Tricarboxylic acid cycle intermediate pool size and estimated cycle flux in human muscle during exercise

Transcription

1 Tricarboxylic acid cycle intermediate pool size and estimated cycle flux in human muscle during exercise MARTIN J. GIBALA, 1 DAVE A. MACLEAN, 1 TERRY E. GRAHAM, 2 AND BENGT SALTIN 1 1 Copenhagen Muscle Research Centre, Rigshospitalet, DK-2200 Copenhagen N, Denmark; and 2 Department of Human Biology and Nutritional Sciences, University of Guelph, Guelph, Ontario, Canada N1G 2W1 Gibala, Martin J., Dave A. MacLean, Terry E. Graham, and Bengt Saltin. Tricarboxylic acid cycle intermediate pool size and estimated cycle flux in human muscle during exercise. Am. J. Physiol. 275 (Endocrinol. Metab. 38): E235 E242, We examined the relationship between tricarboxylic acid (TCA) cycle intermediate (TCAI) pool size, TCA cycle flux (calculated from leg O 2 uptake), and pyruvate dehydrogenase activity (PDH a ) in human skeletal muscle. Six males performed moderate leg extensor exercise for 10 min, followed immediately by intense exercise until exhaustion ( min). The sum of seven measured TCAI ( TCAI) increased (P 0.05) from at rest to after 10 min and to mmol/kg dry wt at exhaustion. TCA cycle flux increased 70-fold during submaximal exercise and was 100-fold higher than rest at exhaustion. PDH a corresponded to 77 and 90% of TCA cycle flux during submaximal and maximal exercise, respectively. The present data demonstrate that a tremendous increase in TCA cycle flux can occur in skeletal muscle despite a relatively small change in TCAI pool size. It is suggested that the increase in TCAI during exercise may primarily reflect an imbalance between the rate of pyruvate production and its rate of oxidation in the TCA cycle. pyruvate dehydrogenase complex; muscle oxygen uptake; amino acids; metabolism THE FACTORS that influence the combined pool size of tricarboxylic acid (TCA) cycle intermediates (TCAI) and the control of cycle activity have been characterized in the isolated, perfused rat heart (31). Because the total concentration of TCAI appears related to the energy state of the myocardium (22), it has been suggested that the size of the TCAI pool serves a regulatory role in cardiac energy metabolism (21). Considerably less information is available regarding the concentrations of TCAI and the factors that regulate TCAI pool size in mammalian skeletal muscle. Indeed, only one investigation has attempted to measure the pool of TCAI in rodent muscle (3), and although several studies have reported changes in specific intermediates in humans (e.g., Refs. 11, 17, 36), only recently has an effort been made to quantify the exercise-induced changes in total TCAI pool size (12, 13). Overall, these studies demonstrated that the total concentration of TCAI can increase severalfold during moderate-to-intense muscle contraction, and peak expansion of the TCAI pool occurs within the initial few minutes of exercise (12). 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. However, the precise physiological significance of changes in TCAI pool size during muscle contraction remains speculative. Flux through the TCA cycle is believed to be primarily regulated through allosteric control of the three nonequilibrium enzymes in the cycle, citrate synthase, isocitrate dehydrogenase, and 2-oxoglutarate dehydrogenase (18, 28, 41). In addition, it has been suggested that an increase in the total concentration of TCAI may be necessary to obtain optimal energy provision under conditions of increased energy demand (25, 36). Several investigators have implied that changes in the total concentration of TCAI are indicative of the capacity for TCA cycle flux in skeletal muscle (35, 36), and two theories that link a decrease in TCAI pool size with peripheral muscle fatigue during prolonged exercise in humans have been proposed (36, 40). Although these hypotheses are interesting, there is in fact little experimental evidence to support them. Indeed, the fundamental relationship between TCAI pool size and TCA cycle flux has not been clearly established for human skeletal muscle. The primary purpose of the present investigation was therefore to examine the relationship between TCAI pool size and estimated TCA cycle flux in human skeletal muscle during exercise. It was hypothesized that a very large increase in TCA cycle flux can occur despite a relatively small increase in TCAI pool size. We utilized the one-legged knee extension exercise model, which essentially restricts work to the quadriceps femoris muscle (1) and permits TCA cycle turnover to be calculated from muscle O 2 uptake based on the Fick principle (6, 9). These methods were recently employed by Blomstrand et al. (6) in a study that compared estimated TCA cycle flux with the maximal in vitro activities of three enzymes in the TCA cycle, citrate synthase, succinate dehydrogenase, and 2-oxoglutarate dehydrogenase. The results from that study were intriguing to us, since the calculated TCA cycle flux was considerably higher than the maximal in vitro activity previously reported for the pyruvate dehydrogenase enzyme complex (PDH) in human skeletal muscle (e.g., Refs. 7, 8, 10, 29, 30). PDH controls the flux of pyruvate-derived acetyl-coa into the TCA cycle (40), and it is generally assumed that, during moderate-tointense exercise when most of the energy is derived from carbohydrate sources, PDH activity is at all times equal to or greater than the rate of flux through the TCA cycle (23). Therefore, a secondary purpose of the present study was to compare the calculated rate of TCA cycle flux with the measured active fraction of /98 $5.00 Copyright 1998 the American Physiological Society E235

