Lactate overload inhibits myogenic activity in C2C12 myotubes

1 Introduction

Skeletal muscle is a dynamic organ system that plays important roles in maintaining whole body metabolism by the uptake and utilization of glucose and fatty acids. Adequate oxygen supply is essential when skeletal muscle uses glucose as an energy source—particularly during exercise — to generate ATP through mitochondrial oxidative phosphorylation. When oxygen requirement exceeds supply, the intracellular NADH/NAD⁺ ratio increases followed by the conversion of pyruvate to lactate (LA). Through the Cori (or lactic acid) cycle, LA is converted to glucose in the liver and is used to supply glucose to other organs, including skeletal muscle. However, when LA conversion is diminished, LA can accumulate in skeletal muscle. During acute- and high-intensity exercise, the rate of production of LA exceeds the conversion rate, so LA concentrations in skeletal muscle are over 10 mM [1, 2]. However, the concentration of LA in skeletal muscle and blood decreases rapidly after exercise with a half-life of <1 h [3]. Therefore, the concentration of LA in skeletal muscle in healthy individuals is rapidly decreased to its baseline concentration of <1 mM in the resting state within a few hours. However, in certain diseases, such as cancers or hepatic disease, LA can be present at high concentrations for longer times [4, 5] and this can induce lactic acidosis; drastic skeletal muscle atrophy is observed in these patients. The skeletal muscle atrophy seen in patients with chronic major diseases might be induced by coexistent inflammation [6, 7, 8]; however, the effects of LA overload on myogenesis has not been elucidated.

The production of LA is closely related to the type, intensity and duration of exercise [2]. During strenuous exercise in healthy persons, 75% of the LA produced is used as an energy source in skeletal muscle and 25% is provided for gluconeogenesis in the liver for continuous glucose supply via the LA shuttle [9, 10]. Skeletal muscle is a dynamic organ system in which various signaling proteins are involved in myogenesis and muscle atrophy. For example, protein kinase B (Akt) is a kinase phosphorylating the protein forkhead box class O 3a (FOXO3a), which is a transcription factor for the muscle atrophic signal, MAFbx [11]. When the activity of Akt is inhibited, dephosphorylated cytosolic FOXO3a is translocated into the nucleus and induces the expression of MAFbx. 5′-Adenosine monophosphate-activated protein kinase (AMPK) is another enzyme regulating myogenesis as well as glucose and fatty acid metabolism. AMPK can inhibit the maintenance of muscle mass through the activation of autophagy signals, but also acts in the inhibition of protein synthesis via the mammalian target of rapamycin (mTOR) signaling pathways [12, 13]. Activation of AMPK phosphorylates regulatory associated protein of mTOR (raptor), a scaffold protein of mTOR complex 1, which subsequently inhibits P70-S6 kinase (P70-S6K) and protein synthesis [14]. It has been reported that AMPK contributes to FOXO-mediated MAFbx expression followed by muscle atrophy, although contradictory results—that AMPK activation mediates the expression of the myogenic markers MyoD and myogenin [15] — have been reported [16]. Mitochondrial function is another potential contributor of myogenesis [17], and it is well known that peroxisome proliferator-activated receptor gamma coactivator 1α (PGC-1α) is a master regulator of mitochondrial biogenesis [18] through the activation by sirtuin1 (Sirt1)-mediated deacetylation [19]. Skeletal muscles are not static organs but respond to external and internal stimuli, and in which the levels of LA are also regulated dynamically. However, the effects of LA on several myogenic regulators are not fully elucidated. In this study, we test if the high concentration of LA that is observed during high-intensity exercise, or in pathological conditions of LA accumulation, regulates myogenic signaling pathways using differentiated C2C12 murine myotubes.

