We examined the effects of increasing acetylcarnitine and acetyl-CoA availability at rest, independent of pyruvate dehydrogenase complex (PDC) activation, on energy production and tension development during the rest-to-work transition in canine skeletal muscle. We aimed to elucidate whether the lag in PDC-derived acetyl-CoA delivery toward the TCA cycle at the onset of exercise can be overcome by increasing acetyl group availability independently of PDC activation or is intimately dependent on PDC-derived acetyl-CoA. Gracilis muscle pretreated with saline or sodium acetate (360 mg/kg body mass) (both n = 6) was sampled repeatedly during 5 min of ischemic contraction. Acetate increased acetylcarnitine and acetyl-CoA availability (both P < 0.01) above control at rest and throughout contraction (P < 0.05), independently of differences in resting PDC activation between treatments. Acetate reduced oxygen-independent ATP resynthesis ∼40% (P < 0.05) during the first minute of contraction. No difference in oxygen-independent ATP resynthesis existed between treatments from 1 to 3 min of contraction; however, energy production via this route increased ∼25% (P < 0.05) above control in the acetate-treated group during the final 2 min of contraction. Tension development was ∼20% greater after 5-min contraction after acetate treatment than in control (P < 0.05). In conclusion, at the immediate onset of contraction, when PDC was largely inactive, increasing cellular acetyl group availability overcame inertia in mitochondrial ATP regeneration. However, after the first minute, when PDC was near maximally activated in both groups, it appears that PDC-derived acetyl-CoA, rather than increased cellular acetyl group availability per se, dictated mitochondrial ATP resynthesis.
- oxygen deficit
- oxidative phosphorylation
- sodium acetate
the readjustment of oxidative (mitochondrial) ATP production to meet the increase in muscular energy demand during the transition from rest to exercise, or the step increase from one workload to another, is delayed and follows an approximately exponential time course (for review, see Ref. 37). During this period of latency, the transient shortfall in mitochondrial ATP production, classically termed the “oxygen deficit” (21, 30), is supplemented by ATP resynthesis from non-oxygen-dependent routes [i.e., ATP and phosphocreatine (PCr) breakdown and glycolysis]. Although ATP production from oxygen-independent routes enables rapid rates of ATP turnover to be achieved, it has only a finite capacity and also results in the accumulation of metabolic byproducts that are deleterious to muscular contractile function (hydrogen ions, lactate ions, and inorganic phosphate; see Ref. 8). Indeed, without the progressive increase in mitochondrial ATP production at the onset of contraction, the onset of muscular fatigue would be markedly accelerated. Classically, the lag in oxidative ATP production and resulting oxygen deficit have been attributed to a lag in muscle blood flow and thereby muscle oxygen delivery, which also follows an approximately exponential time course (21, 30). However, more recently it has been proposed that oxygen availability may not be the sole determinant of the oxygen deficit, at least at submaximal exercise workloads (11, 29), such that the delay in mitochondrial energy production must be attributable, at least in part, to a lag in enzyme activation and/or substrate delivery (metabolic inertia) within energy-producing pathways on the initiation of muscular contraction (10, 12, 18, 37).
A recent series of studies by our group has demonstrated the existence of metabolic inertia at the onset of contraction (27, 28, 33–36). On the basis of evidence from these studies, we believe that the delay in acetyl-CoA provision at the onset of exercise, which we have termed the “acetyl group deficit” (27), is a principal determinant of the oxygen deficit (27, 33, 34). It is our assertion, therefore, that the acetyl group deficit reflects a period at the onset of contraction when acetyl-CoA availability fails to meet the increased demands of the tricarboxylic acid (TCA) cycle, thereby limiting oxidative ATP resynthesis and resulting in an increased contribution by ATP delivery from oxygen-independent routes (27, 33, 34, 36). In our most recent study (27), the acetyl group deficit was clearly characterized by the failure of acetyl-CoA or acetylcarnitine to increase during the first minute of contraction, with strong trends for both acetyl-CoA (P = 0.06) and acetylcarnitine (P = 0.08) to decline during this period. Furthermore, this lag in acetyl group delivery came as a result of, and was mirrored by, a lag in pyruvate dehydrogenase complex activation at the onset of contraction (27).
