A single bout of intense, dynamic exercise in untrained humans induces rapid and marked perturbations in cellular homeostasis within the contracting skeletal muscles, with the magnitude and extent of these disruptions proportional to the prevailing contractile stimulus. Accordingly, during exercise, a multitude of acute responses are initiated to try to match energy supply processes to the increased metabolic demand for adenosine triphosphate (ATP), with the overall goal of minimizing cellular disturbances. On cessation of exercise, restoration of depleted muscle energy stores is given high metabolic priority. Although transcription of several genes that encode for various regulatory and metabolic proteins is elevated immediately after exercise, activation of most of these genes takes place during the initial (1–4 h) of recovery, returning to basal levels within 24 h (5). This transient and short-term increase in the transcription of target genes does not translate into a proportional increase in gene product, raising the possibility that cellular and molecular mechanisms underlying acute transcriptional regulation of gene expression may be linked to the time course of key metabolic processes (i.e., the resynthesis of glycogen) taking place during recovery after exercise (4).
In contrast to the effects of an acute bout of exercise, contractile activity repeated over several weeks or months induces a multitude of metabolic, enzymatic, and morphologic adaptations that function to minimize cellular disturbances during subsequent work bouts, thereby enhancing exercise capacity (2). The biochemical and physiological changes induced by functionally diverse exercise-training regimens have been well characterized, but the unique capacity of skeletal muscle to respond to divergent contractile stimulus and execute the appropriate metabolic and mitogenic responses for specific adaptation and remodeling (i.e., intracellular signaling specificity) is less well defined.
Finally, habitual exercise has differential effects on the activation of several major intracellular signaling cascades (3,14,15). Endurance training downregulates early components of the insulin-signaling cascade and is associated with divergent effects on several mitogen-activated protein kinase (MAPK) pathways (14). Because an acute bout of exercise induces transient increases in skeletal muscle gene transcription (5) even in athletes with a prolonged history of training (15), activation of specific signaling pathways during and after exercise appears to be central to the upregulation of a number of metabolic and mitogenic responses that ultimately induce the specific adaptations in skeletal muscle observed after repeated bouts of exercise. This review focuses on how metabolic responses to exercise impact on subsequent signal transduction in human skeletal muscle. In particular, the role of several parallel contraction-induced signaling cascades that integrate intracellular signals to regulate gene transcription and protein synthesis in muscle are discussed.
INTEGRATION OF THE METABOLIC RESPONSES TO EXERCISE
The metabolic signals believed to play major roles in the activation and coordination of the various energy-producing pathways during muscle contraction can be classified into three main categories: (i) those initiated by changes in Ca2+, (ii) those that respond to changes in the concentrations of metabolites related to the cytoplasmic phosphorylation potential of the muscle cells, and (iii) those that are influenced by the mitochondrial reduction and oxidation (redox) state of nicotinamide adenine dinucleotide (NAD/NADH). Ca2+ release plays a fundamental role in initiating muscle contraction and activating metabolism in a feed-forward or “early warning” manner (Fig. 1). Calcium provides the trigger for force development and ATP hydrolysis and contributes to the activation of glycogenolysis and possibly oxidative phosphorylation. A rise in intracellular Ca2+ contributes to the increased glucose uptake during muscle contraction by activating a signal transduction pathway leading to glucose transporter isoform 4 (GLUT4) translocation (Fig. 2). GLUT4 is of particular importance for maintaining whole body glucose homeostasis because it has been thought to catalyze the rate-limiting step for glucose uptake and metabolism. Several other contraction-induced signaling intermediates (i.e., protein kinase C and the MAPK cascade) are also sensitive to changes in Ca2+ and are likely to mediate exercise-induced responses.
Although the feed-forward Ca2+-activated signals are important for events occurring early on in the excitation-coupling processes, the concentrations of metabolites related to the cytoplasmic phosphorylation potential of muscle cells (i.e., [ATP]/[adenosine diphosphate][free inorganic phosphate]) provide feedback signals necessary to balance ATP production with ATP consumption (Fig. 1). The concentrations of free adenosine monophosphate (AMP) and H+ also have important regulatory functions in metabolism. Changes in the concentrations of metabolites in the phosphagen pool provide the main stimulus for increased oxidative phosphorylation and are necessary (but not sufficient) for full activation of glycolytic ATP synthesis during intense (≥85% of maximal aerobic power [o2max]) sustained exercise.
