Journal of Applied Physiology
Vol. 83, No. 4, pp. 1104-1109, October 1997
EXERCISE AND MUSCLE

Effect of exercise intensity on skeletal muscle malonyl-CoA and acetyl-CoA carboxylase

B. B. Rasmussen and W. W. Winder

Department of Zoology, Brigham Young University, Provo, Utah 84602

ABSTRACT

INTRODUCTION

MATERIALS AND METHODS

RESULTS

DISCUSSION

ACKNOWLEDGEMENTS

FOOTNOTES

REFERENCES

ABSTRACT

Rasmussen, B. B., and W. W. Winder. Effect of exercise intensity on skeletal muscle malonyl-CoA and acetyl-CoA carboxylase. J. Appl. Physiol. 83(4): 1104-1109, 1997.Malonyl-CoA is synthesized by acetyl-CoA carboxylase (ACC) and is an inhibitor of fatty acid oxidation. Exercise induces a decline in skeletal muscle malonyl-CoA, which is accompanied by inactivation of ACC and increased activity of AMP-activated protein kinase (AMPK). This study was designed to determine the effect of exercise intensity on the enzyme kinetics of ACC, malonyl-CoA levels, and AMPK activity in skeletal muscle. Male Sprague-Dawley rats were killed (pentobarbital sodium anesthesia) at rest or after 5 min of exercise (10, 20, 30, or 40 m/min at 5% grade). The fast-twitch red and white regions of the quadriceps muscle were excised and frozen in liquid nitrogen. A progressive decrease in red quadriceps ACC maximal velocity (from 28.6 ± 1.5 to 14.3 ± 0.7 nmol · g¹ · min¹, P < 0.05), an increase in activation constant for citrate, and a decrease in malonyl-CoA (from 1.9 ± 0.2 to 0.9 ± 0.1 nmol/g, P < 0.05) were seen with the increase in exercise intensity from rest to 40 m/min. AMPK activity increased more than twofold. White quadriceps ACC activity decreased only during intense exercise. We conclude that the extent of ACC inactivation during short-term exercise is dependent on exercise intensity.

fatty acid oxidation; quadriceps; adenosine 5-monophosphate-activated protein; coenzyme A

INTRODUCTION

MALONYL-COA IS SYNTHESIZED by acetyl-CoA carboxylase (ACC) and is an inhibitor of fatty acid oxidation in liver and cardiac muscle (1, 10, 19). Exercise induces a decline in skeletal muscle malonyl-CoA levels, which is accompanied by a decrease in the maximal velocity (V_max) of ACC, an increase in the citrate concentration required for half-maximal activation of ACC (K_0.5), and increased activity of AMP-activated protein kinase (AMPK) (30). These changes in kinetic properties mirror those caused by in vitro phosphorylation of ACC by AMPK (30, 32). Carnitine palmitoyltransferase (CPT-1) catalyzes the first step specific to mitochondrial long-chain fatty acid oxidation and is inhibited by malonyl-CoA in skeletal muscle (13). The importance of CPT-1 in regulating the rate of fatty acid oxidation is emphasized by the fact that when palmitate is used as substrate, isolated mitochondria show a seven- to eightfold increase in oxygen consumption when carnitine is added to the reaction mix along with coenzyme A (15). Inactivation of ACC results in a decrease in malonyl-CoA, thus relieving the inhibition of CPT-1. Regulation of ACC and CPT-1 may influence the proportion of intramuscular fatty acids oxidized as an energy source vs. the amount stored as triacylglycerol.

Exercise (28, 30), electrical stimulation (6, 12), and fasting (31) lower malonyl-CoA levels in skeletal muscle. Glucose infusion during exercise attenuates this decrease in malonyl-CoA (8), whereas incubation of soleus muscles or perfusion of the rat hindlimb muscles with insulin and glucose increases malonyl-CoA content (7, 20). Increases in malonyl-CoA are also seen after muscle denervation (20). During fasting or with prolonged submaximal exercise, lipid becomes the preferred energy source (21), and malonyl-CoA decreases (28). Thus ACC and malonyl-CoA may play a vital role in the determination of the substrate oxidation mix in muscle.

