To circumvent problems inherent to the above described preparations, we developed several unique perfused muscle preparations in the rabbit (32). The first two, the glycolytic (gracilis) and oxidative (soleus), are essentially homogeneous for Type IIb and Type I fibers, respectively (Table 1). A third mixed muscle fiber preparation, the triceps surae, consists of all fiber types including Type IIa fibers. Preliminary studies were conducted to determine a common flow rate, ∼0.4 mL·min−1·g−1 muscle, at which both oxygen uptake and lactate release/uptake were observed to be independent of flow. Under such conditions, these preparations were shown to be stable for periods of up to 3 h as assessed by steady-state metabolic parameters, as well as pre- and post-skeletal muscle values for ATP, PCr, protein, and water content. When perfused with both labeled and unlabeled lactate, rates of tracer removal/incorporation could be compared with the net uptake/production of the same metabolites. Thus, these preparations allowed for the quantitative assessment of all major pathways of lactate disposal by the various fiber types at rest.
All three muscle preparations demonstrated net lactate release at rest when perfused with 1 mM lactate, i.e., resting lactate concentration (Fig. 1) (32). Under these conditions, net lactate release was greatest for the glycolytic preparation and lowest for the oxidative preparation (9.7 ± 0.5 vs 3.7 ± 0.2 μmol·100 g−1·min−1). As the lactate concentration in the perfusate was increased, all muscle preparations demonstrated a transition from net lactate release to net uptake (33). For the oxidative muscle preparation, this transition occurred at 1–2 mM lactate, whereas the glycolytic preparation did not switch to uptake until perfusate lactate concentration reached ∼4 mM. It was further observed that at any lactate concentration above 4 mM the oxidative preparation demonstrated greater lactate removal when compared with the glycolytic preparation. Results from the mixed preparation, which demonstrated the highest rates of lactate removal, suggested a substantial capacity for removal by the Type IIa fibers. It should be noted that in the rabbit Type IIa fibers have been shown to have the highest oxidative capacity (34), whereas in the rat Type IIa also retain substantial glyconeogenic capacity (23,37). That oxidative fiber types switch from lactate production to consumption at lower lactate concentrations and maintain higher rates of lactate disposal at elevated lactate concentrations is consistent with the model proposed earlier by Brooks (6), i.e., the shuttling of lactate from glycolytic to oxidative fibers.
Pathways of Lactate Disposal
To assess the relative contribution of various metabolic pathways to the disposal of lactate in different fiber types, we perfused muscles at 8 mM lactate and 5 mM glucose in the presence of 14C-lactate and [3H]glucose for 2 h at rest (33). Net uptake/production values for lactate, glucose, pyruvate, alanine, and oxygen were determined. The uptake of 14C-lactate was quantified, as was the incorporation of 14C into CO2, pyruvate, amino acids, and glycogen. Total tracer recoveries were essentially complete, i.e., 95 ± 2%.
All preparations demonstrated an uptake of 14C-lactate quantitatively consistent with their net extraction of unlabeled lactate. As indicated above, this was greatest for the mixed preparation (reflecting the influence of Type IIa fibers) and lowest for the glycolytic preparation. The primary pathways for lactate disposal were oxidation, glyconeogenesis, and transamination, though their relative significance varied considerably across preparations (Fig. 2). The oxidation of lactate to carbon dioxide was a significant pathway for disposal in all preparations representing 28%, 51%, and 39% of the lactate disposed in the glycolytic, oxidative, and mixed preparations, respectively. Absolute rates of 14CO2 evolution were similar for the oxidative and mixed preparations and substantially higher than that for the glycolytic preparation. The incorporation of 14C into amino acids also constituted a primary pathway for removal in all three preparations, oxidative = 32%, glycolytic = 20%, and mixed = 27%. Tracer estimated rates of transamination based on the lactate specific activity were 80−100% of the measured net production of alanine.
Tracer estimated glyconeogenic rates (μmol·100 g−1·2 h−1) were 227 ± 8 and 175 ± 13 for the glycolytic and mixed preparations, respectively. In contrast the oxidative preparation demonstrated glyconeogenic rates that were only 10% of those observed for the glycolytic preparation. If one assumes that the Type IIb and I fibers of the mixed preparation demonstrate glyconeogenic rates comparable to the glycolytic and oxidative preparations, respectively, then the estimated glyconeogenic rate for the Type IIa fibers would be ∼210 μmol·100 g−1·2 h−1, equivalent to the Type IIb fibers. By contrast, the apparent glyconeogenic rates from glucose (in the absence of insulin) as determined by 3H incorporation were similar across all preparations, i.e., 72–91 μmol·100 g−1·2 h−1. Thus, glyconeogenesis was a significant pathway for lactate disposal only for the Type II fibers and was the dominant pathway for removal in the glycolytic preparation, i.e., Type IIb fibers. These results were consistent with previous studies employing the rat hindlimb (23,37).