The pyruvate dehydrogenase complex (PDC) catalyzes the irreversible reaction that commits pyruvate to its oxidative fate inside the mitochondrion. As a consequence of PDC activation at the onset of muscle contraction (4, 6), acetyl-CoA delivery via the PDC reaction is markedly increased, providing a stream of substrate toward the TCA cycle for subsequent mitochondrial ATP resynthesis via the electron transport chain. Alternatively, PDC-derived acetyl groups can be transferred to the cellular carnitine pool, presumably when acetyl-CoA production exceeds its rate of utilization by citrate synthase (3). Buffering acetyl groups in this way has been said to maintain a viable pool of free reduced coenzyme A (CoASH) for PDC flux to continue and creates a readily available reservoir of substrate (in the form of acetylcarnitine) for the TCA cycle to subsequently utilize (20). After PDC activation during contraction, it would appear that acetyl group delivery is no longer limiting toward TCA cycle flux. This point is exemplified by the recovery of acetyl-CoA to its resting concentration and the almost linear increase in acetylcarnitine concentration during contraction after PDC activation (17, 24, 27).
We have demonstrated that pretreatment of canine skeletal muscle with sodium dichloroacetate, an inhibitor of pyruvate dehydrogenase kinase (32), near maximally activated the PDC and acetylated muscular carnitine and free CoASH pools at rest (27, 34, 35, 36). During subsequent submaximal ischemic contraction (blood flow and, hence, oxygen availability held at its resting state), dichloroacetate overcame the acetyl group deficit, reduced ATP resynthesis from oxygen-independent routes, and improved the maintenance of contractile function over the course of contraction compared with control (27, 34, 35, 36).
Because dichloroacetate both activates the PDC and acetylates the free CoASH and carnitine pools at rest, it was not possible to determine from this series of studies whether the reduction in oxygen-independent energy production during contraction was the direct result of acetyl group delivery via the PDC reaction being increased from the immediate onset of contraction and/or was attributable to the stockpile of acetyl groups generated at rest, providing a readily available pool of substrate for the TCA cycle throughout the rest-to-work transition. With this question in mind, the aim of the present study was to further characterize the acetyl group deficit by investigating whether, in contrast to the effect of dichloroacetate, pharmacologically increasing resting acetyl group availability, independently of PDC activation, could overcome metabolic inertia, reduce oxygen-independent ATP production, and positively affect function in contracting canine skeletal muscle. In particular, we wished to investigate the time course of PDC activation, acetyl group accumulation, substrate utilization, and tension development during 5 min of electrically evoked ischemic muscular contraction. This was done in the absence and presence of pretreatment with sodium acetate, a known PDC independent acetylator of muscle carnitine and free CoASH pools (7, 17, 25).
All in vivo procedures were performed in accordance with United Kingdom Home Office legislation. After an overnight fast, 12 female beagle dogs (Animal Breeding Unit, AstraZeneca Pharmaceuticals, Cheshire, UK; body mass 9.1 ± 0.2 kg) were premedicated with morphine sulfate (10 mg im) 30 min before the induction of anesthesia with pentobarbital sodium (Sagatal; Rhône Merieux, Harlow, UK). Anesthesia was induced as a bolus of pentobarbitone (45.9 ± 0.3 mg/kg body mass), followed by a continuous infusion throughout each experiment (0.11 ± 0.01 mg·kg body mass−1·min−1 iv). The trachea was intubated, and the animals were artificially ventilated with room air (24 cycles/min, tidal volume 13–15 ml/kg body mass; model 16/24, Palmer Bioscience, London, UK). The right carotid artery was cannulated, and systemic blood pressure was recorded with a pressure transducer (PDCR 75; Druck, Barendrecht, The Netherlands). The right antecubital vein was cannulated for the infusion of heparin, saline, and sodium acetate. The left gracilis muscle was surgically isolated, as detailed previously (35), leaving the motor nerve and the arterial and venous blood supply to and from the gracilis muscle intact. The femoral artery, supplying the gracilis, was cannulated proximally and distally and attached to a perfusion pump (Miniplus 3; Gilson, Villiers-Le-Bel, France). Blood flow to the gracilis was fixed at its resting rate and was maintained constant for the duration of the experiment, equating to ∼25% of the normal contraction-induced flow to the gracilis when stimulated under the present parameters (35). The fixing of blood flow at its resting level prevented exercise hyperemia and thereby any increase in oxygen delivery to the muscle during subsequent contraction, enabling us to investigate metabolic inertia largely independently of any alteration in oxygen availability throughout contraction (27, 28, 34–36).