The redox state of the muscle activates numerous reactions in substrate and product or activator and inhibitor capacities. All three types of signal are important in coordinating the upregulation of enzymes that are vital for ATP production in mitochondrial and cytoplasmic compartments during exercise. Along with systemic factors (increased blood flow leading to increased delivery of hormonal factors) and the local release of growth factors and cytokines from contracting muscle, these signals provide the mechanistic framework by which exercise regulates intracellular signal transduction to the transcriptional machinery in the nucleus, modulating gene expression and ultimately protein turnover (Fig. 2).
INTEGRATION OF THE MITOGENIC RESPONSES TO EXERCISE
5′-AMP–Activated Protein Kinase Signaling Cascades
One major sensing signaling pathway in skeletal muscle that responds in an ultrasensitive fashion to changes in the concentrations of metabolites related to the cytoplasmic phosphorylation potential is the AMPK cascade. AMPK is a heterotrimeric protein, composed of one catalytic (α) and two noncatalytic (β and γ) subunits and is activated by cellular stress associated with ATP depletion. All three subunits are required for full activation. Each subunit has several isoforms, with the α and β isoforms being the most frequently characterized in human skeletal muscle (1,3,11,12,15).
The AMPK cascade is turned on by cellular stresses that deplete ATP (and consequently elevate AMP) either by inhibiting ATP production (e.g., hypoxia) or by accelerating ATP consumption (e.g., muscle contraction). Once activated, the AMPK cascade switches on catabolic processes both acutely (by phosphorylation of downstream metabolic enzymes such as acetyl coenzyme A carboxylase) and chronically (by effects of gene expression), while concomitantly switching off ATP-consuming processes. Activation of AMPK probably is responsible for many of the acute effects on skeletal muscle substrate metabolism during exercise and subsequent recovery, in addition to some of the chronic training-induced adaptations such as activation of metabolically related genes. Most studies designed to position the AMPK pathway as central for the regulation of exercise-induced metabolic effects have been correlative in nature. The use of genetically modified animal models to address the role of AMPK in this regard has offered only limited insight.
With regard to the acute effects of exercise, low- to moderate-intensity cycling (≤70% of o2max) in both untrained and moderately trained subjects induces an isoform-specific and intensity-dependent increase in AMPKα2- but not AMPKα1-associated activity (11). Conversely, activity of heterotrimeric complexes containing either AMPKα1 or AMPKα2 is increased in response to sprint exercise (1). Activation of the AMPKα1 isoform after anaerobic exercise is likely related to the rate of fuel use rather than the magnitude of substrate depletion, because no change in AMPKα1 activity is observed after prolonged, continuous, low-intensity (44% of o2max) cycling leading to exhaustion, despite a significant (sixfold) but progressive decrease in the rate of glycogenolysis (12). A surprising finding from that investigation (12) was a gradual rise in AMPKα2 activity over time, so that at exhaustion, there was a significant (2.5-fold) elevation compared with resting basal levels. Interestingly, covalently induced activation of AMPKα2 does not occur in humans undertaking low-intensity exercise of shorter (≤1 h) duration (11).
If AMPK does play a role in the metabolic adaptations to training, then one may expect a blunted exercise-induced response of AMPK to standardized exercise in trained individuals. Recent studies reported attenuation in the rise of AMPKα2 activity in trained versus untrained subjects in response to intense cycling undertaken at the same relative power output (∼80–90% of o2max) (3,15). Despite higher absolute energy requirements during exercise in trained compared with untrained subjects, the phosphorylation potential of the muscle (as reflected by the difference in [PCr]/[PCr + Cr] ratio) was better maintained in trained individuals (3). Besides providing a likely mechanism for the lower AMPKα2 activation in trained subjects (Fig. 3), these investigations (3,15) provide strong evidence that AMPK activity is not a sensitive marker of the total energy flux during exercise, but rather the consequence of the metabolic perturbations induced by such activity.
