Extensor digitorum longus muscles were stimulated to contract to fatigue and allowed to recover for 2 h in the absence or presence of 5.5 or 11 mM extracellular glucose. This was followed by a second fatigue run, which ended when the absolute force was the same as at the end of the first run. During the first fatigue run, the fluorescence ratio for indo 1 increased [reflecting an increase in myoplasmic free Ca2+ concentration ([Ca2+]i)] during the initial tetani, peaking at ∼115% of the first tetanic value, followed by a continuous decrease to ∼90% at fatigue. During the first fatigue run, myofibrillar Ca2+ sensitivity was significantly decreased. During the second run, the number of tetani was 57 ± 6% of initial force in muscles that recovered in the absence of glucose and 110 ± 6 and 119 ± 2% of initial force in muscles that recovered in 5.5 and 11 mM glucose, respectively. Fluorescence ratios during the first, peak, and last tetani did not differ significantly between the first and second fatigue runs during any of the three conditions. Glycogen decreased by almost 50% during the first fatigue run and did not change further after recovery in the absence of glucose. After recovery in the presence of 5.5 and 11 mM glucose, glycogen increased 32 and 42% above the nonstimulated control value (P < 0.01). These data demonstrate that extracellular glucose delays the decrease of tetanic force and [Ca2+]i during fatiguing stimulation and that glycogen supercompensation following contraction can occur in the absence of insulin.
- indo 1
glycogen is the storage form of glucose in muscle and is the major energy source for most forms of physical exercise. At exercise intensities corresponding to 60–80% of maximal oxygen uptake, fatigue is associated with the depletion of muscle glycogen (4, 16). However, the link between glycogen depletion and muscle fatigue is poorly understood. Glycogen depletion could result in substrate limitation, i.e., insufficient production of acetyl coenzyme for the tricarboxylic acid (TCA) cycle. Alternatively, adequate glycogen stores may be required to maintain high levels of TCA cycle intermediates (30,34). High levels of TCA cycle intermediates are considered to be important in attaining optimal aerobic energy transduction (22). The TCA cycle intermediates, together with other activators (e.g., ADP, Ca2+, NAD/NADH), adjust the flux through the TCA cycle to ensure adequate formation of NADH for ATP production in the respiratory chain. The consequence of glycogen depletion, and thus a loss of TCA cycle intermediates, is an accelerated deamination of ATP to inosine 5′-monophosphate (IMP) (30, 34). Carbohydrate ingestion during prolonged submaximal exercise delays the loss of TCA cycle intermediates and formation of IMP (which reflects a stoichiometric loss of ATP) as well as fatigue development (33). This scenario provides a metabolic basis for the role of glycogen in muscle fatigue.
Additionally, recent studies on isolated skeletal muscle preparations suggest a role for glycogen in muscle excitation-contraction coupling, which may not be directly linked to its role in energy metabolism (3, 7, 35). In one of these studies, in which mammalian muscle (mouse) was studied, myoplasmic free Ca2+concentration ([Ca2+]i) was measured in single fibers of the flexor brevis muscle, whereas glycogen measurements were made on bundles of fibers from the same muscle. However, stimulation duration to fatigue was more than fourfold longer in the bundles vs. the single fibers. It was suggested that this difference derived from differences in fiber types (7). Such problems can be avoided by measuring [Ca2+]i, force, and glycogen in a muscle preparation that is sufficient for biochemical analyses and homogenous in terms of fiber composition. Indeed, from a practical standpoint such experiments are easier to perform on whole muscle preparations.
The aim of the present study was to determine 1) whether muscle glycogen affects fatigue development and 2) whether the mechanism of fatigue is qualitatively similar at different glycogen contents. With the assumption that glucose transport limits glycogen synthesis in skeletal muscle (24-26), fatigue development was compared in two fatigue runs. The second fatigue run was elicited after recovery in the presence of different concentrations of extracellular glucose. The mechanism(s) of fatigue was studied by comparing changes in force production and [Ca2+]i during the various conditions with glycogen measurements.
MATERIALS AND METHODS
Male mice (NMRI strain) weighing 25–30 g were housed at room temperature with a 12:12-h light-dark cycle. Food and water were provided ad libitum. Animals were killed by rapid cervical dislocation, and thereafter extensor digitorum longus (EDL) muscles were isolated. All procedures were approved by the local ethical committee.
Mounting, solution, and stimulation.
