for which a clear drift in the steady-state phosphocreatine concentration was observed, associated with a drop in intracellular pH.
The factors affecting the rate of the adjustment of the different energy sources during a sudden change in energy demand have been thoroughly studied (21, 26, 32). During the rest-to-work transients, under the assumption that no other energy sources are involved, changes in PCr concentration and O2 at the muscle level follow, as a first approximation, an exponential time course in a mirrorlike fashion (37, 38, 56), while PCr is considered to provide only an immediate temporal buffer for ATP (53).
A reduced oxygen supply due to a reduction in blood flow led to significantly increased PCr time constants (16, 49). On the contrary, enhanced convective O2 delivery, as well as enhanced peripheral O2 diffusion, did not influence the time constant of muscular O2 in isolated canine muscle in situ at low contraction intensity (21). Consistently, the time constant of the [PCr] kinetics was unaffected, in humans, by the inspired oxygen fraction, which, however, determined the overall percentage fall in PCr (26). These studies provide evidence that at least for moderate exercise intensity the limiting factor for the metabolic adjustments at exercise onset lies in a metabolic inertia within the working muscle (21). The activation of pyruvate dehydrogenase complex (PDC), a key reaction within the complex of oxidative pathways, however, did not affect the O2 kinetics (21). On the contrary, an acute creatine kinase inhibition determined a significant reduction of the mean response time of the fall in intracellular PO2, which reflects the O2 kinetics in Xenopus laevis isolated myocytes (32). Furthermore, the time constant of the oxygen uptake kinetics was shown to increase linearly with the total creatine pool in isolated rat skeletal muscle mitochondria (20). These authors conclude that the oxygen consumption kinetics of muscle subjected to a step change in energy demand are largely attributable to, and adequately explained by, the magnitude of the total creatine pool and the mitochondrial resistance to energy transfer.
Preexercise phosphocreatine concentration. In the present work, the assessment of absolute resting PCr concentration from the 31P-NMR data was based on the assumptions that 1) ATP concentration at rest is 5.72 mmol/kg wet wt (equivalent to 8.54 mM); 2) this value is associated only with the βATP peaks; and 3) to bring back the signal intensities to the completely relaxed condition, appropriate saturation factors can be calculated from the apparent spin-lattice relaxation times (T1) (19) reported by the literature (9).
The comparison with the βATP peak is a common practice, since the other ATP peaks are superimposed on other phosphagen spectral lines. The ATP concentration used in the present investigation (2) was obtained by means of biopsies taken directly from the gastrocnemius muscle and is very close to the most frequently adopted value, i.e., 5.5 mmol/kg wet wt (31). Moreover, the obtained resting muscular PCr concentrations are in the range of values reported by several other authors (31).
The same ATP concentration has been reported in both human fast- and slow-twitch fibers by some authors (30, 46), while others have reported a significantly higher ATP concentration in fast-twitch fibers (39, 42). A higher PCr content has been reported for fast-twitch muscle fibers compared with the slow-twitch ones (50, 53). This is consistent with the observation that the time constant for the rise in oxygen consumption in predominantly type II mouse muscle is appreciably longer than that for predominantly type I muscle (12). Since endurance-trained athletes show higher percentages of slow-twitch fibers compared with sedentary subjects (11, 27), differences in fiber type composition are claimed to explain the variability in intact muscle PCr level (47). Furthermore, despite similar average ATP and PCr concentrations compared with the present work, a linear relationship between resting [PCr] and [ATP] has been reported in a group of soccer players (47). Thus the use of the same ATP concentration as reference value for all our subjects does not take into account the possible physiological variability in the individual ATP concentration. If the variability in ATP concentration is estimated as the ratio between twice the SD and the average value, the data reported by Rico-Sanz et al. (47) allow us to calculate a variability of 40% (i.e., 1.2 x 2/6.0 x 100). As a consequence, in the present work, the resting PCr concentration could be either overestimated by 40% for the lowest figures, or underestimated by the same percentage for the higher values. An absolute quantification performed using a calibrated measurement (31), independent of muscle ATP, could yield a wider range of resting PCr concentrations, which could likely improve the relationship with the [PCr] time constant.
The relatively narrow range of the observed [PCr] values may be considered a limitation of the present study. An easy way to expand the range of preexercise PCr concentration (by decreasing them) would be to perform different intensities of "warm-up" exercises before the investigated constant-load exercise. Consistent with the results of the present work, a faster pulmonary O2 uptake kinetics has been observed after a "priming" exercise separated from the test exercise by some minutes of rest (13, 25). On the contrary, slower kinetics have been reported for continuous work-to-work transitions both within the moderate-intensity domain (7) and for heavy exercise (29). In this latter case, however, PCr did not reach a stable steady-state concentration, and a drop in intracellular pH was observed. Jones et al. (29) suggested that the "size principle" of motor unit recruitment could be responsible, at least in part, for their results; they neglected, however, the fact that lactate production at work onset is reflected by a slower PCr breakdown kinetics (28, 54). As a consequence, to yield comparable conditions (except for the preexercise [PCr]) by means of a prior exercise, the protocol must warrant a similar fiber recruitment pattern and lactate production of the same magnitude (if any). Consecutive exercises, in particular in the heavy domain, as those adopted by Brittain et al. (7) or by Jones et al. (29), are thus not adequate to highlight the dependence of the kinetics from the PCr concentration at the start of the effort.
