Regulation of Glucose Transport and Glycogen Synthesis in L6 Muscle Cells during Oxidative Stress

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 51, Issue of December 17, pp , by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Regulation of Glucose Transport and Glycogen Synthesis in L6 Muscle Cells during Oxidative Stress EVIDENCE FOR CROSS-TALK BETWEEN THE INSULIN AND SAPK2/p38 MITOGEN-ACTIVATED PROTEIN KINASE SIGNALING PATHWAYS* (Received for publication, September 7, 1999, and in revised form, October 4, 1999) Anne S. Blair, Eric Hajduch, Gary J. Litherland, and Harinder S. Hundal From the Department of Anatomy and Physiology, Medical Sciences Institute/Wellcome Trust Biocenter Complex, University of Dundee, Dundee DD1 5EH, United Kingdom We have investigated the cellular mechanisms that participate in reducing insulin sensitivity in response to increased oxidant stress in skeletal muscle. Measurement of glucose transport and glycogen synthesis in L6 myotubes showed that insulin stimulated both processes, by 2- and 5-fold, respectively. Acute (30 min) exposure of muscle cells to hydrogen peroxide (H 2 O 2 ) blocked the hormonal activation of both these processes. Immunoblot analyses of cell lysates prepared after an acute oxidant challenge using phospho-specific antibodies against c-jun N-terminal kinase (JNK), p38, protein kinase B (PKB), and p42 and p44 mitogen-activated protein (MAP) kinases established that H 2 O 2 induced a dose-dependent activation of all five protein kinases. In vitro kinase analyses revealed that 1 mm H 2 O 2 stimulated the activity of JNK by 8-fold, MAP- KAP-K2 (the downstream target of p38 MAP kinase) by 12-fold and that of PKB by up to 34-fold. PKB activation was associated with a concomitant inactivation of glycogen synthase kinase-3. Stimulation of the p38 pathway, but not that of JNK, was blocked by SB or SB203580, while that of p42/p44 MAP kinases and PKB was inhibited by PD and wortmannin respectively. However, of the kinases assayed, only p38 MAP kinase was activated at H 2 O 2 concentrations (50 M) that caused an inhibition of insulin-stimulated glucose transport and glycogen synthesis. Strikingly, inhibiting the activation of p38 MAP kinase using either SB or SB prevented the loss in insulin-stimulated glucose transport, but not that of glycogen synthesis, by oxidative stress. Our data indicate that activation of the p38 MAP kinase pathway plays a central role in the oxidant-induced inhibition of insulin-regulated glucose transport, and unveils an important biochemical link between the classical stress-activated and insulin signaling pathways in skeletal muscle. A key aspect of mammalian physiology involves the regulation of blood glucose levels. Glucose homeostasis is controlled largely by the action of circulating insulin, which facilitates the disposal of blood glucose in the fed state by stimulating its * This work was supported in part by the British Diabetic Association, the Biotechnology and Biological Sciences Research Council, and the Wellcome Trust. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Recipient of a CASE studentship from the Medical Research Council and SmithKline Beecham Pharmaceuticals. To whom correspondence should be addressed. Fax: ; h.s.hundal@dundee.ac.uk. This paper is available on line at uptake into target tissues, primarily skeletal muscle and fat (1 3). Both these tissues contribute toward the lowering of blood glucose, but it is widely accepted that skeletal muscle, by virtue of its large contribution to body mass, represents the major site of insulin-mediated glucose disposal (4). The stimulation in glucose uptake elicited by insulin in both skeletal muscle and fat is achieved principally by the increased recruitment of the insulin regulated glucose transporter, GLUT4, to the blood-facing membranes of these tissues from its intracellular storage pools (5). Reduced insulin sensitivity is a characteristic feature of various pathological conditions such as diabetes (1, 6) and hypertension (7). Insulin resistance in skeletal muscle and fat may result in the progressive dysfunction of important hormonal effects such as glucose and protein homeostasis, and thus it is of considerable interest to understand the molecular pathology of this process. One factor that appears to be important in the progression of insulin resistance in the above conditions, as well as during hypoxia, ischemia/reperfusion injury, and sepsis, is increased oxidant stress (8 12). Prolonged oxidant stress in muscle and fat has been shown to reduce insulin-stimulated glucose transport significantly and to induce a compensatory increase in basal transport, through increased synthesis of GLUT1 (13, 14). The mechanism by which oxidant stress causes insulin resistance, be it a nonspecific effect of chronic exposure to reactive oxygen species or the specific activation of a stress-related signaling pathway, is unknown. However, since various stresses are known to induce a cellular protective response that involves activation of various protein kinases (15), it is possible that this participates in modulating insulin action. Insulin is known to stimulate multiple signaling pathways, but there is general acceptance that phosphoinositide 3-kinase (PI3K) plays a central role in regulating glucose transport and glycogen synthesis (5, 16). The hormonal