Stainless steel hooks were tied with nylon thread to the tendons of the muscle. Muscles were then transferred to a stimulation chamber and mounted between a modified Akers 801 force transducer (SensoNor, Horten, Norway) and an adjustable holder. The chamber was continuously superfused (1.6 ml/min) at room temperature (24°C) with a Tyrode solution of the following composition (in mM): 121 NaCl, 5 KCl, 1.8 CaCl2, 0.5 NaH2PO4, 0.4 MgCl2, 24 NaHCO3, 0.1 EDTA, 5.5 glucose, and 0.1% fetal calf serum. The solution was continuously gassed with 5% CO2-95% O2, resulting in a pH of 7.4. Muscles were stimulated with current pulses (duration 0.5 ms; ∼150% of the voltage required for maximum force response) for 300 ms via platinum plate electrodes lying parallel to the fibers. The muscle was set to optimal length, i.e., the length at which maximum tetanic force was produced.
The fluorescent indicator indo 1 was used to measure [Ca2+]i. The fluorescence of the muscle was measured with a system consisting of a xenon lamp, a monochromator, and two photomultiplier tubes (Photon Technology International, Photo Med, Wedel, Germany). The excitation light was set to 360 nm, and the light emitted at 405 and 490 nm was measured. The autofluorescence of the muscle mounted in the stimulation chamber was measured and subtracted from all subsequent measurements. The muscle was then removed from the stimulation chamber, set at approximately resting length in a homemade holder, and placed in an Eppendorf tube containing 1 ml of the oxygenated Tyrode solution plus 5 μl of Pluronic (20% in DMSO) and 10 μl of 2 mM indo 1 acetoxymethyl ester (indo-AM) in DMSO (Molecular Probes Europe, Leiden, The Netherlands). The tube was placed in a water bath set to 35°C and shaken at 100 oscillations/min for 60 min. The higher temperature during incubation was used to enhance dye loading, but it was subsequently found that the increase in temperature did not enhance loading (data not shown). Thereafter, the muscle was remounted in the stimulation chamber at optimal length, and after 30 min fluorescence was again measured. In ∼20% of the experiments, loading did not result in sufficiently high fluorescence to allow for reliable [Ca2+]i measurements. Typically, however, fluorescence increased three- to fivefold after loading and declined slowly over the course of the experiment (11.2 ± 0.9%/h;n = 5). Altering the concentrations of Pluronic and indo-AM or the duration of incubation did not increase fluorescence any further (data not shown). Changes in [Ca2+]iwere assessed as the ratio of fluorescent light emitted at 405 nm to that at 490 nm. An increase in this ratio reflects an increase in [Ca2+]i (14). The force at 150 Hz was not significantly altered following loading with indo-AM (mean increase of 7.6 ± 10.7%, n = 5,P > 0.05).
Most experiments started by producing a twitch and tetani at 20–150 Hz at 1-min intervals. The force and fluorescence ratio were measured during these contractions, and in this way the relationships among force, fluorescence ratio, and stimulation frequency were established under control conditions. The superfusate was then switched to glucose-free Tyrode. After 10 min in glucose-free Tyrode, a fatigue run was initiated by stimulating muscles for 300 ms at 70 Hz with a tetanic interval of 3 s until 40% of initial force was reached. Thereafter, muscles recovered in Tyrode solution with 0, 5.5, or 11 mM glucose for 110 min. This was followed by an additional 10 min in glucose-free Tyrode for all muscles. Thus the total recovery time for all muscles was 120 min. A second fatigue run was then produced, and this run was stopped at the same absolute force as that reached at the end of the first fatigue run (i.e., 40% of the initial force at 70 Hz). Extracellular glucose was omitted during the fatigue runs to ensure that glycogen was the sole carbohydrate source. In some experiments muscles were frozen in liquid N2 before initiation of the first fatigue run (nonstimulated controls), immediately after the first fatigue run, or after 120 min of recovery.
Force and fluorescence signals were sampled on-line and stored on a desktop computer for subsequent analysis. Tetanic force and fluorescence ratio signals were in each contraction measured as the mean over the last 100 ms of stimulation. For analysis of glycogen, muscles were freeze-dried, extracted in hot 1 M KOH, hydrolyzed enzymatically, and analyzed for free glucose as described elsewhere (30). Glycogen is expressed as micromoles of glucosyl units per gram of dry weight.
Statistically significant differences were determined with Student's paired or unpaired t-tests or with one-way analysis of variance followed by the Newman-Keuls test where appropriate. Values are presented as means ± SE unless stated otherwise.
