INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LOWER BLOOD LACTATE LEVELS during exercise are a distinguishing adaptation after endurance training (8, 19). Because blood lactate concentration is a function of simultaneous lactate production and utilization, blood lactate levels in trained individuals may result from attenuated lactate production (14, 19) or enhanced lactate clearance (8-10). Enhanced lactate clearance after endurance training has been reported both during exercise (8) and at rest during lactate infusions (9, 10). The fate of the enhanced lactate clearance in trained individuals has been attributed primarily to gluconeogenesis and oxidation (9, 24). In support, a recent report from our laboratory has demonstrated an enhanced hepatic gluconeogenic capacity in perfused livers after endurance training (31). This may explain the maintenance of blood glucose levels observed in trained animals during exercise (12). In addition, the oxidation of lactate has been observed to be significantly elevated during exercise after endurance training (24). Lactate clearance via oxidation has also been reported to be elevated in resting trained animals made hyperlactatemic via exogenous infusion (9, 10). However, the precise loci for enhanced lactate oxidation after endurance training remain to be resolved.

Although active skeletal muscle appears to be a significant site for the training-enhanced lactate removal during exercise (2), the importance of inactive muscle remains to be resolved. Training-induced elevations in skeletal muscle oxidative enzyme activities (19) and the shift in the isoenzyme profile for lactate dehydrogenase (34) portend a greater capacity for lactate oxidation. It is also possible that endurance training may improve lactate removal via enhanced skeletal muscle glyconeogenesis. Talmadge and Silverman (32) observed elevations in several key glyconeogenic enzymes in chronically active muscles from mice. However, to date, only one study has specifically examined the potential for training to augment lactate disposal in resting skeletal muscle (5). During strenuous leg exercise, Buckley et al. (5) compared lactate uptake and disposal in the "trained" and "untrained" forearms of competitive squash players. These investigators found no difference in lactate uptake or disposal between forearms. However, the reliance on indirect measurements of blood flow, potential differences in muscle mass, and assumptions inherent to non-steady-state conditions may have masked training differences. Furthermore, their approach did not allow them to assess any adaptations in the pathways for lactate removal.

In the present study, we examined the impact of endurance training on lactate metabolism in quiescent skeletal muscle. Utilizing the isolated hindlimb perfusion technique, we assessed total lactate removal and the primary pathways for lactate removal under conditions of elevated lactate concentration. The glyconeogenic capacity was further resolved in distinct muscles known to contain specific fiber types. The results indicate that endurance training does not alter lactate removal, lactate oxidation, or glyconeogenesis in quiescent skeletal muscle.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiments were performed on female Wistar rats (initially 120-150 g) housed individually in a temperature-controlled room and maintained on a 12:12-h light-dark cycle. Animals were randomly assigned to either the control (n = 5) or endurance-trained (n = 5) group. Trained animals were run on a motorized treadmill 5 days/wk for 8 wk. The running time and speed progressively increased until the animals were running 40 m/min, 10% grade, for 60 min by the sixth week. The animals were maintained at this workload for the remaining 2 wk. Three weeks before the experiments, control animals were acclimatized to treadmill running 2 days/wk for 10 min/day at 13 m/min, 10% grade. Forty-eight hours before the experiment, a final exercise bout was administered to both groups (20 m/min, 10% grade, 60 min) to equate any residual effects from the last training bout. After this exercise session, the animals remained inactive with food and water provided ad libitum before the experiments.