Prior work in our laboratory has shown that these kinetic changes in muscle ACC are associated with contraction (12) rather than hormonally mediated via epinephrine (29). Activation of AMPK with consequent phosphorylation and inactivation of ACC appear to be triggered by events associated with muscle contraction. This study was designed to investigate the effects of increasing running speed on activation of AMPK, inactivation of ACC, and the decrease in malonyl-CoA in skeletal muscle.

MATERIALS AND METHODS

Animal care. Male Sprague-Dawley rats (Sasco, Omaha, NE) were housed in individual cages at a temperature of 19-21°C in a light-controlled (12:12-h light-dark cycle) room and were fed a normal diet (Harlan Teklad rodent diet, Madison, WI) and water ad libitum. Rats were run on a rodent treadmill at various speeds (10-40 m/min) up a 5% grade (10 min/day) for 5-10 days to accustom them to running on the treadmill and to handling procedures. Jugular catheters were implanted, with rats under ether anesthesia, 3 days before the final exercise test. Rats weighed 382 ± 4 g at the time of death.

Exercise test. Rats were anesthetized by injection of pentobarbital sodium via the catheter at rest or after running for 5 min on the treadmill at 10, 20, 30, or 40 m/min up a 5% grade. Previous work in our laboratory has shown that malonyl-CoA drops significantly after 5 min of exercise (28). The 5-min duration was selected so that rats running at 40 m/min could run for the same period of time as rats run at lower speeds. The red and white regions of the quadriceps from both hindlimbs were removed rapidly (90-180 s after anesthetization) and frozen with aluminum block tongs at liquid nitrogen temperature. Blood was removed via the abdominal aorta. An aliquot was heparinized for collection of plasma for determination of free fatty acids (FFA; see Ref. 16). A perchloric acid extract of blood was made for measurement of glucose (2) and lactate (9).

Muscle assays. Muscle samples were kept under liquid nitrogen until analyzed. Red and white quadriceps muscles were ground separately to a powder under liquid nitrogen. For ACC and AMPK assays, the frozen powder was weighed and then homogenized in a buffer containing (in mM) 100 mannitol, 50 NaF, 10 tris(hydroxymethyl)aminomethane, 1 EDTA, 10 -mercaptoethanol, pH 7.5, and proteolytic enzyme inhibitors (5.0 ml/l aprotinin, 5.0 mg/l leupeptin, and 5.0 mg/l anti-trypsin). The homogenate was immediately centrifuged at 48,000 g for 20 min. The ACC and AMPK were precipitated from the supernatant by addition of 144 mg ammonium sulfate/ml and by stirring for 30 min on ice. The precipitate was collected by centrifugation at 48,000 g for 20 min. The pellet was dissolved in 10% of the original volume of the homogenate buffer and was centrifuged again to remove insoluble protein.

The supernatant was used for determination of ACC and AMPK activity. ACC activity was determined at citrate concentrations varied between 0 and 20 mM by measuring the rate of incorporation of [¹⁴C]bicarbonate into malonyl-CoA, an acid-stable compound, at 37°C for 10 min, as previously described (31). The data were fitted to the Hill equation by using the Grafit program (Sigma), which allows for the determination of the maximal activity as a function of citrate concentration (V_max), K_0.5, and the activation constant for citrate (K_a). AMPK activity was determined by using the SAMS peptide by the method described previously (5, 30).

Neutralized perchloric acid extracts of the red and white quadriceps muscles were used for the determination of malonyl-CoA (14). Fatty acid synthetase for malonyl-CoA assays was isolated from livers of fasted/refed rats by the procedure described by Linn (cf. Ref. 28). Glycogen content of both muscles was determined by using the anthrone method (11).

Results are expressed as means ± SE. Analysis of variance and Fisher's least significant difference (as a post hoc test) were used to determine statistical differences (P < 0.05) between treatment groups.

RESULTS

Malonyl-CoA decreased significantly from resting values (P < 0.05) at each treadmill speed in red quadriceps muscle after 5 min of exercise, and no differences were detected in the white quadriceps muscle after 5 min of exercise (Fig. 1).