The force and fluorescence ratio were measured in nonfatigued muscles in response to different stimulation frequencies (Fig.1). The original records in Fig.1 A show that both force and fluorescence ratio increased as the stimulation frequency was increased from 40 to 150 Hz. Mean data of force vs. stimulation frequency show an almost linear increase up to 100 Hz; thereafter, the slope decreases (Fig. 1 B), presumably because the maximum force that the muscle could produce was approached. Mean data of fluorescence ratio vs. stimulation frequency show a similar trend (Fig. 1 C), but the slightly slower increase of the ratio between 100 and 150 Hz may reflect the fact that the fluorescence ratio was approaching the maximum ratio (see Ref.14). The fluorescence ratio in rested muscles was 1.20 ± 0.10 (n = 14), and hence the ratio during a 100-Hz tetanus is almost threefold higher than the resting ratio.
To ensure that movement artifacts were not contributing to changes in the fluorescence ratio, some muscles were loaded with 20 μM succinimidyl N-methylanthranilate, a non-Ca2+-sensitive fluorescent dye (Molecular Probes Europe). Muscles were then stimulated to contract (150 Hz). The ratio decreased insignificantly by 2 ± 4% (n = 3) during contraction. These results are similar to those reported earlier in whole bullfrog semitendinosus muscles (2). Taken together, the data indicate that measurements of changes in the fluorescence ratio reliably reflect changes in [Ca2+]i of the EDL muscle. Indeed, the force and fluorescence ratio records obtained in whole muscle are similar to those obtained in single muscle fibers injected with indo 1 (36,37).
Figure 2 A shows representative records of the fluorescence ratio and force from selected tetani of an initial fatigue run. In this muscle, the tetanic fluorescence ratio increased during the initial tetani of the fatigue run, reaching its peak in the fourth tetanus. Thereafter, the ratio decreased continuously until the point of fatigue. Tetanic force, on the other hand, decreased throughout the fatigue run, but the rate of decrease was largest during the initial tetani, where the ratio increased. Similar results were obtained in the other muscles, and mean data of the tetanic fluorescence ratio and force are shown in Fig. 2,B and C. Figure 3shows the relationship between tetanic force and fluorescence ratio (i.e., force-[Ca2+]i relationship) obtained under control conditions (filled circles) and during fatiguing stimulation (open circles). Data points obtained during fatigue were clearly shifted to the right compared with the control, which indicates that there was a reduced myofibrillar Ca2+ sensitivity in fatigue.
We also measured the half-relaxation time (RT1/2, i.e., the time from end of stimulation until force decreases by 50%), which increased from 26 ± 1 ms during the first 70-Hz tetanus to 50 ± 5 ms during the last tetanus of the first fatigue run (P < 0.001, n = 18). Concomitantly, the basal fluorescence ratio increased from 1.22 ± 0.09 before the first 70-Hz tetanus to 1.30 ± 0.10 after the last tetanus (P < 0.01, n = 16).
The number of tetani required to bring force down to 40% of initial during the first fatigue run averaged 80 ± 6 s for all muscles (n = 18). There was no significant difference in the number of tetani during the first fatigue run between the muscles that subsequently would be exposed to 0 (74 ± 16 tetani), 5.5 (82 ± 6 tetani), or 11 mM glucose (82 ± 10 tetani).
After 120 min of recovery in 0 mM glucose, the first tetanic force during the second fatigue run was decreased to 77 ± 3% of initial (P < 0.001). After recovery in 5.5 mM glucose, the first tetanic force during the second fatigue run was also decreased, but only to 88 ± 1% of initial (P < 0.01). On the other hand, after recovery in 11 mM glucose, the first tetanic force was unchanged (99 ± 4% of initial). Representative continuous force records of the second fatigue run following the three recovery treatments are given in Fig. 4. During the second fatigue run, the number of tetani required to bring force down to 40% of that at the start of the first fatigue run was 57 ± 6% (n = 5) of initial in muscles that recovered in the absence of glucose and 110 ± 6 (n = 7) and 119 ± 2% (n = 6) of initial in muscles that recovered in 5.5 and 11 mM glucose, respectively.
