Biochemical adaptations to training: implications for resisting muscle fatigue"
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
Departments of Biochemistry m d Biophysics, Surgery, and Cardiology, University of Pennsylvania, Philadelphia, PA 19BO$-@@, U.S*A. Received February 2 1, 19% MCCULLY,K. K., CLARK,B. 9., KENT,9. A., WILSON,J., and CHANCE,B. 1991. Biochemical adaptations to training: implications for resisting muscle fatigue. Can. J. Physiol. Phamacol. 69: 274-278. Slceletd muscle activity is invariably associated with a decline in force-generating capacity (fatigue). The build-up sf metabolic by-products such as intracellular H ' and inorganic phosphate (Pi) has been shown to be one of the potential mechani s m of muscle fatigue. The use of phosphorus magnetic resonance spectroscopy is a repeatable and useful tool to study the effect of pH and Pi on force development. When maximal exercise is preceded by submaximal exercise to reduce the starting muscle pH and increase Pi, the degree of muscle fatigue correlates more strongly with H2P0; than pH or Pi alone. However, other studies in humans have found that H,PO; does not always correlate well with fatigue. The use of ramp exercise protocols allow repeatable and sensitive measurement of changes in musele metabolism in response to endurance training. Chronic electrical stimulation in dogs and endurance training in humans results in reduced pH and P, changes at the same exercise intensities. This means that the effect of pH and Pi in depressing force development is reduced, which could partially explain the increased fatigue resistance seen following endurance training. Key words: magnetic resonance spectroscopy (31B-MRS), muscle metabolism, exercise, inorganic phosphate, pH. MCCULLY,K. K., CLARK,B. J., KENT, 9. A., WILSON, J., et CHANCE,B. 1991. Biochemical adaptations to training: implications for resisting muscle fatigue. Can. 9. Physiol. Phamacol. 69 : 274 -278. L'activite du muscle squelettique est invariablement associCe h une diminution de la capacid de production d'une force (fatigue). La formation de sous-produits mitaboliques, tels que le H ' intracellaalaire et le phosphate inorganique (P,), s'est avCrde l'un des micanismes potentiels de la fatigue msculaire. La spectroscopie de risonance magnktique au phosphore produit des risultats ripitables et utiles pour examiner l'effet du pH et du Pi sur le dCveloppement de la force. Lorsqu'un exercice maximal est suivi d'un exercice submximal pour riduire le pH muscu1aire initial et augmenter le Pi, le degri de fatigue musculaire est comili plus fortement au H , B q qu'au pH ou au Pi. Toutefois, d'autres itudes chez les humins ont m n t r i qu'il n'y avait pas toujours une bonne correlation entre le H,PO, et la fatigue. Les prot~colesd'exercices avec rampes pernettrait de faire une mesure sensible et rkpitable des variations du mitabolisme en rkponse h un entrainement d'endurance. Une stimulation Clectrique chronique chez des chiens et un entrainement d'endurance chez des humains se traduisent par des variations reduites de Pi et Be pH aux m6mes intensitks d'exercice. Ceci signifie que l'effet du pH et du Pi sur la diminution du dCveloppement de la force est rCduit, et pourrait expliquer en partie la risistance accrue h la fatigue obsemie aprks aua entrainement d'endurance. Mots clds : spctroscopie de resonance mgnCtique (3'P-MRS), mitabolisme musculaire, exercice, phosphate inorgaHlique, pH. [Traduit par ];a RCdaction]
Introduction how these changes can affect the role of various metabolites in causing muscle fatigue. Skeletd muscle activity is invariably associated with a decline in force-generating capacity (fatigue). While a number Ehe relationship between muscle fatigue and H2P0; of different fatigue mechanisms have been proposed (Edwards A number of metabolic products have been proposed as 1982), it would be logical to assume that at least one of these fatigue mechanisms, including pH, Pi, and ADP. It has been mechanisms would be the result of increased metabolic proposed that the mechanism of metabolic fatigue involves the activity in the myofihrs. This is because skeletal musele has protonated species of Pi (H-$_PO;) (a%'ilkie 1986). To test this very high metabolic rates during contractile activity. In fact, in humans, we performed two series of fatigue tests with the the build-up of metabolic by-products such as intracellular H+ and inorganic phosphate (Pi) have been shown to be a wrist flexor muscles (Wilson et d. 1988). The first test consisted of maximum voluntary contractions (MVC) of 1 s durapotential mechanism of muscle fatigue (Brandt et al. 1982; tion every 5 s for 4 min. In the second protocol the subjects 1987). The diprotonated form Kentish 1986; Nosek et dLZ. performed the same exercise test, only preceded by 2 min of of Pi, H,PO;, has been suggested as the major metabolic submaximal exercise (each contraction was 20 -50 9% of inhibitor s f force development (YVilke 1986). The purpose of MVC). Muscle pH and Pi concentrations were measured this study is to measure changes in muscle pH and ~ P 8 4 with phosphorus magnetic resonance spectroscopy (P-MRS) , using P-MRS (Chance et d. 1981). With maximal exercise there was a linear relationship between the decrease in force which occur in response to endurance training, and to exmine and decreasing pH (r = 8.9 f 8.08 and increasing H,P8; (r = 0.89 & 0.88). The 2 min of submaximal exercise 'This paper was presented at the symposium Cellular Aspects of starting muscle pH and increased N2P0;. During decreased Skeletal Muscle Fatigue, held January 26 and 27, 1990, University the subsequent m a x i d exercise the relationship between s f Guelph, Guelph, Ont., Canada, and has undergone the Journal's force a d W2P0,s remained unchanged, while the force and pH usual peer review. curve shified to the left (Fig. 1 ) . These results as well as those 2Aughor for correspondence at the following address: D501 from other laboratories demonstrate that force development Richards Blvd., Department of Biochemistry and Biophysics, Unicorrelates more strongly with H,PO, than with pH or tot& Pi versity s f Pennsylvamia, Philadelphia, PA 19104-6089, U. S .A. Printed in Canada i Irnprime au Canada
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
MCCULLY ET AL.