activation of PI3K catalyzes the production of 3 -phosphoinositides (e.g. phosphatidylinositol 3,4,5-trisphosphate), which act as key intermediates in the activation of protein kinase B (PKB/Akt), a molecule that is regulated also by growth factors and which plays a crucial role in the control of cell proliferation, differentiation, and survival (17, 18). Activated PKB is known to target glycogen synthase kinase-3 (GSK3), whose phosphorylation results in its inactivation, a step that is considered crucial for the concomitant activation of glycogen synthase (19, 20). In addition, there is growing evidence that expressing constitutively active or dominant negative PKB mutants in muscle and fat cells induce changes in glucose transport and GLUT4 translocation that are consistent with the involvement of PKB in the insulin signaling pathway regulating glucose uptake (21 24). Impaired activation of these early signaling molecules repre-

2 36294 SAPK2/p38 and Insulin Signaling sents a potential mechanism by which insulin sensitivity may be lost in response to increased oxidant stress. This possibility is supported by recent work showing that oxidant stress inhibits the hormonal activation of IRS1, PI3K, and PKB in NIH 3T3 fibroblasts and 3T3-L1 adipocytes (25, 26). However, it is noteworthy that reactive oxygen species such as H 2 O 2 also exert insulin-like effects and can activate PKB (27, 28) and numerous other protein kinases, including components of the classical MAP kinase and stress signaling pathways (11, 27, 29, 30). It is likely that the activation of these pathways plays an important role in mediating some of the biological and possibly pathological responses to environmental stresses. Indeed, in some cell types, activation of the p38 MAP kinase pathway has been implicated with some of the adverse complications associated with raised blood glucose levels and diabetes (31). Whether stress kinases, such as JNK and p38 MAP kinase, participate in modulating insulin action in response to oxidative stress in skeletal muscle is unknown currently. In this study we show that acute oxidative stress activates both JNK and the p38 MAP kinase pathway in cultured rat muscle cells and that activation of the latter results in the inhibition of insulinstimulated glucose transport. EXPERIMENTAL PROCEDURES Materials -Minimal essential media ( -MEM), fetal calf serum, antimycotic/antibiotic solution were from Life Technologies, Inc. (Paisley, Renfrewshire, Scotland). Wortmannin, insulin, cytochalasin B, H 2 O 2, and Kodak X-Omat film were from Sigma-Aldrich Co. Ltd. (Poole, United Kingdom (UK)). SB , SB , and PD were from Calbiochem-Novabiochem Ltd. (Nottingham, UK). Phospho- and dephospho-specific antibodies to p38, PKB, JNK, and p44/42 MAP kinase were from New England Biolabs (Hertfordshire, UK). Horseradish peroxidase-conjugated anti-rabbit IgG and anti-mouse IgG were from SAPU (Lanarkshire, Scotland). Reagents for ECL were from Pierce & Warriner (Chester, UK). Hybond nitrocellulose and uridine diphospho-d-[6-3 H]glucose were from Amersham Pharmacia Biotech (Amersham, UK). 2-Deoxy-[ 3 H]-D-glucose was from NEN Life Science Products and [ 32 P]-ATP from ICN (Costa Mesa, CA). Protein A- and Protein G-Sepharose were from Amersham Pharmacia Biotech (Uppsala, Sweden). Antibodies against PKB, PKB, GSK3, and MAP- KAP-K2 (for immunoprecipitation) and peptide substrates for these kinases were provided kindly by Professor Philip Cohen, MRC Protein Phosphorylation Unit, University of Dundee. ATF-2 peptide, used as a substrate for JNK was from Upstate Biotechnology, Inc. (Lake Placid, NY). Cell Culture Monolayers of L6 muscle cells were cultured to the stage of myotubes as described previously (23) in -MEM containing 2% (v/v) fetal calf serum and antimycotic/antibiotic solution (100 units/ml penicillin, 100 g/ml streptomycin, 250 ng/ml amphotericin B) at 37 C in an atmosphere of 5% CO 2, 95% air. Cells were cultured in six-well plates for uptake studies and in 10-cm dishes for glycogen synthesis or protein kinase assays. Myotubes were deprived of serum routinely for 5 h prior to use in serum-free -MEM plus 25 mm D-glucose. After 4 h, cells were incubated for the final hour in HEPES-buffered saline (HBS; 20 mm HEPES-Na (ph 7.4), 140 mm NaCl, 2.5 mm MgSO 4,5mM KCl, 1 mm CaCl 2 ) containing 25 mm D-glucose. Subsequent additions to the cells were at the times and concentrations indicated in the figure legends. Assay of Immunoprecipitated Kinases from L6 Lysates L6 myotubes were deprived of serum for 4hin -MEM and washed twice with warm HBS. Cells were incubated subsequently at 37 C in HBS plus 25 mm D-glucose for 1 h. During the last hour insulin, H 2 O 2 and kinase inhibitors (wortmannin, PD 98059, SB , or SB ) were added at times and concentrations indicated in the figure legends prior to cell lysis. Myotubes were extracted from 10-cm dishes using ice-cold lysis buffer (50 mm Tris-HCl, ph 7.5, 1 mm EDTA, 1 mm EGTA, 1% (v/v) Triton X-100, 1 mm Na 3 VO 4,10mM sodium -glycerophosphate, 50 mm NaF, 5 mm Na 4 P 2 O 7,1 M microcystin-lr, 0.27 M sucrose, 0.2 mm phenylmethylsulfonyl fluoride, 1 mm benzamidine, 10 g/ml leupeptin, and 0.1% (v/v) 2-mercaptoethanol). PKB and PKB were immunoprecipitated from 100 g of L6 lysate using isoform-specific antibodies and assayed using crosstide, a synthetic peptide (GRPRTSSFAEG) corresponding to the