After the completion of the surgical procedures, each animal was infused with either 30 ml of saline (CON, n = 6) or 360 mg/kg body mass sodium acetate (acetate, n = 6) in 30 ml of saline over a period of 30 min. The dose of acetate used in the present study has been shown previously to near maximally acetylate carnitine and free CoASH pools in human skeletal muscle at rest (7, 17, 25).
Immediately after CON or acetate treatment, the left gracilis muscle was stimulated to contract, under partial ischemia, via electrical stimulation of the obturator nerve (Grass S88 stimulator; Quincy, Medfield, MA). Square-wave impulses of 0.1-ms duration, 3-Hz frequency, and 6-V submaximal voltage were applied for 5 min, resulting in complete muscle fiber recruitment (35). This stimulation protocol produces a workload of ∼80% maximal oxygen uptake (V̇o2 max) within gracilis muscle with normal blood flow intact (35). After the experiment, each animal was killed humanely while still under anesthesia by the infusion of pentobarbitone and saturated potassium chloride.
Muscle sampling and analyses.
Immediately before contraction, a resting muscle biopsy was taken by superficial excision of tissue from the distal end of the left gracilis, using a scalpel blade and forceps. Subsequent muscle samples were excised after 20, 40, 60, 180, and 300 s of continuous contraction. All excised muscle tissue was immediately frozen (<2 s) by submersion in liquid nitrogen. Biochemical and histochemical analyses have shown that the canine gracilis muscle (typically weighing ∼25 g in a 1-yr-old beagle dog) has a fiber type composition of ∼40% type I and ∼60% type IIa fibers (22). The canine gracilis muscle is comparable in fiber type composition to human skeletal muscle, and the model used here allows multiple biopsy sampling without a severe detriment to contractile function (27, 28, 34–36).
All biopsy samples were divided into two pieces under liquid nitrogen. Subsequently, one portion was freeze dried, dissected free from visible blood and connective tissue, and powdered. After extraction in 0.5 M perchloric acid containing 1 mM EDTA, the supernatant was neutralized with 2.2 M KHCO3 and used for the determination of ATP, PCr, creatine, and lactate (13). The extract was also used for the determination of free carnitine, acetylcarnitine, free CoASH, and acetyl-CoA (2). Freeze-dried muscle powder was also used for the determination of muscle glycogen (13). The remaining portion of frozen wet muscle was used to assess PDC activation (5).
Calculations and statistics.
All data are reported as means ± SE. Comparisons between treatments were carried out using two-way analysis of variance (ANOVA) with repeated measures. When a significant F value was obtained (P < 0.05), a least significant difference post hoc test was used to locate any differences (SPSS Base 8.0). With the exception of lactate, the concentrations of all muscle metabolites were adjusted to the mean total creatine concentration within each individual animal (9). Changes in muscle ATP, PCr, and lactate concentrations were summarized as the magnitude of oxygen-independent ATP production, in accordance with the following formula (31) (1)
To minimize the number of animals killed for this work, the CON group was shared with another study performed using the same experimental protocol but with dichloroacetate infusion (27). To alleviate any bias, all experiments were performed on consecutive days, and treatments were randomized between CON, acetate, and dichloroacetate infusions (27).