The fluorescence ratio during the first tetanus, the tetanus with the highest ratio, and the last tetanus obtained during the second fatigue run did not differ significantly from the corresponding values during the first fatigue run regardless of treatment (Fig.5). Furthermore, both RT1/2and the basal fluorescence ratio of the first and last tetani of the second fatigue run did not differ significantly from the corresponding values of the first fatigue run during any of the three treatments (i.e., recovery with 0, 5.5, and 11 mM glucose) (data not shown).
Glycogen averaged 62 ± 6 μmol glucosyl units/g dry wt in nonstimulated muscles (n = 7) and decreased to 35 ± 5 μmol glucosyl units/g dry wt (n = 5) after the first fatigue run (P < 0.01). After 120 min of recovery in the absence of glucose, the glycogen values did not differ significantly from those after the first fatigue run (P> 0.05); therefore, these values have been pooled and average 40 ± 4 μmol glucosyl units/g dry wt (n = 10). After 120 min of recovery in 5.5 and 11 mM glucose, glycogen increased to 82 ± 5 (n = 6) and 88 ± 3 (n = 6) μmol glucosyl units/g dry wt μmol, respectively. These latter two values are significantly higher than the glycogen content in the nonstimulated muscles (P < 0.01) but do not differ significantly from each other. Notably, the rate of glycogen synthesis in isolated and perfused rodent skeletal muscle is linear for at least 3 h, even in the presence of high concentrations of glucose (13 mM) and insulin (20 mU/ml) in the medium (27, 29). This suggests that the glycogen synthesis rates were linear during the 2-h recovery period with glucose in the medium.
Comparing the glycogen contents in muscles immediately before the second fatigue run with the performance of other muscles during the second fatigue run resulted in an excellent relationship (Fig.6). Thus increasing the glucose concentration in the medium from 0 to 5.5 mM resulted in large increases in the muscle glycogen content and fatigue resistance. However, further increasing the glucose to 11 mM did not significantly affect fatigue resistance or glycogen content compared with the results obtained after exposure to 5.5 mM glucose.
The relationship between force and the fluorescence ratio during a fatigue run in whole muscle exhibits several noteworthy phases (Fig.2). At the onset of fatiguing stimulation, the tetanic fluorescence ratio increased while force decreased. This is also a prominent feature in single muscle fibers and has been ascribed to a decrease in cross-bridge force production due to increased inorganic phosphate, which occurs as a consequence of a rapid breakdown of phosphocreatine (8). Thereafter, the tetanic fluorescence ratio and force both showed a gradual decrease. Plotting the force against the fluorescence ratio revealed a marked rightward shift during fatigue compared with control. Again, this finding is similar to results obtained in single muscle fibers and indicates a significant decrease of the force-[Ca2+]i relationship in fatigue (36). Finally, during fatiguing stimulation there is a slowing of relaxation and increase of the resting fluorescence ratio in both whole muscles (present study) and single fibers (36). Thus the changes in fluorescence ratio and its relation to force production observed in the present whole muscles are very similar to those previously described in single muscle fibers (36,37).
A major finding in this study is that fatigue was delayed when muscles were incubated with glucose, and the glucose effect was likely mediated by the increase in glycogen content. The higher glycogen content and fatigue resistance were associated with a slower decline in the tetanic indo 1 fluorescence ratio (i.e., [Ca2+]i). Thus decreases in sarcoplasmic reticulum (SR) Ca2+ release contribute to the loss of force during fatigue (36), and a higher glycogen content could delay fatigue by maintaining adequate SR Ca2+ release. These results and conclusions in whole muscle preparations are similar to those recently derived from studies on isolated single skeletal muscle fibers (7, 20).
It has been suggested that the glycogen effect on SR Ca2+release can be independent of its function as an energy source and could derive from a structural role for the polymer, possibly in binding to the ryanodine receptor or bridging the dihydropyridine/ryanodine receptor region (20, 35). Conclusive evidence for such a structural role remains, however, to be presented.
The current results are consistent with the explanation that SR/T-tubular glycogen is required for local energy requirements (7), since this region has the enzymatic capacity to generate ATP (15). Interestingly, the findings that glycogen is associated with the SR/T-tubular region (10) and that glycogen associated with this region is preferentially depleted in fatigued muscles (11) fit with such an explanation. To differentiate between metabolic and nonmetabolic factors in fatigue, the following may be considered. The finding that myofibrillar Ca2+ sensitivity was decreased and basal [Ca2+]i and RT1/2 were increased during the first fatigue run indicates the involvement of a metabolic component in fatigue, probably an increase in inorganic phosphate or, possibly, reduced pH (13, 36, 37). The finding that myofibrillar Ca2+ sensitivity (same force and fluorescence ratio), basal [Ca2+]i, and RT1/2were altered to the same extent after the second fatigue run compared with the first, regardless of whether the muscle recovered with glucose or not, indicates the same degree of involvement of the fatigue-inducing metabolic component. Thus the rate of production of the fatigue-inducing metabolic component was increased in the muscles that recovered in the absence of glucose and had a low glycogen content at the start of the second fatigue run, whereas the opposite was true in the muscles that recovered in the presence of glucose and had a high glycogen content.
