Journal of Applied Physiology AJP: Gastrointestinal and Liver Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 91: 191-200, 2001;
8750-7587/01 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (13)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Houtman, C. J.
Right arrow Articles by Stegeman, D. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Houtman, C. J.
Right arrow Articles by Stegeman, D. F.
Vol. 91, Issue 1, 191-200, July 2001

pH heterogeneity in tibial anterior muscle during isometric activity studied by 31P-NMR spectroscopy

C. J. Houtman1, A. Heerschap2, M. J. Zwarts1, and D. F. Stegeman1

1 Department of Clinical Neurophysiology, Institute of Neurology and 2 Department of Radiodiagnostics, University Medical Centre, 6500 HB Nijmegen, The Netherlands


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The occurrence of pH heterogeneity in human tibial anterior muscle during sustained isometric exercise is demonstrated by applying 31P-nuclear magnetic resonance (NMR) spectroscopy in a study of seven healthy subjects. Exercise was performed at 30 and 60% of maximal voluntary contraction (MVC) until fatigue. The NMR spectra, as localized by a surface coil and improved by proton irradiation, were obtained at a high time resolution (16 s). They revealed the simultaneous presence of two pH pools during most experiments. Maximum difference in the two pH levels during exercise was 0.40 ± 0.07 (30% MVC, n = 7) and 0.41 ± 0.03 (60% MVC, n = 3). Complementary two-dimensional 31P spectroscopic imaging experiments in one subject supported the supposition that the distinct pH pools reflect the metabolic status of the main muscle fiber types. The relative size of the Pi peak in the spectrum attributed to the type II fiber pool increases with decreasing pH levels. This phenomenon is discussed in the context of the size principle stating that the smaller (type I) motor units are recruited first.

human; size principle; muscle fatigue; sustained isometric exercise; 31P nuclear magnetic resonance spectroscopy


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

HETEROGENEITY of pH in human muscle tissue during exercise and recovery, as monitored by 31P-nuclear magnetic resonance spectroscopy (31P-NMRS), has been addressed in several studies (1, 17-19, 24, 25, 29, 30). A shift of the frequency of the inorganic phosphate (Pi) signal, relative to that of phosphocreatine (PCr) in the 31P-NMR spectrum, corresponds to a change in the intracellular pH. The occurrence of a broadened, or even split, resonance of Pi during exercise is often ascribed to the different metabolic behavior of slow- and fast-twitch fibers involved in the exercise. Slow-twitch (type I) fibers mainly use aerobic energy sources and therefore produce few protons, resulting in a relatively unaffected tissue pH. Fast-twitch (type II) fibers are better equipped for anaerobic glycolysis. This is accompanied by a more extensive production of protons. If this proton production is not balanced by the removal and/or buffering of protons, the pH will decline, resulting in a shift of the Pi peak.

Fiber type-related pH heterogeneity has been reported for wrist flexion muscles (17-19), for calf muscles (1, 24, 25), and for the biceps femoris (29, 30). In these experiments, cyclic concentric exercises were used. Mostly a graded exercise at low frequency (<FR><NU>1</NU><DE>6</DE></FR>-1/4 Hz) was chosen (17-19, 25). This type of exercise ensures the supply of oxygen to muscles contributing to the exercise. Also, cyclic concentric exercise at highest frequency possible (2-3 Hz) was used to guarantee the recruitment of all motor units, the functional building blocks of a muscle (24, 25). Vandenborne et al. (25) applied a NMR localization method to distinguish between the contributions of the three different calf muscles to the sampled signal.

Slow-twitch (type I) and fast-twitch (type II) motor units differ in size. According to the size principle, as ascribed by Henneman (8), motor units are recruited from small to large. Combining the size principle with the development of Pi peaks is an alternative approach to study pH heterogeneity. Because the size principle is disputed in nonisometric exercises (e.g., Ref. 9), we studied pH heterogeneity during sustained isometric exercise at two different loads [30 and 60% maximal voluntary capacity (MVC)]. At first, during a 30% MVC level, only the relative small slow-twitch (type I) motor units are recruited. In the course of the fatiguing contraction at 30% MVC, new motor units, including the relative large fast-twitch (type II), are indispensable. This would induce a changeover to the simultaneous activity of both fiber types. At 60% MVC, the muscle is used under ischemic conditions (14, 22). At this exercise level, motor units of both types are expected to be active from the beginning (5). This leads to the hypothesis that 1) pH heterogeneity will emerge in the course of an exercise at 30% MVC and 2) pH heterogeneity will emerge soon after the start of an exercise at 60% MVC.

