Impaired Muscle Glycogen Synthase in Type 2

0013-7227/01/$03.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 86(9):4307– 4314 Copyright © 2001 by The Endocrine Society

Impaired Muscle Glycogen Synthase in Type 2 Diabetes Is Associated with Diminished Phosphatidylinositol 3-Kinase Activation SVETLANA E. NIKOULINA, THEODORE P. CIARALDI, LESLIE CARTER, SUNDER MUDALIAR, KYONG SOO PARK, AND ROBERT R. HENRY Veterans Affairs San Diego Healthcare System, San Diego, California 92161; and Department of Medicine, University of California, San Diego, La Jolla, California 92093 Insulin signaling pathways potentially involved in regulation of skeletal muscle glycogen synthase were compared in differentiated human muscle cell cultures from nondiabetic and type 2 diabetic patients. Insulin stimulation of glycogen synthase activity as well as phosphorylation of MAPK, p70 S6 kinase, and protein kinase B (Akt) were blocked by the phosphatidylinositol 3-kinase inhibitors wortmannin (50 nM) and LY294002 (10 ␮M). In contrast to lean and obese nondiabetic subjects, where there were minimal effects (15–20% inhibition), insulin stimulation of glycogen synthase in muscle cultures from diabetic subjects was greatly diminished (⬃75%) by low concentrations of wortmannin (25 nM) or LY294002 (2 ␮M). This increased sensitivity of diabetic muscle to impairment of

S

KELETAL MUSCLE GLYCOGEN synthesis plays an important role in blood glucose homeostasis (1–3). The ability of insulin to increase glucose utilization into muscle involves stimulation of glucose transport and activation of glycogen synthase (GS), the rate-limiting enzyme for glycogen synthesis (4, 5). Both the basal activity of GS and its responsiveness to insulin are impaired in skeletal muscle of type 2 diabetic (T2D) subjects (6 –9) and may play a major role in the development of glucose intolerance and insulin resistance. Of the multitude of signaling events initiated after insulin binding to its receptor, three candidate pathways have been proposed to play an important role in activating GS. These include: the ras/MAPK/p90rsk, phosphatidylinositol 3kinase (PI 3-kinase)/protein kinase B (Akt), and PI 3-kinase/ mTOR/p70 S6 kinase pathways (10, 11–17). In several cell lines, structurally unrelated inhibitors of PI 3-kinase (wortmannin and LY294002) inhibit insulin stimulation of GS (18 –22), suggesting that PI 3-kinase-dependent pathways might mediate this insulin response. Events downstream of PI 3-kinase and the relative importance of these pathways in mediating insulin stimulation of GS seem to be both tissueand species-dependent. For example, constitutively active Akt has been shown to stimulate glycogen synthesis and elevate GS activity in L6 myotubes but not in 3T3L-1 adiAbbreviations: ERK, Extracellular regulated kinase; FV, fractional velocity; GDR, glucose disposal rates; GS, glycogen synthase; GSK3, GS kinase 3; IC50, concentration to attain 50% inhibition; LND, lean nondiabetic; OND, obese nondiabetic; PI 3-kinase, phosphatidylinositol 3-kinase; PP1-G, glycogen-associated protein phosphatase-1; T2D, type 2 diabetic subjects; UDP, uridine diphospho.

insulin-stimulated glycogen synthase activity occurs together with diminished insulin-stimulation (by 40%) of IRS-1-associated phosphatidylinositol 3-kinase activity in the same cells. Protein expression of IRS-1, p85, p110, Akt, p70 S6 kinase, and MAPK were normal in diabetic cells, as was insulin-stimulated phosphorylation of Akt, p70 S6 kinase, and MAPK. These studies indicate that, despite prolonged growth and differentiation of diabetic muscle under normal metabolic culture conditions, defects of insulin-stimulated phosphatidylinositol 3-kinase and glycogen synthase activity in diabetic muscle persist, consistent with intrinsic (rather than acquired) defects of insulin action. (J Clin Endocrinol Metab 86: 4307– 4314, 2001)