FIG. I. The relationship is shown of developed force to pH and H,PO; during maximal voluntary exercise consisting of 1-s csntraetions every 5 s. The pre-ex group performed 3 wain of submaximal exercise immediately before the m x i m d exercise (from McCully et al. 1988).
(Miller et al. 1988). This suggests that Pi (H,PO;) is a part of the mechanism that inhibits force development during muscle fatigue in voluntary human exercise. The relationship between tension development and H2PO; is simila in human and in v i m studies (Kentish 1986; Nosek et al. 1987). Mowever, in viva experiments show less sensitivity to low Pi levels than do in vitro preparations (Cady et al. 1989). It has been proposed that increased Pi (H2P0,-).acts to decrease the lifetime and thus the number of cross bridges in the tensiongenerating state (Brandt et al. 1982). However, there are circumstances where H2POr does not appear to correlate with muscle fatigue. At the end of a ramp exercise protocol (described below), no relationship was found between levels of H,PO, and the mount of muscle fatigue when comparing two groups of control subjects and one group of endurance-trained athletes (Fig. 2). The H,PO; levels in the control groups should have been high enough to influence force development. Other studies have reported that the relationship between muscle fatigue and H2PO; was not the same during exercise and recovery and between patients with myophosphorylase deficiency and control subjects (Cady et al. 1989). This means that either H,PO, levels are not directly involved in influencing force development or that other fatigue mechanisms are influencing force development.
Metabolic response to endurance training Chronic electrical stirnubtion One approach to the treatment of heart failure is to surgically make skeletal muscle into a pump to assist the heart. The major problem with this approach is that skeletal muscle does not have the degree of fatigue resistance needed to function as a cardiac assist device. Therefore, chronic electrical stimulation was used to condition the skeletal muscle to the point where it could be usehl (Asker et al. 1986). The right latissimus dorsi muscle of adult male beagles was activated with an implanted stimulator at a rate similar to that of the heart, 25 Hz (duty cycle: 312 rns on, 812 ms off) for 24 h/day for 8 weeks. Following this protocol the muscles showed almost complete conversion from fast to slow muscle fiber types as shown by histochemistry, SDS gel electrophoresis of light chains, and time to peak twitch tension (Clark et al. 1988). The stimulated muscles showed little or no fatigue during electrical stimulation designed to mimic the work rest cycle of the dog heart. With progressive increases in the work to rest cycle, P-MRS showed much less change in Pi/PCr ratios (PCr, phosphwreatine) than uwsthdatd muscle (Fig. 3).
FIG.2. The relationship between the end level of work achieved and the end values of $PO; is shown. Included in this figure are the results sf six endurance-trained subjects (rowers) a d two groups s f control subjects.
The decrease in Pi/P@r ratios with stimulation agreed well with direct measurements of oxygen consumption, which were also lower at any given work level. The increased fatigue resistance seen in the conditioned muscle could be explained by the reduced levels of metabolic by-products such as pH and Pi during stimulation. Voluntary exercise in humans Before the initiation of training studies in human subjects, a series of tests were performed to determine the reproducibility of the P-MRS measurements. A ramp exercise protocol was chosen to provide a larger number of data points in a reasonable period of time. Tests of the repeatability and the influence of the size of the work increment were performed on three n o r d subjects. The subjects were placed in a 1.9 T magnet and the mist flexor muscles were placed over a surface coil (4 cm) tuned to both phosphorus and proton frequencies (Chance et d. 1981). Most of the P-MRS signal was obtained from muscle within 1.5 cm of the skin. The muscles included in this area are the flexor carpi radialis, palmaris longus with some contribution from the flexor d i g i t o m superficialis, and the flexor digitomm profundus muscles. Phosphorus signals were collected every 5 s using an optimal pulse width. The collected signals were averaged into 1-min blocks and relative concentrations of Pi, PCr, and ATP were determined from peak areas. The distance between the Pi and PCr peaks was used to determine the intracellular pH (Taylor et al. 1986). Absolute values of $PO; were calculated
276
CAN. I. PHYSIOL. PHARhqACOL. VOL. 69, 1991 0
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
UB
Control Stimulated
FIG. 3. Shown is tension time integral versus Bi/PCr for control rand chronically stimulated dog muscle. Data points are means f SEM, n = 5 (from Clark et al. 1988).