2 Materials and Methods

3 Results

3.1 Early responses of the AMPK signaling pathway to LA overload

AMPK, a master regulator of cellular energy metabolism, is crucial for myogenesis and protein synthesis [25]. To evaluate the effect of LA on AMPK and its downstream signaling proteins, we evaluated the levels of phosphorylated AMPK at Thr172 (p-AMPK), peroxisome proliferator-activated receptor gamma coactivator-1 alpha (PGC-1α), Sirtuin 1 (Sirt1), phosphorylated raptor at Ser792 (p-raptor) and phosphorylated P70-S6K at Thr421/Ser424 (p-P70-S6K). As shown in Figures 1A and 1B, treatment of C2C12 myotubes with LA at 8 mM for 1 h, no significant difference in the levels of p-AMPK or its downstream signaling proteins was seen when compared with vehicle controls.

Figure 1

Early response of factors associated with myogenesis or atrophy to LA overload. The levels of protein expression in myotubes treated with lactate (LA, 8 mM) for 1 h were similar to the control myotubes (A, C, E). The quantified band intensities showed no significant difference (B, D, F). The mean band intensities quantified from two independent experiments (n=3) were compared, and representative bands are presented.

3.3 Delayed activation of AMPK signalling pathways by LA overload

As shown in Figures 2A and 2B treatment of C2C12 myotubes with LA (8 mM) for 24 h significantly increased the level of p-AMPK but significantly suppressed the expressions of Sirt1 and PGC-1α. The level of p-raptor, a downstream target of AMPK in the mTOR complex, was increased significantly, while the level of p-P70-S6K was suppressed significantly by LA treatment for 24 h.

Figure 2

Delayed response of myogenic or atrophic factors to LA overload. The levels of AMPK-mediated inhibition of mTOR (decreased p-raptor and p-P70-S6K) and of Sirt1 and PGC-1α in myotubes treated with lactate (LA, 8 mM) for 24 h were significantly different when compared to control myotubes (A, B). Myogenic factors (myogenin, MyoD and MHC) in LA-treated myotubes were significantly lower, while atrophic factor (MAFbx) was significantly higher than control myotubes (C, D). Akt-mediated inhibition (phosphorylation) of FOXO3a was significantly blocked by LA treatment for 24 h (E, F). The mean band intensities quantified from two independent experiments (n=3) were compared, and representative bands were presented. *p<0.05, **p<0.01 by unpaired t-test.

3.4 Inhibition of Akt-mediated myogenic signaling pathways by LA overload for 24 h

Treatment of C2C12 myotubes with LA for 24 h significantly inhibited the levels of p-Akt and downstream targets of p-FOXO3a, while the expression of p-mTOR was similar to the controls. Consistent with the suppression of Akt-mediated myogenic pathways, we observed that LA overload significantly suppressed the levels of MyoD, myogenin and MHC, but increased the level of MAFbx, when compared with vehicle-treated controls. Consistent with the results of anti-myogenic effects of LA, we also observed that LA treatment for 24 h decreased the mean myotube diameter (Fig. 3B, 3C). Furthermore, we observed the treatment of LA for 24 h, but not for 1hr, increased the nuclear translocation of FOXO3a, a transcription factor for MAFbx (Fig 3D), indicating that LA-induced nuclear translocation of FOXO3a may be a mechanism of the decreased diameter of myotube by LA.

Figure 3

Reduction of myotube diameters by LA. (A – C) When we treated myotubes with lactate (LA, 8 mM) for 24 h, mean diameter of LA treated myotubes was significantly lower than control myotubes. ***p<0.001 vs. control (Cont) by unpaired t-test. (D) When we prepared the subcellular fraction (for details, see Method 2.4.), LA treatment for 24 h increased the nuclear translocation of FOXO3a, while LA overload for 1 h did not change the level of nuclear FOXO3a.

4 Discussion

In pathological conditions, including but not limited to cellular hypoxia [31] and cancer [5] that induce increased LA production, or hepatic failure [32], diabetes [33] and biguanide intoxication [34] that inhibit hepatic LA clearance, the levels of LA can be elevated in blood and skeletal muscle. LA can be used as an energy source instead of pyruvate; however, LA can also act as a signaling molecule [10, 35, 36]. Here we tested the effects of LA overload on myogenic signaling using differentiated C2C12 myotubes. We found that LA overload for 24 h—but not for 1 h—significantly suppressed the levels of Akt-mediated myogenic signaling proteins p-P70-S6K, MyoD, myogenin, Sirt1 and PGC-1α, while it increased the levels of the anti-myogenic molecules p-AMPK, FOXO3a and MAFbx.