We selected the tibialis anterior muscle (TA) for this study because it has no synergists for its performance, excluding the possibility that the measured force is produced by more than one muscle. The measurements only use coil localization to visualize as many as possible temporal details (temporal resolution of 16 s). In one of the seven subjects, the coil profile and the spatial distribution of the Pi peaks during exercise were investigated.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Subjects. Seven healthy men and women participated in the study, aged 22-48 (Table 1). All subjects were regularly engaged in low to moderate aerobic exercise. They were informed of the purpose of the experiments and gave their written consent. The protocol was approved by the local ethics committee of the Faculty of Medical Sciences of the Nijmegen University.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Subject and force parameters

Exercise. Subjects took a supine position on the investigation table with the left leg slightly bent and supported with vacuum pillows (Fig. 1A). For the two-dimensional phase-encoded 31P spectroscopic imaging (2D 31P SI) measurements, the exercising leg was straightened to enable its horizontal alignment with respect to the static magnetic (B0) field. The angle of the left lower leg and the left foot was always 90°, and the foot was fixed with straps in a pedal. Subjects were supplied with visual feedback of the force and were verbally encouraged during exercise. All subjects performed both exercise protocols, separated by at least 2 days.


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1.   A: experimental setup. The subject lies in a supine position on the table. The U-shaped coil is placed over the tibialis anterior (TA) of the left leg. The left foot is strapped in the ergometer. The left leg is slightly bent and supported by vacuum pillows. The angle between foot and lower leg is 90°. The subject gets visual feedback of the force and is encouraged verbally during exercise. B: example of a 31P nuclear magnetic resonance (NMR) spectroscopy (31P-NMRS) spectrum (2 acquisitions, repetition time = 7 s) measured 68 s after the start of a 60% maximal voluntary contraction (MVC) exercise (subject 6, see also Fig. 2). Compared with a NMR spectrum obtained at rest, the phosphocreatine (PCr) peak is decreased and the inorganic phosphate (Pi) peak is increased and doubled. pH values are derived from the frequency of the Pi peaks relative to that of the PCr peak. The area of a peak is a measure for the tissue level of the corresponding molecule. PME, phosphomonoesters; PDE, phosphodiesterase; pHhigh and pHlow, values derived from, respectively, left (low-field) and right (high-field) Pi peaks.

The ergometer was home built, was specially designed for use in the magnet, and was applicable for both dorsal and plantar ankle flexion. The force signal was digitally stored at a sample rate of 100 Hz. To enable synchronization afterward, a trigger signal of the NMR system was registered on the force system. The NMR data points were assigned to the midpoint of the two 31P-NMR excitation pulses, applied for each measurement.

The exercise consisted of an isometric plantar ankle flexion until fatigue at 30 or 60% MVC. After offset correction, two MVC measurements were done, from which the maximum was used (Table 1). The subject was permitted to deviate ±3% at 30% MVC or ±5% at 60% MVC. If he or she could no longer meet this condition, despite encouragement, exercise was stopped. MVC measurements and exercise were separated by at least 30 min. 31P-NMR data were collected during rest (3 min), exercise (Table 1), and recovery (15 min).

Nuclear magnetic resonance. All experiments were performed on a 1.5-T whole body system (Magnetom SP, Siemens Medical Systems, Erlangen, Germany). The surface coil was home built and specially designed to detect signals from the TA. It consisted of two concentric loops: 25 cm × 81/2 cm (tuned to the 1H frequency) and 20 cm × 31/2 cm (tuned to the 31P frequency). It was carefully placed over the TA and fixed with tape. Before 31P-NMRS data collection, a series of five transversal 1H-NMR images [gradient-recalled echo: echo time = 6 ms, repetition time (TR) = 70 ms, thickness = 10 mm, distance factor = 0.4] were made to validate the correct placement of the surface coil. In case of the 2D 31P SI measurements, this series was extended to 15 transversal 1H-NMR images divided over the length of the coil to control the horizontal position of the lower leg and consequently that of the TA. After the last 2D 31P SI measurement, this series of 15 1H-NMR images was repeated to confirm an unchanged leg position.

The homogeneity of the B0 field was adjusted by using the proton signal from water, resulting in a peak width at half-height of <= 0.5 ppm. To overcome inhomogeneity of the B1 field, an amplitude-modulated adiabatic 90° pulse (sincos), with a length of 2.56 ms, was used for excitation. The spectral excitation of this pulse was constant over a frequency range of 1,000 Hz (~39 ppm), which is sufficient to map all resonances present in skeletal 31P-NMR spectra.

For optimal SNR per unit time, we used a TR of 7 s, according to Ernst and Anderson (6) (TR = 1.25 T1, where T1 is spin-lattice relation time) and to Thomsen et al. (23) [muscle T1 (PCr, Pi, ATP) <=  5.5 s]. Two acquisitions were averaged per measurement. Including data storage, the time resolution became 16 s. The free induction decay (FID) was low-pass filtered at 5 kHz and sampled during 512 ms at a sample frequency of 4 kHz. During data collection, high-level (10 W) WALTZ4 proton irradiation was applied to decouple the 31P-1H spin coupling (15). During the remaining time, low-level (0.6 W) WALTZ4 proton irradiation was applied to amplify the 31P-NMR signal by the nuclear Overhauser effect (2).

For localization of the 31P-NMR signal, 2D 31P SI without slice selection and with a matrix size of 8 × 8 was applied in the transverse direction. Each volume of interest had a nominal resolution of 1.5 × 1.5 cm2. In the third dimension, the view of the surface coil dictated the localization (~20 cm). Acquisition time was 7.57 min (~8 × 8 × 7 s), taking one acquisition for every phase-encoding step.