2 E236 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE PDH (PDH a ) in biopsy samples obtained from the same muscle. METHODS Subjects. Six healthy males, with a mean age, height, and mass of yr, cm, and kg, respectively, volunteered for the investigation. Four subjects were involved in recreational sport activities (e.g., soccer and bicycling), but none were engaged in any form of regular physical training. The subjects were advised of the purposes and associated risks of the study and gave written informed consent. The experimental protocol was approved by the Ethical Committee for Copenhagen and Frederiksberg communities. Preexperimental procedures. The subjects were familiarized with the Krogh ergometer modified for one-legged knee extensor exercise as previously described (2). At least 3 days before the experiment, subjects performed an incremental exercise test with their dominant leg (kicking frequency 60/min) to determine the maximal exercise capacity of the knee extensors. This was defined as the highest workload that could be sustained while the desired kicking frequency was maintained. The mean peak workload for the group was 77 2 W. Subjects were instructed to consume their habitual diet and refrain from exercise or strenuous physical activity for 48 h before the experiment. Experimental protocol. Subjects arrived at the laboratory in the morning after an overnight fast. Teflon catheters were inserted into the femoral artery and vein of one leg, 2 cm proximal and distal to the inguinal ligament, respectively. A thermistor for measurement of venous blood temperature was inserted through the venous catheter, and the tip was advanced 8 cm proximal to the tip of the catheter. Subjects were moved to the exercise apparatus, where they rested supine for 30 min before the initiation of the exercise test. During this time, the area over the vastus lateralis muscle of the leg to be exercised was prepared for the extraction of muscle biopsy samples (5). The exercise protocol consisted of kicking at 60% of the predetermined one-legged maximal exercise capacity for 10 min, followed immediately by intense exercise at 100% of maximum until exhaustion (Exh), i.e., the point at which the subject could no longer maintain the desired kicking frequency of 60/min. The mean time to Exh, after the 10-min period of submaximal work, was 3.8 min (range min). Arterial and venous blood samples were drawn simultaneously at rest, after 5 and 10 min of exercise, and immediately before exhaustion. Measurements of leg blood flow using the thermodilution technique (2) were made several times at rest and immediately before and after each blood sample during exercise. During blood sampling and blood flow measurements, an occlusion cuff positioned just below the knee was inflated to 220 mmhg. Muscle biopsy samples were obtained at rest, after 5 and 10 min of exercise, and at exhaustion. Heart rate, pulmonary O 2 uptake, and ventilation were measured at rest and continually during exercise (MedGraphics CPX System, Klampenborg, Denmark). Blood analyses. Blood samples were drawn with heparinized syringes. O 2 saturation, O 2 content, and hemoglobin were measured using an OSM-2 hemoximeter (Radiometer, Copenhagen, Denmark). Whole blood lactate concentrations were immediately determined with a Yellow Springs lactate analyzer (Yellow Springs, OH). The remainder of the arterial and venous blood samples was centrifuged, and the supernatant was collected and stored at 80 C. Plasma samples were subsequently analyzed for ammonia (4) with a fluorometer (Perkin-Elmer model LS-50) and free amino acids by HPLC (20). Muscle analyses. Biopsy samples were immediately frozen ( 5 s) by plunging the needle into liquid nitrogen, removed from the needle while still frozen, and stored at 80 C. A 15- to 30-mg piece of muscle was chipped from each sample and used for the determination of the active fraction of PDH a by using the method of Constantin-Teodosiu et al. (8), as modified and described by Putman et al. (30). Total creatine concentrations (TCr) were measured in neutralized perchloric acid extracts of the PDH a homogenates by using a spectrophotometer (4). To correct for differences in blood or connective tissue between samples, PDH a values were adjusted to the highest TCr value in all the biopsy samples obtained from each subject, as described by Putman et al. (29). The mean TCr correction for all PDH a determinations was Due to limited tissue in biopsy samples obtained from three subjects at the 10-min time point, complete PDH a data for all six subjects were obtained only for the rest, 5 min, and Exh time points. The remaining portion of each biopsy sample was freeze dried, powdered to dissect out nonmuscle elements, and stored at 80 C. Aliquots of freeze-dried muscle were extracted with 0.5 M perchloric acid (containing 1 mm EDTA), neutralized with 2.2 M KHCO 3, and assayed for citrate, isocitrate, 2-oxoglutarate, succinate, fumarate, malate, and oxalacetate (4, 27, 36) by fluorometric procedures as previously described (13). Succinyl-CoA was not determined because of its extremely low concentration; however, it is likely that the remaining seven intermediates comprise 99% of total TCAI and thus provide a quantitative index of total pool size. A portion of the muscle extract was also used for the fluorometric determination of pyruvate (27). Calculations. Thigh volume was estimated based on Simpson s rule, which includes thigh length, three circumferences of the thigh, and three skinfold measurements (24), and muscle mass was estimated from a regression equation (2). On the basis of these calculations, the estimated active mass of the quadriceps muscles used during exercise was kg. The uptake and/or release of O 2, glucose, lactate, ammonia, and amino acids was calculated by multiplying the blood or plasma flow by the arteriovenous difference in concentration and expressed per kilogram wet weight of active muscle. Flux through the TCA cycle was calculated from the rate of muscle O 2 uptake (6, 9), assuming that carbohydrate was the only substrate being oxidized. Briefly, on the basis of the stoichiometry of the pathway for glucose oxidation, the flux through the TCA cycle is equivalent to one-third of the O 2 uptake. At a temperature of C, 1 mol of O 2 is equivalent to liters according to the general law for gases. Statistics. Cardiorespiratory, blood, and muscle metabolite data were analyzed using a one-factor (1 4; time) repeatedmeasures ANOVA. PDH a data were analyzed using a onefactor (1 3; time) repeated-measures ANOVA. Linear and polynomial regression analyses were used to examine the relationship between TCAI pool size and estimated TCA cycle flux. Statistical significance for all analyses was accepted as P 0.05, and significant main effects were further analyzed using a Tukey honestly significant difference post hoc test. Data are expressed as means SE unless otherwise noted. RESULTS Cardiorespiratory, leg blood flow, muscle O 2 uptake, and TCA cycle flux data. Changes in heart rate, pulmonary O 2 uptake, and ventilation during exercise are

3 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE E237 Table 1. Cardiorespiratory, blood flow, muscle O 2 uptake, and estimated TCA cycle flux data at rest and during exercise Rest 5 min 10 min Exh Heart rate, beats/min * 102 5* 137 4* Pulmonary V O2, l/min * * * Ventilation, l/min * * * Muscle blood flow, l min 1 kg * * * Muscle V O2,mlO 2 min 1 kg * * * TCA cycle flux, mmol min 1 kg * * * Values are means SE; n 6. Muscle blood flow, muscle V O2, and tricarboxylic acid (TCA) cycle flux data are expressed per wet weight of quadriceps muscle. V O2, rate of O 2 consumption; Exh, exhaustion. *P 0.05 vs. rest. P 0.05 vs. 10 min. summarized in Table 1. Muscle blood flow was 27- and 35-fold higher (P 0.05) than at rest during the submaximal exercise period and at Exh, respectively (Table 1). Muscle O 2 uptake and the estimated flux through the TCA cycle increased (P 0.05) 70-fold above rest during the submaximal exercise period and reached values 100-fold higher than rest at Exh (Table 1). Blood and plasma metabolites and flux data. Changes in the arterial concentrations of lactate, ammonia, alanine, glutamine, and glutamate are summarized in Table 2. The arterial lactate concentration and lactate release were higher (P 0.05) at all times during exercise compared with rest, and the respective values at Exh were also higher (P 0.05) compared with values at 5 and 10 min of exercise (Table 2). Ammonia release was higher (P 0.05) at all times during exercise compared with rest; however, the efflux at Exh was approximately fivefold higher (P 0.05) compared with values at 5 and 10 min of exercise (Table 2). The release of alanine and glutamine was also higher (P 0.05) during exercise compared with rest, but the respective values at Exh were not significantly different compared with values at 5 and 10 min of exercise (Table 2). Glutamate was the only amino acid that was taken up in significant amounts during exercise, and Table 2. Arterial concentrations and net uptake/release of metabolites and some plasma amino acids at rest and during exercise Rest 5 min 10 min Exh Arterial concentration Lactate, mm * * * Ammonia, µm * Alanine, µm * Glutamine, µm Glutamate, µm * 41 3* 41 3* Net uptake/release Lactate * * * Ammonia * 18 4* * Alanine * 83 17* * Glutamine * 78 14* 89 11* Glutamate * 11 2* 18 4* Values are means SE; n 6. Net uptake/release values are in µmol min 1 kg wet wt 1 except for lactate which is in mmol min 1 kg wet wt 1. A negative value indicates a release. *P 0.05 vs. rest. P 0.05 vs. 10 min. the uptake at Exh was also higher (P 0.05) compared with values at 5 and 10 min of exercise (Table 2). Intramuscular TCA cycle intermediates. The total intramuscular concentration of the seven measured TCAI ( TCAI) increased (P 0.05) approximately onefold above rest levels during the submaximal exercise period (Fig. 1). The TCAI showed a further increase after the period of maximal work, and the value at Exh was approximately threefold higher (P 0.05) than at rest (Fig. 1). There was a significant positive correlation between total concentration of TCAI and the calculated flux through the TCA cycle (Fig. 2). The explained variance was not different when this relationship was described based on a linear ( y 0.873x 1.024; r ; P ) or polynomial ( y 0.072x x 1.199; r ; P ) regression analysis; however, the pooled data suggest that a curvilinear relationship exists between TCAI pool size and TCA cycle flux (Fig. 3). The changes in individual TCAI during exercise are summarized in Table 3. Citrate, isocitrate, fumarate, malate, and oxalacetate were higher (P 0.05) at all times during exercise compared with rest, whereas succinate was only higher (P 0.05) at Exh compared with rest. The concentrations of malate, fumarate, Fig. 1. Total muscle concentration of 7 measured tricarboxylic acid (TCA) cycle intermediates (TCAI; citrate, isocitrate, 2-oxoglutarate, succinate, fumarate, malate, and oxalacetate) at rest, after 5 and 10 min of submaximal exercise (60% of maximum), and at exhaustion (Exh) after maximal dynamic knee extensor exercise. Values are means SE; n 6. *P 0.05 vs. rest. P 0.05 vs. 10 min.