In skeletal muscle, Akt activates the mTOR/P70-S6K protein synthesis pathway that induces muscle hypertrophy [26, 27]. In addition, Akt inhibits the transcriptional activity of FOXO3a by phosphorylation [26] followed by inhibition of MAFbx expression. AMPK can also phosphorylate FOXO3a, but unlike Akt, AMPK-mediated phosphorylation increases the activity of this target molecule [37, 38]. Here, we observed that LA overload for 24 h significantly increased the activity of AMPK but suppressed Akt activity, indicating that LA accumulation for longer times can initiate the canonical anti-myogenic signaling pathways. Consistent with the effects of LA on AMPK and Akt, we observed that LA induced the expression of MAFbx, a target molecule of FOXO3a, but decreased the myogenic transcription factors, MyoD and myogenin. Phenotypically, myotubes treated with LA for 24 h, but not in cells treated for 1 h, showed the significantly lower mean myotube diameter than control myotubes. The decreased myotube diameter by treatment of LA for 24 h was consistent with the decreased expression of MHC (Fig. 2C and D) and increased nuclear translocation of FOXO3a (Fig. 3D), which further supports the muscle atrophic effects of chronic LA overload. Protein synthesis is essential for increasing skeletal muscle mass [39]. AMPK activation inhibits mTORC1-mediated protein synthesis in which P70-S6K activity is crucial [13, 40, 41]. We observed that LA treatment inhibited the phosphorylation of P70-S6K, consistent with increased levels of p-AMPK and p-raptor, therefore, LA overload might inhibit the protein synthesis pathway required for hypertrophy, although we did not study the protein synthesis rate specifically. We could not detect the change of mTOR phosphorylation by LA treatment, however, LA-induced AMPK activation may inhibit mTORC1-mediated protein synthetic activity through the phosphorylation of raptor at Ser792 rather than direct mTOR phosphorylation at Ser2448. Sirt1, a deacetylase, inhibits the myogenic molecules MyoD and MEF2 by formation of inhibitory complexes [42], even though this has not been properly tested in vivo. However, in contrast, a recent study provided evidence that Sirt1 and PGC-1α can induce the levels of myogenic factors in conjunction with MyoD [17, 43]. Although further studies in the detailed mechanisms and roles of Sirt1 and PGC-1α in LA overload are required, our data suggest that LA overload for long durations can inhibit the myogenic roles of Sirt1 and PGC-1α, in addition to Akt-mediated myogenic signaling. It is unlikely that the LA-mediated activation of AMPK is linked to increased activity of Sirt1 and PGC-1α, which are essential in mitochondrial biogenesis [25]. Skeletal muscle contraction by exercise activates AMPK by increasing AMP/ATP ratios [41]. However, in our experimental conditions we treated myotubes with a high concentration of LA without glucose deprivation, so this did not mimic exercise but rather the accumulation of LA. Otherwise, during exercise, LA build-up might activate AMPK independent of the AMP/ATP ratio. In addition, because LA overload suppressed the expression of both Sirt1 and PGC-1α in this study, the effects of LA on mitochondrial biogenesis should be evaluated further.