pocytes (23). Inhibition of p70 S6 kinase activation blunts insulin stimulation of GS in 3T3-L1 adipocytes (21) and rat skeletal muscle (24) but is ineffective in rat adipocytes (25), PC12 cells (18), skeletal muscle in mice (26), and human myotubes (12). Furthermore, in skeletal muscle of GTG-obese mice, p70 S6 kinase seems to participate in insulin regulation of GS, contrary to the situation in normal mice (26). In light of this tissue and species specificity, it is critical to study impaired insulin signaling and insulin resistance of diabetes in human skeletal muscle. There are several lines of evidence that implicate inherited factors in the impairments of regulation of glycogen synthesis in type 2 diabetes. Abnormal activation of glycogen synthesis by insulin is retained in cultured fibroblasts (27) and skeletal muscle cells from T2D patients (9, 28). The nature of this impairment is still unknown and may involve defects in signaling pathways leading to activation of GS. In support of this conjecture, studies of muscle biopsy samples after in vivo insulin stimulation showed decreased activation of IRS-1associated PI 3-kinase activity in muscle of T2D subjects, in comparison with control nondiabetic subjects (29 –31). As we reported earlier, human skeletal muscle cells, when fully differentiated, display morphological, biochemical, and metabolic properties characteristic of adult skeletal muscle (9, 32). Cells from T2D subjects demonstrate impaired GS activity, in both the absence and presence of insulin (9), that is highly reflective of defects seen in muscle biopsies (33) and thus provides a powerful tool for the investigation of potential intrinsic defects associated with type 2 diabetes, independent of the influence of chronic hyperglycemia, hyperinsulinemia, and other metabolic abnormalities.

4307

4308

The Journal of Clinical Endocrinology & Metabolism, September 2001, 86(9):4307– 4314

To test the hypothesis that the defects in GS activity observed in muscle cultures from T2D subjects could be the result of impaired activation of intermediate steps involved in insulin signal transduction pathways, we compared the effect of inhibitors of the p70 S6 kinase, MAPK, and PI 3kinase pathways on GS activity. In addition, we investigated protein expression and insulin responses of these enzymes. These studies demonstrate that the major portion of insulin’s effect on GS occurs through activation of PI 3-kinase. This activation is impaired in muscle cells from T2D subjects. Persistence of these defects in diabetic muscle cultures under eumetabolic conditions is consistent with intrinsic defects of insulin action. Materials and Methods Materials Glycogen, wortmannin, rapamycin, pepstatin, leupeptin, phenylmethylsulfonylfluoride, and other reagents and chemicals were purchased from Sigma (St. Louis, MO). LY 294002 was from Calbiochem (San Diego, CA). PD098059 was a kind gift from Dr. A. Saltiel, ParkeDavis Pharmaceuticals (Ann Arbor, MI). Antibodies against phosphop44/42 MAPK and phospho-Akt were obtained from New England Biolabs, Inc. (Beverly, MA). Affinity-purified rabbit polyclonal IRS-1, p85, and p70 S6 kinase antibodies were purchased from Upstate Biotechnology, Inc. (Lake Placid, NY), and the anti-Akt and anti-p110 polyclonal antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Sources of other common reagents have been detailed in previous publications (9, 33).

Nikoulina et al. • Insulin Signaling in Human Muscle

rapamycin (50 nm), or PD098059 (50 ␮m) for 30 min, followed by a 30-min (if not otherwise indicated) incubation with 30 nm insulin. Previous studies (9, 28) have established this dose of insulin as that that provides maximal stimulation of metabolic activities in both nondiabetic and diabetic muscle cells. Though human muscle cells also contain IGF-1 receptors, this insulin concentration displaces approximately 1% of specifically bound 125I-IGF-1 (28), making it unlikely that insulin action is mediated by the IGF-1 receptor. Wortmannin is a potent, cell-permeable, and irreversible inhibitor of PI 3-kinase with IC50 ⬵ 5 nm in purified preparations and cell cultures that, at concentrations lower than 100 nm, is selective for PI 3-kinase. LY 294002 is a cell-permeable, potent and specific PI 3-kinase inhibitor (IC50 ⬵ 1.4 ␮m) that acts on the ATP-binding site of the enzyme. PD098059 is a selective and cell-permeable inhibitor of MAP kinase kinase that acts by inhibiting the activation of MAP kinase and subsequent phosphorylation of MAP kinase substrates. Rapamycin is an antifungal immunosuppressant that, acting on mTOR, selectively inhibits the phosphorylation and activation of p70S6 kinase (IC50 ⫽ 50 pm) (35).