Infusion of acetate elevated blood pH at rest compared with the CON group (CON = 7.36 ± 0.01 vs. acetate = 7.53 ± 0.01, P < 0.01). Resting acetyl-CoA was elevated above CON after acetate pretreatment (CON = 12.2 ± 1.7 vs. acetate = 37.2 ± 9.5 μmol/kg dry muscle, P < 0.01; Fig. 1A). Similarly, acetylcarnitine was elevated above CON at rest after acetate (CON = 5.9 ± 0.9 vs. acetate = 21.7 ± 0.5 mmol/kg dry muscle, P < 0.01; Fig. 1B). No difference in the degree of PDC existing in its active “a” form (PDCa) was observed between groups at rest (CON = 0.13 ± 0.05 vs. acetate = 0.18 ± 0.03 mmol acetyl-CoA·min−1·kg wet muscle−1; Fig. 1C).
During the subsequent 5 min of ischemic contraction, the concentration of acetyl-CoA in the acetate group remained elevated above CON at all time points (P < 0.05; Fig. 1A). There was a trend for acetyl-CoA to decline during the first 20 s of contraction (P = 0.10) in the acetate group, after which it declined further from its resting concentration at 5 min (P < 0.05; Fig. 1A). The concentration of acetyl-CoA was unchanged from rest throughout contraction in CON but showed a clear tendency to decline during the first 20 s of contraction (P = 0.06; Fig. 1A). Changes in acetyl-CoA were mirrored by changes in the concentration of free CoASH, with no differences existing in the size of the total CoASH pool between groups at any time point during contraction (Table 1).
Acetylcarnitine concentration was unchanged from rest throughout contraction in the acetate group, remaining significantly higher than CON at all time points during contraction (P < 0.01; Fig. 1B). Acetylcarnitine increased from rest after 180 s of contraction in CON but showed a clear tendency to decline during the first 20 s of contraction (P = 0.08; Fig. 1B). Changes in acetylcarnitine were mirrored by changes in the concentration of carnitine, with no differences in the concentration of the total carnitine pool existing between groups at any time point during contraction (Table 1).
PDCa increased during contraction in both groups, with no significant differences in its activation status existing between treatments at any time point (Fig. 1C). Despite the similarities in activation status, the transformation of PDC to PDCa was faster in CON, increasing significantly from rest after 40 s (CON rest = 0.13 ± 0.05 vs. CON 40 s = 1.10 ± 0.25 mmol acetyl-CoA·min−1·kg wet muscle−1; P < 0.05) compared with 60 s in the acetate group (acetate rest = 0.18 ± 0.03 vs. acetate 60 s = 1.00 ± 0.27 mmol acetyl-CoA·min−1·kg wet muscle−1; P < 0.05).
No differences in muscle glycogen concentration existed between the treatment groups at rest or at any time point during contraction (Table 1). Both groups showed a similar profile of degradation during contraction, with the concentration of glycogen falling from rest after 3 min (P < 0.05).
ATP concentrations were maintained during the majority of the period of contraction in both groups (Table 1). However, after 1 min of contraction, the concentration of ATP in the acetate-treated group was significantly better maintained than in the CON group, where it fell from its resting concentration after 5 min of contraction (P < 0.05; Table 1).
Resting PCr concentration was similar between groups (Table 1). After 20 s of contraction, PCr had fallen from its resting concentration in both groups (P < 0.01; Table 1). However, the rate of PCr hydrolysis was markedly reduced in the acetate group over the initial 60 s of contraction, with the concentration of PCr better maintained in the acetate treatment group compared with CON after 20, 40, and 60 s of contraction (P < 0.05; Table 1). No differences existed between groups in PCr concentration after the first minute of contraction (Table 1). Changes in the concentration of PCr mirrored changes in creatine concentration, with no difference in the concentration of the total creatine (sum of PCr and creatine) pool existing within and between groups (Table 1).