GSK3 sequence surrounding the Ser phosphorylated by MAPKAP-K1 (20). GSK3 was immunoprecipitated from 100 g of cell lysate and incubated with or without 25 milliunits/ml PP2A 1 prior to assay using phospho-gs peptide-1 (32). MAPKAP-K2 was immunoprecipitated from 50 g of cell lysate and assayed by monitoring phosphorylation of the peptide (KKLNRTLSVA), derived from glycogen synthase (33). JNK was immunoprecipitated from 100 g of cell lysate, and assayed by monitoring the phosphorylation of ATF-2 peptide. One unit of PKB, MAPKAP-K2, and JNK activity was defined as that amount which catalyzed the phosphorylation of 1 nmol of substrate in 1 min. Protein concentrations were determined using the method of Bradford (34). Assay of Glycogen Synthase Activity L6 lysates were prepared as described above. Glycogen synthase activity in lysates was assayed at ph 6.8 by measuring the incorporation of glucose from uridine diphospho-[6-3 H]-D-glucose into glycogen as described previously (35). Enzyme activity was expressed as a ratio of the activity measured in the absence of glucose 6-phosphate divided by that in the presence of the allosteric activator (20 mm). Glucose Transport L6 myotubes were incubated in serum-free -MEM for 4 h and then placed for 1hinHBScontaining 25 mm D-glucose. During the final hour, H 2 O 2 and kinase inhibitors (wortmannin, PD 98059, SB , or SB ) were added at times and concentrations indicated in the figure legends prior to assaying glucose uptake. Cells were washed rapidly with HBS, and glucose uptake was assayed by incubating cells with 10 M 2-deoxy-[ 3 H]-D-glucose (1 Ci/ ml, 26.2 Ci/mmol) for 10 min in HBS. Carrier-mediated uptake was determined by quantitating cell-associated radioactivity in the presence of 10 M cytochalasin B (an inhibitor of facilitative glucose transport). Radioactive medium was aspirated rapidly followed by three cell washes in ice-cold isotonic saline solution (0.9% NaCl, w/v) prior to lysis in 0.05 M NaOH (23). Cell-associated radioactivity was determined by liquid scintillation counting and protein determined by the method of Bradford (34). Glycogen Synthesis L6 myotubes were deprived of serum for 5 h in -MEM and washed twice with warm HBS prior to pre-incubation at 37 C with kinase inhibitors (wortmannin, 100 nm, 15 min; PD 98059, 50 M, 15 min; 10 M SB or SB for 30 min). After the appropriate pre-treatment period, cells were washed with HBS and incubated with 100 nm insulin, H 2 O 2, and/or inhibitors for another 30 min at 37 C in HBS containing [U 14 C]-D-glucose (0.1 Ci/ml). The incubation was terminated by three washes with ice-cold 0.9% (w/v) NaCl prior to lysis in 60% (w/v) KOH. Cellular glycogen was precipitated from lysates using a method adapted from that described previously (36), and associated radioactivity was determined by liquid scintillation counting. Protein was determined using the method of Bradford (34). SDS-PAGE and Immunoblotting Cell lysates (30 g of protein) were subjected to SDS-PAGE and immunoblotting as described previously (23). Separated proteins were transferred onto nitrocellulose membranes and blocked with Tris-buffered saline (ph 7.0) containing 5% bovine serum albumin and 0.05% (v/v) Tween 20. Membranes were probed with antibodies against the phosphorylated forms of JNK, p38, PKB, and p42 and p44 MAP kinase or that recognizing native PKB (all at a dilution of 1:1000). Following primary antibody incubation, membranes were incubated with horseradish peroxidase-conjugated antirabbit IgG (1:1000) or anti-mouse IgG (1:500) as appropriate. Immunoreactive protein bands were visualized by ECL on Kodak X-Omat film. Statistical Analysis Statistical analysis was carried out using Student s t test. Data were considered statistically significant at p RESULTS Increased Oxidative Stress Impairs Insulin-stimulated Glucose Transport and Glycogen Synthesis in L6 Muscle Cells To assess whether increased oxidant levels modulate key end point responses to insulin action in skeletal muscle, we investigated the effects of H 2 O 2 upon glucose uptake and glycogen synthesis in L6 myotubes. Insulin stimulated glucose uptake by 2-fold (Fig. 1a) and caused a near 5-fold increase in glycogen synthesis (Fig. 1b). However, when muscle cells were incubated simultaneously with insulin and H 2 O 2 at concentrations of 50 M or above, we observed a dramatic inhibition in the hormonal activation of both processes (Fig. 1, a and b). Insulin-stimulated glucose transport was suppressed by 73% in the presence of 50 M H 2 O 2 (Fig. 1a) and was abolished completely in the presence of 1 mm H 2 O 2. This inhibition could not be explained by changes in the basal rate of glucose trans-