Pathway for Skeletal Muscle Glyconeogenesis
The precise pathway by which glyconeogenesis may occur in skeletal muscle has puzzled physiologists for some time. In particular much debate has centered on the formation of phosphoenolpyruvate (PEP) and circumvention of the apparently unfavorable thermodynamics of the pyruvate kinase reaction. In the liver, PEP is formed from pyruvate via a complex set of reactions that traverse the tricarboxylic acid cycle and bypass pyruvate kinase. A characteristic of the hepatic glyconeogenic pathway is that any label introduced on a specific position of the lactate molecule (e.g., 2-14C-lactate) tends to become highly randomized in PEP and glucose/glycogen (16). This is a direct result of entering the TCA cycle and equilibrating with symmetrical intermediates. That such randomization of label apparently does not occur during skeletal muscle glyconeogenesis has led most investigators to propose alternate routes (14,27,30,38). However, these previous studies had failed to demonstrate this lack of randomization in concert with a demonstration of net synthesis. Such observations were criticized by Krebs and Woodford (18), who suggested that the incorporation of label from lactate into skeletal muscle glycogen simply reflected isotopic exchange, not actual synthesis.
We chose to reinvestigate the putative role of the TCA cycle in skeletal muscle glyconeogenesis under conditions of demonstrable net glycogen synthesis (31). As an alternative to determining the randomization of label in glycogen, a laborious process, we compared rates of label incorporation from lactate labeled at different carbons, i.e., 1-14C or 2-14C lactate. If previous studies demonstrating a lack of randomization were accurate, the recovery of 14C in glycogen should be similar for the two labeled molecules. Alternatively, if the molecule did traverse the TCA cycle during glyconeogenesis substantial discrepancies in the recovery of 14C would be expected for 1-14C- versus 2-14C-lactate (16). As this approach required the use of separate preparations for each positional isotopomer, we were compelled to demonstrate comparable conditions across these preparations. Table 2 illustrates that net lactate uptake, 14C-lactate uptake, 14C-lactate specific activity, and 14CO2 yields were not significantly different between muscles perfused with 1-14C- versus 2-14C-lactate. As well, net glycogen deposition was the same across preparations and was not significantly different from the same muscles perfused under similar conditions with U-14C-lactate (Fig. 3). Consistent with previous observations for skeletal muscle, the recovery in glycogen of 14C introduced as lactate was uninfluenced by its original position on the lactate molecule, i.e., no apparent randomization of the label.
The above data strongly suggest that intermediates of the TCA cycle are not involved in skeletal muscle glyconeogenesis. To further confirm this observation, we labeled the TCA pool utilizing 2-14C-acetate. If the TCA cycle were involved in skeletal muscle glyconeogenesis, one would expect to see the label incorporated into glycogen during net glyconeogenesis (39). Glycolytic and mixed muscle preparations were infused with either U-14C-lactate or 2-14C-acetate in the presence of 8 mM lactate and 1 mM acetate. Again, muscles perfused with the different labeled molecules were shown to have similar net rates for lactate and acetate uptake, as well as net glycogen deposition (Fig. 4). Muscles perfused with 14C-lactate demonstrated substantial label incorporation into glycogen. In contrast, muscles perfused with 14C-acetate demonstrated no measurable incorporation of label into glycogen. That the 14C-acetate was incorporated into the TCA cycle pool was demonstrated by the fact that the net extraction of labeled acetate could be quantitatively accounted for as 14CO2. Thus, no measurable exchange occurred between intermediates of the TCA cycle and glyconeogenic intermediates in skeletal muscle.
The above experiment with only acetate and lactate in the perfusate provided the strongest evidence that our tracer measurements of glyconeogenesis reflected net metabolic conversion. In the absence of glucose, lactate was the only significant precursor for glycogen formation. As a result, the net glycogen synthetic rate was 66−75% of what we had previously observed (see Pathways of Disposal above). However, under these conditions, the apparent rates of glyconeogenesis, as determined from total tracer incorporation and the lactate specific activity, were essentially the same as the rate of net glycogen deposition. As a result, our observations are not confounded by dilution of the label as has been the case with studies employing the perfused rat hindlimb (23,37). In those studies (23,37) net glycogen accumulation was observed to exceed the rate of glyconeogenesis as estimated by the tracer incorporation. With lactate as the sole precursor and no interaction with the TCA cycle, the source of this apparent dilution remains unexplained.
