Many studies have demonstrated an increase in markers of the reaction of reactive oxygen and nitrogen species (ROS) with lipid, proteins, or DNA in blood and tissues of humans and animals during and after exercise (e.g., ref (3)). This increase in ROS activity appears to be in major part because of their generation by contracting skeletal muscle. Recently, specific reactive oxygen species have been demonstrated to increase in the extracellular fluid during contractile activity of skeletal muscle. These include superoxide (10,12), hydroxyl radical (11), and nitric oxide (1).
There are a number of potential sites for the generation of ROS within the exercising muscle and mechanisms by which free radicals may be increased in muscle secondary to damage induced by other processes. Primary sources of free radicals may include mitochondria, xanthine oxidase enzymes, prostanoid metabolism, and NAD(P)H oxidases (3). Oxidative metabolism in mammals involves the reduction of molecular oxygen in mitochondria. As part of the process of ATP production, molecular oxygen generally undergoes a four-electron reduction catalyzed by cytochrome oxidase. This process accounts for 95% to 98% of the total oxygen consumption, but the remainder may undergo one-electron reduction with the production of superoxide (O2·). Further reduction of superoxide produces hydrogen peroxide (H2O2). An increase in aerobic metabolism during exercise therefore may lead to an increased mitochondrial production of superoxide and hydrogen peroxide as a result of the increased electron flux through the electron transport chain.
Recently McArdle et al. (10) demonstrated that the extracellular superoxide released from electrically stimulated skeletal muscle myotubes does not seem to be derived from mitochondria and may be derived from contraction-activated, membrane-bound oxidoreductase(s) (Fig. 1). One potential site may be a membrane-bound NAD(P)H oxidase that recently has been reported to be expressed in skeletal muscle (5). Other authors have shown that hydroxyl radical activity in muscle extracellular fluid is increased during contractile activity and that this increase in activity occurs by an iron-dependent process (11). Nitric oxide (NO) also has become recognized as being generated continuously by skeletal muscle, and NO production is increased in contracting muscle (1). Skeletal muscle expresses both the neuronal isoform of NO synthase (nNOS or type I NOS) and the endothelial isoform of NO (eNOS or type III NOS). nNOS is localized to the sarcolemma, strongly expressed in fast-twitch muscle fibers, and can be modulated by skeletal muscle activity (7), whereas eNOS localizes to the muscle mitochondria (8). Generation of both superoxide and nitric oxide can result in the formation of peroxynitrite (−ONOO), which can be a more oxidizing species than NO or superoxide alone. A schematic diagram of the ROS generated by skeletal muscle and their potential sources is shown in Figure 2.
The precise pattern of ROS generation by contracting skeletal muscle is still under intensive investigation, but the multiple potential sources of the ROS and their subcellular locations suggest that the pattern and magnitude of their production is likely to be influenced by the nature of the contractile activity. We hypothesize that activities increasing mitochondrial activity, such as sustained aerobic exercise, are likely to induce release of relatively large amounts of hydrogen peroxide (and hence extracellular hydroxyl radical activity), but that mechanical distentions (contraction or stretch) are the prime activators of the plasma membrane-located systems for superoxide and NO generation. Thus, these latter systems will be activated in situations in which muscle is both fully oxygenated and relatively hypoxic.
It is also likely that there are other sources of ROS within muscle during damaging exercise, including radical generation by phagocytic white cells. After substantial injury to muscle fibers, invasion of the area by macrophages and other phagocytic cells from the blood and interstitium follows. These cells are essential in providing fast, effective regeneration of the tissue. However, as part of the phagocytic process, substantial amounts of free radicals are released to aid in the breakdown of necrotic areas, but also contribute to damage to surrounding viable tissue (3).
In addition to these multiple sites of generation of ROS, skeletal muscle has a well-developed protective system to prevent potentially deleterious effects of these substances. These include both mitochondrial and cytosolic isoforms of superoxide dismutase, catalase and glutathione peroxidase enzymes, and a number of direct scavengers of free radical species, including glutathione, vitamin E, and ascorbate. In general, slow twitch, mitochondria rich (type I) fibers have an increased content of these protective systems in comparison with fast (type II) fibers.