Data analysis. The data-analyzing method VARPRO (27) was used as fitting procedure. The first three data points of each FID were not included in the Fourier transformation to avoid the broad baseline component arising from the tibia bone. Starting values for the peak position and line width of PCr and Pi were given. The peaks were assumed to have a Lorentzian line shape. If the Pi resonance split into two resonances, equal line widths were not assumed because the origin of this phenomenon is part of the study. To avoid improper use of prior knowledge, a broadened Pi peak was analyzed in two ways: as one peak and as two independent peaks. In the evaluation of the results, the lower bound of the theoretical statistical errors, the Cramer-Rao (CR) lower bound, was used. The result with the lowest CR lower bound in the calculated parameters (peak area, line width, and frequency) was accepted as the best fit. Sometimes, both results came up with CR lower bounds of more than 60% of the parameter values. Then the Pi peak was excluded from the fit procedure.

Intracellular pH was calculated from the chemical shift of Pi based on the equation pH = 6.75 + log(delta  - 3.26)/log(5.75 - delta ), where delta  equals the chemical shift of the Pi peak (in ppm), relative to PCr. The curves of the pH were fitted by piecewise linear regression. Continuity was assumed, except for the transition of one pH to two pHs and vice versa. The distribution of data points over the successive regression lines was optimized by minimizing the sum of the residuals of the lines. This method is described by Vieth (28) for the combination of two regression lines. We extended this method to an arbitrary number of line pieces.

The 2D 31P SIs were analyzed with LUISE, a data-analyzing program supplied by Siemens. No zero filling, filtering, or offset correction was used in the k space (the spacial frequency space). The shift of the matrix grid was equal for both 2D 31P SIs (rest and exercise). The first 2D 31P SI was depicted on 1H-NMR images measured before exercise, and the last 2D 31P SI was depicted on 1H-NMR images measured after exercise.

Preprocessing of the spectra shown in this paper (Figs. 1B, 2B, 3B, and 4B) was as follows. Before Fourier transformation, the first three data points of the FID were skipped, an asymmetric Gauss window (center 25 ms, width 100 ms) was applied, and zero filling was performed to 4,096 data points.


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2.   Results of pH analysis and 31P-NMR spectra of subject 6, performing a 60% MVC exercise. A: pH. Start (at 0 s) and end of exercise (at 96 s) are marked with vertical lines. One (×) or two (open circle ) Pi peaks were analyzed. Lines through data points are fitted as described in the text. Horizontal bar, just over the time scale, indicates the time interval corresponding to the spectral selection shown in B for the region with Pi peaks. B: Pi peak region of sequential spectra. Numbers below the spectra represent time (in s). Note the gradual appearance of 2 Pi peaks during exercise leading to 2 different pH levels (A and B), the decrease of the 2 pH levels and simultaneous increase of the relative size of the right Pi peak (A and B), and the relative quick recovery (= disappearance) of the left Pi peak (B).



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 3.   Results of pH analysis and 31P-NMR spectra of subject 6, performing a 30% MVC exercise. A: pH. Start (at 0 s) and end of exercise (at 580 s) are marked with vertical lines. One (×) or two (open circle ) Pi peaks were analyzed. Lines through data points are fitted as described in the text. Horizontal bar, just over the time scale, indicates the time interval corresponding to the spectral selection shown in B for the region with Pi peaks. B: Pi peak region of sequential spectra. Numbers below the spectra represent time (in s). Note the appearance of 2 different pH levels during exercise corresponding to 2 Pi peaks (A and B), the accelerated decrease of pHhigh and the increase of the relative size of the right Pi peak between 415 and 510 s (A and B), and the merging of both Pi peaks between t = 510 and 573 s, leading to 1 low pH level (A and B).



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 4.   Results of pH analysis and 31P-NMR spectra of subject 3, performing a 30% MVC exercise. The same exercise was performed by this subject to obtain the 2D 31P SIs (Fig. 5). A: pH. Start (at 0 s) and end of exercise (at 660 s) are marked with vertical lines. One (×) or two (open circle ) Pi peaks were analyzed. Lines through the data points are fitted as described in the text. Horizontal bar, just over the time scale, indicates the last 8 min of exercise and corresponds with the spectral selection shown in B for the region with Pi peaks. B: Pi peak region of sequential spectra. Numbers below the spectra represent time (in s). Every second spectrum is shown for clarity. Note the visibility of 2 stable Pi peaks leading to 2 constant pH levels during the last 61/2 min of the exercise.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Exercise. MVC ranged between 207 and 336 N. The duration of the exercises varied between 80 and 167 s for 60% MVC and between 348 and 660 s for 30% MVC (Table 1). Although toe extension muscles did not contribute to the delivered force, subjects tended to extend their toes when it became hard to maintain the desired force. This means that the extensor digitorum longus (EDL) and the extensor hallucis longus were also activated and could potentially contribute to the 31P-NMR signal. However, the extensor hallucis longus is situated distal to the TA and therefore did not contribute to the 31P-NMR signal. The influence of the EDL was studied with the 2D 31P SI results (see Two-dimensional phase-encoded 31P spectroscopic imaging below).