assuming a disassociation constant of Pi = 6.75 (Wilson et d. 1988). Wrist flexion exercise was performed by depressing a handle attached to the lever arm of a Cybex I1 ergometer. The Bevel of work was controlled by having the subject press the handle to match a target level of force. The distance traveled and the velocity s f contraction were held constant so that each contraction was 0.5 s in duration. Before testing, the subject's arm was warmed in a heating blanket for 18 min. Preliminary studies have shown that prewarming the muscle to be tested helps to remove temperature variations and reduces test variability (McCully et al. 1987). Before the exercise protocol started, the subject was allowed to become familiar with matching the target levels of force at the low work levels and the MVC was determined. During the ramp exercise test the subject performed 12 contractionslmin, gated so that each contraction occurred just after the MRS pulse was collected (one every 5 s). The force level in the B st min was approximately 5 - 10 % of MVC. In each of the following minutes the force of contraction was increased by 5 - 18% of MVC until the subjects could not achieve the target value, at which time the test was stopped. To test the effect of varying the rate, in which the work level was increased, of the ramp test, the rate of increase in work was varied in two of the control subjects from 30% MVC every minute to 5 % MVC every 3 min. Work vdues were normalized to each subject's MVC. Total duration of the ramp exercise test was approximately 1 h. Increasing work resulted in increased Pi and decreased PCr. The relationship between work and the ratio of Bi/PCr appeared to be hyperbolic with the initial portion approximating a straight line. A high degree of reproducibility was seen during repeat tests on the same subject (Fig. 4). The variation in the calculated initid slope of the work versus PilP@rcurve averaged 5% of the mean for three subjects. En control subjects the initial slope was independent of the rate of increase in work up to a rate of 15 % MVClrnin. Increasing work increments above 15% MVC/rnin resulted in Bower Pi/P@r and higher pH vdues. At high levels of work ( >40% MVC) Pi/P@r and pH varied with work rate, indicating nonsteadystate metabolism. The end level of work appeared to be unaffected by the rate of change in the work level.
=
m
8 Ea
6.8
€2
BB
e
e 8 0
6.6 0
20
40 60 Power (% MVC)
8Q
10Q
FIG.4. (A) Work versus Pi/PGr and (B) pH versus work for four ramp protocol tests on a control subject. Work was expressed as the percent of MVC s f each contraction with 12 contractions being per-
fsmed each minute. Abmpt transitions in work levels result in a lag of oxidative metabolism behind the changes in work (oxygen deficit). The ramp exercise protocol makes use of numerous small increments of work to avoid this oxygen deficit. Thus, while the level sf work is constantly changing, the increase in work levels of 5 - 10 % MVCImin is sufficiently smdl so h t the initial portion of the ramp exercise protocol can be considered representative of a m e h h l i c steady state. Previous studies have fit the results of steady-state exercise tests to a rectangular hyperbola, assuming the relationship between Pi/PCr a d work represents Michaelis -Menten finetics (Chance et al. 1986). While changes in pH at higher levels of work make this cdculation invalid, the calculation of a 'Vm,'' should be a g o d indicator of the shape of the curve and of the metabolic capacity of the muscle.
Metabolic response 80 training in humans To determine the metabolic changes in human muscle following voluntary exercise, two training studies were performed. In the first study seven subjects performed 20 min of wrist flexion exercise using weights (Kent et d. 1988). The training consisted of lifting 50% MVC every 5 s and was done 5 dayslweek for 8 weeks. Before and 2 days after the training program ended the sub~ectsperformed an endurance test consisting of repeated maximum contractions of 1 s duration
9
Pretraining
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
s l h postexercise m 5 days posttraining
e 14 days posttraining 1
lism was impaired, as demonstrated by a right shift of the work versus Pi/PCr curve. While there may have been a slight increase in Vm, at the end of the training period, there was a dramatic improvement in Vm, after the training stopped. In this subject it reached a maximum 3 -5 days after training stopped. The improvement with training gradually disappeared to pretraining vdues over the next 2 weeks. The time course of training adaptation and deadaptation is similar to that reported by Dudley et alal.(1982) for endurance-trained rats. The improvement in M,, following cessation of exercise was complementary to the concept of "taperingq ' athletic training before an important competition. The reduction in pH and Pi during exercise with the endurance-training studies were similar to that seen following chronic stimulation in the dog.
Summary Muscle fatigue appears to be associated with the build-up of metabolic products such as pH and Pi; however, the actual mechanisrn(s) and the role of H,PO; are not clearly known at present. The ramp exercise prstwsl using P-MRS of Iocalized muscle is reproducible and a sensitive method of monitoring changes in muscle metabolism. It can be applied to both human and animal studies. Endurance training in both dogs and humans results in reduced pH and Pi changes at any given level of work. This means that the effect of pH and Pi in depressing force development is reduced, which could part i d y explain the increased fatigue resistance seen following endurance training.
Time (days) R o . 5. Shown are the results from an endurance training program of the wrist flexor muscles for one subject. (A) Work level versus P,IPCr. Work level is plotted as the force of contraction as a percent of a single repetition maximum (12 contractionslmin) ((fromMcCully et d. 1989b). (B) Calculated y-asymptote (Vm,). Day - 15, pretraining; days - 14 to 0, during training; days 1-20, posttraining.