Figure 2 shows the effect of various exercise intensities (speed in m/min) on the citrate dependence of red quadriceps ACC. Red quadriceps ACC is progressively inhibited by increases in exercise intensity, with all values being significantly different from resting (P < 0.05). The K_0.5 and K_a for citrate in the ACC reaction also progressively increase as exercise intensity increases in red quadriceps muscle (Table 1). All values, excluding 10 m/min for K_0.5, are significantly different from those in resting rats, with the values for rats exercising at 40 m/min significantly different from all other speeds (P < 0.05). The complete set of enzyme kinetic values for red quadriceps is listed in Table 1. Figure 3 shows that all reaction velocities of ACC at a physiological citrate concentration (0.2 mM) are significantly different from resting values (P < 0.05).

Table  1.   Effect of exercise intensity on enzyme kinetics of ACC in red quadriceps muscle Speed, m/min Vmax, nmol · g1 · min1 K0.5, mM citrate Ka, mM citrate 0 28.6 ± 1.5  2.2 ± 0.1  4.6 ± 0.3  10 23.8 ± 1.2  2.8 ± 0.3  6.8 ± 0.7  20 18.0 ± 1.4  3.0 ± 0.2  8.8 ± 1.3  30 16.6 ± 0.8  3.4 ± 0.2  10.6 ± 1.0  40 14.3 ± 0.7  4.2 ± 0.3  13.4 ± 1.4 Values are means ± SE from 7-11 rats at each speed. ACC, acetyl CoA-carboxylase; Vmax, maximum velocity; K0.5, citrate concentration required for half-maximal activation of ACC; Ka, activation constant. All values are significantly different from those found in resting rats (0 m/min), excluding K0.5 at 10 m/min (P < 0.05).

Figure 4 shows the effect of various exercise intensities (speed in m/min) on the citrate dependence of white quadriceps ACC. White quadriceps ACC is not inactivated with increases in exercise intensity, although there appears to be a tendency toward a decrease in V_max of ACC at 40 m/min (P = 0.06). In Table 2, the K_0.5 and K_a in white quadriceps muscle showed significant increases only during the most intense of exercise speeds (P < 0.05). The complete set of enzyme kinetic values for white quadriceps is listed in Table 2. Figure 3 shows that reaction velocities of white quadriceps ACC at a physiological citrate concentration (0.2 mM) are significantly different from resting (P < 0.05) for only the 30 and 40 m/min work rates.

Table  2.   Effect of exercise intensity on enzyme kinetics of ACC in white quadriceps muscle Speed, m/min Vmax, nmol · g1 · min1 K0.5, mM citrate Ka, mM citrate 0 19.3 ± 1.7  2.3 ± 0.1  5.0 ± 0.3  10 18.2 ± 1.2  2.4 ± 0.1  5.8 ± 0.5  20 18.5 ± 1.2  2.4 ± 0.1  5.4 ± 0.7  30 17.4 ± 1.1  2.4 ± 0.1  5.9 ± 0.6  40 15.0 ± 1.7 2.8 ± 0.2* 7.6 ± 0.8* Values are means ± SE from 9-11 rats at each speed. * Significantly different from those found in resting rats (0 m/min) (P < 0.05); significant at P = 0.06.

Red quadriceps AMPK activity also increased with the increase in exercise intensity (Fig. 5). The 30 and 40 m/min speeds were significantly different from those at rest (P < 0.05), whereas the 20 m/min speed was significant at P = 0.06. AMPK activity in white quadriceps was not significantly altered in response to exercise at any of these work rates (Fig. 5).

Glycogen in both muscle types decreased significantly from the values in the resting rats (P < 0.05) in response to each increase in exercise intensity, excluding 30 m/min in the white quadriceps (Fig. 6).

Blood lactate progressively increased with the increases in work rate. Values from rats running at 30 and 40 m/min were significantly different (P < 0.05) from resting values (Fig. 6).

Blood glucose levels were significantly increased (P < 0.05) from rest at each work rate level (Fig. 6).

There was no significant increase (P < 0.05) in plasma FFA after 5 min of exercise at each work rate (Fig. 6).