The [PCr] time constant.
The time constant of [PCr] kinetics was determined initially on perfused dog gastrocnemius in situ by Piiper et al. (44), who found a time constant of 24 s. Subsequently, the time constant of PCr breakdown at the onset of exercise was obtained on humans by several authors (3, 4, 38, 48). Whereas Barstow et al. (3) and Binzoni et al. (4) reported values close to the present one (26.3 and 23.3 s), the data obtained by McCreary et al. (38) and by Rossiter et al. (48) are substantially greater (47 and 35 s). In these two latter cases, however, it is rather likely that some lactate production occurred. Indeed, during high-intensity exercise, muscle energy requirement is partially covered by lactate production, leading to the sparing of a certain amount of PCr at work onset, which is reflected by a slower PCr breakdown kinetics (28, 54). In the present investigation, the ergometer used allowed the subjects to control accurately the frequency and amplitude of the ankle flexions thanks to the LED bar fixed at eye level (17). This helped the volunteers to attain the target exercise work rate, which was set below the intracellular lactate threshold (38). In fact, the PCr concentration at steady state in our subjects was reached after 90 s of exercise and ranged from 70 to 90% of the resting value. Since the O2 at steady state is linearly related to the corresponding PCr concentration, it can be estimated from the O2/PCr relationship obtained in our previous studies under similar experimental conditions (14) that the highest steady-state O2 amounted to 92.7 ml·kg–1·min–1. This is 25% of the maximal O2 consumption values obtained by Blomstrand and coworkers (5) for the knee extensor muscles by means of a specific ergometer (which amounted to 350 ml·kg–1·min–1). In addition, intracellular pH, as determined from 31P chemical shift, 1) increased significantly during the first minute of exercise, 2) during steady-state exercise did not differ significantly from rest, and 3) did not change during the first minute of recovery (data not shown). These small pH changes are compatible with those attributable to the hydrolysis of PCr (28). Finally, as stated in METHODS, the splitting of the Pi peak was never observed, indicating that the pH was the same throughout the investigated muscle mass (52). Even if Blomstrand and coworkers (5) and the present data were obtained on different muscle groups, taken together the above observations support the view that exercise intensity was indeed in the moderate domain and that there was no substantial lactate production at the start of exercise.
Influence of ADP concentration on oxidative phosphorylation. The amount of PCr hydrolyzed at steady-state exercise is linearly related to power output (4, 18), thus likely yielding equivalent increases in ADP levels for the same exercise intensity. In turn, ADP concentration is commonly thought to be one of the main feedback signals controlling mitochondrial respiration (10, 55). This view is supported by the relationship between ADP concentration and PCr recovery rate observed at the end of isometric contractions of different force level in normal subjects (6). In contrast, however, in the present investigation no relationship was found between the estimated change in ADP concentration and the time constant of [PCr] kinetics at exercise onset (Fig. 3). Nevertheless, differential regulation of mitochondrial respiration clearly occurs among slow/oxidative and fast/glycolytic muscle fiber types (24). A simple Michaelis-Menten-type relationship between ADP content and oxygen consumption has been observed essentially in fast-twitch fibers (35). Mitochondrial respiration in these fibers is not stimulated by activation of the mitochondrial creatine and adenylate kinase reactions, and it participates essentially in the recovery of the PCr level (23). On the contrary, mitochondria within slow-twitch fibers exhibit an efficient coupling between ATP production and mitochondrial kinases located in the intermembrane space. This allows a fast matching between ATP demand and oxidative synthesis essentially by means of a direct channeling of adenine nucleotides between myosin-ATPase and mitochondria, which progressively overcomes the classical creatine kinase-catalyzed phosphotransfer system (23).
Conclusions. Despite the small range of resting [PCr] values and of the few subjects examined for the present work, the regression between PCr concentration at rest and the time constant of [PCr] kinetics is statistically significant. These data support the hypothesis that also in humans PCr concentration at the onset of a sudden increase in energy demand sets the time course of the oxidative metabolism in skeletal muscle during the transition, as already demonstrated in rats (41). PCr hydrolysis is one of the main controllers of oxidative phosphorylation. It is governed by creatine kinase in a process dependent on fiber types and metabolite signals arising from the contractile sites, which control the permeability of the outer mitochondrial membrane to ADP (23).
|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
O2 on-response and early blood lactate. J Appl Physiol 47: 761–769, 1979.O2) on-kinetics in skeletal muscles. Biochem J 379: 703–710, 2004.[CrossRef][Web of Science][Medline]This article has been cited by other articles:
|
|
 |
|
  A. M. Jones, D. P. Wilkerson, and J. Fulford Influence of dietary creatine supplementation on muscle phosphocreatine kinetics during knee-extensor exercise in humans Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2009; 296(4): R1078 - R1087. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vanhatalo and A. M. Jones Influence of prior sprint exercise on the parameters of the \#8216;all-out critical power test' in men Exp Physiol, February 1, 2009; 94(2): 255 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Kemp Interpreting the phosphocreatine time constant in aerobically exercising skeletal muscle J Appl Physiol, January 1, 2009; 106(1): 350 - 350. [Full Text] [PDF]
|
|
|
|
|
|
M. P. Francescato, V. Cettolo, and P. E. di Prampero Reply to Kemp J Appl Physiol, January 1, 2009; 106(1): 351 - 351. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