3 SAPK2/p38 and Insulin Signaling FIG. 1. Effects of insulin and H 2 O 2 on glucose transport and glycogen synthesis in L6 myotubes. a, L6 cells were exposed to 100 nm insulin and/or H 2 O 2 ( M) for 30 min prior to assaying glucose uptake. Glycogen synthesis (b) was assayed after treating cells with 100 nm insulin and/or 50 M H 2 O 2 for 30 min. c, in some experiments ( / ) L6 myotubes were pre-incubated for 30 min with 100 nm insulin and 1 mm H 2 O 2. They were then washed three times with HBS prior to re-stimulation for 30 min with 100 nm insulin and assaying glucose uptake. Values are the mean S.E. for up to six experiments, each performed in triplicate. Asterisks indicate a significant change from the appropriate control value (p 0.05). port, which was unaffected in cells incubated with 1 mm H 2 O 2 (Fig. 1a). Interestingly, the loss in insulin-stimulated glucose transport elicited by H 2 O 2 was reversed when cells pre-treated with 1 mm H 2 O 2 and insulin were washed with saline and subsequently subjected to a second round of insulin stimulation (for 30 min), but in the absence of H 2 O 2 (Fig. 1c). This latter finding suggests that the inhibition in insulin action may be mediated via rapid changes in the activity of key molecules involved in insulin signaling. Protein Kinases Modulated by Oxidative Stress To determine how H 2 O 2 may have inhibited two different cellular responses to insulin, we investigated its effects on the protein kinases JNK, p38 MAP kinase, p42/p44 MAP kinases, and PKB. Immunoblot analysis using phospho-specific antibodies indicated that H 2 O 2 induced a dose-dependent activation of all five protein kinases (Fig. 2a). Interestingly, the H 2 O 2 -induced phosphorylation of PKB correlated very closely with the in vitro activity of the kinase (Fig. 2b). PKB was activated by H 2 O 2 concentrations above 100 M and was stimulated by 27-fold when muscle cells were incubated with 1 mm H 2 O 2 (Fig. 2b). As indicated above, both JNK and p38 MAP kinase were phosphorylated by H 2 O 2 treatment of cells (Fig. 2a). However, when L6 myotubes were pre-incubated with either SB or its sister compound SB (37), phosphorylation of p38 MAP kinase, but not of JNK, was prevented (Fig. 2c). Consistent with this finding, neither compound inhibited the H 2 O 2 -induced activation of JNK (Fig. 2d) in vitro, whereas that of MAPKAP-K2 (the downstream target of p38 MAP kinase) was reduced significantly (Fig. 2e). The 12-fold activation elicited in MAPKAP-K2 activity by H 2 O 2 was not suppressed by wortmannin or PD 98059, which block PI3K and the classical MAP kinase pathway, respectively (Fig. 2e). Although PD was ineffective in suppressing the H 2 O 2 -induced activation of the p38/mapkap-k2 pathway (Fig. 2e), it did inhibit the phosphorylation of p42/p44 MAP kinases. This suggests that activation of the MAP kinase pathway occurred at the level of, or upstream of, MEK (data not shown). The observation that PKB was phosphorylated upon H 2 O 2 treatment at concentrations above 50 M indicates that, despite inhibiting insulin-stimulated glucose transport and glycogen synthesis (Fig. 1), this stress agent can exert insulin-mimetic effects on proteins involved in proximal insulin signaling. To establish the mechanism underpinning this effect, we immunoprecipitated PKB and PKB from L6 myotubes following incubation with insulin and/or H 2 O 2 in the absence or presence of wortmannin and/or SB Insulin stimulated the activity of PKB and PKB by 27- and 18-fold, respectively, and both isoforms were also activated to similar levels following incubation of L6 muscle cells with 1 mm H 2 O 2 (Fig. 3, a and b). Both the insulin- and H 2 O 2 -induced activation of PKB and PKB were inhibited by wortmannin but not by SB (or SB ), suggesting that both PKB isoforms were stimulated in a PI3K-dependent fashion, and that the p38 pathway was not involved in their activation. One of the known physiological targets of PKB is GSK3 (20). Since H 2 O 2 caused activation of both PKB and PKB, we determined whether there was an associated inactivation of GSK3. Fig. 4 shows that both insulin and H 2 O 2, either alone or in combination, induced a significant inhibition of GSK3 activity. This inactivation was suppressed by wortmannin, consistent with our observation that both insulin and H 2 O 2 stimulate the upstream inactivator of GSK3 (i.e. PKB) in a PI3K-dependent fashion (Fig. 3). Effects of H 2 O 2 on the Hormonal Activation of Glycogen Synthase Since insulin and H 2 O 2 both activate PKB and inhibit GSK3, we assessed their effects on glycogen synthase activity. Treatment of L6 myotubes with insulin caused a 6.5-fold increase in glycogen synthase activity, which was prevented by pre-treatment of muscle cells with wortmannin (Fig. 5). However, despite the observed effects of H 2 O 2 on cellular PKB and GSK3 activity, 1 mm H 2 O 2 failed to elicit any stimulation in glycogen synthase activity. Moreover, when muscle cells were incubated simultaneously with insulin and 1 mm H 2 O 2, the stimulatory effect of insulin was abolished (Fig. 5). This loss in activation cannot be explained by an effect of H 2 O 2 on the insulin that was delivered to these cells, given that the hormone was capable of phosphorylating PKB in the presence of 50 M H 2 O 2 (Fig. 5b), a concentration that has no effect on PKB phosphorylation per se, but which nonetheless causes a profound inhibition in insulin-stimulated glucose transport (Fig. 1a) and glycogen synthesis (Fig. 1b). The loss in hormonal activation of glycogen synthase was also evident at this lower (50 M) H 2 O 2 concentration (data not shown). Inhibition of p38 MAP Kinase Suppresses the H 2 O 2 -induced