GS activity The activity of GS in response to hormone and/or drug treatment was measured as described in detail previously (9). Differentiated human muscle cells were incubated with inhibitory agents and insulin in media containing 5 mm glucose and solubilized as described previously (36). GS activity was assayed at a physiologic substrate concentration (0.3 mm UDP-[14C] glucose) in parallel incubations with 0.1 or 10 mm glucose6-phosphate. GS activity is expressed as nmol UDP-glucose incorporated into glycogen/min⫺1.mg total protein or as fractional velocity (FV; the ratio of activity at 0.1 mm G-6-P/10 mm G-6-P). Previous studies have indicated that the FV, as measured in this assay, is an indicator of the activation state of the enzyme (33, 36). Total cellular protein was determined by the Bradford method (37), with BSA as a standard.

Subjects, human muscle cell culture, and cell treatment Muscle biopsies for skeletal muscle cultures were obtained from 11 healthy lean, 7 obese nondiabetic (OND), and 11 T2D patients. All subjects underwent a 2-h 75-g oral glucose tolerance test. Normal glucose tolerance was defined as a fasting plasma glucose level less than 110 mg/dl and a 2-h glucose level less than 140 mg/dl (34). Insulin action or resistance was determined in each subject, using a 3-h maximal insulin-stimulated hyperinsulinemic (300 mU/m2䡠min) euglycemic (5.0 mm) glucose clamp. Glucose disposal rates (GDR) were measured during the last 30 min of the clamp procedure. The experimental protocol was approved by the Committee on Human Investigation of the University of California, San Diego. Informed written consent was obtained from all subjects after explanation of the protocol. Subject characteristics are summarized in Table 1. The method for skeletal muscle cell isolation and growth has been described in detail previously (28). Cells were grown and differentiated in media containing normal levels of glucose (5 mm) and insulin (⬃20 pm) for 4 – 6 wk from the time of biopsy to actual analysis. All studies were carried out on first-passage differentiated cultures. On the day of assay, differentiated cells were placed in serum-free ␣-MEM media for 1 h before further incubation. Cells were then incubated with varying concentrations of wortmannin (25 or 50 nm), LY294002 (2 or 10 ␮m),

Determination of IRS-1-associated PI 3-kinase activity The assay performed was a modification of that described by Kelly and Ruderman (38). Cells were lysed in buffer containing 20 mm TrisHCl, 145 mm NaCl, 10% glycerol, 5 mm EDTA, 1% Triton X-100, 0.5% NP-40, 1 mm sodium orthovanadate, 10 mm sodium pyrophosphate, 200 ␮m phenylmethylsulfonylfluoride, 1 ␮m leupeptin, 1 ␮m pepstatin, 10 mg/ml aprotinin, pH 7.5. Overnight incubation of lysates (equal amounts of protein) with IRS-1 antibody was followed by 2 h of incubation with protein A-agarose; all steps were performed at 4 C. The immunoprecipitates were washed, and PI 3-kinase activity was measured, as described (39), and quantitated by scanning densitometry. Preliminary studies found that only a small portion of insulin-stimulated PI 3-kinase activity was associated with IRS-2 (⬍10% of that associated with IRS-1), so attention was focused on the component associated with IRS-1.