Low initial muscle glycogen contents result in an accelerated rate of IMP accumulation during both moderate and heavy aerobic exercise, and this reflects a greater ATP breakdown (32, 34). On the basis of data in these reports, the accumulation of inorganic phosphate can be calculated from the changes in phosphocreatine, ATP, hexosemonophosphates, and glycerol 3-phosphate, which represent the major changes in phosphate-containing metabolites during exercise (31). Such calculations show that at a given workload and exercise duration, accumulation of inorganic phosphate is ∼20% lower when the muscle glycogen content at the onset of exercise is high (32, 34). Thus glycogen can delay fatigue by attenuating the accumulation of inorganic phosphate. A decreased inorganic phosphate concentration in the high glycogen condition can explain both the delayed decrease in myofibrillar Ca2+ sensitivity (13) and the better maintenance of SR Ca2+release (9, 12). Ultimately, however, it is likely that the same inorganic phosphate concentration is reached at the point of fatigue in the low and high glycogen states.
Glycogen supercompensation (the increase in glycogen above basal) during recovery from exercise is a well-established phenomenon (5). It occurs only when carbohydrate is administered (18) and is associated with an increase in plasma insulin concentration. Insulin increases both glucose transport and glycogen synthase activity, and contraction increases the sensitivity of muscle to insulin (17). To our knowledge, glycogen supercompensation following contraction in skeletal muscle has not been observed under in vitro conditions. Thus a novel, important finding of the present study is that glucose increases the glycogen content of previously stimulated muscles to levels significantly higher than in nonstimulated control muscles within 2 h despite the fact that insulin was not present. This finding demonstrates that glycogen supercompensation following contraction can occur in the absence of insulin.
It has been suggested that glucose transport is rate limiting for glycogen synthesis in skeletal muscle (24-26). This conclusion is essentially based on the finding that overexpressing glucose transporters in mouse skeletal muscle results in large increases in muscle glycogen that are not associated with increases in glycogen synthase activity. In rodent skeletal muscle theK m of the glucose transport system ranges between 4 and 10 mM (19, 28). Thus, if glucose transport is rate limiting for glycogen formation, increasing the ambient glucose concentration from 5.5 to 11 mM would substantially increase glucose transport and should increase glycogen synthesis proportionally. The finding that the glycogen contents did not differ significantly between 5.5 and 11 mM glucose treatments indicates that glucose transport is not rate limiting for glycogen biogenesis under the conditions studied. Others have also failed to observe significant increases in glycogen content or [14C]glucose incorporation into glycogen in perfused hindlimb or isolated muscles from rodents when studying similar increases in extracellular glucose (6, 21). Furthermore, in contrast to the glucose transporter overexpression studies (25, 26), others have shown that overexpression of glycogen synthase in mouse skeletal muscle results in large increases in muscle glycogen that are not associated with increases in glucose transport (1, 23). Thus the proposal that glucose transport is rate limiting for glycogen synthesis in the presence of physiological glucose concentrations requires further investigation.
In conclusion, these data demonstrate that extracellular glucose delays fatigue in isolated muscle likely via an increase in glycogen, which delays the decrease in myoplasmic Ca2+ concentration and myofibrillar Ca2+ sensitivity. Furthermore, glycogen supercompensation following contraction can occur in the absence of insulin. Finally, glucose transport does not limit glycogen synthesis under the conditions studied.
This research was supported by grants from the Swedish Medical Research Council (no. 10842), the Swedish National Center for Sports Research, the Novo Nordisk Foundation, and funds at the Karolinska Institutet.
Address for reprint requests and other correspondence: A. Katz, Dept. of Physiology and Pharmacology, Von Eulers väg 4, Karolinska Institute, 171 77 Stockholm (E-mail:).
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.
First published January 16, 2002;10.1152/ajpcell.00490.2001
- Copyright © 2002 the American Physiological Society