4 36296 SAPK2/p38 and Insulin Signaling FIG. 2.Effect of insulin and H 2 O 2 on JNK, p42/p44 MAP kinases, SAPK2/P38 MAP kinase, and PKB in L6 myotubes. a, L6 myotubes were treated for 10 min with 100 nm insulin or H 2 O 2 at the concentration indicated prior to cell lysis and immunoblotting. Panel shows representative immunoblots showing dose-dependent activation of JNK, SAPK2/p38, p44/42 MAP kinases, and PKB-Ser 473 using phospho-specific antibodies. Equal loading of cell lysate protein was confirmed by use of an antibody that recognized the C-terminal domain of native PKB. b, dose-dependent activation of PKB assayed using crosstide. c, muscle cells were pre-treated for 30 min with 10 M SB (SB a )or10 M SB (SB b ) prior to stimulation with 1 mm H 2 O 2 for 10 min. Cells were then lysed and immunoblotted using phospho-specific antibodies against p38 and JNK. d, JNK was immunoprecipitated from lysates used in c and in vitro kinase activity determined using ATF-2. e, for analysis of MAPKAP-K2 activity, muscle cells were pre-treated for 15 min with PD (50 M) or wortmannin (100 nm) or for 30 min with SB (10 M) prior to stimulation with 1 mm H 2 O 2 (10 min) and kinase assay. Values are the mean S.E. for three to six experiments each carried out in triplicate. Asterisks indicate statistically significant changes from the appropriate control value (p 0.05). Loss in Insulin-stimulated Glucose Transport Given that H 2 O 2 can promote a loss in insulin sensitivity at low concentrations (50 M) and that only p38 MAP kinase was activated at this concentration, we hypothesized that this stress pathway may play a role in modulating insulin action. To test this hypothesis we performed two separate experiments. The first was based on our observation that removal of H 2 O 2 from the extracellular bathing solution reinstated insulin sensitivity within 30 min of washing the cells free of the stress agent (Fig. 1c). Since it is plausible that this restoration was as a result of the concurrent inactivation of the stress kinase, we determined whether removal of H 2 O 2 from the bathing solution resulted in a change in phosphorylation of p38 MAP kinase, as well as that of JNK. Fig. 6a shows that p38 phosphorylation decreased rapidly in L6 myotubes during the post-wash period and was virtually undetectable after 20 min. In contrast, phosphorylation of JNK was sustained at a significant level over the same period of time (Fig. 6a), and JNK remains active in in vitro assays under these circumstances (data not shown). Thus, phosphorylation of p38 MAP kinase (but not that of JNK) correlates inversely with the ability of insulin to stimulate glucose uptake following removal of H 2 O 2. The second experimental approach assessed the effects of the pyridinyl imidazole compounds SB and SB on insulin-stimulated glucose transport and glycogen synthesis in the presence of H 2 O 2. Consistent with the data shown in Fig. 1, H 2 O 2 caused a loss in the hormonal activation of both processes (Fig. 6, b and c). Strikingly however, when muscle cells were pre-exposed to either 10 M SB or 10 M SB , a concentration that blocks the activation of p38 and MAP- KAP-K2 by H 2 O 2 (Fig. 2, c and e), but not that of JNK (Fig. 2, c and d), the observed loss in insulin-stimulated glucose transport was effectively abolished (Fig. 6b). In these experiments we found that both SB compounds reduced basal glucose uptake by up to 30%. However, the net stimulation in glucose influx elicited by insulin in the presence of these inhibitors and H 2 O 2 was similar to that seen in muscle cells exposed to insulin alone (Fig. 6b). We have made very similar observations in isolated rat muscle strips in which 1 mm H 2 O 2 completely blocked insulin-stimulated glucose uptake and SB largely prevented this effect (data not shown). In contrast to their effects on glucose transport, pre-incubation of muscle cells with either SB or SB was unable to prevent the inhibition by H 2 O 2 of insulin-stimulated glycogen synthesis (Fig. 6c). In these experiments, neither wortmannin nor PD was capable of suppressing the inhibition in the hormonal activation of glucose uptake or glycogen synthesis induced by H 2 O 2 (Fig. 6, b and c). Effects of SB and Wortmannin on Insulin and H 2 O 2 - regulated Glucose Transport We postulated that if the H 2 O 2 - induced activation of the p38 pathway was exerting an inhibitory effect on the insulin signaling pathway, then using SB to suppress this activation should enable H 2 O 2 to activate glucose transport independently of insulin in a wortmannin-sensitive fashion. Indeed, in the presence of 1 mm H 2 O 2 and 10 M SB , where activation of the p38 pathway, but not that of PKB, was blocked (Fig. 6d), glucose uptake was stimulated to a level comparable to that seen by insulin (Fig. 6b). This H 2 O 2 -induced stimulation in glucose uptake was inhibited by wortmannin under conditions when activation of the p38 pathway was inhibited by SB (Fig. 6b). DISCUSSION The present study has shown that subjecting muscle cells to acute oxidative stress (using H 2 O 2 ) results in a dramatic loss in insulin-stimulated glucose transport and glycogen synthesis. Interestingly, the loss in insulin sensitivity can be reversed rapidly, suggesting that H 2 O 2 is likely to modulate insulin action by altering the activity of signaling molecules that mediate the biological effects of the hormone. This notion is sup-