Measurements of p85, p110, and IRS-1 protein levels Protein levels of IRS-1, and the p85 and p110 subunits of PI 3-kinase were measured in the same cell lysates as were used for determination of PI 3-kinase activity. Western blot analysis was performed by the

TABLE 1. Clinical characteristics of experimental subjects Nondiabetic

Age (yr) Body mass index (kg/m2) Glycosylated hemoglobin A1c (%) Fasting plasma glucose (mM) Fasting serum insulin (pM) GDR (mg/kg䡠min)

Lean

Obese

41 ⫾ 3 24.3 ⫾ 0.5 5.6 ⫾ 0.5 5.0 ⫾ 0.1 30.0 ⫾ 6.6 13.4 ⫾ 1.3

43 ⫾ 3 30.1 ⫾ 1.0a 5.1 ⫾ 0.2 4.9 ⫾ 0.1 47.3 ⫾ 9.9a 8.0 ⫾ 0.6a

T2D

47 ⫾ 2 31.3 ⫾ 1.2a 8.1 ⫾ 0.9a,b 8.4 ⫾ 1.0a,b 141.6 ⫾ 30.2a,b 5.4 ⫾ 0.6a,b

Insulin action was determined by a 3-h hyperinsulinemic (300 mU/m2䡠min) euglycemic (5.0 –5.5 mM) clamp. GDR were determined during the last 30 min of the clamp. a P ⬍ 0.05, compared with lean nondiabetic subjects. b P ⬍ 0.05, compared with obese nondiabetic subjects.


The Journal of Clinical Endocrinology & Metabolism, September 2001, 86(9):4307– 4314 4309

method of Burnette (39). After separation of proteins by SDS/PAGE (10%), proteins were transferred to nitrocelluose membranes and blocked to reduce nonspecific binding with TBS containing 5% nonfat dry milk and 0.05% Tween 20. In all cases, the secondary antibody was antirabbit IgG conjugated with horseradish peroxidase. Proteins were visualized with SuperSignal-enhanced chemiluminescence reagent (Pierce Chemical Co., Rockford, IL) and exposed to autoradiograph film (Hyperfilm, Amersham Pharmacia Biotech, Arlington Heights, IL). The intensity of the specific protein bands was quantified by densitometry (NIH Image).

Determination of MAPK, p70 S6 kinase, and Akt phosphorylation Changes in the phosphorylation state of MAPK, in response to insulin and drug treatments, were detected using phospho-specific antibodies that selectively interact with two different phosphorylated forms of MAPK: extracellular regulated kinase (ERK)-1 (44 kDa) and ERK-2 (42 kDa) (New England Biolabs, Inc.). Phosphorylation of Akt as a marker of enzyme activation was detected using an antibody raised against a phosphopeptide (Ser473) sequence conserved in Akt (1/2/3) isoforms (New England Biolabs, Inc.). The activation state of p70 S6 kinase was evaluated using antibodies specific to phospho-Thr389 and Ser411Thr421 (New England Biolabs, Inc.). Lysates from human skeletal muscle cells were analyzed by Western blot after separation of proteins by SDS/PAGE (8 –10%). The secondary antibody was antirabbit IgG conjugated with horseradish peroxidase. Proteins were visualized with enhanced chemiluminescence and exposed to autoradiograph film. The intensity of the bands was quantified by densitometry. After blotting with phospho-specific antibodies, membranes were stripped and reprobed with antibodies against the total protein, to ensure equal loading and transfer of protein.

Calculations and statistical methods All data and calculated results are expressed as means ⫾ sem. All calculations on cells were performed using paired controls. Paired t test analysis was performed for comparisons of exposures to different inhibitors and/or insulin in the same sets of cells. Significance was accepted at the P ⬍ 0.05 level. Because of limitations in tissue availability, not all studies were performed on cells from all subjects. The number of individuals involved in each experiment is indicated in the figure legends.