Glycogen content in skeletal muscle may modulate the AMPK response to contraction. Certainly during low-intensity prolonged exercise to exhaustion, AMPKα2 activity peaks at a time when glycogen content and glucose uptake across the working limb reached their nadir (12). However, this relationship is uncoupled at rest: AMPKα2 activity returns to basal levels after 1 h of recovery from fatiguing exercise even in the face of low glucose delivery and persistently low muscle glycogen content (12). Accordingly, during exercise, AMPKα2 activity appears to be more responsive to small decreases in the PCr-to-Cr ratio than changes in muscle glycogen content, blood glucose uptake, or both. Well-trained subjects have been studied under conditions of low and high glycogen concentration (∼160 vs. ∼900 mmol·kg−1 d.w.), at rest and subsequently during 1 h cycling at 70% o2max (10). At rest, AMPKα1 and AMPKα2 activities were elevated 2.5- to 2.6-fold in the low- versus high-glycogen states. Low preexercise glycogen content also increased AMPKα2 activity and acetyl coenzyme A carboxylase β phosphorylation during subsequent submaximal cycling. However, because the low-glycogen state also was accompanied by significantly increased leg glucose and net fatty acid uptake, as well as increased plasma concentrations of catecholamines compared with the high glycogen condition, AMPK activity during exercise seems to be regulated by both fuel availability (i.e., the prevailing muscle glycogen content) and possibly by humoral factors.
Many of the chronic training-induced adaptations in skeletal muscle have been proposed to involve AMPK. Chronic pharmacologic activation of AMPK (an “exercise-like” effect) enhances the protein expression of GLUT4, hexokinase, and several oxidative enzymes, as well as increasing mitochondrial density and muscle glycogen content. One mechanism by which AMPK potentially can regulate gene expression is through increased DNA binding. In vitro myocyte enhancer factor 2 (MEF2) sequence-specific binding activity is increased in skeletal muscle from lean and ob/ob diabetic mice after 7 d of pharmacologic activation of AMPK using 5-aminoimidazole-4-carboxamide ribonucleoside (7). The MEF2 DNA binding site seems to be essential for GLUT4 expression, because deletions or point mutations within the MEF2 consensus binding sequence of the human GLUT4 promoter completely prevents tissue-specific and hormonal or metabolic regulation of GLUT4 and induction of the muscle differentiation phenotype. Interestingly, MEF2–DNA complex binding was increased in nuclear extracts prepared from skeletal muscle obtained from healthy men after running a marathon (13). Thus, increased MEF2 sequence-specific binding activity may confer exercise-specific changes in gene expression, presumably via an AMPK-mediated pathway. AMPK also may activate downstream effectors such as members of the MAPK family, providing cross-talk between these pathways (Fig. 2).
The Mitogen-Activated Protein Kinase Signaling Cascade
The MAPK signal transduction cascade has been identified as a candidate system that may convert contraction-induced biochemical perturbations in skeletal muscle into appropriate intracellular responses (Fig. 2). Exercise is a powerful stimulator of several parallel MAPK pathways, including the extracellular signal-regulated kinases (ERK) 1/2 and the two stress-activated protein kinases, p38 MAPK and the c-Jun NH2-terminal kinase, in untrained and well-trained subjects (8,13,15). A single bout of exercise in untrained subjects leads to a dramatic and early increase in ERK 1/2 phosphorylation, with the greatest effect observed after ∼30 min of submaximal cycling, and not further enhanced by prolonging the duration of exercise (8). The increase in ERK 1/2 phosphorylation is a local rather than systemic effect, with activity suppressed to baseline levels within 1 h of cessation of contraction, suggesting that exercise does not have a persistent effect on ERK 1/2 phosphorylation (8). Exercise-induced phosphorylation of ERK 1/2 is intensity dependent (9). Changes in the intracellular milieu, such as increasing Ca2+ or muscle temperature, development of local hypoxia, mechanical stress (particularly muscle damage caused by eccentric contractions), and energy depletion, are all factors known to activate MAPK.