DISCUSSION

Skeletal muscle ACC has been characterized (23) and appears to be regulated differently from liver and adipose tissue ACC (30, 32). Previous studies have shown that muscle malonyl-CoA decreases in response to exercise (28, 30) and that this decrease is not dependent on epinephrine (29). Fasting (31) and electrical stimulation (12) also decrease muscle malonyl-CoA. Skeletal muscle expresses the gene for AMPK to a high extent (26), and phosphorylation of ACC by AMPK inactivates the enzyme (30, 32). Inactivation of muscle ACC during exercise is accompanied by a decrease in malonyl-CoA (30). A similar response is shown in this study. Although the correlation is not exact, we acknowledge that the precise time when ACC activity decreases is not known, and there is probably a time lag between the decrease in ACC activity and the decrease in malonyl-CoA. During the isolation procedure, allosteric activators and inhibitors are not retained in the final ACC preparation, and this may contribute to the lack of a high correlation between decreased ACC activity and the decrease in malonyl-CoA. It seems plausible that the initial signal for inactivation of ACC is associated with muscle contractile activity (12), and we believe that the data from this study support this proposition. A postulated sequence of events is that stimulation of the muscle and the subsequent release of calcium and contraction of the muscle activate a specific kinase for AMPK, similar to that reported in liver (27). It is possible that this kinase may be activated by calcium and/or AMP. Once AMPK is phosphorylated and becomes activated, it then phosphorylates and inactivates ACC, with the result being a decrease in malonyl-CoA. A drop in malonyl-CoA would be expected to relieve the inhibition of CPT-1 and thus allow FFA to be taken up and oxidized in the mitochondria.

The data from the present study suggest that fast-twitch oxidative fibers (type IIA), which are the predominant fiber type of the red quadriceps muscle, respond to increases in exercise intensity by inactivating ACC. This inactivation, with the resulting decrease in malonyl-CoA, likely removes the inhibition on CPT-1, which would then allow FFA to be utilized as an energy source during exercise. Previous work (25) has shown that when exercise is done at such an intensity to elicit a large increase in lactate that the muscle utilizes carbohydrate as the principle source of energy. This leads to the important question of the role of ACC inactivation and the consequent decrease of malonyl-CoA during intense exercise. Some possible reasons for this include 1) preparation of the muscle for FFA utilization after exercise; 2) the fact that the entire system simply responds to muscle contraction and thus prepares the muscle to utilize FFA when they become available; and 3) oxidation of intramuscular triacylglycerols.

A recent study (22) has shown that energy expenditure is increased for 24 h after exercise at a high vs. low intensity. Although there were no differences between the two groups in 24-h lipid oxidation, the authors suggest that more lipid was oxidized in the postexercise period after high-intensity exercise. More lipid would have been utilized as an energy source during the low-intensity exercise bout, whereas carbohydrates would be used to restore the depleted glycogen levels associated with intense exercise. Wolfe et al. (34) has reported that total fatty acid oxidation is increased above resting values during 2 h of postexercise recovery. The present study shows that the extent of ACC inactivation is dependent on exercise intensity in rat fast-twitch oxidative muscle fibers. Carbohydrate is most likely the primary source of energy for the muscle during the intense exercise in our study; therefore, the inactivation of ACC and decrease in malonyl-CoA may serve the role of preparing the muscle for FFA when they become available. This is probably very important during postexercise recovery after intense exercise, since the muscle needs ATP to restore glycogen reserves. Oxidation of FFA may play an important role in this restoration.

White quadriceps muscle (predominantly fast-twitch glycolytic fibers; type IIB) changes in ACC occur only during the higher intensity bouts of exercise. This may imply that white quadriceps muscle is not recruited until the exercise is at a certain intensity level. The present finding provides further evidence that contraction is the key initiator of the intracellular signaling cascade that inactivates muscle ACC. Because white quadriceps has relatively few mitochondria, the decrease in glycogen would be expected to exceed that of the red quadriceps if both were contracting at the same rate. The observation that only small decreases in glycogen were observed in white quadriceps provides evidence that the fibers are relatively inactive during exercise of the lower intensities for the 5-min duration. A previous study (28) demonstrated a decrease in malonyl-CoA in this region of the quadriceps with longer duration exercise.