4 E238 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE Fig. 2. Relationship between TCAI pool size and calculated TCA cycle flux. Individual data are plotted for each subject at rest and during exercise. Values are means SE; n 6. Explained variance was not different when relationship was described based on a linear (r ) or polynomial (r ) regression analysis. See RESULTS for further explanation. succinate, and isocitrate were also higher (P 0.05) at Exh compared with 5 and 10 min of exercise. 2-Oxoglutarate was the only TCAI that decreased during exercise and was lower (P 0.05) at all times compared with rest. The overall changes in TCAI pool size during exercise were primarily due to increases in the concentration of malate. This intermediate accounted for over one-half of the increase in pool size during each transition in workload and, at Exh, was more than sixfold higher than at rest (Table 3). Succinate and fumarate also demonstrated relatively large concentration changes during exercise, and together these three TCAI accounted for a progressively larger portion of the total pool size at each work intensity; i.e., the total concentration of malate, fumarate, and succinate comprised 61 4% of total pool size at rest, but this proportion increased to 67 3 and 68 2% after 5 and 10 min of Fig. 3. Relationship between TCAI pool size and TCA cycle flux at rest and during exercise. Values are means SE; n 6. submaximal exercise, respectively, and was 81 2% at Exh. Thus, in both an absolute and relative sense, most of the anaplerotic carbon that entered TCA cycle was directed toward increasing the TCAI in the second span of the cycle, i.e., the span from 2-oxoglutarate to oxalacetate (Fig. 4). PDH a and intramuscular pyruvate. PDH a increased progressively during exercise and was 3.6- and 7.1-fold higher than at rest after 5 min of exercise and at Exh, respectively (Fig. 5). PDH a at rest ( mmol kg wet wt 1 min 1 ) was severalfold higher than TCA cycle flux; however, PDH a after 5 min of exercise ( ) and at Exh ( ) corresponded to 77 and 90% of the estimated TCA cycle flux, respectively. Intramuscular pyruvate was higher (P 0.05) at all times during exercise compared with rest, and the value at Exh was also higher (P 0.05) compared with 5 and 10 min of exercise (Fig. 6). DISCUSSION The present results demonstrate that there is a significant, positive correlation between TCAI pool size and estimated TCA cycle flux in human skeletal muscle during dynamic leg extensor exercise. Although a previous study (35) demonstrated that the total intramuscular concentration of citrate, malate and fumarate increased progressively during incremental cycle exercise, this is the first study to directly compare changes in TCAI pool size with estimates of TCA cycle flux in mammalian skeletal muscle. The relative changes in these variables were very different, and the present data illustrate that a tremendous increase in TCA cycle flux can occur despite only a modest elevation in the total concentration of TCAI. For example, during the transition from rest to submaximal exercise, the rate of TCA cycle flux increased by 70-fold above rest, whereas the total concentration of TCAI only doubled. In addition, these data suggest that the relationship between TCAI pool size and TCA cycle flux is not linear (Fig. 3), since the ratio of the change in TCAI pool size to change in cycle flux (i.e., TCAI pool size/ cycle flux) was larger during the second transition in workload; i.e., when the exercise intensity was increased from the submaximal to maximal workload, TCA cycle flux increased another 30-fold above rest values (or one-half of initial relative increase during first transition in workload), and yet the TCAI pool size increased a further 2-fold. It must be emphasized that a correlation between two variables does not imply cause and effect, and these data should not be interpreted to suggest that an increase in the total concentration of TCAI is necessary to increase cycle flux. Indeed, the mechanisms that control steady-state flux through the TCA cycle are extremely complex (18, 28, 41), and it remains speculative whether the size of the TCAI pool plays an important physiological role in this regard. Several investigators (35, 36, 39) have implied that changes in the total concentration of TCAI may be indicative of the capacity for TCA cycle flux in human skeletal muscle during exercise. However, although there is evidence to suggest that anaplerotic reactions may play a role in the

5 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE E239 Table 3. Intramuscular concentrations of individual TCA cycle intermediates at rest and during exercise Rest 5 min 10 min Exh Citrate * * * Isocitrate * * * 2-Oxoglutarate * * * Succinate * Fumarate * * * Malate * * * Oxalacetate * * * Values are means SE in mmol/kg dry wt; n 6. *P 0.05 vs. rest. P 0.05 vs. 10 min. maintenance of contractile function in isolated rat hearts oxidizing acetoacetate (32, 33), no such data have been presented for mammalian skeletal muscle. To properly investigate this question, it will be necessary to manipulate levels of TCAI and to determine what effect, if any, this has on TCA cycle flux, oxidative energy metabolism, and skeletal muscle function. Nonetheless, the present findings demonstrate that there is a strong correlation between changes in TCAI pool size and TCA cycle