The Pathway for Phosphoenolpyruvate Formation in Skeletal Muscle
The above findings indicate that skeletal muscle glyconeogenesis must proceed via an extramitochondrial path. Two pathways have been proposed, reversal of the pyruvate kinase reaction and synthesis of PEP via the malic enzyme. Although the formation of PEP from pyruvate via pyruvate kinase is generally considered to be energetically unfavorable (3,7,18), Dyson et al. (10) calculated that it could sustain glyconeogenic flux from lactate in skeletal muscle. Bendall and Taylor (3) suggested an alternative pathway employing malic enzyme, malate dehydrogenase, and phosphoenolpyruvate carboxykinase (PEPCK). This latter hypothesis, which appears to have been more widely accepted, is not without its own limitations. As pointed out by Shiota et al. (37), the standard free energy for the malic enzyme reaction favors decarboxylation (i.e., formation of pyruvate from malate), and muscle appears to lack a significant source of cystolic NADPH essential for carboxylation. Although hydroxymalonate, an inhibitor of the malic enzyme, has been shown to inhibit lactate uptake in the perfused rabbit hindlimb, it was not observed to impact upon glycogen synthesis (12) Thus, it is not clear that either pathway for the formation of PEP is substantially more favorable.
A limited number of studies have employed mercaptopicolinic acid (MPA), a potent inhibitor of PEPCK, to assess the relative contribution of the malic enzyme pathway for glyconeogenesis. Connett (7) reported a 33% suppression of glyconeogenesis from lactate in the frog sartorius muscle in the presence of MPA. However, measured rates of glyconeogenesis in that study relied entirely upon tracer incorporation, as net glycogen deposition could apparently not be demonstrated. In contrast, Shiota et al. (37) did demonstrate net glyconeogenesis in addition to tracer incorporation but found no evidence of suppression with MPA. As this latter study involved a very limited sample size (N = 2), we sought to reinvestigate this issue with our glycolytic and mixed muscle preparations (31). As had been shown previously, when these preparations are perfused with 8 mM lactate and 14C-lactate they demonstrate tracer incorporation into glycogen, which is quantitatively consistent with the net glycogen synthesis. Under such conditions, the addition of MPA, 150 and 500 μM, failed to inhibit glyconeogenic rates (Fig. 5). Although the impact of MPA on PEPCK was not directly assessed in that study, the concentrations employed were substantially above those required to completely suppress hepatic gluconeogenesis. Further, the concentration of MPA employed was comparable to that used by Connett (7). These finding support the concept that PEPCK is not involved in mammalian skeletal muscle glyconeogenesis. Taken together, the above studies suggest the “reversal” of pyruvate kinase as the likely pathway for skeletal muscle glyconeogenesis.
That PEPCK is not involved in skeletal muscle glyconeogenesis is not entirely surprising. Early studies demonstrated marked differences in the level of PEPCK activity depending upon the type of assay employed. Those demonstrating levels of PEPCK activity capable of sustaining observed rates of glyconeogenesis relied almost exclusively upon spectrophotometric assays (8,23,29). These spectrophotometric assays were subsequently shown to yield spuriously high values for PEPCK (9). Duff and Snell (9) concluded that levels of PEPCK activity were at least an order of magnitude less than previously reported (8,23,29) and unable to support glyconeogenic flux. The lack of any significant PEPCK and pyruvate carboxylase activity in skeletal muscle has been confirmed more recently (5).
The primary objection to the formation of PEP via pyruvate kinase is that this reaction is purported to be displaced far from equilibrium and therefore irreversible (28). That pyruvate kinase catalyzes a nonequilibrium reaction has been based on the disparity between the Keq, 6.5 × 103 (24), and reported values for the mass action ratio (MAR) in skeletal muscle of 35–40 (2). These calculated values for the MAR were based on chemically measured values of ATP and ADP, which yield an ATP:ADP ratio of ∼8:1 (2). However, it is now recognized that the majority of ADP in skeletal muscle is bound to myosin and that the free concentration is considerably lower, ∼8 μM (19). In Type II, muscle at rest the free ATP:ADP ratio is now reported to be on the order of 1000:1 (19). If one substitutes the free ATP and ADP concentrations into the MAR equation for the pyruvate kinase reaction, a value of 4.7 × 103 is calculated, essentially equivalent to the Keq. Thus, the pyruvate kinase reaction in mammalian Type II skeletal muscle is most likely near equilibrium and does not constitute a huge energetic barrier to the formation of PEP. A similar proposal has been put forward for the pyruvate kinase reaction in white muscle from fish (36). That pyruvate kinase may catalyze a near equilibrium reaction in skeletal muscle is consistent with other known aspects of this enzyme, e.g., the very high enzyme activity and apparent lack of regulation (28).
