INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MISMATCH BETWEEN ATP HYDROLYSIS and synthesis rates associated with intense short-term muscle contractions leads to a reduction of total adenine nucleotides (AdN = ATP + ADP + AMP) within the myocyte. The action of AMP deaminase (AMP  IMP + N₃) accounts for the elevation in blood NH₃ (actually NH₄ at plasma pH) and muscle IMP observed after very intense exercise (5, 16). IMP has little-known adverse effects, remains within the cell (18), and is reaminated back to AdN via the purine nucleotide cycle (17, 22). If, however, any IMP becomes dephosphorylated to inosine by the enzyme 5'-nucleotidase, there can be a net loss of purine nucleotides from the myocyte (1, 23). Purine nucleosides and bases, in contrast to nucleotides, can cross the muscle's sarcolemmal membrane and efflux from the cell. This is most evident in humans after bouts of intense exercise where a 15-20% decline in ATP concentration (7, 19) is due to loss of ATP degradation products from the muscle (10, 20). In fact, Hellsten et al. (7) have shown that inosine and purine base efflux accounts for 40-50% of the muscle ATP decline after intense cycle exercise.

Once the ATP-derived purine molecule is lost from the cell, recovery of the AdN pool must depend on de novo purine synthesis and/or the purine salvage metabolic pathways. These pathways normally proceed at relatively slow absolute rates (3, 24). The inadequacy of these pathways to recover muscle ATP concentration in a timely manner may be illustrated in the studies of Hellsten et al. (8, 9) and Stathis et al. (19). Repeated days of short-term intense exercise in humans led to a significant reduction (~15-20%) in resting ATP concentration in the previously exercised muscles. This implies that an overnight rest period was insufficient to permit full recovery of muscle ATP concentration, unless there is a specific adaptation within the active muscle to "reset" the muscle's ATP concentration lower. Accumulated days of deficits would lead to the measured decline in ATP concentration observed after a few days. This may be an "overtraining" phenomenon and would perhaps compromise subsequent high-intensity exercise performance.

Although the pathways for AdN synthesis proceed at relatively slow rates in skeletal muscle, a dramatic increase in their rates occurs with ribose supplementation. Perfusion of resting muscle with ~4 mM ribose increases the rate of de novo AdN synthesis by three- to fivefold (24) and the rates of adenine and hypoxanthine salvage by three- to eightfold (3), depending on the skeletal muscle fiber type. Furthermore, continuous infusion of adenine and ribose for 5 h restored rat myocardial ATP content that had been decreased by isoproterenol, whereas adenine or ribose alone had no effect (26). This demonstrates that an increased rate of adenine salvage can positively impact adenine nucleotide levels. It is unclear, however, whether the high rates of purine salvage in the presence of ribose are sustained in skeletal muscle after intense contractions (i.e., when the myocyte is recovering its resting metabolic status). Thus the purpose of this study was to measure the rates of purine salvage with and without ribose supplementation during recovery after short-term intense contractions. In anticipation that accelerated purine salvage could assist in AdN recovery, we also evaluated whether subsequent performance during intense contraction conditions could be improved.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental design. After surgical preparation of the rat hindquarter and establishment of steady-state perfusion conditions, the calf muscles of the left limb were subjected to an intense contraction sequence (known to produce a large loss of ATP to IMP). The muscles of the right limb of all animals were never stimulated to contract. To substantiate the contraction-induced reduction in ATP and increase in IMP, the muscles from both legs of six animals were removed immediately after the initial 3-min stimulation period. In the remaining animals, the contraction sequence was followed by a 1-h perfusion with 2.0 mM [³H]adenine to measure purine salvage (cf. Fig. 1). During this 1-h recovery period, 12 animals were perfused without ribose and 12 animals were perfused with ~4 mM ribose to assess the impact of ribose on adenine salvage rates. The muscles from six animals of each group were quick-frozen at the end of the 60-min recovery period. The other six animals from each group were used to assess the impact of ribose supplementation during recovery on subsequent muscle performance. The calf muscles of the left limb were again subjected to the same intense contraction protocol as before the recovery period, and the muscles of both limbs were frozen immediately after stimulation.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Experimental design. Muscles of both hindlimbs were quick-frozen from different groups of animals at 3 time points: 1) immediately after 3-min stimulation of left calf muscle (n = 6), 2) immediately after 60 min of perfusion with ~4 mM ribose (n = 6) or without ribose (n = 6), and 3) immediately after a second 3-min stimulation of left calf muscle of animals perfused either with (n = 6) or without (n = 6) ribose.