flux. There appears to be a curvilinear relationship between these variables so that, at high work intensities, there is a disproportionate increase in TCAI pool size relative to the change in TCA cycle flux (Fig. 3). One interpretation of these data is that, during the transition from rest to moderate exercise, only a small increase in TCAI (relative to resting concentrations) is necessary to accommodate a relatively large change in TCA cycle flux. However, at more intense workloads, the TCAI pool must expand to a larger extent to sustain smaller increments in TCA cycle flux. Alternatively, it must also be considered that the increase in TCAI may not represent an important regulatory signal but may simply be a consequence of the huge increase in metabolic flux which occurs during exercise. Therefore, another possible explanation for anaplerosis is that the increase in TCAI during contraction primarily represents a sink for pyruvate when its rate of formation from glycolysis exceeds its rate of oxidation in the TCA cycle. This interpretation is analogous to the mass action theory proposed to explain lactate accumulation in skeletal muscle during exercise (for review, see Ref. 14). An increase in the concentration of pyruvate appears necessary for anaplerosis, since many of the reactions that could lead to a net influx of TCAI are directly or indirectly dependent on the level of pyruvate, including those catalyzed by alanine aminotransferase, phosphoenolpyruvate carboxykinase, pyruvate carboxylase, and malic enzyme. Although the precise mechanisms responsible for anaplerosis have not been completely resolved, the alanine aminotransferase reaction (pyruvate glutamate = 2-oxoglutarate alanine) appears quantitatively most important for the increase in TCAI at the onset of exercise in humans (12, 36). It therefore might be expected that an increase in TCAI should occur only under conditions in which the rate of pyruvate production from glycolysis exceeds its rate of oxidation in the TCA cycle. Sahlin and coworkers (34, 35) have presented evidence, in separate studies, suggesting that an increase in TCAI only takes place at exercise intensities which result in an elevation of muscle pyruvate. When healthy subjects cycled at a very low workload [ 25% maximum O 2 consumption (V O2max )] that did not cause an increase in intramus- Fig. 4. Total concentration of TCAI in span I of TCA cycle [i.e., citrate (Cit), isocitrate (Iso), and 2-oxoglutarate (2OG)] and span II of cycle [i.e., succinate (Suc), fumarate (Fum), malate (Mal), and oxalacetate (Oxa)] at rest, after 5 min of submaximal exercise (60% of maximum), and at exhaustion after maximal dynamic knee extensor exercise. Values are means SE; n 6. Fig. 5. Measured active fraction of pyruvate dehydrogenase (PDH a ) and estimated flux through TCA cycle at rest, after 5 min of submaximal exercise (60% of maximum), and at exhaustion after maximal dynamic knee extensor exercise. Values are means SE; n 6. *P 0.05 vs. rest. P 0.05 vs. 5 min.

6 E240 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE Fig. 6. Intramuscular pyruvate at rest and during exercise. Values are means SE; n 6. *P 0.05 vs. rest. P 0.05 vs. 10 min. cular pyruvate (34), there was no significant increase in the intramuscular concentration of citrate, malate, or fumarate (35). However, when subjects cycled at workloads corresponding to 50 and 80% of V O2max, there were significant elevations in both pyruvate (34) and the three measured TCAI (35). These authors have also shown that patients with McArdle s disease, who lack the enzyme glycogen phosphorylase and cannot produce pyruvate from glycogen, display markedly attenuated increases in these TCAI during exercise (35). Finally, Spencer et al. (37) demonstrated that epinephrine infusion caused a significant elevation in pyruvate in resting human muscle, and this was associated with a doubling of intramuscular citrate, malate, and fumarate. Two spans of TCA cycle. The relative distribution of the individual TCAI during exercise also deserves comment. Malate consistently demonstrates the largest quantitative change of any measured TCAI during exercise (12, 13, 35, 36) and, in the present study, accounted for more than one-half of the net increase in TCAI pool size during each transition in workload. In addition to malate, fumarate and succinate also demonstrated relatively large concentration changes during exercise, and together these three TCAI accounted for a progressively larger portion of total pool size at each work intensity. Although the precise mechanisms responsible for the large increase in these three TCAI are not clear, the present data confirm the findings of our previous investigations (12, 13) and demonstrate that the vast majority of anaplerotic carbon which enters the TCA cycle during exercise is directed toward increasing the concentrations of the intermediates in the second span of the cycle (Fig. 4). The disproportionate increase in the concentrations of malate, fumarate, and succinate can be reconciled if one considers the TCA cycle to consist of two physiological pathways: the span from acetyl-coa to 2-oxoglutarate and the span from 2-oxoglutarate to oxalacetate (26, 31). As noted by Newsholme and Leech (26), in skeletal muscle during sustained exercise, this division of the cycle is only academic, since the flux through the two pathways must be identical