The hindlimb was surgically isolated, essentially as described by Ruderman et al. (29), with modifications. Briefly, after anesthetization (ketamine, xylazine, and acepromazine), an incision was made on the ventral surface of the right hindlimb and the skin was pulled back to expose the ventral and dorsal aspects of the leg musculature. The superficial epigastric vessel was then ligated near its branch point from the femoral artery and vein. The abdomen was opened, the bladder was ligated and drained, and the mesenteric, hypogastric, ovarian, uterine, and iliolumbar blood vessels were ligated near their branch points. Ligatures were also placed and secured around the tail and the residual portions of the uterus and descending colon. Unlike the surgical preparation outlined by Ruderman et al., loose ligatures were then placed around the iliac artery, the common iliac blood vessels of the contralateral hindlimb, and the inferior vena cava near the common iliac branch point. Before the cannulation of the right iliac artery, the tip of the cannula was filled with heparinized saline. The arterial cannula was then inserted and secured, and flow was immediately established. The venous cannula was then rapidly inserted and secured, and the contralateral hindlimb was ligated. The soleus (Sol), plantaris (Plt), red gastrocnemius (RG), and white gastrocnemius (WG) were removed from the contralateral hindlimb and freeze-clamped with aluminum tongs precooled in liquid nitrogen. The remaining musculature (vastus, gracilis, biceps femoris, and so on) was then quickly dissected and freeze-clamped.

All perfusions were single-pass with a perfusion medium consisting of Krebs-Henseleit buffer with 25 mM sodium bicarbonate, dialyzed bovine serum albumin (40 g/l; fraction V, Sigma Chemical), and 2.5 mM calcium chloride. Fresh bovine erythrocytes were obtained and thoroughly washed before their addition to the Krebs-Henseleit bicarbonate buffer. Neutralized lactate and [U-¹⁴C]lactate were added to the perfusate and placed in a circulating water bath maintained at 37°C with continual mixing. The final lactate concentration of the perfusate was 10 mM with a [¹⁴C]lactate specific activity of 3,000 dpm/µmol. The final hematocrit (Hct) of the perfusate was ~30%, and the pH was checked to ensure a range of 7.1-7.4. Once the right hindlimb isolation was achieved and the left hindlimb muscles were dissected, the animal was placed in a Plexiglas perfusion chamber apparatus as previously described (28). Humidified air was circulated and maintained at 37°C. Before entering the hindlimb, the perfusate was filtered through a nylon mesh, oxygenated, and passed through a bubble trap. Arterial pressure was monitored in-line within the perfusion chamber. To minimize any alteration in the perfusion chamber temperature, all sampling occurred externally via tubing extensions.

After the surgical isolation and the adjustment of flow to ~5 ml/min, the perfusate was allowed to pass through the preparation with no sample collection for 10 min. This "wash out" period allowed for equilibration between the perfusate and tissue. Preliminary work (unpublished findings) demonstrated the attainment of steady-state values by minute 10 that were observed to vary randomly by less than 0.5%/min thereafter. Arterial and venous blood samples were collected at minutes 10, 20, 30, and 40. Blood samples for ¹⁴CO₂ evolution, Hct, and blood pH/gas analysis (Radiometer BMS3 Mk2) were collected anaerobically at minutes 15, 25, and 35. After the 40-min perfusion, the hindlimb was exposed, and the Sol, Plt, WG, RG, and remaining musculature were rapidly excised and freeze-clamped as described for the contralateral hindlimb. All tissue samples were stored at 70°C for subsequent analyses. Preliminary experiments (n = 4) were performed to establish the experimental protocol used above. In these preliminary experiments, dye (India ink) was injected into the arterial line to confirm the isolation of the hindlimb and the musculature perfused. The weight of the muscle mass perfused was determined in a separate group of control (n = 5) and trained (n = 5) animals undergoing a similar training protocol.