OXIDANTS AS SIGNALING MOLECULES
Reactive oxygen and nitrogen species may act to stimulate changes in cell function, differentiation, and proliferation and may play a role in cell damage and death. It has become increasingly evident that the redox balance within cells is important in the regulation of gene expression. Reactive oxygen species from both extracellular and intracellular sources may initiate changes in cell signaling. Well over 100 mammalian genes have been identified that can be regulated by cellular redox state. This area has recently been comprehensively reviewed by Jackson et al. (5).
Reactive oxygen species can act through several different pathways of signal transduction, making use of signaling cascades such as protein tyrosine kinases (especially mitogen-activated protein (MAP) kinase-related pathways), and protein tyrosine phosphatases, such as serine and threonine kinases, phospholipases, and calcium. Transcription factors such as nuclear factor-kappa B (NF-κB), activated protein-1, and p53 are affected by redox conditions (Table 1). NF-κB is seen as the classical model of a redox-sensitive transcription factor. A substantial body of evidence links NF-κB activity to cellular oxidative status, although the mechanism by which NF-κB is activated by ROS is unknown. It is thought, however, that oxidizing conditions in the cytoplasm favor translocation of NF-κB to the nucleus, but that reducing conditions are required within the nucleus for NF-κB DNA binding (5).
Most cells respond to changes in environmental exposure to oxidative stress and to endogenous free radical production by increasing or decreasing proliferation, changing immune response, or by induction of apoptotic cell death, all processes regulated by closely integrated signaling pathways (5). In contrast, extreme levels of oxidative stress may lead to increased cell death as a result of necrosis.
EVIDENCE FOR A ROLE OF OXIDANTS IN SKELETAL MUSCLE SIGNALING AND ADAPTIVE RESPONSES
The dramatic changes in oxidant release during contractile activity and their known functions in other cell types suggest that these substances may play a role in stimulating changes in gene expression resulting from contractile activity. Numerous publications indicate that muscle contractile activity leads to activation of cell signaling systems such as the MAP kinases and the adenosine monophosphate activated kinase (see refs. (2) and (13) for reviews). The stimuli for induction of these pathways are unclear, but Wretman et al. (14) reported that contraction-induced changes in the skeletal muscle mitogen-activated protein kinaseerk1/2 signaling pathway are mediated by oxidants. Some data from our laboratory indicate that contraction-induced stimulation of stress protein expression is mediated by oxidants (4–6), and in an unrelated area, Tidball et al. (13) have shown that some adaptive responses of the skeletal muscle cytoskeleton in response to passive stretch are mediated by NOS activity.
OXIDANTS AS DAMAGING AGENTS TO SKELETAL MUSCLE
There is little consensus about the importance of oxidants in the pathogenesis of exercise-induced muscle damage. Most data indicate that there is increased free radical activity in muscle during exhaustive exercise where the muscle is contracting in a primarily concentric manner, but this type of exercise does not normally lead to significant muscle damage, implying that the muscle antioxidant system is capable of preventing cell damage. There is much less evidence for increased free radical activity when exercise is of the type to cause muscle damage (i.e., usually involving eccentric or lengthening contractions). There is, however, evidence for the involvement of nonmuscle-derived oxidants in the delayed damage that may follow lengthening contractions. This, so-called secondary phase of muscle damage is associated with infiltration of phagocytic cells, an increased oxidation of muscle cell components (9) (Fig. 3), and may be prevented by pretreatment of the animals with polyethylene glycol-tagged superoxide dismutase (15).
It is clear that ROS are generated by skeletal muscle in increased amounts during contractile activity. The main potential sources seem to be mitochondria and membrane-associated oxidoreductase systems. Furthermore, we hypothesize that the nature of the different generating systems argues that different types of contractile activity (aerobic or anaerobic or shortening or lengthening) may lead to generation of different patterns or magnitude of ROS generation. It also seems likely that ROS stimulate changes in cell signaling and gene expression that may contribute some aspects of the adaptive responses to contractile activity including changes in stress protein expression and upregulation of some cytoskeletal proteins. In contrast, an excessive increase in ROS generation may lead to damage to skeletal muscle. However, the role of muscle-derived ROS in contraction-induced muscle damage has not been substantiated.