No difference in muscle lactate concentration existed between treatment groups at rest. Muscle lactate concentration remained unchanged from rest during the initial 20 s of contraction in the acetate group but more than doubled in concentration over this time in the CON group (P < 0.05; Table 1). However, after 5 min of contraction, muscle lactate concentration was significantly greater in the acetate treatment group compared with CON (P < 0.05; Table 1).
The contribution of oxygen-independent ATP production toward energy production over the entire 5 min of contraction was no different between treatment groups (CON = 122 ± 9 vs. acetate = 148 ± 14 mmol ATP/kg dry muscle). The calculated rate of oxygen-independent ATP production was significantly lower during the first minute of contraction after acetate infusion (CON = 60 ± 2 vs. acetate = 40 ± 8 mmol ATP·min−1·kg dry muscle−1, P < 0.05; Fig. 2). From 1 to 3 min, no difference existed between groups (Fig. 2), and during the final 2 min of contraction, ATP production via this route was greater after acetate treatment (CON = 13 ± 2 vs. acetate = 32 ± 6 mmol ATP·min−1·kg dry muscle−1, P < 0.05; Fig. 2).
No differences in resting isometric tension (CON = 936 ± 158 vs. acetate = 1,078 ± 53 g/100 g wet muscle) or maximum isometric tension (CON = 4,539 ± 361 vs. acetate = 5,220 ± 220 g/100 g wet muscle) existed between treatment groups. Isometric tension fell from maximum during contraction in both groups, with tension falling by ∼57% from maximum in CON at 5 min and by ∼35% in acetate (P < 0.05; Fig. 3). Acetate treatment attenuated the decline in contractile function after 3 min of contraction onward compared with CON (P < 0.05; Fig. 3).
We investigated in the present study whether inertia in mitochondrial ATP production at the onset of muscle contraction is attributable to acetyl group availability per se limiting substrate delivery to the TCA cycle or, alternatively, can be attributed to a lag in PDC activation and thereby flux on the initiation of contraction. Increasing the resting provision of acetyl groups (in the form of acetylcarnitine and acetyl-CoA), independently of PDC activation, through the infusion of sodium acetate transiently delayed muscle lactate accumulation during the first 20 s of contraction and reduced PCr degradation and oxygen-independent ATP resynthesis during the first 60 s of contraction. Despite this initial fall, no difference in oxygen-independent ATP resynthesis was observed between treatments from 1 to 3 min of contraction, as we had seen previously after dichloroacetate administration in this animal model (27, 34, 36), and, during the final 2 min of contraction, energy production via this route was increased above CON in the acetate-treated group. Isometric tension development was significantly greater at 3 and 5 min of contraction after acetate treatment compared with CON. Collectively, these findings support the contention that, at the immediate initiation of contraction, when the PDC was largely inactive, increasing cellular acetyl group availability overcame inertia in oxygen-dependent ATP regeneration. However, after the first minute of contraction, when the PDC was activated to the same extent in both groups, it appears that PDC-derived acetyl-CoA, rather than increasing cellular acetyl group availability per se, was primarily responsible for acetyl group delivery to the TCA cycle and therefore regulated the contribution of mitochondrial ATP resynthesis toward the energy demands of the cell.
Sodium acetate and acetyl group availability.
In agreement with previous studies, acetate increased muscle acetyl-CoA and acetylcarnitine concentrations at rest (P < 0.01) and throughout contraction (P < 0.05) in CON (Fig. 1, A and B; see Refs. 7, 17, 25). Furthermore, and in accordance with our aims, acetate acetylated the carnitine and free CoASH pools to the same extent as pretreatment with dichloroacetate within this animal model (27, 34, 36) and, in contrast to dichloroacetate, achieved this independently of any alteration in PDC activation at rest (Fig. 1C).