31P-NMR spectroscopy with only surface coil localization, 60% MVC. An example of a 31P-NMR spectrum of the TA obtained during a 60% MVC exercise is shown in Fig. 1B. A doubling of the Pi peak is clearly visible. Such a doubling of the Pi NMR signal was evident in three subjects. The other four subjects showed a clear broadening of the line width of the Pi peak during exercise and early recovery. In three of these subjects, a second Pi peak was visible in the first spectrum after the end of exercise.

The data of three subjects in which doubled Pi peaks were analyzed are summarized in Table 2. The values were derived from the fitted line pieces through the data points as described. The pH levels derived from the left (low field) and right (high field) Pi peaks are designated as pHhigh and pHlow, respectively. The slopes of pHhigh and pHlow, the maximum difference between pHhigh and pHlow during exercise, and the pH levels at the end of exercise are given. For pHhigh and pHlow, the mean ± SD slopes are (-4.2 ± 4.4) × 10-3 and (-8.4 ± 5.0) × 10-3 pH units/s, respectively, and the mean of the maximum difference in pH levels during exercise is 0.41 ± 0.03 pH units. The mean ± SD values at the end of exercise are, respectively, 6.69 ± 0.26 and 6.28 ± 0.24. No slopes are given for subject 5 because two Pi peaks could only be analyzed just before the end of exercise and during recovery.

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   60% MVC: pH slopes, maximum differences, and end-of-exercise values

Figure 2A shows the results for the pH analysis before, during, and after an exercise at 60% MVC performed by subject 6. Figure 2B shows stack plots of spectra for the Pi peak region for the same period. During exercise, two Pi peaks gradually appear. With lowering of the pH levels, the relative size of the right (high-field) Pi peak increases (Fig. 2, A and B). This Pi peak corresponds to pHlow. The subsequent recovery of the left (low-field) Pi peak is remarkably quicker than the recovery of the right Pi peak (Fig. 2B).

31P-NMR spectroscopy with only surface coil localization, 30% MVC. All seven subjects showed doubled Pi peaks during the 30% MVC exercise. The results of the pH analysis for all subjects are summarized in Table 3. The values were derived from fitted line pieces through the data points as described. The slopes of pHhigh and pHlow, the maximum difference in pHhigh and pHlow during exercise, and the pH levels at the end of exercise are given for each subject. If pHhigh or pHlow consisted of two line pieces, the slope from the first line piece is given. The mean ± SD slopes for pHhigh and pHlow are, respectively, (-0.7 ± 0.6) and (-1.4 ± 1.1) × 10-3 pH units/s, and the mean of the maximum difference in pH levels is 0.40 ± 0.07 pH units. The mean pH values (± SD) at the end of exercise are 6.90 ± 0.16 and 6.54 ± 0.11 for pHhigh and pHlow, respectively, and 6.34 ± 0.09 if only one Pi peak is left at the end of exercise.

                              
View this table:
[in this window]
[in a new window]
  		Table 3.   30% MVC: pH slopes, pH differences, and end-of-exercise values

Figure 3A shows the results for the pH analyses during rest, exercise, and recovery at 30% MVC as analyzed from the data of subject 6. Figure 3B shows the Pi peak region of the spectra, corresponding to the most remarkable changes in peak area distribution over both Pi peaks and to the course of the peak after the end of exercise. pH heterogeneity appears during a large part of the exercise but is absent in the first and last spectra during exercise (Fig. 3A). The relative size of the right Pi peak (corresponding to pHlow) increases as the highest pH level decreases (Fig. 3, A and B, between 415 and 510 s). pHhigh declines faster than pHlow so that the two Pi peaks merge into one large peak, corresponding to one low pH level (Fig. 3, A and B, starting after 510 s).

Figure 4A shows the results of pH analyses of rest, exercise, and recovery at 30% MVC, performed by subject 3. Figure 4B shows the Pi region of the spectra during the final 8 min of exercise. For clarity, every second available spectrum is skipped in this presentation. Again two Pi peaks appear. In this subject, these two peaks last until the end of exercise and, in addition, the peaks and the calculated pH values stabilize during a long period compared with the other subjects. This behavior was the reason to select this subject for the spectral imaging session presented below, which has a much lower time resolution (8 min).

Two-dimensional phase-encoded 31P spectroscopic imaging. The 2D 31P SI measurement required an acquisition time of 8 min. Too much shifting of the Pi peaks during these 8 min would cause a spread and flattening of the Pi peaks. Therefore, the 2D 31P SI measurements were performed on subject 3, who showed two stable pH levels during 30% MVC in the measurement with only surface coil localization (Fig. 4A). Two 2D 31P SI measurements were acquired in one session. The first was measured during rest, immediately preceding the exercise. The signal intensities reflect the sensitivity profile of the 31P coil (Fig. 5A). Four of the 64 voxels contained almost all signal intensity. These voxels cover the TA (31/2 voxel) and part of the EDL (1/2 voxel). Because the signal intensity is higher in the most ventral voxels, more than <FR><NU>7</NU><DE>8</DE></FR> of the total signal intensity originates from the TA. The signal intensity of PCr is highest in the medial ventral (left upper) voxel.