All mammalian skeletal muscle fiber types are net producers of lactate at resting blood concentrations of lactate, ∼1 mM. As the blood lactate concentration rises, all fiber types revert from net production to net consumption. This transition occurs at lower lactate concentrations for the oxidative fiber types, i.e., Type I and IIa, when compared with glycolytic fibers, IIb. For Type I and IIa fibers oxidation appears to be the primary route of disposal accounting for approximately 50% of the lactate removed. Oxidation is also a major pathway of removal for Type IIb fibers but accounts for only 28% of the lactate disposal. For all fiber types, transamination is a significant pathway for the disposal of lactate carbon when muscles are perfused with 8 mM lactate. Glyconeogenesis is the primary pathway for disposal in Type IIb fibers accounting for over 40% of the lactate removed. Quantitatively, the glyconeogenic capacity appears similar for Types IIa and IIb but is negligible for Type I fibers. The pathway for glyconeogenesis in skeletal muscle contrasts markedly with hepatic glyconeogenesis in that it does not involve intermediates of the TCA cycle nor apparently PEPCK. Evidence from our lab and others strongly supports the formation of PEP via “reversal” of the pyruvate kinase reaction. The above findings are restricted to mammalian skeletal muscle and may not be applicable to animals from other classes, e.g., reptiles and amphibians.
The authors would like to thank Dr. Wayne Willis for his valuable critique of this manuscript.
1. Baldwin, K. M., A. M. Hooker, and R. E. Harrick. Lactate oxidative capacity in different types of muscle. Biochem. Biophys. Res. Commun. 83:151–157, 1978.
2. Beis, I., and E. A. Newsholme. The contents of adenine nucleotides, phosphagens and some glycolytic intermediates in resting muscles from vertebrates and invertebrates. Biochem. J. 152:23–32, 1975.
3. Bendall, J. R., and A. A. Taylor. The Meyerhof quotient and the synthesis of glycogen from lactate in frog and rabbit skeletal muscle. Biochem. J. 118:887–893, 1970.
4. Bonen, A., J. C. McDermott, and M. H. Tan. Glycogenesis and glyconeogenesis in skeletal muscle: effects of pH and hormones. Am. J. Physiol. 258:E693–E700, 1990.
5. Brodal, B., and K. Hjelle. Synthesis of phosphoenolpyruvate from pyruvate in rat skeletal muscle. Int. J. Biochem. 22:753–758, 1990.
6. Brooks, G. A. The lactate shuttle during exercise and recovery. Med. Sci. Sports Exerc. 18:360–368, 1986.
7. Connett, R. Glyconeogenesis from lactate in frog striated muscle. Am. J. Physiol. 237:C231–C236, 1979.
8. Crabtree, B., S. J. Higgins, and E. A. Newsholme. The activities of pyruvate carboxylase, phosphoenolpyruvate carboxykinase and fructose diphosphatase in muscles from vertebrate and invertebrates. Biochem. J. 130:391–396, 1972.
9. Duff, D. A., and K. Snell. Limitations of commonly used spectrophotometric assay methods for phosphoenolpyruvate carboxykinase activity in crude extracts of muscle. Biochem. J. 206:147–152, 1982.
10. Dyson, R. D., J. M. Cardenas, and R. J. Barsotti. The reversibility of skeletal muscle pyruvate kinase and an assessment of its capacity to support glyconeogenesis. J. Biol. Chem. 250:3316–3321, 1975.
11. Gorski, J., D. Hood, and R. Terjung. Blood flow distribution in tissues of perfused rat hindlimb preparations. Am. J. Physiol. 250:E441–E448, 1986.
12. Gutierrez, G., E. Fernandez, F. Hurtado, et al. Hydroxymalonate inhibits lactate uptake by the rabbit hindlimb. J. Appl. Physiol. 76:2735–2741, 1994.
13. Hermansen, L., and O. Vaage. Lactate disappearance and glycogen synthesis in human muscle after maximal exercise. Am. J. Physiol. 233:E422–E429, 1977.
14. Hiatt, H., M. Goldstein, J. Lareau, and B. Horecker. The pathway of hexose synthesis from pyruvate in muscle. J. Biol. Chem. 231:303–307, 1958.
15. Jorfeldt, L., A. Juhlin-Dannfelt, and J. Karlsson. Lactate release in relation to tissue lactate in human skeletal muscle during exercise. J. Appl. Physiol. 44:350–352, 1978.