Studies that have investigated acetyl group utilization during moderate-to-intense skeletal muscle contraction after sodium acetate pretreatment have reported no reduction in the requirement for oxygen-independent ATP production at any contraction time point (7, 17, 25), implying that elevating acetyl group availability, independently of PDC activation, cannot overcome the period of metabolic inertia and thereby accelerate the onset of mitochondrial ATP production. These previous findings are rather at odds with those reported in the present study; however, this lack of concordance can be accounted for, and on several levels. First, previous studies have failed to examine changes in acetyl group availability at any time point during contraction before PDC activation (7, 17, 25), i.e., to examine the functional role of increasing substrate reserve (in the form of acetylcarnitine and acetyl-CoA) during the period of contraction when they are known to be limiting toward the demands of the TCA cycle. Second, and perhaps more important, previous studies appear to have utilized exercise workloads that are either too low (7, 25) or too intense (17) to optimally assess the functional and metabolic consequences of increasing substrate reserve through acetate infusion. We can say this, as it would appear that metabolic inertia, at least in the form of an acetyl group deficit, only exists over a narrow and predictable range of exercise intensities (between ∼65 and 90% V̇o2 max; see Ref. 27), above and below which increasing resting acetyl group availability (in both the absence and presence of PDC activation) will be ineffective at reducing oxygen-independent ATP production and thereby improving contractile function (19, 27). By way of example, at workloads below ∼65% V̇o2 max, it appears that PDC flux and/or fat-mediated acetyl group delivery is sufficient to match the energy demands at all contraction time points (7). This is substantiated by the absence of any changes in muscle acetylcarnitine and acetyl-CoA concentrations during 30 s of submaximal exercise (65% V̇o2 max) after sodium acetate and saline (control) infusion in healthy volunteers, despite PDC activation increasing significantly from rest during this period (7). Similarly, at workloads above ∼90% V̇o2 max, where the PDC is activated within as little as 5 s of contraction (1), it appears that acetyl-CoA availability is at no point limiting toward the demands of the TCA cycle, typified by the almost linear accumulation of acetylcarnitine and acetyl-CoA throughout contraction (15, 16). Therefore, on scrutiny of previously published literature, it remains unclear whether increasing the provision of substrate toward the TCA cycle at rest, independently of PDC activation, will reduce the reliance on ATP resynthesis from oxygen-independent routes and improve contractile function. The present study has addressed this issue by using an exercise workload and including contraction time points at which an acetyl group deficit is known to exist.
First minute of contraction.
Previous work in this model, where dichloroacetate has been used to increase acetyl group availability, found that oxygen-independent ATP synthesis is markedly reduced during the first 3 min of submaximal ischemic contraction and can be quantitatively accounted for by increased acetyl group utilization after dichloroacetate infusion (27). It would be interesting, therefore, to calculate whether the reduction in oxygen-independent ATP resynthesis seen during the first 60 s of contraction after acetate pretreatment in the present study can be accounted for by the increased availability of acetyl groups.
According to the in vitro-assessed Michaelis-Menten constant (Km) of citrate synthase for acetyl-CoA (2–500 μmol/l; Ref. 23), the increased availability of acetyl groups after acetate pretreatment (∼5.5 mmol/l intracellular water) would, if freely available, markedly promote acetyl-CoA entry into the TCA cycle, via citrate synthase, on the initiation of contraction compared with CON. Indeed, with the assumption that 1 mmol of acetyl groups will produce 12 mmol of ATP equivalents, the sequestering of an additional ∼1.7 mmol acetyl groups/kg dry muscle by the TCA cycle would be required to account for the difference in oxygen-independent ATP resynthesis (∼20 mmol ATP equivalents/kg dry muscle) observed between treatment groups during the first minute of contraction (P < 0.05; Fig. 2). Because the mean concentration of acetylcarnitine appeared to fall by 2.0 mmol/kg dry muscle during the first 60 s of contraction in the acetate group (Fig. 1B), the results appear to indicate that the increased concentration of acetylcarnitine, with a modest contribution from acetyl-CoA (Δ8.0 μmol/kg dry muscle; Fig. 1A), overcame an acetyl group deficit during this period and thereby reduced the demand for ATP production from oxygen-independent routes (Fig. 2).