Results In vivo and in vitro insulin action in experimental subjects

Subjects who provided muscle tissue for culture were matched for age. Obese nondiabetic and T2D subjects were also matched for body mass index. Diabetic subjects displayed elevated fasting glucose, insulin, and glycosylated hemoglobin A1c levels (Table 1). Insulin resistance in that group was confirmed by lower GDR, compared with both nondiabetic groups, during a 3-h hyperinsulinemic euglycemic clamp (Table 1). Obese nondiabetic subjects displayed moderate insulin resistance, indicated by intermediate GDR in response to insulin, compared with lean nondiabetic (LND) and T2D subjects (Table 1). The activity of GS in muscle cells, expressed as FV, behaved much like whole-body glucose disposal. Both basal and insulin-stimulated GS FV in cells of diabetic subjects were significantly impaired, compared with the LND group (basal: 0.059 ⫾ 0.008 vs. 0.103 ⫾ 0.008, P ⬍ 0.05; insulinstimulated: 0.109 ⫾ 0.019 vs. 0.183 ⫾ 0.019, P ⬍ 0.05), in agreement with previously published results (9). The insulinstimulated increment in FV (Fig. 1), calculated as the difference in GS FV measured in basal and insulin-stimulated conditions, was reduced by approximately 30 – 40% in cells of diabetic patients (0.084 ⫾ 0.01 vs. 0.049 ⫾ 0.01, nondiabetic and T2D, respectively; P ⬍ 0.05). In OND cells, both basal and insulin-stimulated GS FV were intermediate between LND and diabetic groups (basal: 0.087 ⫾ 0.020; insulin-stimulated: 0.137 ⫾ 0.021), as well as the insulin-stimulated increment in GS FV (0.061 ⫾ 0.015). Effects of PI 3-kinase inhibition on GS activity

Treatment of cells, from nondiabetic and diabetic patients, with 50 nm wortmannin did not affect basal total activity of GS (measured at a saturating concentration of the allosteric activator glucose-6-phosphate) in either group of subjects (2.63 ⫾ 0.37 nmol/mg䡠min vs. 2.49 ⫾ 0.30 in LND, 3.12 ⫾ 0.41 vs. 3.02 ⫾ 0.42 in OND, and 2.73 ⫾ 0.31 vs. 2.64 ⫾ 0.33 in

FIG. 1. Effect of wortmannin on the activation of GS by insulin in muscle cultures. Skeletal muscle cultures obtained from LND, obese diabetic, and T2D subjects were incubated in the absence or presence of 30 nM insulin for 30 min (C-control). Wortmannin [25 nM (W25) or 50 nM (W50)] was added 30 min before insulin. Extracts were prepared, and GS FV were measured, as described in Materials and Methods. Results are expressed as increments in GS FV after insulin stimulation over basal for each individual, mean ⫾ SEM. A, Independent experiments in duplicate were performed in cultures from eight LND patients; B, results from muscle cells of six OND subjects; C, results from cells of nine T2D subjects. *, P ⬍ 0.01, compared with paired control value; †, P ⬍ 0.05, compared with the same condition in LND cultures.

4310


diabetic groups). There was also no effect of wortmannin on total activity of GS after insulin stimulation. The same treatment caused a significant reduction of basal activity of GS, measured at 0.1 mm G-6-P, reflected in a reduction of GS FV in all three groups (diabetic: 0.042 ⫾ 0.011; OND: 0.047 ⫾ 0.012; LND: 0.068 ⫾ 0.011). This concentration of wortmannin inhibited insulin stimulation of GS FV equally in nondiabetic (71% and 80% inhibition of insulin effect in lean and obese subjects, respectively) and diabetic (82%) muscle cells (Fig. 1). These results indicate that wortmannin does not affect GS activity directly but rather inhibits the activation state of the enzyme by impairing signal transduction through inhibition of PI 3-kinase. Treatment of cells with a second, structurally unrelated inhibitor of PI 3-kinase inhibitor, LY 294002 (10 ␮m), resulted in inhibition of insulin activation of GS similar (80 –90%) to that caused by wortmannin. Incubation of muscle cells with a lower (25 nm) wortmannin dose only modestly decreased basal or insulin-stimulated GS activity in the LND or obese groups (0.147 ⫾ 0.027 and 0.16 ⫾ 0.015, respectively) but significantly diminished insulin-stimulated GS FV in T2D cells (0.068 ⫾ 0.010). There was also a significantly greater reduction of the insulin increment of synthase FV in muscle cells of diabetic (71 ⫾ 11%), compared with lean and OND subjects (24 ⫾ 7 and 33 ⫾ 14, Fig. 1). Impaired insulin activation of PI 3-kinase but normal protein expression in muscle cells of T2D subjects