View larger version (73K):
[in this window]
[in a new window]
  		Fig. 5.   A: 31P 2-dimensional phase-encoded spectroscopic imaging (2D 31P SI) of the left lower leg of subject 3 during rest, preceding a 30% MVC exercise. In each voxel, that part of the spectrum containing Pi and PCr is shown. TA is outlined with a broken line. Note that 4 of the 64 voxels contain most of the 31P-NMR signal (outlined with a solid line). These 4 voxels cover mainly the TA and partly the extensor digitorum longus. The highest PCr peak occurs in the medial ventral (left upper) side of these 4 voxels. Large white area on left of the 1H NMR image is the inner side of the tibia bone. Small white spot is the inner side of the fibula bone. Long white area on right is an artifact of the 1H coil (high-flux area). B: 2D 31P SI of the same left lower leg as in A measured during the last 8 min of the fatiguing 30% MVC exercise after measuring A. TA is outlined with a broken line. In each voxel, the same part of the spectrum as in A is shown; however, the y-axis is rescaled. Note that the Pi peak is increased in 5 voxels (outlined with a solid line). The PCr peak is highest in the lateral ventral (right upper) voxel, and the Pi peak is broadened in the medial (left) voxels covering the TA. C: enlargements of 5 voxels covering the TA (outlined with a solid line in B). Dashed lines originate from the 2D 31P SI measured during rest; solid lines are from the 2D 31P SI measured during 30% MVC exercise. Scaling of the x- and y-axes is the same for all. Numbers at the Pi peaks are the corresponding pH levels. Note the doubled Pi peaks in the medial (left) voxels. Furthermore, note the pH gradient from lateral (right) side to medial (left) side of the TA.

The second 2D 31P SI (Fig. 5B) was measured during the last 8 min of an exercise at 30% MVC. In this 2D 31P SI, the highest intensity of PCr is now observed for the lateral ventral (right upper) voxel. The medial ventral (left upper) voxel, in which PCr is most declined, shows a doubled Pi peak. Five voxels clearly reveal increased Pi peaks. Fig. 5C shows enlargements of the corresponding spectra obtained during rest and during exercise. Obviously, a pH gradient from lateral (pH level of 7.0 in the lateral voxels) to medial (lowest pH levels of 6.3 and 6.7 in the most medial voxel) is present. A summation of all five spectra (not shown) during exercise results in two Pi peaks at pH 7.0 and 6.5, which is similar to the pH levels measured with only surface coil localization in the same subject (Fig. 4A). As expected, there is a good reproducibility within one subject (16).


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

pH heterogeneity within the TA. This study reveals pH heterogeneity in the TA during sustained isometric exercise at both sides of the anaerobic threshold (30 vs. 60% MVC). This pH heterogeneity is not caused by the contributions of different muscles in the leg because the only other muscle within the field of view of the coil (the EDL), which was sometimes activated by the subjects, contributes to less than <FR><NU>1</NU><DE>8</DE></FR> of the total 31P-NMR signal.

The combination of the two 2D 31P SIs, measured during rest and during exercise at 30% MVC, respectively, shows that during exercise PCr is more decreased at the medial side than at the lateral side of the TA. In this medial part of the muscle, a doubling of the Pi peak occurs during the last 8 min of exercise. Whether the resulting two Pi peaks were present simultaneously during the 8 min acquisition time or one after the other cannot be concluded from these data. The 31P-NMRS data, acquired with only surface coil localization, however, show that the two pH levels occur simultaneously during the last 61/2 min of the exercise, in this subject. Therefore, the doubled Pi peak must be ascribed to two pH compartments in the medial part of the TA. The size of these compartments has to be smaller than the nominal voxel size of 1.5 × 1.5 × 20 cm3. As argued by Vandenborne et al. (25), differences between intra- and extracellular pH, or intracellular pH differences, are unlikely to be the origin of the distinct Pi peaks. Remaining sources are difference in activation or a partial limitation of the oxygen supply, both within a voxel size of 1.5 × 1.5 × 20 cm3. We have no anatomic indications for two distinct areas of blood supply within this voxel. Neither do we have anatomic indications for two distinct areas of muscle activation otherwise than on the level of muscle fiber types. Therefore, the following discussion proceeds from the most probable interpretation linking the left (pHhigh) peak to the slow-twitch (type I) motor units in the muscle and the right (pHlow) one to the fast-twitch (type II) motor units.

pH gradient within the TA. The fact that the pH shows a declining gradient from lateral to medial within the TA is possibly caused by differences in blood supply (20). The compliance of the tibia bone is smaller than the compliance of the membrane surrounding the anterior compartment. During a sustained isometric exercise, this could possibly lead to a gradient in intramuscular pressure. If it is assumed that the mean arterial blood pressure and the local metabolic vasodilatation are homogeneously distributed over the whole TA, then blood flow is lowest closest to the tibia bone.