every 3 s for 18 min. The subjects increased the total mount of work done in 10 min from 45.0 13.4 J before training to 51.7 18.4 J following training. The subjects also performed a ramp exercise protocol before and after training. The Pi/PCr vdues at the highest exercise level decreased from 2 -59 f 1.39 before training to 1.37 f 8-51 after training. Peak power at the highest exercise level was not different between before and after training. Muscle pH was d s o lower at any given work level following endurance training. The decrease in Pi/PCr ratios with training was consistent with results of endurance-training athletes compared with control subjects (McCully et al. 198%~). Four subjects performed a more strenuous training program consisting of lifting 25% MVC at a rate of 180 contractions/min for B h/day for 14 straight days (McCully et al. 1989b). This study was based on a strenuous short-term training program of Costill et d.(1985). Training was done in the afternoon with subsequent testing the following morning, which dlowed the m i m u m time for recovery after each training bout. The metabolic response to exercise following endurance training was quite interesting. Figure 542 shows the work versus Pi/PCr curves and Fig. 5b shows the extrapolated maximum work levels (V,,,) for one subject. For several hours after the first training protocol muscle metabo-
+
+
W. 19". An autologous pump motor. Thorac. Cnrdiovasce Surg. 92: 733-746* BRAN^, P. \V., Cox, R. N., KAMI, M., and ROBINSON, T. 1982. Regulation of tension in skinned muscle fibers: effect of crossbridge kinetics on apparent Ca2+ sensitivity. J. @en. Physiol. 79: 997 - 1016. CADY,E. B., JONES,D. A., LYNN,J., and NEWHAM,D. J. 1989. Changes in force and intracellular metabolites during fatigue of human skeletal muscle. 3. Physisl. (London), 418: 3 11-325. D., and SAPEGA, CHANCE, B., ELEPF,S., LEIGH,J. S., JR., SOKOLW, A. 1981. Mitochondrial regulation of phssphocreatinelinorganic phosphate ratios in exercising humam muscle: a gated "P NMR study. Prw. NatB. Acad. Sci. U.S.A. '78: 6714-6718. CHANCE, B., LEIGH,J. S., JR., KENT,J., MCCULLY, K., NHOKA, S., CLARK, B. J., MARJS,J., and GRAHAM, T. 1986. Control of owidative metabolism and oxygen delivery in human skeletal mussle: multiple controls of oxidative metabolism in living tissues as ~ O ~ S resonance. Proc. Natl. Acad. Sci. studied by ~ ~ Q S P magnetic U.S.A. 83: 9458-9462. CLARK,B. J., ACKER,hf. A., MCCULLY,K. K., ~ W B R A M A N ~ A N , H. V., HAMMOND, B., SALMONS, S., CHANCE,B., and STEPHENSON,L. 1988. In vivo 31-P NMR spectroscopy sf chronically stimulated canine skeletal muscle. Am. J. Bhysiol. 254: C258 C2M. COSTILL,D* L., KING, D. S., R-IBMAS,Ha., and HARGREAVES, M. 2985. Effects of reduced training on muscular power in swimmers. Physician Sportsmed. f 3: 95 - 101. DUDLEY,G. A., ABRAHAM, W. M., a d TEWUNG,W. L. 1982. Influence of exercise intensity and duration on biochemical adaptations in skeletal muscle. J. Appl. Physiol. 53: 844 - 850. EDWARDS, R. H. T. 1982. Biochemical bases of fatigue in exercise performance: catastrophe theory of muscular fatigue. International Series of Sports Science. Vol. 13. Biochemistry of exercise. Edited by H.G. Knuatgen, J. H. Vogel, and J. H. Poortmms. pp. 3 -28. K. K., and CHANCE,B. 1988. Metabolic KENT,J. A., MCCWLLY,
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
278
CAN. 5. PHYSIOL. PHAIWACOL. V 8 L . 69, I991
effects of training in man: a 31-P NMR study. M d . Sci. Sports Exercise, 2O(Suppl.): %a. KENTESH, J. C. 1986. The effects of inorganic phosphate and creatine phosphate on force production in skinned muscle from rat vvetricle. J . Phy siol. (London), 370: 5 85 -604. MCCULLY, K., BBDEN,B., D O N ~ N E.,, and CHANCE, B. 1987. The ""warm up" effect during exercise studied with 3 1-P MRS. Fed. Proe. Fed. Am. Soc. Exp. Biol. 46:820. MCCWLLY, K. K., KENT,J. A., and CHANCE, B. 1988. Application of 31-P magnetic resonance spectroscopy to the study of athletic performance. Sports Med. 5: 312-321. MCCULLY, K. K., BODEN,B. P*, WCHLEW, M., FOUNTAIN, M.,and CHANCE,B. 1989a. The wrist flexor muscles of elite rowers measured with magnetic resonance spectroscopy. J. Appl. Physiol. 67: 926-932. MCCWLLY~ K. K., KENT J. A., and CHANCE,B. 1989b. Muscle injury and stress measured with 3%-Pmagnetic resonance spectroscopy. Muscle energetics. Progress in clinical and bio1ogicd research. Vol. 315. Edited by R. J. Paul, K. Y m d a , and G. Elzinga. A. R. Eiss, New York. pp. 197-207.
MILLER,R. a., Bos~sa,h'f. D., MOUSSAVI, R. S., CARSON,p. JJ., ~EINER K., We 1988. 3'P nuclear magnetic resonance studies of high energy and phosphates and pH in h u m n muscle fatigue: comparison of aerobic and anaerobic exercise. J. Clin. Invest. 81: 1190- 1196. NOSEK,TaM.,FENDER, K. Y., and Gom9R. E. 1987. It is diprotonat& inorganic phossphate that depresses force in skinned skeletal muscle fibres. Science (Washington, D. C .), 236: 19B - 193. TAYLOR,D. J., STYLES,Pe, MATHEWS, P. M.,ARNOLD,D. A , GADIAN, D. G . , BORE,P., and RADDA, G. K. 1986. Energetics of humn muscle: exercise-induced ATP depletion. Magn. Reson.