and regulated in a concerted manner. Nonetheless, this division might explain how it is possible for carbon skeletons to feed into the cycle, e.g., at the level of 2-oxoglutarate through the alanine aminotransferase reaction (12), and accumulate at the level of malate, fumarate, and succinate. In principle, the carbon that enters could also be removed from the cycle at the level of malate via the reaction catalyzed by malic enzyme or at the level of oxalacetate via phosphoenolpyruvate carboxykinase (3, 38). However, the net accumulation of TCAI during exercise indicates that the removal of intermediates through these pathways was much slower than the rate of influx into the cycle. This could be due to the fact that the potential egress of TCAI through the reactions catalyzed by malic enzyme and phosphoenolpyruvate carboxykinase was prevented by the elevated intramuscular pyruvate concentration during exercise (Fig. 6). PDH a and TCA cycle flux. A second point that emerged from the present study was that PDH a at Exh after maximal leg extensor exercise, when the enzyme complex was likely fully transformed to its more active form (29), corresponded to only 90% of the calculated TCA cycle flux. This finding confirms the observation, made a priori based on data from separate studies, that the calculated maximal rate of flux through the TCA cycle in vivo (6) is higher than the maximal reported activity for PDH in vitro (e.g., Refs. 7, 8, 10, 29, 30). These observations are perplexing, given that, during moderate-to-intense exercise when most of the energy is derived from carbohydrate sources, it is generally assumed that PDH a is at all times equal to or greater than the rate of flux though the TCA cycle in skeletal muscle (23). Indeed, the explanation for the welldocumented exercise-induced rise in acetylcarnitine concentration (7, 19, 30, 34) is that PDH flux exceeds TCA cycle flux, and carnitine functions to buffer the excess formation of acetyl groups generated from pyruvate through PDH. In doing so, carnitine serves to prevent the depletion of the mitochondrial CoASH pool, which would otherwise inhibit flux though the TCA cycle at the level of 2-oxoglutarate dehydrogenase (7, 19). Although this important role for carnitine is widely accepted, as noted earlier, this theory is not well supported by existing literature values for PDH a and TCA cycle flux in human muscle during maximal exercise. The present study is the first attempt to directly compare estimates of TCA cycle flux with measurements of PDH a in biopsy samples obtained from the same muscle, and our data are consistent with existing literature values for these variables. The rate of TCA cycle flux that we calculated at Exh was 10% lower than that reported by Blomstrand et al. (6) for maximal dynamic knee extensor exercise, mainly because our estimated quadriceps muscle mass (based on anthropometric measurements) was higher than that reported in their study (which utilized computer tomography). The values that we obtained for PDH a at Exh are similar to (10) or slightly higher than those previously reported for human skeletal muscle during intense dynamic exercise (29, 30), obtained with the same

7 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE E241 analytical procedures. The present data must therefore be considered valid and highlight the fact that although the reported changes in acetylcarnitine concentration (7, 19, 30, 34) suggest otherwise, in vitro measurements of PDH a in human muscle during maximal exercise are nonetheless lower than calculated rates of TCA cycle flux based on muscle O 2 uptake. There are a number of explanations that could account for this discrepancy, including the accuracy of our analytical techniques. It is possible that the combined experimental variability associated with the measurement of both leg blood flow (which is used to derive muscle O 2 uptake and TCA cycle flux) (2) and the determination of PDH a (8) is equal to or greater than the 10% difference observed between these variables in the present study. A second possibility is that measurements of PDH a are made under optimal in vitro conditions and in some situations may not represent the actual in vivo flux, since the latter is also determined in part by the availability of substrates and accumulation of products (30, 40). For example, when Putman et al. (30) attempted to calculate the in vivo rate of flux through PDH a during intense cycle exercise from measurements or estimates of glycogen degradation, glucose uptake, lactate production, leg blood flow, and active muscle mass, the value which they obtained was 19% higher than their in vitro measurement of PDH a. It could also be argued that the use of leg O 2 uptake may slightly overestimate the rate of TCA cycle flux in the quadriceps muscles, particularly during maximal work, since a substantial fraction of the limb blood flow is perfusing not only the knee extensors but also the hamstring and gluteal muscles (which may not be completely relaxed). Our subjects practiced the knee extensor exercise before the experiment to minimize this possibility, and it has previously been shown that familiarized subjects demonstrate little or no electromyographic activity in their noninvolved leg muscles (1). In addition, it has recently been demonstrated that derived measurements of muscle temperature reveal no elevation in the