Blood samples were collected in chilled test tubes containing sodium fluoride and heparin. A portion of the blood was deproteinized in 8% perchloric acid and neutralized with KOH in preparation for lactate analysis (18) and ion exchange chromatography (28) for the determination of lactate specific activities (LSA). Pyruvate assays (22) were performed on the remaining blood samples after deproteinization with perchloric acid and neutralization with K₃PO₄ · 7 H₂O. Perfusate PO₂, PCO₂, and pH were determined from anaerobically drawn samples utilizing a blood-gas analyzer (Radiometer BMS3 Mk2). The hematocrit was measured, and hemoglobin (Hb) content was determined by using a regression equation, Hb = 0.421(Hct)  0.683, established in our laboratory for bovine erythrocytes (unpublished findings). Oxygen consumption (O₂) was then calculated from PO₂ and Hb values and corrected for oxygen saturation, as previously described (26). The perfusate bicarbonate concentration was determined from the Henderson-Hasselbach equation utilizing measured values for pH and PCO₂. CO₂ production was calculated from the arterial-venous difference and perfusate flow rate, with bicarbonate assumed to account for 60% of CO₂ transport. ¹⁴CO₂ was determined by utilizing the method of Chan and Dehaye (6). Glycogen content (13) and glycogen specific activity (7) were determined from specific muscle samples after homogenization and solubilization in KOH (16). The remaining musculature was pulverized under liquid nitrogen and deproteinized in ice-cold perchloric acid. After centrifugation, aliquots were neutralized with KOH. Ion-exchange chromatography was applied to aliquots of the supernatant. The lactate, pyruvate, and amino acid eluants were collected, and the radioactivity was determined via liquid scintillation counting. In addition, a portion of the remaining musculature from both the perfused and nonperfused contralateral limb was homogenized for the determination of protein (23), ATP (20), and creatine phosphate (17) concentration. A different portion of the musculature was weighed before and after heating in an oven at 80°C for 48 h to determine the water content. Skeletal muscle succinate dehydrogenase (SDH, EC 1.3.99.1) activity was determined only on the remaining musculature of the contralateral hindlimb (33).

Rates of lactate uptake and O₂ were calculated as the product of arterial-venous difference (µmol/ml) and flow (ml/min). ¹⁴C activities (dpm/ml) were normalized to an arterial LSA of 3,000 dpm/µmol. [¹⁴C]lactate uptake (dpm/min) was calculated as the product of the normalized arterial-venous [¹⁴C]lactate difference and the flow. Tracer estimated lactate removal (µmol/min) was calculated as the [¹⁴C]lactate uptake divided by the normalized venous LSA. ¹⁴CO₂ was calculated as the product of venous-arterial difference (dpm/ml) and flow. Finally, tracer estimated glycogen synthesis (µmol C6 · 40 min¹ · g¹) was calculated as the [¹⁴C]glycogen counts (dpm/g) incorporated from [¹⁴C]lactate divided by two times the normalized venous LSA where C6 represents glucosyl units. Steady-state values represent an average of four serial samples that varied randomly by <0.5%/min. All differences between conditions as well as pre- and postperfusion measurements were compared by utilizing a Student's t-test. Where appropriate, comparisons of specific muscles from the experimental and contralateral leg between pre- and postgroups, as well as the levels of perfusate pressure and lactate uptake examined over the perfusion time, were analyzed by using an analysis of variance and Tukey's post hoc test. Statistical significance was accepted at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The body weights for trained animals (279.6 ± 8.6 g) were not significantly different from controls (286.1 ± 6.5 g) after the 8 wk of training. Preliminary perfusion experiments (n = 4) confirmed the isolation of the hindlimb musculature. The perfused muscle mass, determined on a separate group of animals with body weights of 282 ± 7 and 275 ± 9 g for trained and control animals, respectively (P = 0.45), was not significantly different between control (15.2 ± 0.5 g) and trained (15.6 ± 0.9 g) animals.

There were no significant differences in perfusate flow between trained (5.64 ± 0.58 ml/min) and control hindlimbs (5.43 ± 0.56 ml/min). In addition, no significant differences were observed between groups in perfusate Hct and pH (Table 1). Perfusate pressure was not significantly different between the groups, nor did it vary significantly during the experiment (Fig. 1). However, skeletal muscle SDH activity was significantly elevated in trained (4.50 ± 0.34 µmol · min¹ · g¹) compared with control animals (2.94 ± 0.27 µmol · min¹ · g¹). After the hindlimb perfusions, there was no visual evidence of swelling. Finally, there were no significant differences in skeletal muscle protein, water, ATP, and creatine phosphate between the nonperfused contralateral hindlimb and the hindlimb perfused for 40 min (Table 2).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Arterial pressure (mean ± SE) during isolated hindlimb perfusion. Each point represents the arterial pressure from 5 perfusions/group.