Md.3: 44. WILKIE,D. R. 1986. Muscular fatigue: effects of hydrogen ions and inorganic phosphate. P d . Prsc. Fed. Am. SOC.Exp. Biol. 45: 292 1 -2923. WILSON,J. R., MCCULLY,K. K., MANCINI,D. M., BOBEN,B. P., and CHANCE, B. 1988. Relationship between muscle fatigue to pH and diprotonat& Pi in humans: a 31-P NMR study. J. Appl. Physid. 6%: 2333 -2339.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
Departments of Biochemistry m d Biophysics, Surgery, and Cardiology, University of Pennsylvania, Philadelphia, PA 19BO$-@@, U.S*A. Received February 2 1, 19% MCCULLY,K. K., CLARK,B. 9., KENT,9. A., WILSON,J., and CHANCE,B. 1991. Biochemical adaptations to training: implications for resisting muscle fatigue. Can. J. Physiol. Phamacol. 69: 274-278. Slceletd muscle activity is invariably associated with a decline in force-generating capacity (fatigue). The build-up sf metabolic by-products such as intracellular H ' and inorganic phosphate (Pi) has been shown to be one of the potential mechani s m of muscle fatigue. The use of phosphorus magnetic resonance spectroscopy is a repeatable and useful tool to study the effect of pH and Pi on force development. When maximal exercise is preceded by submaximal exercise to reduce the starting muscle pH and increase Pi, the degree of muscle fatigue correlates more strongly with H2P0; than pH or Pi alone. However, other studies in humans have found that H,PO; does not always correlate well with fatigue. The use of ramp exercise protocols allow repeatable and sensitive measurement of changes in musele metabolism in response to endurance training. Chronic electrical stimulation in dogs and endurance training in humans results in reduced pH and P, changes at the same exercise intensities. This means that the effect of pH and Pi in depressing force development is reduced, which could partially explain the increased fatigue resistance seen following endurance training. Key words: magnetic resonance spectroscopy (31B-MRS), muscle metabolism, exercise, inorganic phosphate, pH. MCCULLY,K. K., CLARK,B. J., KENT, 9. A., WILSON, J., et CHANCE,B. 1991. Biochemical adaptations to training: implications for resisting muscle fatigue. Can. 9. Physiol. Phamacol. 69 : 274 -278. L'activite du muscle squelettique est invariablement associCe h une diminution de la capacid de production d'une force (fatigue). La formation de sous-produits mitaboliques, tels que le H ' intracellaalaire et le phosphate inorganique (P,), s'est avCrde l'un des micanismes potentiels de la fatigue msculaire. La spectroscopie de risonance magnktique au phosphore produit des risultats ripitables et utiles pour examiner l'effet du pH et du Pi sur le dCveloppement de la force. Lorsqu'un exercice maximal est suivi d'un exercice submximal pour riduire le pH muscu1aire initial et augmenter le Pi, le degri de fatigue musculaire est comili plus fortement au H , B q qu'au pH ou au Pi. Toutefois, d'autres itudes chez les humins ont m n t r i qu'il n'y avait pas toujours une bonne correlation entre le H,PO, et la fatigue. Les prot~colesd'exercices avec rampes pernettrait de faire une mesure sensible et rkpitable des variations du mitabolisme en rkponse h un entrainement d'endurance. Une stimulation Clectrique chronique chez des chiens et un entrainement d'endurance chez des humains se traduisent par des variations reduites de Pi et Be pH aux m6mes intensitks d'exercice. Ceci signifie que l'effet du pH et du Pi sur la diminution du dCveloppement de la force est rCduit, et pourrait expliquer en partie la risistance accrue h la fatigue obsemie aprks aua entrainement d'endurance. Mots clds : spctroscopie de resonance mgnCtique (3'P-MRS), mitabolisme musculaire, exercice, phosphate inorgaHlique, pH. [Traduit par ];a RCdaction]
Introduction how these changes can affect the role of various metabolites in causing muscle fatigue. Skeletd muscle activity is invariably associated with a decline in force-generating capacity (fatigue). While a number Ehe relationship between muscle fatigue and H2P0; of different fatigue mechanisms have been proposed (Edwards A number of metabolic products have been proposed as 1982), it would be logical to assume that at least one of these fatigue mechanisms, including pH, Pi, and ADP. It has been mechanisms would be the result of increased metabolic proposed that the mechanism of metabolic fatigue involves the activity in the myofihrs. This is because skeletal musele has protonated species of Pi (H-$_PO;) (a%'ilkie 1986). To test this very high metabolic rates during contractile activity. In fact, in humans, we performed two series of fatigue tests with the the build-up of metabolic by-products such as intracellular H+ and inorganic phosphate (Pi) have been shown to be a wrist flexor muscles (Wilson et d. 1988). The first test consisted of maximum voluntary contractions (MVC) of 1 s durapotential mechanism of muscle fatigue (Brandt et al. 1982; tion every 5 s for 4 min. In the second protocol the subjects 1987). The diprotonated form Kentish 1986; Nosek et dLZ. performed the same exercise test, only preceded by 2 min of of Pi, H,PO;, has been suggested as the major metabolic submaximal exercise (each contraction was 20 -50 9% of inhibitor s f force development (YVilke 1986). The purpose of MVC). Muscle pH and Pi concentrations were measured this study is to measure changes in muscle pH and ~ P 8 4 with phosphorus magnetic resonance spectroscopy (P-MRS) , using P-MRS (Chance et d. 1981). With maximal exercise there was a linear relationship between the decrease in force which occur in response to endurance training, and to exmine and decreasing pH (r = 8.9 f 8.08 and increasing H,P8; (r = 0.89 & 0.88). The 2 min of submaximal exercise 'This paper was presented at the symposium Cellular Aspects of starting muscle pH and increased N2P0;. During decreased Skeletal Muscle Fatigue, held January 26 and 27, 1990, University the subsequent m a x i d exercise the relationship between s f Guelph, Guelph, Ont., Canada, and has undergone the Journal's force a d W2P0,s remained unchanged, while the force and pH usual peer review. curve shified to the left (Fig. 1 ) . These results as well as those 2Aughor for correspondence at the following address: D501 from other laboratories demonstrate that force development Richards Blvd., Department of Biochemistry and Biophysics, Unicorrelates more strongly with H,PO, than with pH or tot& Pi versity s f Pennsylvamia, Philadelphia, PA 19104-6089, U. S .A. Printed in Canada i Irnprime au Canada
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
MCCULLY ET AL.