The amount of PDC existing in its active form (PDCa) increased substantially in both groups during the first minute of contraction, but no difference in the magnitude of activation existed between treatments at any time point during contraction (Fig. 1C). However, despite these similarities in activation status, the rate of transformation of the PDC to its active moiety was greater in CON, such that PDCa was significantly increased above rest after 40 s of contraction in CON compared with 60 s after acetate pretreatment. Although activation is a prerequisite for an increase in flux through the PDC, differences in the rate of activation between groups during the first minute of contraction do not necessarily equate to differences in flux, which is further controlled by the availability of various reaction substrates and products as well as numerous cofactors (39). The slower transformation of PDC to its active form after acetate is likely to be due to the effect of the high resting acetyl-CoA concentration on the ratio of the two key PDC regulatory enzymes: pyruvate dehydrogenase kinase (PDK) and pyruvate dehydrogenase phosphatase (PDP). The antagonist action of these enzymes determine the proportion of PDC existing in its active, dephosphorylated form. A high acetyl-CoA-to-free CoASH ratio, as observed after acetate treatment, is a positive allosteric modulator of PDK activity and would result in a high degree of PDC phosphorylation and inactivation at rest (39). This would have the effect of rendering the PDC more difficult to dephosphorylate and thereby reactivate at the immediate onset of contraction by the intrinsic PDP. Therefore, it is envisaged that resting PDK activity would have been greatly increased after acetate infusion, resulting in the slower transformation of PDC that was observed during the first 60 s of contraction.
Between 1 and 3 min of contraction.
The amount of PDCa increased substantially in both groups during the first minute of contraction. The concentration of acetylcarnitine and acetyl-CoA remained significantly elevated above CON throughout contraction after acetate (Fig. 1, A and B). Despite this increased availability of acetyl groups, the reduction in oxygen-independent ATP regeneration observed during the first minute of contraction in the acetate group was not maintained (Fig. 2), which is in contrast to what we previously have observed after dichloroacetate administration (27, 34). For example, in the study of Roberts et al. (27), which was conducted under experimental conditions identical to the present study, acetylation of carnitine and free CoASH pools at rest, brought about by activation of the PDC with dichloroacetate, reduced oxygen-independent ATP resynthesis by ∼20 mmol/kg dry muscle between 1 and 3 min of contraction compared with CON. Collectively, these findings imply that an abundance of substrate was available to the TCA cycle after acetate infusion but not utilized. To comprehend why, we need to understand the mechanism by which acetate infusion increases cellular acetyl-CoA and acetylcarnitine concentrations at rest (Fig. 4).
The infusion of acetate results in the equilibrium formation of acetic acid and sodium bicarbonate within the blood (Fig. 4; Ref. 38). The increased available acetic acid, from the action of acetyl-CoA synthase and consumption of ATP, is converted into acetyl-CoA in the cytosol of the muscle cell (Fig. 4; Ref. 38). The increased concentration of cytosolic acetyl-CoA, through the sequential action of cytosolic carnitine acetyltransferase, translocase, and mitochondrial carnitine acetyltransferase (mCAT), would transfer acetyl groups across the mitochondrial membrane and increase the concentration of mitochondrial acetyl-CoA, secondary to an increase in the concentration of cytosolic acetylcarnitine (Fig. 4). It is worth noting that this route of acetylation of the muscular free CoASH and carnitine pools, although resulting in similar resting concentrations of acetyl-CoA and acetylcarnitine, is markedly different from that of dichloroacetate (27). Dichloroacetate-stimulated activation and flux through PDC would first increase mitochondrial acetyl-CoA availability before increasing mitochondrial and then cellular acetylcarnitine concentration. Therefore, the infusion of acetate can be viewed as increasing the cellular availability of acetyl groups, compared with dichloroacetate predominantly increasing their mitochondrial availability. If we accept this point, we can begin to understand why no fall in muscle acetyl-CoA concentration was observed from 1 to 3 min of contraction after acetate infusion, when the TCA cycle was capable of greater flux. The absence of any