hamstrings, whereas the knee extensor muscles increase 1.0 C during this type of exercise (15). Finally, if we assume our measurements of PDH a and leg O 2 uptake are precise indicators of in vivo fluxes, then it must be considered that, even during maximal leg extensor exercise, a small portion of the acetyl-coa which entered the TCA cycle was derived from noncarbohydrate sources (e.g., intramuscular triglycerides or amino acid catabolism). Ammonia metabolism during exercise. A final observation from the present study concerned the relative changes in the flux of ammonia and the amino carriers, alanine and glutamine. The efflux of all three compounds was higher after 5 and 10 min of submaximal exercise compared with rest; however, the arterial concentrations did not change, indicating that their clearance (e.g., by liver) was well matched to muscle release. In addition, the increase in ammonia efflux was much smaller than the relative changes in glutamine and alanine release during the submaximal exercise period. However, at exhaustion after the period of maximal work, the arterial ammonia concentration doubled and ammonia release increased another fourfold, whereas the arterial concentrations and efflux of alanine and glutamine were unchanged compared with the respective values during submaximal exercise. The relative changes in the efflux of glutamine and ammonia at the two workloads may be due to the energetics of the glutamine synthase reaction (glutamate ammonia ATP = glutamine ADP P i ). Although this reaction is quantitatively important for the clearance of muscle ammonia during submaximal exercise (16), the data from the present investigation suggest that this route of clearance was inhibited during maximal exercise. This could be due to the fact that the glutamine synthase reaction is energy requiring or possibly that there were other demands on the intramuscular glutamate pool. Conclusions. In summary, the results from the present investigation demonstrate there is a significant, positive correlation between the total concentration of TCAI and estimated TCA cycle flux in human skeletal muscle during exercise. The relative changes in TCAI pool size and cycle turnover rate were very different, however, and these data demonstrate that a tremendous increase in TCA cycle flux can occur despite a relatively small change in the total concentration of TCAI. The measured active fraction of PDH corresponded to 77 and 90% of the estimated flux through the TCA cycle during submaximal and maximal knee extensor exercise, respectively. Most of the anaplerotic carbon that entered the cycle during exercise was directed toward increasing the concentration of malate, and, overall, the TCAI in the second span of the TCA cycle (i.e., malate, fumarate, and succinate) accounted for a progressively larger portion of the TCAI pool at each work intensity. It remains speculative whether the increase in TCAI during exercise is important to augment cycle flux or is simply a consequence of the increase in pyruvate concentration that occurs when its rate of formation from glycolysis exceeds its rate of oxidation in the TCA cycle. The authors thank Karin Juel, Charlotte Mortensen, Carsten Nielsen, Maureen Odland, Lynda Powell, Premila Sathasivam, and Hana Villmusen for excellent technical assistance. We also thank our subjects for time and tremendous effort. This project was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Danish National Research Foundation. M. J. Gibala was the recipient of an NSERC Postgraduate Scholarship, and D. A. MacLean was supported by a Medical Research Council of Canada Postdoctoral Fellowship. Address for reprint requests: M. J. Gibala, Copenhagen Muscle Research Centre, Rigshospitalet, Section 7652, Tagensvej 20, DK Copenhagen N, Denmark. Received 10 February 1998; accepted in final form 21 April REFERENCES 1. Anderson, P., R. P. Adams, G. Sjogaard, A. Thorboe, and B. Saltin. Dynamic knee extension as model for study of isolated exercising muscle in humans. J. Appl. Physiol. 59: , Anderson, P., and B. Saltin. Maximal perfusion of skeletal muscle in man. J. Physiol. (Lond.) 366: , 1985.

8 E242 ANAPLEROSIS AND TCA CYCLE FLUX IN HUMAN SKELETAL MUSCLE 3. Aragón, J. J., and J. M. Lowenstein. The purine nucleotide cycle. Comparison of the levels of citric acid cycle intermediates with the operation of the purine nucleotide cycle in rat skeletal muscle during exercise and recovery from exercise. Eur. J. Biochem. 110: , Bergmeyer, H. U. Methods of Enzymatic Analysis. New York: Academic, Bergström,J.Percutaneous needle biopsy of skeletal muscle in physiological and clinical research. Scand. J. Clin. Lab. Invest. 35: , Blomstrand, E., G. Rådegran, and B. Saltin. Maximum rate of oxygen uptake by human skeletal muscle in relation to maximal activities of enzymes in the Krebs cycle. J. Physiol. (Lond.) 501: , Constantin-Teodosiu, D., J. J. Carlin, G. Cederblad, R. Harris, and E. Hultman. Acetyl group accumulation and pyruvate dehydrogenase activity in human muscle during incremental exercise. Acta Physiol. Scand. 143: , Constantin-Teodosiu, D., G. Cederblad, and E. Hultman. A sensitive radioisotopic assay of pyruvate dehydrogenase complex in human muscle tissue. Anal. Biochem. 