FIG. I. The relationship is shown of developed force to pH and H,PO; during maximal voluntary exercise consisting of 1-s csntraetions every 5 s. The pre-ex group performed 3 wain of submaximal exercise immediately before the m x i m d exercise (from McCully et al. 1988).
(Miller et al. 1988). This suggests that Pi (H,PO;) is a part of the mechanism that inhibits force development during muscle fatigue in voluntary human exercise. The relationship between tension development and H2PO; is simila in human and in v i m studies (Kentish 1986; Nosek et al. 1987). Mowever, in viva experiments show less sensitivity to low Pi levels than do in vitro preparations (Cady et al. 1989). It has been proposed that increased Pi (H2P0,-).acts to decrease the lifetime and thus the number of cross bridges in the tensiongenerating state (Brandt et al. 1982). However, there are circumstances where H2POr does not appear to correlate with muscle fatigue. At the end of a ramp exercise protocol (described below), no relationship was found between levels of H,PO, and the mount of muscle fatigue when comparing two groups of control subjects and one group of endurance-trained athletes (Fig. 2). The H,PO; levels in the control groups should have been high enough to influence force development. Other studies have reported that the relationship between muscle fatigue and H2PO; was not the same during exercise and recovery and between patients with myophosphorylase deficiency and control subjects (Cady et al. 1989). This means that either H,PO, levels are not directly involved in influencing force development or that other fatigue mechanisms are influencing force development.
Metabolic response to endurance training Chronic electrical stirnubtion One approach to the treatment of heart failure is to surgically make skeletal muscle into a pump to assist the heart. The major problem with this approach is that skeletal muscle does not have the degree of fatigue resistance needed to function as a cardiac assist device. Therefore, chronic electrical stimulation was used to condition the skeletal muscle to the point where it could be usehl (Asker et al. 1986). The right latissimus dorsi muscle of adult male beagles was activated with an implanted stimulator at a rate similar to that of the heart, 25 Hz (duty cycle: 312 rns on, 812 ms off) for 24 h/day for 8 weeks. Following this protocol the muscles showed almost complete conversion from fast to slow muscle fiber types as shown by histochemistry, SDS gel electrophoresis of light chains, and time to peak twitch tension (Clark et al. 1988). The stimulated muscles showed little or no fatigue during electrical stimulation designed to mimic the work rest cycle of the dog heart. With progressive increases in the work to rest cycle, P-MRS showed much less change in Pi/PCr ratios (PCr, phosphwreatine) than uwsthdatd muscle (Fig. 3).
FIG.2. The relationship between the end level of work achieved and the end values of $PO; is shown. Included in this figure are the results sf six endurance-trained subjects (rowers) a d two groups s f control subjects.
The decrease in Pi/P@r ratios with stimulation agreed well with direct measurements of oxygen consumption, which were also lower at any given work level. The increased fatigue resistance seen in the conditioned muscle could be explained by the reduced levels of metabolic by-products such as pH and Pi during stimulation. Voluntary exercise in humans Before the initiation of training studies in human subjects, a series of tests were performed to determine the reproducibility of the P-MRS measurements. A ramp exercise protocol was chosen to provide a larger number of data points in a reasonable period of time. Tests of the repeatability and the influence of the size of the work increment were performed on three n o r d subjects. The subjects were placed in a 1.9 T magnet and the mist flexor muscles were placed over a surface coil (4 cm) tuned to both phosphorus and proton frequencies (Chance et d. 1981). Most of the P-MRS signal was obtained from muscle within 1.5 cm of the skin. The muscles included in this area are the flexor carpi radialis, palmaris longus with some contribution from the flexor d i g i t o m superficialis, and the flexor digitomm profundus muscles. Phosphorus signals were collected every 5 s using an optimal pulse width. The collected signals were averaged into 1-min blocks and relative concentrations of Pi, PCr, and ATP were determined from peak areas. The distance between the Pi and PCr peaks was used to determine the intracellular pH (Taylor et al. 1986). Absolute values of $PO; were calculated
276
CAN. I. PHYSIOL. PHARhqACOL. VOL. 69, 1991 0
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
UB
Control Stimulated
FIG. 3. Shown is tension time integral versus Bi/PCr for control rand chronically stimulated dog muscle. Data points are means f SEM, n = 5 (from Clark et al. 1988).