In skeletal muscle, Akt activates the mTOR/P70-S6K protein synthesis pathway that induces muscle hypertrophy [26, 27]. In addition, Akt inhibits the transcriptional activity of FOXO3a by phosphorylation [26] followed by inhibition of MAFbx expression. AMPK can also phosphorylate FOXO3a, but unlike Akt, AMPK-mediated phosphorylation increases the activity of this target molecule [37, 38]. Here, we observed that LA overload for 24 h significantly increased the activity of AMPK but suppressed Akt activity, indicating that LA accumulation for longer times can initiate the canonical anti-myogenic signaling pathways. Consistent with the effects of LA on AMPK and Akt, we observed that LA induced the expression of MAFbx, a target molecule of FOXO3a, but decreased the myogenic transcription factors, MyoD and myogenin. Phenotypically, myotubes treated with LA for 24 h, but not in cells treated for 1 h, showed the significantly lower mean myotube diameter than control myotubes. The decreased myotube diameter by treatment of LA for 24 h was consistent with the decreased expression of MHC (2C, and D), and increased nuclear translocation of FOXO3a (Fig. 3D), which further supports the muscle atrophic effects of chronic LA overload. Protein synthesis is essential for increasing skeletal muscle mass [39]. AMPK activation inhibits mTORC1-mediated protein synthesis in which P70-S6K activity is crucial [13, 40, 41]. We observed that LA treatment inhibited the phosphorylation of P70-S6K, consistent with increased levels of p-AMPK and p-raptor, therefore, LA overload might inhibit the protein synthesis pathway required for hypertrophy, although we did not study the protein synthesis rate specifically. We could not detect the change of mTOR phosphorylation by LA treatment, however, LA-induced AMPK activation may inhibit mTORC1-mediated protein synthetic activity through the phosphorylation of raptor at Ser792 rather than direct mTOR phosphorylation at Ser2448. Sirt1, a deacetylase, inhibits the myogenic molecules MyoD and MEF2 by formation of inhibitory complexes [42], even though this has not been properly tested in vivo. However, in contrast, a recent study provided evidence that Sirt1 and PGC-1α can induce the levels of myogenic factors in conjunction with MyoD [17, 43]. Although further studies in the detailed mechanisms and roles of Sirt1 and PGC-1α in LA overload are required, our data suggest that LA overload for long durations can inhibit the myogenic roles of Sirt1 and PGC-1α, in addition to Akt-mediated myogenic signaling. It is unlikely that the LA-mediated activation of AMPK is linked to increased activity of Sirt1 and PGC-1α, which are essential in mitochondrial biogenesis [25]. Skeletal muscle contraction by exercise activates AMPK by increasing AMP/ATP ratios [41]. However, in our experimental conditions we treated myotubes with a high concentration of LA without glucose deprivation, so this did not mimic exercise but rather the accumulation of LA. Otherwise, during exercise, LA build-up might activate AMPK independent of the AMP/ATP ratio. In addition, because LA overload suppressed the expression of both Sirt1 and PGC-1α in this study, the effects of LA on mitochondrial biogenesis should be evaluated further.

Figure 4

Summarized effects of chronic LA overload on the signal proteins regulating myogenesis and muscle atrophy. Red-colored lines and black lines indicate the inhibitory and activating effects, respectively.

There are several limitations of our study; first, the use of a murine cell-line rather than use of human primary cells to evaluate the effects of LA is a limitation. C2C12 differentiated myotube model is a widely used experimental tool for the experiment of muscle biology. Nevertheless, our results should be confirmed in human primary cell culture model. Second, we did not perform experiments for the effects of LA at intermediate time (e.g., 4 h, 12 h). However, the aim of our study has focused on the acute effects of LA during short duration that mimic the strenuous exercise and chronic effects of LA that mimic a pathologic LA accumulating condition. Finally, we could not confirm the effects of LA-induced activation of AMPK on myogenic signaling, but we speculate LA-mediated inhibition of myogenesis through AMPK-mTORC1-P70-S6K pathways. Fu et al. reported contradictory result that knockdown of AMPK-α1 inhibited myogenic signaling [15, 17], the effects of LA-mediated AMPK activation on the myogenenic signaling should be further investigated, although we recently reported that inflammation activated AMPK and inhibition of the expression of myogenic signaling [44].

In conclusion, LA overload for 24 h inhibited myogenic pathways through the activation of AMPK but also via suppression of Akt, which is consistent with decreased levels of myogenic factors (MyoD and myogenin) and increased levels of the atrophic factor MAFbx. The LA-mediated anti-myogenesis, at least in part, is likely to induce transcriptional activity of FOXO3a. Our study was performed under in vitro experimental conditions using C2C12 myotubes, so further in vivo evidence is needed to support our results.