| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
INVITED REVIEW
HIGHLIGHTED TOPIC
Free Radical Biology in Skeletal Muscle
Department of Physiology and Pharmacology, Karolinska Institutet, Stockholm, Sweden
 |
ABSTRACT |
|---|
 TOP ABSTRACT ROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
exercise; adenosine 5'-monophosphate-activated protein kinase
40% of body weight), are GLUT-1 and -4. GLUT-1 resides in the cell membrane, is constitutively active, and controls basal glucose transport. GLUT-4, which under basal conditions resides primarily in intracellular compartments, is recruited to the plasma membrane/t-tubules in response to physiological stimuli, such as insulin, exercise, and hypoxia (10, 28, 47, 59, 68). Several recent reviews have focused on the regulation of glucose transport in skeletal muscle during exercise/contraction (29, 34, 58). The present review will focus on the role of reactive oxygen species (ROS) in modulation of glucose transport into skeletal muscle. |
ROS AND GLUCOSE TRANSPORT |
|---|
|
TOPABSTRACTROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
–), hydroxyl and hydroperoxyl radicals, and hydrogen peroxide (H2O2). The parent molecule, O2–, is highly reactive. Dismutation of O2–, as well as dehydration of H2O2, occurs spontaneously. Additionally, O2– can be enzymatically dismutated to H2O2 by superoxide dismutase (SOD, catalyzes 2O2– + 2H+ 
H2O2 + O2). H2O2 can, in turn, be oxidized in a reaction catalyzed by glutathione peroxidase (H2O2 + 2GSH GSSG + H2O, where GSH and GSSG are reduced and oxidized glutathione, respectively) and catalase (2H2O2 2H2O + O2). There are several cellular sites where O2– can be produced (53), but it is generally accepted that mitochondria serve as a major site of production. Measurements of ROS formation in skeletal muscle are not trivial and are normally performed with the use of fluorescent dyes, although other methods (e.g., electron spin resonance spectroscopy, cytochrome c reduction, glutathione oxidation) are also used (53). Limitations to these techniques include the inability to quantify the extent of ROS production, as well as to identify the species of ROS produced. Furthermore, during conditions of electrical stimulation of muscles, the repeated current pulses may generate molecular species that result in artifactual fluorescence increases of the dye, independent of muscle ROS formation (Sandström M, Westerblad H, Katz A, unpublished data). This requires carefully designed experiments to correct for such potential artifacts (60). Additional information on free radical biology can be obtained elsewhere (53). Early studies demonstrated that thiols exerted insulin-like effects in isolated fat cells (13, 42). Subsequently, it was shown that H2O2 was formed in the reaction of thiols (14) and that exogenous H2O2 accelerated glucose transport in adipocytes (15). Such findings were extended to other preparations (17, 27), including skeletal muscle (11, 65). Other compounds (e.g., polyamines) were subsequently also found to act through generation of H2O2 (43). Initially, the mechanism whereby H2O2 accelerated glucose transport was unclear. H2O2 stimulated the uptake of D-glucose, which requires GLUTs for entry, but not L-glucose, which enters cells via diffusion. Moreover, the H2O2-mediated effect appeared to occur at a step subsequent to hormone (insulin)-receptor interaction and was sensitive to N-ethylmaleimide (15), an inhibitor of GLUT-4 translocation. Later studies demonstrated that H2O2 increased tyrosine phosphorylation and intracellular Ca2+ concentrations ([Ca2+]i) in cells, and that this occurred as a consequence of activation of tyrosine kinases and inhibition of tyrosine phosphatases (24, 25, 62). The involvement of [Ca2+]i in trafficking/docking of GLUT-4 is now well established and will not be elaborated here (69).
Whereas it is firmly established that exogenous H2O2 accelerates glucose transport in skeletal muscle (11, 65), the mechanism whereby this occurs is unclear. Studies on intact single fibers from a mouse foot muscle have shown that exposure of the fiber to H2O2 (100–300 µM) increases [Ca2+]i (50% above basal) (1), albeit not to levels that initiate contraction. Elevations of [Ca2+]i have long been implicated in accelerating glucose transport in skeletal muscle (28), even at levels that are insufficient to induce contraction/force generation (76). Thus one mechanism whereby H2O2 can increase glucose transport is via an increase in [Ca2+]i. Precisely how the rise in [Ca2+]i would result in an increased translocation and insertion of GLUT-4 proteins into surface membranes is not known, but several possibilities are discussed below.