assuming a disassociation constant of Pi = 6.75 (Wilson et d. 1988). Wrist flexion exercise was performed by depressing a handle attached to the lever arm of a Cybex I1 ergometer. The Bevel of work was controlled by having the subject press the handle to match a target level of force. The distance traveled and the velocity s f contraction were held constant so that each contraction was 0.5 s in duration. Before testing, the subject's arm was warmed in a heating blanket for 18 min. Preliminary studies have shown that prewarming the muscle to be tested helps to remove temperature variations and reduces test variability (McCully et al. 1987). Before the exercise protocol started, the subject was allowed to become familiar with matching the target levels of force at the low work levels and the MVC was determined. During the ramp exercise test the subject performed 12 contractionslmin, gated so that each contraction occurred just after the MRS pulse was collected (one every 5 s). The force level in the B st min was approximately 5 - 10 % of MVC. In each of the following minutes the force of contraction was increased by 5 - 18% of MVC until the subjects could not achieve the target value, at which time the test was stopped. To test the effect of varying the rate, in which the work level was increased, of the ramp test, the rate of increase in work was varied in two of the control subjects from 30% MVC every minute to 5 % MVC every 3 min. Work vdues were normalized to each subject's MVC. Total duration of the ramp exercise test was approximately 1 h. Increasing work resulted in increased Pi and decreased PCr. The relationship between work and the ratio of Bi/PCr appeared to be hyperbolic with the initial portion approximating a straight line. A high degree of reproducibility was seen during repeat tests on the same subject (Fig. 4). The variation in the calculated initid slope of the work versus PilP@rcurve averaged 5% of the mean for three subjects. En control subjects the initial slope was independent of the rate of increase in work up to a rate of 15 % MVClrnin. Increasing work increments above 15% MVC/rnin resulted in Bower Pi/P@r and higher pH vdues. At high levels of work ( >40% MVC) Pi/P@r and pH varied with work rate, indicating nonsteadystate metabolism. The end level of work appeared to be unaffected by the rate of change in the work level.
=
m
8 Ea
6.8
€2
BB
e
e 8 0
6.6 0
20
40 60 Power (% MVC)
8Q
10Q
FIG.4. (A) Work versus Pi/PGr and (B) pH versus work for four ramp protocol tests on a control subject. Work was expressed as the percent of MVC s f each contraction with 12 contractions being per-
fsmed each minute. Abmpt transitions in work levels result in a lag of oxidative metabolism behind the changes in work (oxygen deficit). The ramp exercise protocol makes use of numerous small increments of work to avoid this oxygen deficit. Thus, while the level sf work is constantly changing, the increase in work levels of 5 - 10 % MVCImin is sufficiently smdl so h t the initial portion of the ramp exercise protocol can be considered representative of a m e h h l i c steady state. Previous studies have fit the results of steady-state exercise tests to a rectangular hyperbola, assuming the relationship between Pi/PCr a d work represents Michaelis -Menten finetics (Chance et al. 1986). While changes in pH at higher levels of work make this cdculation invalid, the calculation of a 'Vm,'' should be a g o d indicator of the shape of the curve and of the metabolic capacity of the muscle.
Metabolic response 80 training in humans To determine the metabolic changes in human muscle following voluntary exercise, two training studies were performed. In the first study seven subjects performed 20 min of wrist flexion exercise using weights (Kent et d. 1988). The training consisted of lifting 50% MVC every 5 s and was done 5 dayslweek for 8 weeks. Before and 2 days after the training program ended the sub~ectsperformed an endurance test consisting of repeated maximum contractions of 1 s duration
9
Pretraining
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
s l h postexercise m 5 days posttraining
e 14 days posttraining 1
lism was impaired, as demonstrated by a right shift of the work versus Pi/PCr curve. While there may have been a slight increase in Vm, at the end of the training period, there was a dramatic improvement in Vm, after the training stopped. In this subject it reached a maximum 3 -5 days after training stopped. The improvement with training gradually disappeared to pretraining vdues over the next 2 weeks. The time course of training adaptation and deadaptation is similar to that reported by Dudley et alal.(1982) for endurance-trained rats. The improvement in M,, following cessation of exercise was complementary to the concept of "taperingq ' athletic training before an important competition. The reduction in pH and Pi during exercise with the endurance-training studies were similar to that seen following chronic stimulation in the dog.
Summary Muscle fatigue appears to be associated with the build-up of metabolic products such as pH and Pi; however, the actual mechanisrn(s) and the role of H,PO; are not clearly known at present. The ramp exercise prstwsl using P-MRS of Iocalized muscle is reproducible and a sensitive method of monitoring changes in muscle metabolism. It can be applied to both human and animal studies. Endurance training in both dogs and humans results in reduced pH and Pi changes at any given level of work. This means that the effect of pH and Pi in depressing force development is reduced, which could part i d y explain the increased fatigue resistance seen following endurance training.
Time (days) R o . 5. Shown are the results from an endurance training program of the wrist flexor muscles for one subject. (A) Work level versus P,IPCr. Work level is plotted as the force of contraction as a percent of a single repetition maximum (12 contractionslmin) ((fromMcCully et d. 1989b). (B) Calculated y-asymptote (Vm,). Day - 15, pretraining; days - 14 to 0, during training; days 1-20, posttraining.