198: , Cooney, G. J., H. Taegtmeyer, and E. A. Newsholme. Tricarboxylic acid cycle flux and enzyme activities in the isolated working rat heart. Biochem. J. 200: , Dyck, D. J., S. J. Peters, P. S. Wendling, A. Chesley, E. Hultman, and L. L. Spriet. Regulation of muscle glycogen phosphorylase activity during intense aerobic cycling with elevated FFA. Am. J. Physiol. 270 (Endocrinol. Metab. 33): E116 E125, Essén, B., and L. Kaijser. Regulation of glycolysis in intermittent exercise in man. J. Physiol. (Lond.) 281: , Gibala, M. J., D. A. MacLean, T. E. Graham, and B. Saltin. Anaplerotic processes in human skeletal muscle during brief dynamic exercise. J. Physiol. (Lond.) 502: , Gibala, M. J., M. A. Tarnopolsky, and T. E. Graham. Tricarboxylic acid cycle intermediates in human muscle at rest and during prolonged cycling. Am. J. Physiol. 272 (Endocrinol. Metab. 35): E239 E244, Gladden, L. B. Lactate transport and exchange during exercise. In: Handbook of Physiology. Exercise: Regulation and Integration of Multiple Systems. Bethesda, MD: Am. Physiol. Soc., 1996, sect. 12, chapt. 14, p Gonzáles-Alonso, J., J. Bangsbo, P. Krustrup, B. Quistorff, and B. Saltin. Mechanical efficiency alterations during intense dynamic exercise in humans (Abstract). FASEB J. 12: A1115, Graham, T. E., B. Kiens, M. Hargreaves, and E. A. Richter. Influence of fatty acids on ammonia and amino acid flux from active human muscle. Am. J. Physiol. 261 (Endocrinol. Metab. 24): E168 E176, Graham, T. E., and B. Saltin. Estimation of the mitochondrial redox state in human skeletal muscle during exercise. J. Appl. Physiol. 66: , Hansford, R. G. Control of mitochondrial substrate oxidation. Curr. Top. Bioenerg. 10: , Harris, R. C., C. V. L. Foster, and E. Hultman. Acetylcarnitine formation during intense muscular contraction in humans. J. Appl. Physiol. 63: , Heinrickson, R. L., and S. C. Meredith. Amino acid analysis by reverse-phase high-performance liquid chromatography: precolumn derivatization with phenylisothiocyanate. Anal. Biochem. 136: 65 74, Hiltunen, J. K., and E. J. Davis. The disposition of citric acid cycle intermediates by isolated rat heart mitochondria. Biochim. Biophys. Acta 678: , Hiltunen, J. K., and I. E. Hassinen. Energy-linked regulation of the citric acid cycle and the pool size of the cycle intermediates in the isolated perfused rat heart. J. Biochem. 8: , Hultman, E. Pyruvate dehydrogenase as a regulator of substrate utilization in skeletal muscle. In: Biochemistry of Exercise IX, edited by R. J. Maughan and S. M. Sherreffs. Champaign, IL: Human Kinetics, 1996, p Jones, P. R. M., and J. Pearson. Anthropometric determination of leg fat and muscle plus bone volumes in young male and female adults (Abstract). J. Physiol. (Lond.) 204: 36P, Lee, S.-H., and E. J. Davis. Carboxylation and decarboxylation reactions. Anaplerotic flux and removal of citrate cycle intermediates. J. Biol. Chem. 254: , Newsholme, E. A., and A. R. Leech. Biochemistry for the Medical Sciences. New York: Wiley, Passoneau, J. V., and O. H. Lowry. Enzymatic Analysis: A Practical Guide. Totowa, NJ: Humana, Peuhkurinen, K. J. Regulation of the tricarboxylic acid cycle pool size in heart muscle. J. Mol. Cell. Cardiol. 16: , Putman, C. L., L. L. Spriet, E. Hultman, D. J. Dyck, and G. J. F. Heigenhauser. Skeletal muscle pyruvate dehydrogenase activity during acetate infusion in humans. Am. J. Physiol. 268 (Endocrinol. Metab. 31): E1007 E1017, Putman, C. L., L. L. Spriet, E. Hultman, M. I. Lindinger, L. C. Lands, R. S. McKelvie, G. Cederblad, N. L. Jones, and G. J. F. Heigenhauser. Pyruvate dehydrogenase activity and acetyl group accumulation during exercise after different diets. Am. J. Physiol. 265 (Endocrinol. Metab. 28): E752 E760, Randle, P. J., P. J. England, and R. M. Denton. Control of the tricarboxylate cycle and its interactions with glycolysis during acetate utilization in rat heart. Biochem. J. 117: , Russell, R. R., and H. Taegtmeyer. Changes in citric acid cycle flux and anaplerosis antedate the functional decline in isolated rat hearts utilizing acetoacetate. J. Clin. Invest. 87: , Russell, R. R., and H. Taegtmeyer. Pyruvate carboxylation prevents the decline in contractile function of rat hearts oxidizing acetoacetate. Am. J. Physiol. 261 (Heart Circ. Physiol. 30): H1756 H1762, Sahlin, K. Muscle carnitine metabolism during incremental dynamic exercise in humans. Acta Physiol. Scand. 138: , Sahlin, K., L. Jorfeldt, K.-G. Henriksson, S. R. Lewis, and R. G. Haller. Tricarboxylic acid cycle intermediates during incremental exercise in healthy subjects and in patients with McArdle s disease. Clin. Sci. Lond. 88: , Sahlin, K., A. Katz, and S. Broberg. Tricarboxylic acid cycle intermediates in human muscle during prolonged exercise. Am. J. Physiol. 259 (Cell Physiol. 28): C834 C841, Spencer, M. K., A. Katz, and I. Raz. Epinephrine increases tricarboxylic acid cycle intermediates in human muscle. Am. J. Physiol. 260 (Endocrinol. Metab. 23): E436 E439, Spydevold, Ø., E. J. Davis, and J. Bremer. Replenishment and depletion of citric acid cycle intermediates in skeletal muscle. Eur. J. Biochem. 71: , Wagenmakers, A. J. M., J. H. Coakley, and R. H. T. Edwards. Metabolism of branched-chain amino acids and ammonia during exercise: clues from McArdle s disease. Int. J. Sports Med. 11: S101 S113, Wieland, O. H. The mammalian pyruvate dehydrogenase complex: structure and regulation. Rev. Physiol. Biochem. Pharmacol. 96: , Williamson, J. R., and R. H. Cooper. Regulation of the citric acid cycle in mammalian systems. FEBS Lett. 117: K73 K85, 1980.