Another factor to consider is AMP-activated protein kinase (AMPK), a sensor of metabolic stress (20, 21). It was recently demonstrated that H2O2 activates AMPK in various cell types (12, 54, 73), and it is now well established that AMPK activation results in accelerated glucose transport into skeletal muscle (34, 70). Subsequently, it was demonstrated that exogenous H2O2 and a superoxide-generating system activated AMPK, as well as glucose transport, in intact, isolated rat muscle preparations. The ROS-mediated activation of AMPK was independent of marked changes in high-energy phosphates (66). Moreover, the stimulatory effects of ROS on AMPK and glucose transport were to a large degree blocked by a general antioxidant, N-acetylcysteine (NAC). These findings led the authors to propose that oxidative stress is an important stimulus of AMPK and glucose transport during contraction (see below). As with the rise in [Ca2+]i, the link between increased AMPK activity and an increased translocation and insertion of GLUT-4 proteins into surface membranes is not known. Moreover, whether there is a link between the H2O2-mediated increases in [Ca2+]i and AMPK activity is also not clear.
It should be noted that, while addition of exogenous ROS to cells/tissues can result in metabolic/functional responses, this does not prove that such responses are necessarily mediated by ROS under physiological conditions. To directly test the possibility that endogenously produced ROS are involved in a cellular response (e.g., glucose transport), experiments wherein antioxidants (exogenous and endogenous) block ROS formation and the cellular response are required. Such approaches could be complemented by experiments based on genetic manipulations, where ROS formation/buffering is altered.
|
ROS, INSULIN, AND GLUCOSE TRANSPORT |
|---|
|
TOPABSTRACTROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
The findings that ROS exhibited insulin-mimetic effects (see above) led to the investigation of the involvement of ROS in insulin action. Insulin accelerates glucose transport in skeletal muscle by a pathway that involves a series of phosphorylation/dephosphorylation steps, ultimately resulting in the translocation of GLUT-4 transporters to the surface membranes (9). One of the indications for ROS involvement in insulin signaling derived from the observation that both H2O2 and insulin enhanced phosphorylation of the insulin receptor in isolated rat adipocytes (23). Early studies showed that both insulin and H2O2 activated insulin receptor tyrosine kinase activity, but only H2O2 inhibited the activity of protein tyrosine phosphatase (PTP) (23, 25). However, subsequent studies, using more developed techniques, did demonstrate that insulin resulted in inactivation of PTPs (also PTP-1B) (46). The involvement of PTPs in insulin signaling was supported by the observation that nondiabetic PTP-1B-deficient mice exhibited improved whole body glucose homeostasis, as well as enhanced skeletal muscle glucose uptake and insulin signaling following insulin administration (16, 40).
PTP-1B contains a critical cysteine residue (cys-215) required for catalysis, and oxidation or disulfide conjugation of this residue inactivates the enzyme (18). That ROS can induce such changes is thus clear. But does insulin do this, and, if so, how? Early studies by Czech et al. (15) presented evidence for sulfhydryl oxidation in the regulation of fat cell glucose transport by insulin. Subsequently, it was demonstrated that insulin stimulated H2O2 production in adipocytes (48). Apparently, this occurs first via the activation of an NADPH oxidase (Nox; likely the Nox4 isoform), which catalyzes the following reaction: 2O2 + NADPH 2O2– + NADP+ + H+, followed by SOD-mediated production of H2O2 (18). Noteworthy is that inhibition of insulin-stimulated H2O2 production results in loss of several key signaling steps, including insulin-mediated activation of phosphatidyl inositol 3-kinase and Akt (45). In support of a role for Nox4 in insulin action are the findings that Nox4-deficient cells exhibit reduced insulin signaling and reduced insulin-mediated glucose transport (44). For a more exhaustive review of the role of ROS in insulin action, the reader is referred elsewhere (18).