Intramuscular triacylglycerol use during exercise was recently reviewed (25). The authors concluded that it has been difficult to quantify how much intramuscular triacylglycerols are utilized during exercise, but their availability to be hydrolyzed and then oxidized is apparent. Thus inactivation of ACC with the subsequent decrease in malonyl-CoA would allow muscle to oxidize FFA from this source. This could prove to be a very important mechanism during the onset of exercise.

Blood lactate and glucose were significantly elevated with the progressive increases in exercise intensity. The high lactate concentration provides indirect evidence that exercise at 40 m/min was intense. Brooks and White (4) reported oxygen uptake determinations of rats running at various speeds and treadmill grades. Extrapolating from their data, we estimated that rats in our study are running at ~55, 65, 75, and 85% of maximal oxygen uptake (O_{2 max}) at exercise speeds of 10, 20, 30 and 40 m/min, respectively.

The "crossover concept," as proposed by Brooks and Mercier (3), states that there is a specific crossover point, as exercise intensity increases, at which energy from carbohydrates predominates over lipid oxidation on a relative basis. Although the relative contribution of fat oxidation decreases with increases in exercise intensity, the absolute amount of fat that is oxidized is still increased compared with resting. For example, in humans under resting conditions, ~60% of energy comes from lipid oxidation. Romijn et al. (18) has shown that the relative amount of fat oxidation increases to nearly 90% during 30 min of exercise at 25% O_{2 max}. As the intensity of exercise increases, this value decreases to ~50% at 65% O_{2 max} and to 25-30% during exercise at 85% O_{2 max}. Their study reports that at 65% O_{2 max} the total fatty acid oxidation rate in trained human subjects was 42.8 µmol · kg¹ · min¹, which is a ninefold increase above resting values. As exercise increases to 85% O_{2 max}, total fatty acid oxidation decreases to 29.6 µmol · kg¹ · min¹. Although we see a decrease in the rate of total fatty acid oxidation with the increase in exercise intensity, the rate is still sixfold higher than at rest. In the rat, palmitate oxidation increased 10-fold (0.025 to 0.25 µmol · g¹ · h¹) after 15 min of electrical stimulation in hindlimbs perfused with 600 µmol palmitate and 6 mM glucose (24). This increase in absolute rate of fat oxidation during higher intensity exercise may be facilitated by the inactivation of ACC and decrease in malonyl-CoA as shown in this study.

One study has reported that malonyl-CoA does not decrease in humans after exercise (17). Possible reasons for this discrepancy may include species differences and use of different measurement techniques. A recent study has identified and characterized ACC in human skeletal muscle (33). The authors report the molecular mass for human ACC as 275 kDa. This is similar to what our laboratory has published for rat skeletal muscle (23). Further research is needed to determine whether the regulation of human skeletal muscle ACC is similar to that of rats and whether inhibition of ACC causes malonyl-CoA to decrease in human skeletal muscle.

In summary, the extent of inactivation of ACC is dependent on exercise intensity in fast-twitch oxidative muscle fibers. The inactivation of ACC was accompanied by increases in AMPK activity and decreases in malonyl-CoA. Exercise altered ACC kinetics in fast-twitch glycolytic fibers only during exercise of the highest intensity. This provides additional evidence that inactivation of ACC seen during exercise is a result of muscle contraction. It is plausible that inactivation of ACC during intense exercise may have the function of preparing muscle to oxidize FFA on their availability and/or the utilization of lipid in the postexercise state. Energy expenditure after exercise provides a means for ATP synthesis for the restoration of lost energy reserves.

ACKNOWLEDGEMENTS

This research was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant AR-41438.

FOOTNOTES

Address for reprint requests: W. W. Winder, Dept. of Zoology, 545 WIDB, Brigham Young Univ., Provo, UT 84602 (E-mail: winderw{at}acd1.byu.edu).

Received 9 December 1996; accepted in final form 16 May 1997.

REFERENCES