Animal care. Adult male Sprague-Dawley rats (~350 g) were housed two per cage in a temperature-controlled room with a 12:12-h day-night cycle and fed commercial rat chow and water ad libitum. The management and care of animals and the experimental protocol used in this study were approved by the University of Missouri Animal Care Committee.

Hindquarter preparation. The surgical preparation and physical system for the isolated perfused hindquarter has been described in detail previously (4, 12). With establishment of flow into the abdominal aorta and out of the vena cava, the perfusion medium was initially discarded to wash out the rat's red blood cells and then recirculated. During the steady-state flow conditions of 60 ml/min, the aortic perfusion pressure was 35-45 mmHg.

Perfusion medium. The perfusion medium consisted of Krebs-Henseleit bicarbonate buffer containing 5 g/100 ml bovine serum albumin, 100 µU/ml bovine insulin, amino acids typical for rat plasma (14), 6 mM glucose (maintained with periodic additions of glucose), 2 mM adenine at 0.2 µCi/ml using [2,8-³H]adenine (Moravek Biochemicals), and, when appropriate, 4.75 mM ribose. The medium was continuously gassed with 95% O₂-5% CO₂.

Muscle stimulation. Maximal tetanic contractions were elicited with supramaximal square-wave pulses (6-8 V, 0.1-ms duration) at 100 Hz delivered in 100-ms trains to the sciatic nerve. Resting muscle length was adjusted to produce maximal active tension. Muscle force was recorded using a Cambridge force transducer connected to a MacLab recording system. We used a stimulation protocol of 60 tetani/min for 3 min because this produces extensive ATP reduction (17).

Muscle sampling. At the appropriate time, the superficial medial gastrocnemius (predominantly fast-twitch white fibers), the deep lateral gastrocnemius (predominantly fast-twitch red fibers), and plantaris (mixed-fiber type) muscle sections were excised and quick-frozen in tongs cooled in liquid nitrogen. Samples were stored at 80°C until analyzed. Values are given as micromoles per gram wet weight.

Metabolite analyses. Metabolites from muscle and perfusate samples were extracted in perchloric acid as done previously (24). AdN, purine nucleosides, and bases were quantified via HPLC, as described previously (24). Adenine, ATP, ADP, and IMP peaks were collected and counted in a Beckman LS8000 scintillation counter. Phosphocreatine (PCr) and creatine (Cr) were measured by HPLC as described by Wiseman et al. (25). Muscle glycogen was determined using the anthrone method (6). Glucose and lactate were measured using a Yellow Springs Instrument model 2700 autoanalyzer. Ribose was measured using a refractive-index detector (Varian Star 9040) as described by Masson et al. (15).

Salvage rate calculation. The salvage rates were calculated from the rate of labeled adenine incorporated into ATP, ADP, and IMP (dpm · g¹ · h¹) divided by the adenine specific activity measured in each muscle (dpm/nmol).