The studies of ROS involvement in insulin action have mostly been performed on adipocytes (18). The extent to which ROS participate in insulin action in skeletal muscle is unclear. The extrapolation of results from fat to muscle cells must be made with caution, since insulin signaling pathways differ in the two tissues (33). Antioxidants improve insulin-mediated glucose transport in skeletal muscle and whole body glucose homeostasis of insulin resistant/diabetic animals and diabetic patients (6, 26), which supports the idea that prolonged, excessive levels of ROS interfere with insulin action. What is less clear, however, is whether insulin acutely increases ROS production in intact skeletal muscle preparations of normal muscle preparations. To the author's knowledge, measurements of insulin-mediated O2– or H2O2 production in intact skeletal muscle preparations have not been reported. If insulin-mediated ROS production occurs to a significant extent, then addition of antioxidants would be expected to attenuate insulin-mediated glucose transport. However, NAC does not alter insulin-mediated glucose transport in isolated muscle preparations (60). Similarly, prolonged treatment of rats (20 days) with L-buthionine-[S,R]-sulfoximine (BSO; inhibitor of 
-glutamylcysteine synthetase), to deplete tissue glutathione pools, did not affect insulin-mediated glucose transport in isolated muscle preparations (39). If insulin-mediated H2O2 production were important for the activation of glucose transport, one would have expected an increased insulin-mediated glucose transport in BSO-treated muscle, since the accumulation of H2O2 should be enhanced in the presence of lower GSH levels. Thus convincing evidence in support of endogenous ROS production in insulin-mediated glucose transport in healthy skeletal muscle preparations is lacking at present, but additional research is required to shed more light on this issue.
|
ROS, HYPOXIA, AND GLUCOSE TRANSPORT |
|---|
|
TOPABSTRACTROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
There is a paucity of information regarding the involvement of ROS in hypoxia-mediated glucose transport. The first question is: does hypoxia increase ROS production? Whereas there is good evidence that ROS production is increased in tissues during the reoxygenation phase following hypoxia (64, 79), what occurs during the hypoxia phase is less clear. Some studies indicate that ROS production decreases during hypoxia (49, 50), whereas others indicate that ROS production increases during hypoxia (4, 67).
Often, when measuring the effect of hypoxia on glucose transport, the actual measurement of glucose transport following the exposure to hypoxia occurs under aerobic conditions (8, 10, 51, 63). This allows for direct comparison of transport rates between the hypoxic and aerobic conditions. However, this raises the possibility that the ROS produced during reoxygenation results in the activation of glucose transport. Early studies demonstrated that, when measurements of transport were made under hypoxic conditions, the transport rate was still markedly elevated (41, 55, 77), and recent studies have confirmed this (60). Thus the stimulatory effect of hypoxia on glucose transport is not a consequence of the reoxygenation step.
Direct measurements of ROS during hypoxia in skeletal muscle are scarce. Recent studies indicate an increased ROS production in rat diaphragm muscle strips during hypoxia (78). This increase was fully blocked by the glutathione peroxidase mimetic ebselen. The increase in ROS production, assessed by oxidation of dihydrofluorescein-diacetate, amounted to 5% above baseline (78). Whether such an increase is sufficient to stimulate glucose transport is not known. Perhaps a more sensitive approach to detect the extent of the oxidative stress during hypoxia is to measure the glutathione status. Indeed, studies on isolated coronary arterial rings have shown that short-term hypoxia results in significant decreases in GSH and increases in GSSG (19). If hypoxia caused a change in glutathione status, then an antioxidant should block this change.
In this context, it is noteworthy that recent studies on isolated mouse EDL muscles have shown that hypoxia-mediated activation of glucose transport was not significantly affected by the general antioxidant NAC (60). Thus these data do not support a role for ROS in hypoxia-mediated glucose transport in the preparation studied. However, additional studies are required to more fully investigate the role of ROS in hypoxia-mediated glucose transport in skeletal muscle.