every 3 s for 18 min. The subjects increased the total mount of work done in 10 min from 45.0 13.4 J before training to 51.7 18.4 J following training. The subjects also performed a ramp exercise protocol before and after training. The Pi/PCr vdues at the highest exercise level decreased from 2 -59 f 1.39 before training to 1.37 f 8-51 after training. Peak power at the highest exercise level was not different between before and after training. Muscle pH was d s o lower at any given work level following endurance training. The decrease in Pi/PCr ratios with training was consistent with results of endurance-training athletes compared with control subjects (McCully et al. 198%~). Four subjects performed a more strenuous training program consisting of lifting 25% MVC at a rate of 180 contractions/min for B h/day for 14 straight days (McCully et al. 1989b). This study was based on a strenuous short-term training program of Costill et d.(1985). Training was done in the afternoon with subsequent testing the following morning, which dlowed the m i m u m time for recovery after each training bout. The metabolic response to exercise following endurance training was quite interesting. Figure 542 shows the work versus Pi/PCr curves and Fig. 5b shows the extrapolated maximum work levels (V,,,) for one subject. For several hours after the first training protocol muscle metabo-
+
+
W. 19". An autologous pump motor. Thorac. Cnrdiovasce Surg. 92: 733-746* BRAN^, P. \V., Cox, R. N., KAMI, M., and ROBINSON, T. 1982. Regulation of tension in skinned muscle fibers: effect of crossbridge kinetics on apparent Ca2+ sensitivity. J. @en. Physiol. 79: 997 - 1016. CADY,E. B., JONES,D. A., LYNN,J., and NEWHAM,D. J. 1989. Changes in force and intracellular metabolites during fatigue of human skeletal muscle. 3. Physisl. (London), 418: 3 11-325. D., and SAPEGA, CHANCE, B., ELEPF,S., LEIGH,J. S., JR., SOKOLW, A. 1981. Mitochondrial regulation of phssphocreatinelinorganic phosphate ratios in exercising humam muscle: a gated "P NMR study. Prw. NatB. Acad. Sci. U.S.A. '78: 6714-6718. CHANCE, B., LEIGH,J. S., JR., KENT,J., MCCULLY, K., NHOKA, S., CLARK, B. J., MARJS,J., and GRAHAM, T. 1986. Control of owidative metabolism and oxygen delivery in human skeletal mussle: multiple controls of oxidative metabolism in living tissues as ~ O ~ S resonance. Proc. Natl. Acad. Sci. studied by ~ ~ Q S P magnetic U.S.A. 83: 9458-9462. CLARK,B. J., ACKER,hf. A., MCCULLY,K. K., ~ W B R A M A N ~ A N , H. V., HAMMOND, B., SALMONS, S., CHANCE,B., and STEPHENSON,L. 1988. In vivo 31-P NMR spectroscopy sf chronically stimulated canine skeletal muscle. Am. J. Bhysiol. 254: C258 C2M. COSTILL,D* L., KING, D. S., R-IBMAS,Ha., and HARGREAVES, M. 2985. Effects of reduced training on muscular power in swimmers. Physician Sportsmed. f 3: 95 - 101. DUDLEY,G. A., ABRAHAM, W. M., a d TEWUNG,W. L. 1982. Influence of exercise intensity and duration on biochemical adaptations in skeletal muscle. J. Appl. Physiol. 53: 844 - 850. EDWARDS, R. H. T. 1982. Biochemical bases of fatigue in exercise performance: catastrophe theory of muscular fatigue. International Series of Sports Science. Vol. 13. Biochemistry of exercise. Edited by H.G. Knuatgen, J. H. Vogel, and J. H. Poortmms. pp. 3 -28. K. K., and CHANCE,B. 1988. Metabolic KENT,J. A., MCCWLLY,
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by San Francisco (UCSF) on 01/01/15 For personal use only.
278
CAN. 5. PHYSIOL. PHAIWACOL. V 8 L . 69, I991
effects of training in man: a 31-P NMR study. M d . Sci. Sports Exercise, 2O(Suppl.): %a. KENTESH, J. C. 1986. The effects of inorganic phosphate and creatine phosphate on force production in skinned muscle from rat vvetricle. J . Phy siol. (London), 370: 5 85 -604. MCCULLY, K., BBDEN,B., D O N ~ N E.,, and CHANCE, B. 1987. The ""warm up" effect during exercise studied with 3 1-P MRS. Fed. Proe. Fed. Am. Soc. Exp. Biol. 46:820. MCCWLLY, K. K., KENT,J. A., and CHANCE, B. 1988. Application of 31-P magnetic resonance spectroscopy to the study of athletic performance. Sports Med. 5: 312-321. MCCULLY, K. K., BODEN,B. P*, WCHLEW, M., FOUNTAIN, M.,and CHANCE,B. 1989a. The wrist flexor muscles of elite rowers measured with magnetic resonance spectroscopy. J. Appl. Physiol. 67: 926-932. MCCWLLY~ K. K., KENT J. A., and CHANCE,B. 1989b. Muscle injury and stress measured with 3%-Pmagnetic resonance spectroscopy. Muscle energetics. Progress in clinical and bio1ogicd research. Vol. 315. Edited by R. J. Paul, K. Y m d a , and G. Elzinga. A. R. Eiss, New York. pp. 197-207.
MILLER,R. a., Bos~sa,h'f. D., MOUSSAVI, R. S., CARSON,p. JJ., ~EINER K., We 1988. 3'P nuclear magnetic resonance studies of high energy and phosphates and pH in h u m n muscle fatigue: comparison of aerobic and anaerobic exercise. J. Clin. Invest. 81: 1190- 1196. NOSEK,TaM.,FENDER, K. Y., and Gom9R. E. 1987. It is diprotonat& inorganic phossphate that depresses force in skinned skeletal muscle fibres. Science (Washington, D. C .), 236: 19B - 193. TAYLOR,D. J., STYLES,Pe, MATHEWS, P. M.,ARNOLD,D. A , GADIAN, D. G . , BORE,P., and RADDA, G. K. 1986. Energetics of humn muscle: exercise-induced ATP depletion. Magn. Reson.
Md.3: 44. WILKIE,D. R. 1986. Muscular fatigue: effects of hydrogen ions and inorganic phosphate. P d . Prsc. Fed. Am. SOC.Exp. Biol. 45: 292 1 -2923. WILSON,J. R., MCCULLY,K. K., MANCINI,D. M., BOBEN,B. P., and CHANCE, B. 1988. Relationship between muscle fatigue to pH and diprotonat& Pi in humans: a 31-P NMR study. J. Appl. Physid. 6%: 2333 -2339.