To investigate whether differences in the effects of wormannin and LY 294002 between diabetic and nondiabetic cells could be attributable to differences in activities and expression of PI 3-kinase proteins, we measured IRS-1associated PI 3-kinase activity and protein abundance of the p85␣- and p110␤-subunits of PI 3-kinase. Incubation of nondiabetic muscle cells, for 30 min, with insulin increased IRS-1-associated PI 3-kinase activity by more than 7-fold (Fig. 2). In cells of T2D subjects, insulin activation of PI 3-kinase activity was significantly diminished (4.3 ⫾ 0.7 vs. 7.2 ⫾ 0.8-fold, compared with LND, P ⬍ 0.05). The response of PI 3-kinase to insulin was intermediate in cells of OND subjects (5.7 ⫾ 0.9-fold, Fig. 2B). Western blot analyses showed that expression of p85␣ and p110␤ proteins was similar in nondiabetic and diabetic cultures (Fig. 3, A and C). Because insulin activation of PI 3kinase can involve binding to IRS-1, we also measured IRS-1 protein levels in the same cultures of nondiabetic and diabetic subjects. There were no significant differences in IRS-1 protein levels between nondiabetic and diabetic subjects (Fig. 3, B and C). Thus, similar to insulin activation of GS, diminished stimulation of PI 3-kinase activity was retained in skeletal muscle cell cultures of T2D subjects. This defect was not associated with impaired expression of proteins involved in its activation. Possible involvement of Akt, MAPK, and p70 S6 kinase in insulin signaling to GS

Because Akt, MAPK, and p70 S6 kinase may all be affected by wortmannin and lie downstream of PI 3-kinase (19, 40, 41), we investigated the activation and expression of these ki-


FIG. 2. IRS-1-associated PI 3-kinase activity in cultured muscle cells of T2D, LND, and OND subjects. A, Representative autoradiogram of PI 3 kinase activity measured in IRS-1 immunoprecipitates from cells incubated without or with insulin (30 nM), as described under Materials and Methods. The position of phosphatidylinositol 3-phosphate (PI3P) is indicated on the left. B, IRS-1-associated PI 3-kinase activity was quantitated in cultures incubated without or with insulin (30 nM) and presented as increase in PI 3-kinase activity over basal after insulin stimulation. Data are mean ⫾ SEM of muscle cultures from nine lean and eight T2D subjects and six cultures from OND subjects. *, P ⬍ 0.05 vs. control cells from lean subjects.

nases in muscle cell cultures of diabetic subjects, in an attempt to reveal intermediate steps of insulin signaling leading to activation of GS. Insulin (30 nm) treatment of muscle cells resulted in the appearance of a phosphorylated form of Akt (Fig. 4A). Insulin-induced phosphorylation of Akt was blocked by preincubation of cells with wortmannin (50 nm) or LY294002 (10 ␮m). Rapamycin and/or PD098059 inhibited neither PI 3kinase activity (not shown) nor Akt phosphorylation (Fig. 4C), suggesting that PI 3-kinase, but not p70 S6 kinase or MAPK, is involved in the insulin signaling pathway leading to Akt phosphorylation. Unlike the situation for IRS-1 associated PI 3-kinase activation, insulin-stimulated phosphorylation of Akt and Akt protein expression were similar in muscle cells from all groups (Fig. 4, A and B). Insulin stimulation of MAPK phosphorylation was transient, with maximum effect attained within 10 min and dissipating by 30 min (not shown). The time course of MAPK activation in muscle cells, indicated by phosphorylation of the enzyme, preceded that of GS, as GS FV was maximal after 30 min of insulin stimulation and remained elevated for at least 1 h. The levels of phosphorylated and total MAPK1 and MAPK2 were similar in cells from nondiabetic and T2D subjects (Fig. 5A) in both the absence and presence of insulin. Treatment of muscle cells with PD098059 (25 ␮m), 30 min before insulin treatment (10 min), totally blocked the insulin