The initial lactate concentration in the perfusate was not significantly different between trained (10.59 ± 0.18 mM) and control conditions (10.35 ± 0.25 mM) (Table 1). During the 40 min of perfusion, no significant differences in skeletal muscle lactate uptake were observed between trained (7.31 ± 0.20 µmol/min; 0.47 ± 0.01 µmol · min¹ · g¹) and control hindlimbs (6.98 ± 0.43 µmol/min; 0.46 ± 0.03 µmol · min¹ · g¹), nor did it significantly vary during the perfusion (Fig. 2). In support, no significant differences were observed for [¹⁴C]lactate uptake (22,776 ± 370 vs. 21,924 ± 1,373 dpm/min) and ¹⁴CO₂ production (2,203 ± 185 vs. 2,098 ± 155 dpm/min), trained vs. control, respectively (Fig. 3). The O₂, 4.93 ± 0.18 and 4.92 ± 0.13 µmol/min (0.32 ± 0.01 and 0.32 ± 0.01 µmol · min¹ · g¹), and respiratory quotient, 0.88 ± 0.03 and 0.87 ± 0.05, were not significantly different for trained and control hindlimbs, respectively. Tracer-estimated lactate removal was not significantly different between trained (7.86 ± 0.24 µmol/min) and control hindlimbs (7.43 ± 0.47 µmol/min). In addition, tracer-estimated lactate removal rates were not significantly different from the measured lactate uptake rates for trained and control groups.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Lactate uptake (mean ± SE) during the isolated hindlimb perfusion. Each point represents lactate uptake from 5 perfusions/group.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Effect of endurance training on skeletal muscle lactate metabolism. LAup, lactate uptake (µmol/min); O2, oxygen consumption (µmol/min); 14C-LA, [14C]lactate uptake (dpm/min); 14CO2, 14CO2 production (dpm/min); Con, control; Tr, trained. Radioactive values are normalized to 3,000 dpm/µmol. Values are means ± SE for 5 animals/group.

Significant increases were observed for both groups in glycogen concentration in the Plt and RG after the perfusion (Table 3). Concomitant, although nonsignificant, increases were also observed in the Sol, the WG (Table 3), and the remaining musculature of the contralateral hindlimb vs. the perfused hindlimb for both the trained (6.16 ± 0.30 to 6.61 ± 0.48 mg/g) and control animals (5.30 ± 0.69 to 5.78 ± 0.32 mg/g). Although the muscles from trained animals tended to have higher skeletal muscle glycogen levels before and after the perfusion, there were no significant differences between groups in the relative increase in glycogen deposition in any specific muscle examined (Table 3). No differences were observed for [¹⁴C]lactate incorporation into [¹⁴C]glycogen between trained and control hindlimbs (Fig. 4). When the plantaris was used as a representative sample for the muscle fiber composition of the perfused hindlimb (25), no significant differences were observed for net glycogen synthesis, 7.1 ± 0.4 vs. 6.5 ± 0.3 µmol · 40 min¹ · g¹ for trained vs. control, respectively. Similarly, tracer-estimated glycogen synthesis was not significantly different between trained (2.43 ± 0.10 µmol · 40 min¹ · g¹) and control hindlimbs (2.26 ± 0.11 µmol · 40 min¹ · g¹). The amount of [¹⁴C]lactate incorporated into glycogen amounted to ~34% of the actual glycogen synthesis calculated from pre- and postglycogen levels for both trained and control hindlimbs.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Effect of endurance training on the deposition of radioactive carbons in glycogen ([14C]glycogen; in dpm/g) in the soleus (Sol), plantaris (Plt), red gastrocnemius (RG), and white gastrocnemius (WG). Values are means ± SE for 5 animals/group.