In contrast, exercise leads to a smaller increase in p38 MAPK phosphorylation, an effect that is maintained throughout exercise and is observed in both exercised and nonexercised limbs (8). Unlike ERK 1/2 phosphorylation, p38 MAPK phosphorylation is mediated by both local and systemic factors associated with exercise. The precise role for a systemic response rather than a predominantly local induction of p38 MAPK is not known. Systemic factors have been hypothesized to be important for amplifying signaling responses (9).
Activity of several downstream substrates of ERK 1/2 and p38 MAPK signaling cascades, such as MAPK-activated protein kinase 1 and 2, as well as the mitogen and stress-activated kinase (MSK) 1 and 2, are increased immediately after acute sprint and endurance exercise. Using specific inhibitors of ERK 1/2 and p38 (PD98059 and SB203580, respectively), downstream components of MAPK signaling cascade in response to exercise and contraction have been validated in rat epitrochlearis muscle (6). Contraction-induced induction of MAPK-activated protein kinase 1 and MAPK-activated protein kinase 2 occurs via separate pathways, reflecting ERK 1/2 and p38 MAPK stimulation, respectively. In contrast, induction of MSK1 and MSK2 require simultaneous activation of ERK 1/2 and p38 MAPK. One area of future research will be directly to link ERK 1/2 and p38 phosphorylation and their downstream targets to changes in gene expression after exercise.
MAPK substrates can mediate alterations in the chromatin environment of specific genes by direct phosphorylation, acetylation of nucleosomal and chromatin proteins, or both. Histone H3 phosphorylation is associated with MAPK signaling to p90 rsk and MSK1, thereby directly linking the MAPK pathway to induction of immediate early response gene expression. Phosphorylation of histone H3 is increased in skeletal muscle in both untrained and trained individuals in response to intense cycling undertaken at the same relative (85% of o2max) intensity (15). However, activation of signaling intermediates was generally greater in muscle from untrained subjects, suggesting that a greater stimulus is required to activate signal transduction through MAPK pathways in individuals with a history of training (15). Activation of MSK1 in response to physical exercise provides a putative signaling mechanism to immediate early response genes and alterations in the nucleosome and chromatin structure through phosphorylation of histone H3, although this remains to be firmly established.
SUMMARY AND DIRECTIONS FOR FUTURE RESEARCH
The biggest challenge for exercise physiologists in the forthcoming years will be directly to link AMPK and MAPK signaling cascades to defined metabolic responses and specific changes in gene expression in skeletal muscle that occur after exercise. This will be complicated by the fact that these pathways are not linear, but rather they constitute a complex network, with a high degree of cross-talk, feedback regulation, and transient activation. An increased understanding of these pathways through studies of comparative genomics between humans, rodents, and model organisms, coupled with greater knowledge of the functional relevance of the signaling intermediates, will facilitate the understanding of mechanisms by which exercise and stress alters gene expression. Ultimately, many of the components of these exercise-responsive networks will be validated through the use of in vitro models and inhibitors to repress directly, and systematically, components of the defined cascades. Through genetically modified rodent models and gene silencing through in vivo application of siRNA, studies of the complex interaction between central and systemic factors that influence exercise responses on these networks can be revealed. Together with translational studies in humans, these combined approaches will facilitate efforts to assign the physiologic role for these targets in the regulation of exercise-mediated responses on metabolism and gene expression.
1. Chen, Z. P., McConell, G. K. Mitchell, B. J. Snow, R. J. Canny, B. J. and Kemp. B. E. AMPK signalling in contracting human skeletal muscle: Acetyl CoA carboxylase and NO synthase phosphorylation. Am. J. Physiol. Endocrinol. Metab. 279: E1202–E1206, 2000.
2. Hawley, J. A. Adaptations of skeletal muscle to prolonged, intense endurance training. Clin. Exp. Pharmacol. Physiol. 29: 218–222, 2002.