FIG. 3. Protein expression of the p85 and p110 subunits of PI 3kinase and IRS-1 in human muscle cultures. A, Representative immunoblots of p85 regulatory subunit and p110 catalytic subunit p110 of PI 3 kinase. B, IRS-1 protein in myotubes obtained from nondiabetic lean and obese, and T2D subjects. Each lane contains a sample from a different subject. C, Quantitation of the protein abundance of p85, p110, and IRS-1. Results are expressed as mean ⫾ SE for 6 –11 subjects per group.

response, as well as inhibiting basal MAPK phosphorylation (Fig. 5B). Wortmannin treatment (50 nm) also reduced (⬃70% inhibition) insulin stimulation of MAPK phosphorylation. Insulin-stimulated GS FV was reduced (diabetic: 0.077 ⫾ 0.019, nondiabetic obese: 0.086 ⫾ 0.016, LND: 0.133 ⫾ 0.029, all P ⬍ 0.05 vs. control), and the insulin increment was diminished by approximately 30 – 40% after PD098059 treatment in cells of diabetic and nondiabetic subjects (Fig. 6). Insulin (30 nm) incubation generated phosphorylated forms of p70 S6 kinase (Fig. 5C) in a time-dependent manner, with the maximum effect after 30 min of incubation (a time course similar to that of GS activation). Preincubation of cells with 50 nm rapamycin completely prevented the appearance of the phosphorylated p70 S6 kinase (Fig. 5D). This concentration of rapamycin has been previously shown to inhibit (by ⬎60%) insulin stimulation of p70 S6 kinase activity under similar conditions (24, 26, 30). Rapamycin (50 nm) treatment had no influence on insulin-induced MAPK phosphorylation, Akt phosphorylation, or PI 3-kinase activation (not shown). There were no significant differences in insulin-stimulated Thr389 phosphorylation of p70 S6 kinase between diabetic and nondiabetic groups (Fig. 5C). Protein expression of p70

FIG. 4. Insulin stimulation of Akt phosphorylation in muscle cultures from T2D and nondiabetic subjects. A, Representative immunoblots using antibody that recognizes only phosphorylated Ser473 form of Akt isoforms1–3 (p-Akt) and antibody that recognizes total (both nonphosphorylated and phosphorylated) Akt in muscle cells of LND and obese (OND), and diabetic (T2D) subjects incubated ⫾ 30 nM insulin for 30 min. B, Quantitation of p-Akt as a ratio of phosphorylated to total Akt in muscle cultures of six diabetic, six lean, and six OND subjects. C, Representative immunoblot of p-Akt after treatment with indicated inhibitors followed by insulin, as described in Materials and Methods.

S6 kinase was also similar in all subjects (Fig. 5C). Rapamycin treatment was without significant effect on basal GS activity in all groups. Insulin-stimulated GS FV was reduced after incubation with rapamycin in OND subjects (0.110 ⫾ 0.021, P ⬍ 0.01), which resulted in a decrease in the insulin increment of GS FV (Fig. 6). In LND and diabetic subjects, the effect of rapamycin on insulin-stimulated GS was less pronounced (Fig. 6) and not statistically significant (0.152 ⫾ 0.029, P ⫽ 0.08 by paired analysis in LND cells; 0.077 ⫾ 0.019, P ⫽ 0.08 in diabetic cells). Activities of GS measured at 10 mm of G-6-P, an indicator of the fully activated enzyme, were unaffected by treatment in both nondiabetic (2.67 ⫾ 0.25 vs. 2.92 ⫾ 0.26 nmol UDPG/min䡠mg protein, P ⫽ not significant)

4312



FIG. 6. Effect of selective inhibitors on the activation of GS by insulin in human skeletal muscle cells. Cultured human myotubes were incubated ⫾ 30 nM insulin (Ins) for 30 min. In parallel, some plates of the cultures from the same subject were pretreated for 30 min with 50 nM rapamycin (R); 50 ␮M PD98059 (PD), or 50 nM wortmannin (W). Results are presented as percentage of maximal insulin-stimulated GS FV calculated for each culture separately and expressed as mean ⫾ SEM for cultures from 6 –11 subjects per group. *, P ⬍ 0.05 statistical significance, compared with the value obtained in cells treated by insulin alone. †, P ⬍ 0.05, compared with the same condition in LND cultures.