|
ROS, EXERCISE/CONTRACTION, AND GLUCOSE TRANSPORT |
|---|
|
TOPABSTRACTROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
Direct experiments to investigate the role of ROS in contraction-mediated glucose transport were recently completed (60). The initial and key finding was that NAC inhibited 50% of contraction-mediated glucose uptake in isolated mouse fast-twitch muscle. This observation agreed well with the observation that NAC also inhibited 50% of contraction-mediated activation and phosphorylation of AMPK in the same preparations. Moreover, repeated contractions resulted in increased ROS formation, and this was abolished by NAC. On the other hand, NAC did not significantly affect basal, hypoxia-, or insulin-mediated glucose transport. These data supported the idea that endogenous ROS production plays a key role in contraction-mediated glucose transport and that the ROS effect is mediated via AMPK. To further study the species of ROS involved, experiments were performed with the glutathione peroxidase mimetic, ebselen, which catalyzes H2O2 removal in the presence of GSH (60). Ebselen inhibited contraction-mediated glucose uptake by almost 60%, i.e., similar to the effect of NAC. This result implicated H2O2, or its derivatives, as the signaling molecule. In an alternative approach to investigate the role of H2O2, muscles were studied from mice overexpressing Mn2+-dependent SOD (expressed in mitochondria). Theoretically, these muscles should have accelerated rates of H2O2 production, owing to an enhanced capacity to convert O2– to H2O2, and consequently an increased contraction-mediated glucose uptake. Indeed, these muscles exhibited a 25% increase in contraction-mediated glucose uptake vs. wild-type muscles. In contrast, basal and insulin-stimulated glucose uptake were not different between SOD overexpressing and wild-type muscles (60). In a follow-up study, using an inhibitor of myosin ATPase and hence an inhibitor of cross-bridge function, isolated mouse fast-twitch muscle was stimulated to perform repeated contractions. Cross-bridge inhibition decreased force production to 5% of control, but had little effect on contraction-mediated glucose transport or contraction-mediated changes in oxidative stress (61).
The experiments described above were all performed on fast-twitch muscles (60). This raises the question of whether the results are also applicable to slow-twitch muscles. Initial experiments demonstrate that contraction-mediated glucose transport is higher in soleus muscle from mice in which Mn2+-dependent SOD is overexpressed (1.97 ± 0.12 vs. 1.37 ± 0.16 µmol·20 min–1·ml intracellular water–1 for wild-type muscles; P < 0.05), whereas there is no significant difference in transport under basal conditions or after insulin stimulation (Sandström M, Zhang SJ, Silva IP, Westerblad H, Katz A, unpublished observations). These findings are consistent with the idea that ROS is also a significant component in contraction-mediated glucose transport in slow-twitch muscle. Taken together, the results indicate that endogenously produced ROS, possibly H2O2 or its derivatives, play an important role in contraction-mediated activation of glucose transport.
One question that arises as a consequence of these experiments is what level of ROS/H2O2 is required to initiate the glucose transport response to contraction? The answer is not known, since there are no reliable data regarding ROS/H2O2 concentrations in contracting muscle. What is clear is that the oxidative stress (as judged by the extent of glutathione oxidation) exhibited by isolated fast-twitch muscle in response to 3 mM exogenous H2O2 is markedly greater than that exhibited in response to 10 min of repeated contractions (60). Interestingly, the increase in glucose uptake was greater after repeated contractions than following exposure to exogenous H2O2. One interpretation of these results is that excessive oxidative stress can have negative effects on glucose transport, which would be consistent with the data regarding excessive oxidative stress and the development of insulin resistance and diabetes (see above). Another interpretation would be that low or physiological levels of ROS/H2O2, such as those achieved during repeated contractions, are sufficient to achieve maximal rates of glucose uptake. This would be consistent with the idea that such ROS levels are requisite for optimal cell signaling (18, 53). Indeed, the extent of glutathione oxidation was markedly smaller after repeated contractions than after exposure to 3 mM H2O2 (60). Yet a third possibility is that endogenous ROS production affects sites involved in glucose uptake that are not accessed by exogenous ROS. Clearly, additional research is required to address the optimal ROS/H2O2 concentrations required to elicit the increase in muscle glucose transport in response to contraction.