3. Nielsen, J. N., Mustard, K. J. W. Graham, D. A. Yu, H. MacDonald, C. S. Pilegaard, H. Goodyear, L. J. Hardie, D. G. Richter, E. A. and Wojtaszewski. J. F. P. 5′-AMP-activated protein kinase activity and subunit expression in exercise-trained human skeletal muscle. J. Appl. Physiol. 94: 631–641, 2003.
4. Pilegaard, H., Keller, C. Steensberg, A. Helge, J. W. Karlund Pedersen, B. Saltin, B. and Neufer. P. D. Influence of pre-exercise muscle glycogen content on exercise-induced transcriptional regulation of metabolic genes. J. Physiol. 541: 261–271, 2002.
5. Pilegaard, H., Ordway, G. A. Saltin, B. and Neufer. P. D. Transcriptional regulation of gene expression in human skeletal muscle during recovery from exercise. Am. J. Physiol. Endocrinol. Metab. 279: E806–E814, 2000.
6. Ryder, J. W., Fahlman, R. Wallberg-Henriksson, H. Alessi, D. R. Krook, A. and Zierath. J. R. Effect of contraction on mitogen-activated protein kinase signal transduction in skeletal muscle: Involvement of the mitogen- and stress-activated protein kinase. J. Biol. Chem. 275: 1457–1462, 2000.
7. Song, X. M., Fiedler, M. Galuska, D. Ryder, J. W. Fernstrom, M. Chibalin, A. V. Wallberg-Henriksson, H. and Zierath. J. R. 5-Aminoimidazole-4-carboxamide ribonucleoside treatment improves glucose homeostasis in insulin-resistant diabetic (ob/ob) mice. Diabetologia 45: 56–65, 2002.
8. Widegren, U., Jiang, X. J. Krook, A. Chibalin, A. V. Bjornholm, M. Tally, M. Roth, R. A. Henriksson, J. Wallberg-Henriksson, H. and Zierath. J. R. Divergent effects of exercise on metabolic and mitogenic signalling pathways in human skeletal muscle. FASEB J. 12: 1379–1389, 1998.
9. Widegren, U., Wretman, C. Lionikas, A. Hedin, G. and Henriksson. J. Influence of exercise intensity on ERK/MAP kinase signalling in human skeletal muscle. Pflügers Arch. 441: 317–322, 2000.
10. Wojtaszewski, J. F., MacDonald, C. Nielsen, J. N. Hellstend, Y. Hardie, G. Kemp, B. E. Kiens, B. and Richter. E. A. Regulation of 5′AMP-activated protein kinase activity and substrate utilization in exercising human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 284: E813–822, 2003.
11. Wojtaszewski, J. F. P., Nielsen, P. Hansen, B. F. Richter, E. A. and Kiens. B. Isoform-specific and exercise intensity-dependent activation of 5′-AMP-activated protein kinase in human skeletal muscle. J. Physiol. 528: 221–226, 2000.
12. Wojtaszewski, J. F. P., Mourtzakis, M. Hillig, T. Saltin, B. and Pilegaard. H. Dissociation of AMPK activity and ACCβ phosphorylation in human muscle during prolonged exercise. Biochem. Biophys. Res. Comm. 298: 309–316, 2002.
13. Yu, M., Blomstrand, E. Chibalin, A. V. Krook, A. and Zierath. J. R. Marathon running increases ERK1/2 and p38 MAP kinase signalling to downstream targets in human skeletal muscle. J. Physiol. 536: 273–282, 2001.
14. Yu, M., Blomstrand, E. Chibalin, A. V. Wallberg-Henriksson, H. Zierath, J. R. and Krook. A. Exercise-associated differences in an array of proteins involved in signal transduction and glucose transport. J. Appl. Physiol. 90: 29–34, 2001.
15. Yu, M., Stepto, N. K. Fryer, L. G. D. Carling, D. Krook, A. Hawley, J. A. and Zierath. J. R. Metabolic and mitogenic signal transduction in skeletal muscle after intense cycling exercise. J. Physiol. 546: 327–335, 2003.