5 SAPK2/p38 and Insulin Signaling FIG. 5.Effects of insulin and H 2 O 2 on glycogen synthase and PKB. a, L6 myotubes were pre-incubated for 15 min with 100 nm wortmannin prior to stimulation with 100 nm insulin and/or 1 mm H 2 O 2 (10 min). Glycogen synthase activity was expressed as a ratio of that observed in the absence of the allosteric activator glucose 6-phosphate divided by that in its presence. b, a representative immunoblot of PKB phosphorylation. Muscle cells were incubated for 10 min with 50 M H 2 O 2 and/or 100 nm insulin and cell lysates immunoblotted using an antibody against Ser 473. Values in a are the mean S.E. for three to six experiments each carried out in triplicate. Asterisks indicate statistically significant changes from the appropriate control value (p 0.05). FIG. 3.Effects of insulin and H 2 O 2 on protein kinase B (PKB) activity. L6 myotubes were pre-incubated for 15 min with wortmannin (100 nm) or for 30 min with SB (10 M) prior to stimulation with 100 nm insulin and/or 1 mm H 2 O 2 (10 min). The upper panel of a shows the same cell lysates used for in vitro PKB activity analyses, immunoblotted with a phospho-specific antibody to PKB-Ser 473. In vitro kinase analysis of PKB (a) and PKB (b) were performed by assessing phosphorylation of crosstide. Values are the mean S.E. for three to six experiments each carried out in triplicate. Asterisks indicate statistically significant changes from the appropriate control value (p 0.05). FIG. 4. Effects of insulin and H 2 O 2 on GSK3 activity. Muscle cells were pre-incubated for 15 min with 100 nm wortmannin prior to stimulation with 100 nm insulin and/or 1 mm H 2 O 2 (10 min) and kinase assay. GSK3 activity was expressed as a re-activation ratio (i.e. GSK3 activity measured without PP2A 1 treatment divided by GSK3 activity after PP2A 1 treatment). Values are the mean S.E. for three to six experiments each carried out in triplicate. Asterisks indicate statistically significant changes from the appropriate control value (p 0.05). ported by recent work showing that fibroblasts and 3T3-L1 adipocytes subjected to oxidative stress show a marked decline in insulin sensitivity that stems from a disruption in the activation of IRS1, PI3K, and PKB (25, 26). However, our findings indicate that such a mechanism is unlikely to explain the inhibition in insulin-stimulated glucose transport and glycogen synthesis in L6 myotubes as both processes were inhibited by H 2 O 2 concentrations (50 M) that had no apparent effect on PKB activation by insulin (Fig. 5b). This latter finding implies that the upstream insulin signaling events are fully functional. Furthermore, at higher (1 mm) concentrations, H 2 O 2 acts as an insulin mimetic, in that it activates both PKB and and inhibits GSK3, a finding that is in line with similar observations reported using HEK 293 cells (27). However, despite the changes in PKB and GSK3 activity that take place in the presence of 1 mm H 2 O 2, insulin fails to elicit any stimulation in glucose uptake or glycogen synthesis, signifying that the loss in insulin action in L6 cells is likely to occur by a mechanism different to that reported in fat cells (26). In addition to its insulin-like effects on PKB and GSK3, H 2 O 2 also activated two separate stress signaling pathways, JNK and p38 MAP kinase. Activation of p38 has been suggested to be important in mediating some of the harmful effects associated with hyperglycaemia in smooth muscle cells (31), and we believe that this pathway may also participate in the regulation of insulin-stimulated glucose transport in L6 myotubes during acute oxidative stress. This tenet is based on three separate lines of evidence. First, the loss in insulin-stimulated glucose transport was elicited at a H 2 O 2 concentration (50 M) that also induced activation of the p38 MAP kinase pathway. At this H 2 O 2 concentration, none of the other kinases assayed showed any detectable changes in phosphorylation and/or activity. Second, we were able to restore insulin-stimulated glucose transport within 30 min of removing H 2 O 2 from the extracellular bathing solution. The ability to recover the insulin response correlated with the rapid dephosphorylation of p38 MAP kinase observed during the post-wash period. Finally, inhibiting the H 2 O 2 -induced activation of p38 MAP kinase, using SB and SB , prevented the loss in insulinstimulated glucose transport caused by H 2 O 2. In contrast, SB and SB failed to inhibit JNK activation, consistent with previous reports showing that, at micromolar concentrations, both compounds target p38 MAP kinase selectively (for review see Ref. 15). Moreover, our finding that both SB

6 36298 SAPK2/p38 and Insulin Signaling FIG. 6.Time course of JNK and p38 dephosphorylation and the effects of SB , SB , PD 98059, and wortmannin on glucose transport and glycogen synthesis. a, L6 myotubes were pre-incubated for 30 min with 1 mm H 2 O 2. Cells were then washed three times with HBS and lysed either immediately or allowed to recover for the times indicated in HBS plus 5 mm D-glucose prior to cell lysis. L6 lysates were immunoblotted using phospho-specific antibodies against SAPK2/p38 and JNK. b, muscle cells were treated for 30 min with 100 nm insulin and/or 1 mm H 2 O 2 having been pre-incubated for 15 min with PD (50 M), wortmannin (100 nm), or 30-min pre-incubation with either SB (10 M) or SB (10 M). Subsequently, 2-deoxyglucose uptake was assayed. c, glycogen synthesis in L6 myotubes was assayed using [ 14 C]-D-glucose after the same pre-treatment regime described for b. The immunoblots shown in d depict the phosphorylation status of both PKB and SAPK2/p38 in lysates prepared from cells incubated