Comparing 60% and 30% MVC. Directly after the start of exercise, a small transient increase of pH is visible in all experiments with only coil localization (also in Fig. 2A, 3A and 4A). This reflects the proton consumption of PCr, which initially is the main energy source in these experiments (4).

In both exercise levels, the decrease of the pHlow is roughly twice as fast as that of pHhigh (Tables 2 and 3). These pH slopes can be used as a measure of glycolytic activity (13). The factor of two corresponds well with the difference of phosphofructokinase activity in fiber types I and II, also a measure of glycolytic activity (7).

Recovery of both Pi peaks could only be followed in two cases because the Pi peaks tend to disappear quickly in the noise after exercise. In these two cases (both after 60% MVC), the Pi peak ascribed to the type I fibers disappears much faster than the one ascribed to the type II fibers. This can be explained by the pH dependency of the recovery of Pi (10) and the higher oxidative capacity of the type I fibers (21) and is in agreement with the observations of others (1, 18, 19, 24, 29, 30).

The pH heterogeneity is more difficult to measure during 60% MVC than during 30% MVC exercise, although the pH differences at the end of exercise are similar (Tables 2 and 3). An obvious explanation is that fewer data points are available for a 60% MVC exercise. Moreover, a faster shifting of both Pi peaks limits proper analysis.

Distribution of peak areas of Pi. A number of issues are relevant in relating the distribution of peak areas to the recruitment of motor units.

1) Johnson et al. (12) and Polgar et al. (20) investigated fiber type distribution and sizes in young, healthy, male subjects. In the superficial region of the TA, they found a mean percentage of type I fibers of 73% and a mean cross section of type I and type II fibers of 2.4 × 10-9 m2 and 3.4 × 10-9 m2, respectively. Therefore, it is assumed that 64% of the TA cross section consists of type I fibers.

2) With increasing central neural drive, motor units are recruited according to the size principle of Henneman (8), whereby smaller type I motor units become active before larger type II motor units.

3) Vandenborne et al. (26) showed that, in contrast to results of experiments on animals, human muscle with predominantly type I fibers had the same PCr content as muscle with predominantly type II fibers. Thus, it is assumed that type I fibers and type II fibers have the same PCr content in rest.

4) PCr consumption might be higher in type II fibers because PCr also acts as a proton buffer (31).

5) If local ischemia arises, a part of the Pi is trapped by mitochondria and becomes invisible for 31P-NMRS (11). Already recruited motor units will especially "lose" some of their Pi signal. From the size principle, it can be predicted that the peak area of type I fibers is most affected.

In the theoretical case that all motor units will be equally recruited during the whole isometric exercise, around or somewhat less than 64% of the PCr consumption is attributed to type I motor units (see above, issues 1, 3, and 4). If two Pi peaks can be discerned, one expects a somewhat larger left peak. If ischemia occurs, a decrease of Pi in all active fibers is expected, leading to a proportional decrease of both Pi peaks (issue 5). The uptake of Pi in the mitochondria is possibly larger in type I fibers because of a larger mitochondrial density in that fiber type. In that case, a larger decrease of the left Pi peak is expected.

During sustained isometric exercise at 30 or 60% MVC, the prediction is more complicated because only a part of the motor units is recruited at the start of exercise. Fatigue and/or a deterioration of blood supply forces the system to recruit more motor units. Even at sustained isometric exercise at 30% MVC, deterioration of blood supply may occur, caused by an increase of intramuscular pressure (3). Two Pi peaks appear as soon as most type I units and a part of the type II units are recruited (issue 2) and intracellular protons accumulate. A relatively larger left Pi peak and a smaller right Pi peak are expected. The difference between the peak areas depends on the number of type II motor units that is necessary at that time and on the level of ischemia (issue 5). The more type II units involved, the larger the right Pi peak; the more ischemia, the smaller the left Pi peak.

Our results (Figs. 2, 3, and 5C) nicely illustrate that an increase of the relative size of the right Pi peak coincides with a notable decrease of pHhigh and pHlow. This is particularly clear during the last part of the exercise at 30% MVC, where (in most of our subjects) a striking increase of the right Pi peak occurs in coincidence with an accelerated decline of pHhigh. All together, this points to a deterioration of blood supply combined with anaerobic glycolysis. This fits well with the mechanisms sketched above: the lower the blood flow, the less efficient type I fiber contributions and thus the more type II motor units have to be recruited, according to the size principle, to maintain the expected force.

Yoshida and Watari (29) showed the development of Pi peaks in one subject during a progressive dynamic exercise at an intermediate frequency (0.8 Hz) of the biceps femoris. Their results (Fig. 5) show that a substantial increase of the low pH peak area coincides with a shift of both Pi peaks toward the PCr peak (after 3-31/2 min of exercise). Although their exercise is nonisometric, and thus the size principle is disputed (e.g., Ref. 9), the development of Pi peak areas and positions suggests an orderly recruitment of motor units.