Finally, the normalized venous LSA effluent was not significantly different between trained (2,893 ± 36 dpm/µmol) and control perfusions (2,937 ± 41 dpm/µmol). Glycogen synthesis accounted for ~23% of the total [¹⁴C]lactate uptake in both trained and control hindlimbs (Table 4). Furthermore, no significant differences were observed between groups in the relative amount of radioactive compounds from the venous effluent of the hindlimb nor recovered in the perfused muscle extract (Table 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Despite evidence that endurance training augments skeletal muscle lactate uptake (1, 3) and whole body lactate clearance (4, 9), resting muscle does not appear to be a locus for enhanced lactate removal. Hindlimbs from trained and control animals demonstrated not only similar net lactate uptake but also equivalent rates of [¹⁴C]lactate uptake and ¹⁴CO₂ efflux. This suggests that, despite the training-induced elevation in skeletal muscle oxidative capacity, no augmentation in skeletal muscle lactate removal or oxidation is observed at rest. Furthermore, training did not alter rates of glyconeogenesis as determined by net glycogen deposition and [¹⁴C]glycogen accumulation in any of the muscles studied. As such, in the presence of elevated lactate concentration, endurance training does not affect lactate removal or any of the pathways for lactate removal in quiescent skeletal muscle.

The present results support the only other study to date examining the role of endurance training on resting skeletal muscle lactate disposal. Buckley et al. (5) recently reported lactate removal in the resting trained and untrained forearms of competitive squash players during strenuous leg exercise yielding arterial lactate concentrations of 11 mM. Under the non-steady-state conditions of that study, a significant portion of the lactate uptake is not metabolized, and lactate uptake will not equal true lactate removal, i.e., metabolic conversion. Although inactive muscle was shown to be a significant site for both lactate uptake and disposal, no difference was observed between the trained and untrained forearms (5). Although Buckley et al. (5) did not examine the pathways for lactate disposal, e.g., oxidation or glyconeogenesis, the present findings indicate that endurance training does not alter these pathways in resting skeletal muscle.

Several previous observations had led us to suspect quiescent skeletal muscle as a potential site for enhanced lactate removal. When lactate levels were elevated at rest via exogenous infusion, trained animals demonstrated enhanced lactate clearance (9, 10). In those studies ¹⁴CO₂ was the primary fate for [¹⁴C]lactate, indicative of enhanced lactate clearance via oxidation. The elevation in skeletal muscle oxidative enzyme activities (19) as well as the alterations in lactate dehydrogenase toward the heart isoform (34) after endurance training suggested skeletal muscle as a site for enhanced lactate oxidation (15, 24). Indeed, at elevated lactate levels, our laboratory has previously observed greater lactate oxidation by an oxidative muscle preparation than by a glycolytic muscle preparation using a perfused rabbit hindlimb model (28). There are several possible explanations for the apparent discrepancy between these previous observations and the present findings. Lactate oxidation in resting muscle may be limited by the low overall rate of O₂ by this tissue, which was not significantly different between control and trained hindlimbs. At a respiratory quotient of 0.87, lactate oxidation as measured by ¹⁴CO₂ accounted for ~80% of nonlipid O₂ in the present study. Even at 100% of the nonlipid O₂, the oxidation of lactate could have accounted for no more than 12% of the lactate removed in the present preparation. Our previous observation of greater lactate oxidation by an oxidative vs. glycolytic muscle preparation (26) reflected a difference in the cytochrome c concentration of approximately fivefold, considerably greater than the twofold increase observed with endurance training (19). In that study, the oxidative preparation also demonstrated a significantly greater resting O₂ rate than the glycolytic preparation (26), something that was not observed after training. That trained animals at rest demonstrate enhanced whole body lactate clearance via oxidation may be explained by adaptations in other tissues, e.g., liver. In support of this, we have previously observed increased lactate oxidation in perfused livers from trained animals (31). Finally, it is possible that the increase in ¹⁴CO₂ yield during exogenous lactate infusion was the result of increased gluconeogenesis in trained animals (31) not oxidation. Gluconeogenesis from [¹⁴C]lactate will generate ¹⁴CO₂ as a result of isotopic exchange at the level of oxaloacetate. The generation of ¹⁴CO₂ by isotopic exchange could potentially have accounted for a substantial fraction of the total ¹⁴CO₂ produced in our earlier study (21). If this were the case, the apparent enhanced lactate clearance via oxidation might actually be a reflection of enhanced clearance via gluconeogenesis.