under similar conditions to those used for assaying 2-deoxyglucose uptake (b). Values in b and c are the mean S.E. of three to six experiments each carried out in triplicate. Asterisks signify statistically significant changes from the appropriate control value (p 0.05) and SB prevent the H 2 O 2 -induced phosphorylation of p38 MAP kinase is fully consistent with reports showing that these compounds inhibit the agonist-induced phosphorylation and activity of p38 (38, 39). Collectively, the above findings support the view that the oxidant-induced activation of the p38 MAP kinase pathway plays a role in preventing the hormonal stimulation in glucose transport in skeletal muscle cells. The concept that there may be some element of cross-talk between the p38 MAP kinase pathway and the insulin signaling pathway is not without precedent. Evidence exists in the literature showing that insulin can, depending on cell type, stimulate or inhibit the activity of p38 MAP kinase (40, 41), and more recently it was reported that incubation of both 3T3-L1 adipocytes and L6 myotubes with SB inhibited insulin-stimulated glucose transport in these two cell lines (42). In the latter study the inhibitor was purported to induce a reduction in the functional activity of glucose transporters in the plasma membrane rather than blocking their acquisition from intracellular stores in response to insulin. However, we have been unable to observe any inhibition in the insulinstimulated influx of glucose following pre-incubation of muscle cells with either SB (Fig. 6b) or SB (data not shown). The precise reason for this discrepancy remains unclear, but it may result plausibly from differences in experimental design and the duration of cell pre-treatments. Previous work has reported that JNK activation in response to anisomycin (an environmental stress agent) can stimulate glycogen synthase in a manner similar to insulin in skeletal muscle (40). The stimulation in glycogen synthase elicited by anisomycin was attributed to an activation of PP-1 (the phosphatase that dephosphorylates and stimulates glycogen synthase; Ref. 43) and a parallel inactivation of GSK3. In our study, however, although incubation of muscle cells with 1 mm H 2 O 2 caused JNK activation and an inhibition of GSK3, we did not detect any stimulation in glycogen synthase activity or glycogen synthesis. Moreover, in the presence of H 2 O 2, insulin fails to stimulate glycogen synthase. One possible explanation for this is that H 2 O 2, unlike anisomycin, either inhibits or fails to activate PP-1. If so, then increased oxidant levels may also interfere with the hormonal regulation of the phosphatase pathway, by an unknown mechanism. It is highly unlikely that the p38 MAP kinase pathway participates in the regulation of glycogen synthase or PP-1 given that, unlike their effects on glucose transport, neither SB nor SB suppressed the H 2 O 2 -induced loss in insulin-stimulated glycogen synthesis. With one exception (44), work from a number of laboratories, including our own, has suggested that PKB is likely to be an integral component of the insulin signaling pathway regulating glucose transport (21 24). If this is correct, then agents that activate PKB should enhance glucose uptake in a manner similar to insulin. However, no increase in glucose transport was observed (Fig. 1a) under circumstances when both PKB and PKB were activated 34- and 12-fold, respectively (Fig. 3, a and b) byh 2 O 2. We speculated that one explanation for this observation was that, like insulin, the effects of H 2 O 2 might be inhibited downstream of PKB by a mechanism that involved the parallel activation of the p38 MAP kinase pathway. Indeed, when activation of the p38 MAP kinase pathway was inhibited using SB , H 2 O 2 stimulated glucose transport to a level comparable to that seen in muscle cells treated with insulin. Importantly, the increase in glucose transport elicited by H 2 O 2 was sensitive to wortmannin, consistent with the observation that H 2 O 2 activates PKB in L6 myotubes in a PI3K-dependent fashion. These findings therefore support the view that PKB is a key component of the insulin-signaling pathway regulating glucose transport. Taken together, our findings indicate the existence of an important biochemical link between the p38 MAP kinase pathway and the insulin-signaling cascade that regulates glucose transport in skeletal muscle cells. Activation of the former in

7 SAPK2/p38 and Insulin Signaling response to oxidative stress modulates the hormonal regulation of glucose transport at a point downstream of PKB. Such a mechanism may be significant physiologically and might help explain the reduced insulin sensitivity that is often observed after bouts of strenuous muscle exercise (45, 46), during which there is increased production of reactive oxygen species (for review, see Ref. 47) and a stimulation of the p38 MAP kinase pathway (48, 49). Precisely how activation of the p38 pathway modulates insulin-stimulated glucose transport remains poorly understood, but possible downstream effectors include: MAP- KAP-K2/K3 (50, 51), p38-regulated/activated kinase (PRAK) (52), MAP kinase interacting kinases (MNK1/2) (53), and the mitogen- and stress-activated kinase (MSK1) (54). However, we believe that MSK1 is unlikely to be involved, based on preliminary data showing that the protein kinase