Statistical procedures. All data are expressed as means ± SE. Statistical differences between groups were determined using analysis of variance. When statistical significance (P < 0.05) was detected, Tukey's post hoc test was used, as appropriate (21); n equals the number of rats per group.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adenine salvage rates. Rates of adenine salvage for resting muscle in the presence of 2.0 mM adenine, as summarized in Fig. 2 and Table 1 for the mixed-fiber plantaris muscle (~45 nmol · h¹ · g wet wt¹), the fast-twitch red gastrocnemius section (~55 nmol · h¹ · g wet wt¹), and the fast-twitch white gastrocnemius section (~35 nmol · h¹ · g wet wt¹), are typical of values obtained in previous work (3). Note that adenine salvage in the absence of ribose supplementation is lowest in the fast-twitch white fiber section and highest in the fast-twitch red muscle section.

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Adenine salvage rates in resting and postcontraction recovering muscle with (+) and without () ~4 mM ribose in the perfusion medium. Values are the average (means ± SE) of all muscles that were resting or recovering for 60 or 63 min (see Experimental design). *Significantly less than rested white gastrocnemius + ribose (paired comparison), P < 0.001.

1

1

1

1

1

1

Muscle performance. Initial force development was similar among the treatment groups (3-min stimulation = 2,941 ± 144 g; no ribose 1-h recovery = 2,505 ± 174 g; no ribose 1-h recovery + 3-min stimulation = 2,952 ± 127 g; ribose 1-h recovery = 2,583 ± 183 g; ribose 1-h recovery + 3-min stimulation = 2,584 ± 199 g). As illustrated in Fig. 3, force development of the calf muscle group declined precipitously over the first 1-1.5 min, as expected for this intense contraction sequence. All groups exhibited a similar fatigue pattern.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Decline in force development of the calf muscles during the first and second contraction sequence at 60 tetani/min. Initial force for the first contractions sequence was not different across groups and averaged 2,713 ± 78 g (n = 30) for the 5 groups. Values are means ± SE. Some SE bars are hidden behind the symbols.

Muscle metabolite contents. Tables 2, 3, and 4 contain metabolite levels measured in plantaris, red gastrocnemius, and white gastrocnemius muscle sections, respectively. Because these values were remarkably similar among the muscles of the right leg, we have pooled these data for simplicity in Tables 2-4. However, because glycogen content in the right leg plantaris muscle increased over time, we have included only the value obtained after the 3-min stimulation period in Table 2. Also, PCr content in the right leg white gastrocnemius section decreased over time; thus we have given the value obtained after the initial 3-min stimulation in Table 4. For all metabolites, statistical comparisons were made only between right and left muscles in corresponding treatment groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adenine nucleotide synthesis rates from the de novo pathway (24) and the salvage pathway (3) are relatively low in skeletal muscle but differ among the muscle fiber phenotypes. The rates of adenine salvage, among the three fast-twitch muscle sections evaluated in this study, were not reduced after intense contractions in the absence of ribose (Fig. 2). This is in contrast to the rates of de novo synthesis during muscle contractions for these same muscles (24). This implies either a fundamental difference in the control of de novo synthesis and the salvage pathways or, more likely, a meaningful difference in the cellular environment favoring anabolic processes. It is likely that during the postcontraction recovery period muscles did not experience the relatively high energy demands that occurred during contractions. Thus energy-requiring processes should be favored to occur at a higher rate, possibly accounting for the similar rates of salvage in resting and recovering muscle.