The ability of skeletal muscle to convert lactate to glycogen has previously been observed with the laboratory rat used as a model of glyconeogenesis (25, 30). As such, the present results are consistent with previous reports of lactate incorporation into glycogen primarily by fast-twitch white and fast-twitch red muscle. This also supports previous work by our laboratory demonstrating quantitatively greater glyconeogenic rates from lactate in glycolytic and mixed muscle preparations than from slow oxidative muscle fibers in the rabbit (28). Unlike in previous reports (25, 30), glycogen-depleted muscles were not used in the present study to augment the absolute glyconeogenic rates and may account for the lower rates of glycogen deposition, as well as the absence of any training-induced alteration. However, in agreement with these earlier reports, we observed similar metabolic fates of the radioactive lactate with a significant amount (i.e., 23%) of the tracer incorporated into glycogen. Also, in accord with a previous report (25), only a small fraction of the [¹⁴C]lactate removal was observed in the form of ¹⁴CO₂.

Recent studies (1, 3) have reported increases in monocarboxylate transporters (MCT) after a program of endurance training. Although we did not measure MCT-1, we suspect that levels were increased in our study owing to the fact that our training protocol consisted of a greater workload, sustained over a longer period of time, than these earlier studies. Increases in skeletal muscle MCT-1 after training have been associated with increased lactate uptake by muscle strips (1, 3). However, in those studies, lactate uptake was measured over a very short time interval, ensuring that the measurement reflected primarily lactate transport. In contrast, the present study examined lactate removal under steady-state conditions, over a considerable time period, in which lactate removal reflects loss due to metabolic conversion, e.g., to glycogen or CO₂. At steady state, we have shown that muscle lactate concentration equilibrates with the perfusate lactate concentration during in situ perfusions (27, 28). Thus our experimental design would be unlikely to reveal any differences in lactate transport capacity between groups.

Although tracer-estimated lactate uptake was not significantly different from net lactate uptake in the present study, tracer-estimated glycogen synthesis was significantly lower than the measured rates of net glycogen deposition. This discrepancy between tracer-estimated and net rates of glycogen synthesis appears unique to studies employing the rat hindlimb (25, 30) and was not observed by our laboratory in perfused rabbit muscles (28). McLane and Holloszy (25) offered the loss of tracer by isotopic exchange as a putative explanation. However, there is little evidence that oxaloacetate and the tricarboxylic acid cycle are involved in skeletal muscle glyconeogenesis (11). No other explanation has been offered, and the discrepancy between net and tracer determined glyconeogenic rates in the rat hindlimb remains to be resolved.

In summary, the present results indicate that lactate removal by quiescent skeletal muscle is not augmented after endurance training. Lactate removal by either oxidation or glyconeogenesis is also unaffected by endurance training in quiescent skeletal muscle. As such, inactive trained skeletal muscle does not appear to be a site for enhanced lactate removal. Under exercise conditions, contracting skeletal muscle appears to be the likely candidate for enhanced lactate clearance (2). Under resting conditions, alternative tissues, e.g., the liver, appear better candidates as the loci for enhanced lactate clearance after endurance training.