16. Katz, J. Determination of gluconeogenesis in vivo with 14
C-labeled substrates. Am. J. Physiol. 248:R391–R399, 1985.
17. Krebs, H. A., R. Hems, and M. J. Weidemann. The fate of isotopic carbon in kidney cortex synthesizing glucose from lactate. Biochem. J. 101:242–249, 1966.
18. Krebs, H. A., and M. Woodford. Fructose 1,6-diphosphatase in striated muscle. Biochem. J. 94:436–445, 1965.
19. Kushmerick, M. J., T. S. Moerland, and R. W. Wiseman. Mammalian skeletal muscle fibers distinguished by contents of phosphocreatine, ATP, and Pi. Proc. Natl. Acad. Sci. USA 89:7521–7525, 1992.
20. Landau, B. R. Gluconeogenesis and pyruvate metabolism in rat kidney, in vitro. Endocrinology 67:744–751, 1960.
21. Landau, B. R., and J. Wahren. Nonproductive exchanges: the use of isotopes gone astray. Metabolism 41:457–459, 1992.
22. McDermott, J. C., and A. Bonen. Glyconeogenic and oxidative lactate utilization in skeletal muscle. Can. J. Physiol. Pharmacol. 70:142–149, 1992.
23. McLane, J. A., and J. O. Holloszy. Glycogen synthesis from lactate in the three types of skeletal muscle. J. Biol. Chem. 254:6548–6553, 1979.
24. McQuate, J. T., and M. F. Utter. Equilibrium and kinetic studies of pyruvate kinase. J. Biol. Chem. 234:2151–2157, 1959.
25. Meyerhof, O. Über die Energieumwandlunge in Muskel. II. Schicksal der Erholungsperiode des Muskels. Pfluegers Arch. 182:284–317, 1920.
26. Meyerhof, O., K. Lohman, and R. Meier. Über die Synthes des Kohlehydrats im Muskel. Biochemie Zeitschrift 157:459–491, 1925.
27. Moorthy, K., and M. Gould. Synthesis of glycogen from glucose and lactate in isolated rat soleus muscle. Arch. Biochem. Biophys. 130:399–407, 1969.
28. Newsholme, E. A., and C. Start. Regulation in Metabolism. New York: John Wiley & Sons, 1973, pp. 98–104.
29. Newsholme, E. A., and T. Williams. The role of phosphoenolpyruvate carboxykinase in amino acid metabolism in muscle. Biochem. J. 176:623–626, 1978.
30. Omachi, A., and N. Lifson. Metabolism of isotopic lactate by the isolated perfused dog gastrocnemius. Am. J. Physiol. 185:35–40, 1956.
31. Pagliassotti, M. J., and C. M. Donovan. Glycogenesis from lactate in rabbit skeletal muscle fiber types. Am. J. Physiol. 258:R903–R911, 1990.
32. Pagliassotti, M. J., and C. M. Donovan. Influence of cell heterogeneity on skeletal muscle lactate kinetics. Am. J. Physiol. 258:E625–E634, 1990.
33. Pagliassotti, M. J., and C. M. Donovan. Role of cell type in net lactate removal by skeletal muscle. Am. J. Physiol. 258:E635–E642, 1990.
34. Peter, J., R. Barnard, R. Edgerton, C. Gillespie, and K. Stempel. Metabolic profile of three fiber types of skeletal muscle in guinea pig and rabbits. Biochemistry 11:2627–2633, 1972.
35. Pette, D. Metabolic heterogeneity of muscle fibers. J. Exp. Biol. 115:179–189, 1985.
36. Schulte, P. M., C. D. Moyes, and P. W. Hochachka. Integrating metabolic pathways in post-exercise recovery of white muscle. J. Exp. Biol. 166:181–195, 1992.
37. Shiota, M., S. Golden, and J. Katz. Lactate metabolism in the perfused rat hindlimb. Biochem. J. 222:281–292, 1984.
38. Warnock, L. G., R. Koeppe, N. Inciardi, and W. Wilson. L(+) and D(-) lactate as precursors of muscle glycogen. Ann. N.Y. Acad. Sci. 119:1048–1060, 1965.
39. Weinman, E. O., E. H. Strisower, and I. L. Chaikoff. Conversion of fatty acids to carbohydrate: applications of isotopes to this problem and role of the Krebs cycle as a synthetic pathway. Physiol. Rev. 37:252–272, 1957.
40. Wickler, S. J., and T. T. Gleeson. Lactate and glucose metabolism in the mouse (mus musculus) and reptile (anolis carolinensis) skeletal muscle. Am. J. Physiol. 264:R487–R491, 1993.
Chair: L. Bruce Gladden