C inhibitor Ro , which also inhibits strongly the stress-mediated activation of MSK1 but not that of MAPKAP-K2 (54), fails to halt the H 2 O 2 -induced loss in insulin-stimulated glucose uptake (data not shown). No inhibitors exist currently that inactivate MAPKAP-K2/K3, MNK1/2, or PRAK specifically, but assessing their potential role as modulators of insulin action during oxidant stress remain important issues for future study. Acknowledgment We are grateful to Sir Philip Cohen (MRC Protein Phosphorylation Unit, University of Dundee) for providing reagents and antibodies for analyses of PKB, GSK3, and MAPKAP-K2. REFERENCES 1. Kahn, B. B. (1996) Diabetes 45, Holman, G. D., and Cushman, S. W. (1994) BioEssays 11, Galuska, D., Ryder, J., Kawano, Y., Charron, M. J., and Zierath, J. R. (1998) Adv. Exp. Med. Biol. 441, DeFronzo, R. A., Bonadonna, R. C., and Ferrannini, E. (1992) Diabetes Care 15, Holman, G. D., and Kasuga, M. (1997) Diabetologia 40, DeFronzo, R. A., and Ferrannini, E. (1991) Diabetes Care 14, Ferrari, P., and Weidmann, P. (1990) J. Hypertens. 8, Baynes, J. W., and Thorpe, S. R. (1999) Diabetes 48, Parik, T., Allikmets, K., Teesalu, R., and Zilmer, M. (1996) J. Cardiovasc. Risk 3, Duranteau, J., Chandel, N. S., Kulisz, A., Shao, Z., and Schumacker, P. T. (1998) J. Biol. Chem. 273, Clerk, A., Fuller, S. J., Michael, A., and Sugden, P. H. (1998) J. Biol. Chem. 273, Taylor, D. E., and Piantadosi, C. A. (1995) J. Crit. Care 10, Kozlovsky, N., Rudich, A., Potashnik, R., Ebina, Y., Murakami, T., and Bashan, N. (1997) J. Biol. Chem. 272, Rudich, A., Kozlovsky, N., Potashnik, R., and Bashan, N. (1997) Am. J. Physiol. 35, E935 E Cohen, P. (1997) Trends Cell Biol. 7, Shepherd, P. R., Withers, D. J., and Siddle, K. (1998) Biochem. J. 333, Marte, B. M., and Downward, J. (1997) Trends Biochem. Sci. 22, Kolesnick, R. N., and Kronke, M. (1998) Annu. Rev. Physiol. 60, Cohen, P., Alessi, D. R., and Cross, D. A. E. (1997) FEBS Lett. 410, Cross, D. A. E., Alessi, D. R., Cohen, P., Andjelkovic, M., and Hemmings, B. A. (1995) Nature 378, Kohn, A. D., Summers, S. A., Birnbaum, M. J., and Roth, R. A. (1996) J. Biol. Chem. 271, Tanti, J. F., Grillo, S., Gremeaux, T., Coffer, P. J., Vanobberghen, E., and Lemarchandbrustel, Y. (1997) Endocrinology 138, Hajduch, E., Alessi, D. R., Hemmings, B. A., and Hundal, H. S. (1998) Diabetes 47, Wang, Q., Somwar, R., Bilan, P. J., Liu, Z., Jin, J., Woodgett, J. R., and Klip, A. (1999) Mol. Cell. Biol. 19, Hansen, L. L., Ikeda, Y., Olsen, G. S., Busch, A. K., and Mosthaf, L. (1999) J. Biol. Chem. 274, Tirosh, A., Potashnik, R., Bashan, N., and Rudich, A. (1999) J. Biol. Chem. 274, Shaw, M., Cohen, P., and Alessi, D. R. (1998) Biochem. J. 336, Ushio-Fukai, M., Alexander, R. W., Akers, M., Yin, Q., Fujio, Y., Walsh, K., and Griendling, K. K. (1999) J. Biol. Chem. 274, Wang, X., Martindale, J. L., Liu, Y., and Holbrook, N. J. (1998) Biochem. J. 333, Ihara, Y., Toyokuni, S., Uchida, K., Odaka, H., Tanaka, T., Ikeda, H., Hiai, H., Seino, Y., and Yamada, Y. (1999) Diabetes 48, Igarashi, M., Wakasaki, H., Takahara, N., Ishii, H., Jiang, Z. Y., Yamauchi, T., Kuboki, K., Meier, M., Rhodes, C. J., and King, G. L. (1999) J. Clin. Invest. 103, Cross, D. A. E., Alessi, D. R., Vandenheede, J. R., McDowell, H. E., Hundal, H. S., and Cohen, P. (1994) Biochem. J. 303, McDowell, H. E., Eyers, P. A., and Hundal, H. S. (1998) FEBS Lett. 441, Bradford, M. M. (1976) Anal. Biochem. 72, Parker, P. J., Caudwell, F. B., and Cohen, P. (1983) Eur. J. Biochem. 130, Lawrence, J. C. (1977) J. Biol. Chem. 252, Kramer, R. M., Roberts, E. F., Um, S. L., Borsch-Haubold, A. G., Watson, S. P., Fisher, M. J., and Jakubowski, J. A. (1996) J. Biol. Chem. 271, Cohen, P. (1999) Curr. Opin. Chem. Biol. 3, Frantz, B., Klatt, T., Pang, M., Parsons, J., Rolando, A., Williams, H., Tocci, M. J., O Keefe, S. J., and O Neill, E. A. (1998) Biochemistry 37, Moxham, C. M., Tabrizchi, A., Davis, R. J., and Malbon, C. C. (1996) J. Biol. Chem. 271, Heidenreich, K. A., and Kummer, J. L. (1996) J. Biol. Chem. 271, Sweeney, G., Somwar, R., Ramlal, T., Volchuk, A., Ueyama, A., and Klip, A. (1999) J. Biol. Chem. 274, Dent, P., Lavoinne, A., Nakielny, S., Caudwell, F. B., Watt, P., and Cohen, P. (1990) Nature 348, Kitamura, T., Ogawa, W., Sakaue, H., Hino, Y., Kuroda, S., Takata, M., Matsumoto, M., Maeda, T., Konishi, H., Kikkawa, U., Kasuga, M., Lee, A. D., Hansen, P. A., and Holloszy, J. O. (1998) Mol. Cell. Biol. 18, Kirwan, J. P., Hickner, R. C., Yarasheski, K. E., Kohrt, W. M., Wiethop, B. V., and Holloszy, J. O. (1992) J. Appl. Physiol. 72, Asp, S., Daugaard, J. R., and Richter, E. A. (1995) J. Physiol. 482, Sen, C. K. (1995) J. Appl. Physiol. 79, Goodyear, L. J., Chang, P. Y., Sherwood, D. J., Dufresne, S. D., and Moller, D. E. (1996) Am. J. Physiol. 34, E403 E 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. (1998) FASEB J. 12, Rouse, J., Cohen, P., Trigon, S., Morange, M., Alonso-Liamazares, A., Zamanillo, D., Hunt, T., and Nebreda, A. R. (1994) Philos. Trans. R. Soc. Lond. Ser. B 78, Clifton, A. D., Young, P. R., and Cohen, P. (1996) FEBS Lett. 392, New, L., Jiang, Y., Zhao, M., Liu, K., Zhu, W., Flood, L. J., Kato, Y., Parry, G. C., and Han, J. (1998) EMBO J. 17, Waskiewicz, A. J., Flynn, A., Proud, C. G., and Cooper, J. A. (1997) EMBO J. 16, Deak, M., Clifton, A. D., Lucocq, L. M., and Alessi, D. R. (1998) EMBO J. 17,