If ROS does participate in contraction-mediated glucose transport, then how does this occur? In Fig. 1, a hypothetical scheme is provided to link different signaling events that result in acceleration of glucose transport during contraction. Under normal physiological conditions, the process begins with an action potential that leads to a release of Ca2+ from the sarcoplasmic reticulum into the myoplasm. Myoplasmic Ca2+ may play a central role in the initiation of glucose transport via several potential mechanisms. Ca2+ can result in the activation of calmodulin-dependent protein kinase (CaMK) II during contraction, which appears to be involved in the activation of glucose transport (56, 57, 75). The Ca2+ effect on CaMK II is likely a direct one (via calmodulin) (30, 57). Ca2+ can also activate CaMK kinase (CaMKK), which can phosphorylate and activate AMPK (22, 32, 71), and CaMKK protein is expressed in skeletal muscle (36, 57). However, the expression of CaMKK in skeletal muscle is low, and it is not clear whether CaMKK plays a role in AMPK phosphorylation/activation in this tissue (5). Available data would suggest that CaMKK probably does not phosphorylate and activate AMPK in skeletal muscle to any significant extent, because incubation of isolated muscle preparations with a subcontraction concentration of caffeine (increases myoplasmic Ca2+) resulted in the activation of glucose transport and the phosphorylation of CaMK II. The caffeine effect on glucose transport was blocked by inhibitors of CaMKs. However, the same caffeine concentration did not alter the degree of AMPK phosphorylation (75). Alternatively, the caffeine-mediated increase in [Ca2+]i may not have been sufficient to activate CaMKK. The rise in myoplasmic Ca2+ can also result in contraction (activation of myosin-ATPase) and activation of Ca2+ pumps (Ca2+-ATPase). This will alter the concentrations of high-energy phosphates, which can result in direct activation of AMPK. The changes in high-energy phosphates can stimulate mitochondrial respiration and ROS production; Ca2+ can also activate respiration directly by activating mitochondrial dehydrogenases (60). ROS can then activate AMPK, which leads to translocation of GLUT-4 from intracellular sites to the plasma membrane, followed by an increase in glucose transport.
|
When discussing glucose uptake during exercise, it is worthwhile noting that this glucose is not always metabolized for ATP production. As mentioned earlier, glucose uptake by human skeletal muscle can increase up to 50-fold at an intensity corresponding to 100% of maximal oxygen uptake, which can be maintained for 5 min (37). Despite this large increase, the contribution of extracellular glucose to muscle carbohydrate utilization is negligible at this exercise intensity, probably owing to inhibition of hexokinase, the enzyme that phosphorylates glucose to glucose-6-phosphate. The inhibition is caused by glucose-6-phosphate that accumulates as a consequence of glycogenolysis. This results in a large increase in intracellular glucose (37). However, during lower intensity, prolonged exercise, the contribution of extracellular glucose to muscle carbohydrate utilization becomes more significant, probably because the inhibition of hexokinase is relieved (38). The implication of these findings is that glucose supplements are of little benefit for muscle carbohydrate utilization and hence performance during short-term, intense bouts of physical activity. If, however, liver carbohydrate stores are depleted under such conditions, then glucose supplements would be beneficial, but probably for the purpose of maintaining substrate for the central nervous system (37).
|
CONCLUDING COMMENTS |
|---|
|
TOPABSTRACTROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
The recent findings implicating ROS in contraction-mediated glucose transport should be viewed as initial observations, and many additional questions remain. Does the evidence for ROS also hold for other mammalian species or during conditions of submaximal exercise? What is the ROS species that is responsible for signaling, and in which compartment/site is it generated? How is the signal transmitted? Which AMPK isoforms are regulated by ROS during exercise? These questions await further study.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTROS AND GLUCOSE TRANSPORTROS, INSULIN, AND GLUCOSE...ROS, HYPOXIA, AND GLUCOSE...ROS, EXERCISE/CONTRACTION, AND...CONCLUDING COMMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. R. Blair, K. Funai, G. G. Schweitzer, and G. D. Cartee A myosin II ATPase inhibitor reduces force production, glucose transport, and phosphorylation of AMPK and TBC1D1 in electrically stimulated rat skeletal muscle Am J Physiol Endocrinol Metab, May 1, 2009; 296(5): E993 - E1002. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Natalicchio, F. De Stefano, S. Perrini, L. Laviola, A. Cignarelli, C. Caccioppoli, A. Quagliara, M. Melchiorre, A. Leonardini, A. Conserva, et al. Involvement of the p66Shc protein in glucose transport regulation in skeletal muscle myoblasts Am J Physiol Endocrinol Metab, February 1, 2009; 296(2): E228 - E237. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Bashan, J. Kovsan, I. Kachko, H. Ovadia, and A. Rudich Positive and Negative Regulation of Insulin Signaling by Reactive Oxygen and Nitrogen Species Physiol Rev, January 1, 2009; 89(1): 27 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Lecarpentier Physiological role of free radicals in skeletal muscles J Appl Physiol, December 1, 2007; 103(6): 1917 - 1918. [Full Text] [PDF]
|
|
|
|
|
|
M. B. Reid Editorial J Appl Physiol, April 1, 2007; 102(4): 1299 - 1300. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
HEP, changes in high-energy phosphates; CAMK II, calmodulin-dependent protein kinase II; 
O2, oxygen consumption; O2