As observed previously (3), in the presence of ribose supplementation (~4 mM) exceptionally high rates of adenine salvage occur in skeletal muscle (cf. Fig. 3). The increases are approximately five- to sevenfold, depending on the muscle section. This implies that ribose availability, presumably by the provision of 5-phosphoribosyl-1-pyrophosphate, is rate limiting in controlling the purine salvage pathway. Interestingly, the high rate of adenine salvage in the rested fast-twitch white muscle section (262 ± 10.5 nmol · h¹ · g¹; n = 12) was not achieved in the postcontraction recovering white muscle section (215 ± 14.2 nmol · h¹ · g¹; n = 12) of the contralateral limb. It is unclear what caused this modest ~20% difference, because this was not observed in the other muscle sections of the same animals. One confounding factor may be the absence of full metabolic recovery in this fast-twitch muscle section. For example, ATP remained below and IMP remained above that of the rested white gastrocnemius muscle of the contralateral limb. This is in contrast to the high rate of adenine salvage with ribose and the full recovery apparent in the fast-twitch red muscle sections. The plantaris exhibited similar high rates of adenine salvage in the presence of ribose despite not showing full recovery of ATP during the 60-min recovery period. This mixed result may be due to the plantaris being composed of both red and white fast-twitch fibers. It is conceivable that the red fibers showed full recovery, as in the red gastrocnemius, and that the plantaris white fibers showed a similar response to those of the white gastrocnemius. This difference between muscle fiber types in the rate and/or extent of metabolic recovery is similar to that found in vivo after intense treadmill running (22). The relatively meager metabolic and vascular flow capacities of the white gastrocnemius fiber section are likely responsible, especially when considering the extreme metabolic deficit caused by the initial contraction condition. The fast-twitch white muscle fibers have the greatest need for ATP recovery and yet appear to be the slowest in achieving the return to its resting metabolic status.

Even though ribose established a marked increase in the rates of adenine salvage, there was little discernable impact on the adenine nucleotide content within the muscle. This was expected, because even the high rates of adenine salvage of 200-250 nmol · h¹ · g¹ are relatively small compared with the large pool of adenine nucleotides in the muscle (8,000-9,000 nmol/g). Thus the short time of this experiment of only 1 h limits our ability to expand the adenine nucleotide pool within the muscle. More importantly, however, there was no meaningful loss of the adenine nucleotides out of the muscle after the intense contraction sequence. Note that the total adenine nucleotide + IMP pool size did not significantly decrease in any of the muscle samples in this experiment (cf. Table 2-4). Rather, reamination of IMP via the purine nucleotide cycle was effective at recovering the cell's adenine nucleotide pool. This response for rat skeletal muscle appears to be quite different from that observed in humans, where a significant efflux of adenine nucleotide degradation products is observed from muscle during and after intense exercise (7). This likely contributes to the reduction in ATP content within the active muscle after days of intense exercise (9, 11, 19). It should also be pointed out that contrasts among in fiber type characteristics, i.e., oxidative capacity and maximal blood flow, are much greater in rats than in humans. Accelerated rates of adenine salvage induced by ribose supplementation could be effective at improving recovery of ATP contents, if our results are applicable to human muscle.

The absence of any apparent influence of ribose supplementation on muscle performance during the second contraction sequence could be predicted by the reasons mentioned above. The difference in force production between the first and second contraction sequences, irrespective of the presence of ribose, is likely due to contractile and/or metabolic failure of the fast-twitch white muscle fibers. During the first contraction sequence IMP content increased ~4 µmol/g in these fibers, whereas in the second contraction sequence the IMP increase was only 0.5-1.0 µmol/g. This indicates that the mismatch in rates of ATP supply to ATP demand was greater in the first vs. the second contraction sequence. The simplest explanation for this outcome is that the white gastrocnemius muscle fibers developed less force and thereby experienced less energy demand during the second stimulation protocol. On the contrary, the high-oxidative red muscle fibers showed a similar level of energy imbalance in the first and second contraction sequences. The increase in the IMP content of these fibers was 1.5 µmol/g in each contraction protocol, suggesting these fibers experienced a similar energy demand, or force production, during both sets of contractions. Recall that the initial force of the second contraction sequence began at 60% of the initial force observed earlier during the first contraction sequence (Fig. 1). Because fast-twitch white muscle fibers comprise 65-75% of the calf muscle fibers (2), it is likely that most of these fibers were not contributing much force during the second stimulation protocol. This, of course, assumes that the high-oxidative red fibers were responding normally.

In conclusion, our results demonstrate that the inherent adenine salvage rates in skeletal muscle were unaltered during recovery after intense muscle contractions. These salvage rates were increased markedly by ribose supplementation and sustained during recovery, except for a small decrease in the low aerobic capacity white gastrocnemius muscle.