In conclusion, this study reveals pH heterogeneity in the TA during sustained isometric exercise below and above anaerobic threshold. The fact that the spatial pH distribution shows a declining gradient from lateral to medial within the TA (one subject) is attributed to intramuscular differences in blood supply. The pH dependency of relative sizes of both Pi peaks, in the temporally and spatially characterized 31P-NMRS data, can be explained by the size principle of motor unit type-related orderly recruitment of motor units.


    ACKNOWLEDGEMENTS

The indispensable technical support from H. J. van den Boogert, A. J. van den Bergh, and J. P. van Dijk is particularly acknowledged.


    FOOTNOTES

Address for reprint requests and other correspondence: C. J. Houtman, Dept. of Clinical Neurophysiology, Institute of Neurology, 314, Univ. Medical Centre, Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands (E-mail: c.houtman{at}czzoknf.azn.nl).

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.

Received 9 November 1999; accepted in final form 15 February 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Achten, E, van Cauteren M, Willem R, Luypaert R, Malaisse W, van Bosch G, Delanghe G, de Meirleir K, and Osteaux M. 31P-NMR spectroscopy and the metabolic properties of different muscle fibers. J Appl Physiol 68: 644-649, 1990[Abstract/Free Full Text].

2.   Brown, TR, Stoyanova R, Greenberg T, Srinivasan R, and Murphy-Boesch J. NOE enhancements and T1 relaxation times of phosphorylated metabolites in human calf muscle at 1.5 Tesla. Magn Reson Med 33: 417-421, 1995[ISI][Medline].

3.   Crenshaw, AG, Karlsson S, Gerdle B, and Friden J. Differential responses in intramuscular pressure and EMG fatigue indicators during low- vs. high-level isometric contractions to fatigue. Acta Physiol Scand 160: 353-361, 1997[ISI][Medline].

4.   DeGroot, M, Massie BM, Boska M, Gober J, Miller RG, and Weiner MW. Dissociation of [H+] from fatigue in human muscle detected by high time resolution 31P-NMR. Muscle Nerve 16: 91-98, 1993[ISI][Medline].

5.   Enoka, RM. Morphological features and activation patterns of motor units. J Clin Neurophysiol 12: 538-559, 1995[ISI][Medline].

6.   Ernst, RR, and Anderson WA. Application of Fourier transform spectroscopy to magnetic resonance. Rev Sci Instrum 34: 449-455, 1966.

7.   Essen, B, Jansson E, Henriksson J, Taylor AW, and Saltin B. Metabolic characteristics of fiber types in human skeletal muscle. Acta Physiol Scand 95: 153-165, 1975[ISI][Medline].

8.   Henneman, E. Recruitment of motor units: the size principle. In: Motor Unit Types, Recruitment and Plasticity in Health and Disease, edited by Desmedt JE.. Basel: Karger, 1981, p. 26-60.

9.   Howell, JN, Fuglevand AJ, Walsch ML, and Bigland-Ritchie B. Motor unit activity during isometric and concentric-eccentric contractions of the human first dorsal interosseus muscle. J Neurophysiol 74: 901-904, 1995[Abstract/Free Full Text].

10.   Iotii, S, Lodi R, Frassineti C, Zaniol P, and Barbiroli B. In vivo assessment of mitochondrial functionality in human gastrocnemius muscle by 31P MRS. NMR Biomed 6: 248-253, 1993[ISI][Medline].

11.   Iotti, S, Lodi R, Gottardi G, Zaniol P, and Barbiroli B. Inorganic phosphate is transported into mitochondria in the absence of ATP biosynthesis: an in vivo 31P-NMR study in the human skeletal muscle. Biochem Biophys Res Commun 225: 191-194, 1996[ISI][Medline].

12.   Johnson, MA, Polgar J, Weightman D, and Appleton D. Data on the distribution of fiber types in thirty-six human muscles. An autopsy study. J Neurol Sci 18: 111-129, 1973[ISI][Medline].

13.   Kemp, GJ, and Radda GK. Quantitative interpretation of bioenergetic data from 31P and 1H magnetic resonance spectroscopic studies of skeletal muscle: an analytical review. Magn Reson Quart 10: 43-63, 1994.

14.   Kent-Braun, JA, Miller RG, and Weiner MW. Phases of metabolism during progressive exercise to fatigue in human skeletal muscle. J Appl Physiol 75: 573-580, 1993[Abstract/Free Full Text].

15.   Luyten, PR, Bruntink G, Sloff FM, Vermeulen JWAH, van der Heijden JI, den Hollander JA, and Heerschap A. Broadband proton decoupling in human 31P-NMR spectroscopy. NMR Biomed 1: 177-179, 1989[Medline].

16.   Miller, RG, Carson PJ, Moussavi RS, Green A, Baker A, Boska MD, and Weiner MW. Factors which influence alterations of phosphates and pH in exercising human skeletal muscle: measurement error, reproducibility, and effects of fasting, carbohydrate loading, and metabolic acidosis. Muscle Nerve 18: 60-67, 1995[ISI][Medline].

17.   Mizuno, M, Horn A, Secher NH, and Quistorff B. Exercise-induced 31P-NMR metabolic response on human wrist flexor muscles during partial neuromuscular blockade. Am J Physiol Regulatory Integrative Comp Physiol 267: R408-R414, 1994[Abstract/Free Full Text].