Recent data indicate that the systems regulating ROS activity (i.e., the antioxidant defense enzymes and proteins) in skeletal muscle are efficient and adaptable. It seems likely that we are beginning to obtain an understanding of a tightly regulated system whereby the balance between whether ROS induce changes in gene expression or muscle damage after a period of contractile activity is dictated by the magnitude and pattern of the ROS generated and by cellular content and activity of defense systems (Fig. 4). However, the ability of highly reactive ROS to stimulate changes in gene expression ensures that the muscle rapidly adapts (or “trains”) to cope with the rise in ROS generation in common with its ability to train to deal with other undesirable effects of unaccustomed or excessive exercise.
1. Balon, T. W., and Nadler. J. L. Nitric oxide release is present from incubated skeletal muscle preparations. J. Appl. Physiol. 77: 2519–2521, 1994.
2. Jackson, M. J. Free radical mechanisms in exercise-related muscle damage. In: Oxidative stress in skeletal muscle. MCBU, edited by A. Z. Reznick, L. Packer, C. K. Sen, J. O. Holloszy, and Jackson M. J. Berlin: Birkhauser, pp 75–87, 1998.
3. Jackson, M. J., Edwards, R. H. T. and Symons. M. C. R. Electron spin resonance studies of intact mammalian skeletal muscle. Biochim. Biophys. Acta. 847: 185–190, 1985.
4. Jackson, M. J., Khassaf, M. Esanu, C. Vasilaki, A. Brodie, D. A. and McArdle. A. Vitamin C supplements suppress the stress response in human muscle. Free Radic. Biol. Med. 27: S37, 1999.
5. Jackson, M. J., Papa, S. Bolanos, J. Bruckdorfer, R. Carlsen, H. Elliot, R. M. Flier, J. Griffiths, H. R. Heales, S. Holst, B. Lorusso, M. Lund, E. Moskaug, J. O. Moser, U. Di Paola, M. Polidori, M. C. Signorile, A. Stahl, W. Vian-Ribes, J. and Astley. S. B. Antioxidants, reactive oxygen and nitrogen species, gene induction and mitochondrial function. Mol. Aspects Med. 23: 209–285, 2002.
6. Khassaf, M., Child, R. B. McArdle, A. Brodie, D. A. Esanu, C. and Jackson M. J. Time course of responses of human skeletal muscle to oxidative stress induced by nondamaging exercise J. Appl. Physiol. 90: 1031–1035, 2001.
7. Kobzik L., Reid, M. B. Bredt, D. S. and Stamler. J. S. Nitric oxide in skeletal muscle. Nature. 372: 546–548, 1994.
8. McArdle, A., Maglara, A. Appleton, P. Watson, A. J. Grierson, I. and Jackson M. J. Apoptosis in multinucleated skeletal muscle myotubes. Lab Invest. 79: 1069–1076, 1999a.
9. McArdle, A., van der Meulen, J. H. Catapano, M. Symons, M. C. Faulkner, JA. and Jackson. M. J. Free radical activity following contraction-induced injury to the extensor digitorum longus muscles of rats. Free Radic. Biol. Med. 26: 1085–1091, 1999b.
10. McArdle, A., Pattwell, D. Vasilaki, A. Griffiths, R. D. and Jackson. M. J. Contractile activity-induced oxidative stress: Cellular origin and adaptive responses. Am. J. Physiol. Cell Physiol. 280: C621–627, 2001.
11. O’ Neill, C., Stebbens, C. L. Bonigut, S. Halliwell, B. and Longhurst. J. C. Production of hydroxyl radicals in contracting skeletal muscle of cats. J. Appl. Physiol. 81: 1197–1206, 1996.
12. Reid, M. B., Shoji, T. Moody, M. R. and Entman. M. L. Reactive oxygen in skeletal muscle II. Extracellular release of free radicals. J. Appl. Physiol. 73: 1805–1809, 1992.
13. Tidball, J. G., Spencer, M. J. Wehling, M. and Lavergne. E. Nitric-oxide synthase is a mechanical signal transducer that modulates talin and vinculin expression. J. Biol. Chem.12; 274: 33155–33160, 1999.
14. Wretman, C., Lionikas, A. Widegren, U. Lannergren, J. Westerblad, H. and Henriksson. J. Effects of concentric and eccentric contractions on phosphorylation of MAPK(erk1/2) and MAPK(p38) in isolated rat skeletal muscle. J. Physiol.15; 535: 155–164, 2001.
15. Zerba, E., Komorowski, T. E. and Faulkner. J. A. Free radical injury to skeletal muscles of young, adult, and old mice. Am. J. Physiol. 258: C429–435, 1990.