significant fall in acetyl-CoA concentration from 1 to 3 min of contraction after acetate is masked by the continuous drive of acetyl groups from the elevated acetic acid concentration in the plasma, leading toward mitochondrial acetyl-CoA formation by mCAT (Fig. 4). This observation indicates that mitochondrial acetyl-CoA production, independent of that coming through the PDC, is secondary to acetylation of the cytosolic and mitochondrial carnitine pool. Alternatively, it could be due to the majority of the acetylcarnitine reserve existing outside the mitochondria, therefore preventing its rapid utilization by the mCAT. Given that no difference existed in oxygen-independent ATP regeneration between CON and acetate groups from 1 to 3 min of contraction in the present study (Fig. 2), it would appear that flux through the PDC was limiting mitochondrial ATP resynthesis in both experimental groups during this period. Evidence in support of this statement comes from the observation that the area under the curve for PDCa during 1–3 min of contraction in an experiment performed under identical experimental conditions as the present study (but using pretreatment with dichloroacetate as opposed to acetate) was considerably greater after dichloroacetate (4.62 mmol acetyl-CoA/kg wet muscle; Ref. 27) compared with acetate (2.79 mmol acetyl-CoA/kg wet muscle; present study, P < 0.01). This potential 2 mmol acetyl-CoA/kg wet muscle increase in acetyl group production from 1 to 3 min of contraction after dichloroacetate could, if sequestered by the TCA cycle, account for 80% of the 30 mmol ATP equivalents/kg dry muscle differences in oxygen-independent ATP production observed between dichloroacetate and acetate treatments from 1 to 3 min of contraction.
Between 3 and 5 min of contraction.
During the final 2 min of contraction, no differences in PDCa and glycogen degradation existed between groups, with the availability of acetyl-CoA and acetylcarnitine in CON being only slightly lower than that seen in the acetate group after 5 min of contraction (Fig. 1, A and B). However, despite these similarities, ATP delivery from oxygen-independent routes was ∼40 mmol/kg dry muscle (or 20 mmol·kg dry muscle−1·min−1) greater in the acetate group during this period (Fig. 2). This observation could reflect differences in work performed by the muscle during this period (Fig. 3) and/or a reduced contribution of PDC-derived acetyl-CoA toward mitochondrial ATP resynthesis compared with the CON group.
The most striking observation during this period of contraction was the ∼20% improvement in the maintenance of contractile function observed in the acetate group at 5 min compared with CON. The absence of any net reduction in oxygen-independent ATP resynthesis from CON during the 5 min of contraction after acetate would, in accordance with our previous work (27), be expected to reduce contractile function to the same extent as CON at 5 min. It would therefore appear that the reduction in oxygen-independent ATP resynthesis during the first minute of contraction influences subsequent function or, alternatively, that acetate confers some beneficial effect on muscular function, independently of changes in acetyl group availability, i.e., a significant metabolic alkalosis at rest (Fig. 4; Ref. 28). Indeed, bicarbonate treatment is known to be associated with an enhancement in muscular performance, especially during periods of high-intensity exercise of short duration (14), as has previously been described within this animal model (28).
The present study demonstrated that increasing cellular acetyl group availability at the immediate onset of contraction, independently of any alteration in PDC activation status, overcame inertia in mitochondrial ATP production during the first minute of contraction. However, after this first minute, when the PDC was activated to the same extent in both groups, it appears that PDC-derived acetyl-CoA, i.e., flux through the PDC reaction, rather than increased cellular acetyl group availability per se, was primarily responsible for acetyl group delivery to the TCA cycle and therefore regulated the magnitude of contribution of mitochondrial ATP resynthesis to total energy delivery. Acetate administration was also associated with a reduction in the magnitude of isometric tension loss after 3 min of contraction compared with CON.
This work was supported by the United Kingdom Medical Research Council. P. A. Roberts was supported by a Medical Research Council Studentship and by a supplementary research award from the Gatorade Sports Science Institute.
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.
- Copyright © 2005 by American Physiological Society