18.   Mizuno, M, Secher NH, and Quistroff B. 31P-NMR spectroscopy, rsEMG and histochemical fiber types of human wrist flexor muscles. J Appl Physiol 76: 531-538, 1994[Abstract/Free Full Text].

19.   Park, JH, Brown RL, Park CR, McCully K, Cohn M, Haselgrove J, and Chance B. Functional pools of oxidative and glycolytic fibers in human muscle observed by 31P magnetic resonance spectroscopy during exercise. Proc Natl Acad Sci USA 84: 8976-8980, 1987[Abstract/Free Full Text].

20.   Polgar, J, Johnson MA, Weightman D, and Appleton D. Data on fiber size in thirty-six human muscles. An autopsy study. J Neurol Sci 19: 307-318, 1973[ISI][Medline].

21.   Quistorff, B, Johansen L, and Sahlin K. Absence of phosphocreatine resynthesis in human calf muscle during ischemic recovery. Biochem J 291: 681-686, 1992.

22.   Sjøgaard, G, Savard G, and Juel C. Muscle blood flow during isometric activity and its relation to muscle fatigue. Eur J Appl Physiol 57: 327-335, 1988.

23.   Thomsen, C, Jensen KE, and Henriksen O. In vivo measurements of T1 relaxation times of 31P-metabolites in human skeletal muscle. Magn Reson Imaging 7: 231-234, 1989[ISI][Medline].

24.   Vandenborne, K, McCully K, Kakihira H, Prammer M, Bolinger L, Detre JA, de Meirleir K, Walter G, Chance B, and Leigh JS. Metabolic heterogeneity in human calf muscle during maximal exercise. Proc Natl Acad Sci USA 88: 5714-5718, 1991[Abstract/Free Full Text].

25.   Vandenborne, K, Walter G, Leigh JS, and Goelman G. pH heterogeneity during exercise in localized spectra from single human muscles. Am J Physiol Cell Physiol 265: C1332-C1339, 1993[Abstract/Free Full Text].

26.   Vandenborne, K, Walter G, Ploutz-Snyder L, Staron R, Fry A, de Meirleir K, Dudley GA, and Leigh JS. Energy-rich phosphate in slow and fast human skeletal muscle. Am J Physiol Cell Physiol 268: C869-C876, 1995[Abstract/Free Full Text].

27.   Van der Veen, JWC, de Beer R, Luyten PR, and van Ormondt D. Accurate quantification of in vivo 31P-NMR signals using the variable projection method and prior knowledge. Magn Reson Med 6: 92-98, 1988[ISI][Medline].

28.   Vieth, E. Fitting piecewise linear regression functions to biological responses. J Appl Physiol 67: 390-396, 1989[Abstract/Free Full Text].

29.   Yoshida, T, and Watari H. Exercise-induced splitting of the inorganic phosphate peak: investigation by time-resolved 31P-nuclear magnetic resonance spectroscopy. Eur J Appl Physiol 69: 465-473, 1994.

30.   Yoshida, T, Watari H, and Tagawa K. Effects of active and passive recoveries on splitting of the inorganic phosphate peak determined by 31P-nuclear magnetic resonance spectroscopy. NMR Biomed 9: 13-19, 1996[ISI][Medline].

31.   Zandt, HJA in 't, Oerlemans F, Wieringa B, and Heerschap A. Effects of ischemia on skeletal muscle energy metabolism in mice lacking creatine kinase monitored by in vivo 31P nuclear magnetic resonance spectroscopy. NMR Biomed 12: 327-334, 1999[ISI][Medline].

J APPL PHYSIOL 91(1):191-200
8750-7587/01 $5.00 Copyright © 2001 the American Physiological Society


This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
B. M. Damon and J. C. Gore
Physiological basis of muscle functional MRI: predictions using a computer model
J Appl Physiol, January 1, 2005; 98(1): 264 - 273.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
H. E Kan, W. K. J. Renema, D. Isbrandt, and A. Heerschap
Phosphorylated guanidinoacetate partly compensates for the lack of phosphocreatine in skeletal muscle of mice lacking guanidinoacetate methyltransferase
J. Physiol., October 1, 2004; 560(1): 219 - 229.
[Abstract] [Full Text] [PDF]

Home page
 Rheumatology (Oxford)Home page
B. Pfleiderer, J. Lange, K.-D. Loske, and C. Sunderkotter
Metabolic disturbances during short exercises in dermatomyositis revealed by real-time functional 31P magnetic resonance spectroscopy
Rheumatology, June 1, 2004; 43(6): 696 - 703.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
C. J. Houtman, D. F. Stegeman, J. P. Van Dijk, and M. J. Zwarts
Changes in muscle fiber conduction velocity indicate recruitment of distinct motor unit populations
J Appl Physiol, September 1, 2003; 95(3): 1045 - 1054.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (13)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Houtman, C. J.
Right arrow Articles by Stegeman, D. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Houtman, C. J.
Right arrow Articles by Stegeman, D. F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online