FIG. 5. Insulin stimulation of MAPK and p70 S6 kinase phosphorylation in muscle cultures of nondiabetic and T2D patients. A, Representative immunoblot of phosphorylated forms of MAPK/ERK1 (44 kDa) and MAPK/ERK2 (42 kDa) ⫾ insulin incubation (30 min). B, Representative immunoblot of p-ERK after treatment with indicated inhibitors followed by insulin, as described in Materials and Methods. C, Representative immunoblots of phosphorylated form of p70 S6 kinase using specific anti-phospho-Thr389 p70 S6K (P-p70 S6K) and total p70 S6 kinase (p70 S6K). The nitrocellulose membrane was initially blotted with anti-P-p79 S6K and then stripped and reprobed with anti-p70 S6K. D, Representative immunoblot of P-p70 S6K after treatment with indicated inhibitors followed by insulin.

and diabetic cultures (2.74 ⫾ 0.18 vs. 2.76 ⫾ 0.18, P ⫽ not significant), indicating that the effects of rapamycin did not involve direct inhibition of synthase activity. Discussion

Both the activity of GS and its responsiveness to insulin have been found to be impaired in skeletal muscle biopsy samples of T2D subjects (8, 33), differences that may play a major role in the development of glucose intolerance and insulin resistance. It is possible that these defects involve both acquired and genetic components, considering that impairments of basal and insulin-stimulated activity of GS are retained in cultured skeletal muscle cells obtained from T2D subjects cultured under normoglycemic and -insulinemic conditions (9).

With regard to possible insulin stimulation of GS, potentially the most significant signaling pathways are the ras/ MAPK/p90rsk, mTOR/p70 S6 kinase, and PI 3-kinase/protein kinase B (Akt) pathways. There is considerable cross-talk and overlap of these pathways, because activity of the ras/ MAPK and mTOR/p70 S6 kinase pathways can be controlled in both a PI 3-kinase-dependent and -independent manner (19, 40 – 43). Evidence has accumulated to support a key role for PI 3-kinase in mediating insulin action on glucose transport and GS (44, 45). Akt has been designated as a crucial downstream effector of PI 3-kinase (13), but the evidence is mixed concerning whether Akt is essential for control of glucose metabolism (46). Further complicating the question of insulin signaling to GS are discrepancies in the literature, suggesting that the relative importance of different pathways may vary with the species and tissue studied (10, 14, 21, 24, 47). Possible cellular specificity of insulin signaling is one reason why it is important to investigate this question in systems closely resembling the behavior of skeletal muscle, the most significant insulin target tissue in humans. We have employed differentiated myotubes, which display signaling pathways and effector systems similar to those that exist in mature skeletal muscle. Studies in human muscle cultures are also relevant to the in vivo situation, because cultured muscle cells from T2D subjects manifest defects in GS activity and insulin action that are highly reflective of the behavior measured in vivo and in muscle biopsies (33). Though attention has properly focused on insulin regulation of GS, it is important to emphasize that synthase activity in the absence of added insulin, the basal state, is also the result of a balance between enzyme phosphorylation and dephosphorylation. In all groups, wortmannin, but not PD098059 or rapamycin, was able to significantly decrease basal FV, indicating that insulin-independent control of GS also involves PI 3-kinase-dependent processes. Because basal



skeletal muscle GS activity is reduced in type 2 diabetes, both in biopsies (8) and cultured myotubes (9), these events involving PI 3-kinase take on added importance. The current data obtained in differentiated human myotubes are also supportive of a central role for PI 3-kinase in insulin stimulation of GS. The primary objective of the current work was to identify which steps in insulin signaling might be altered in diabetic muscle, contributing to impaired insulin action. There were two significant differences observed between muscle cells from T2D subjects and those from nondiabetic subjects: a greater sensitivity to wortmannin for inhibition of insulin signaling to GS (Fig. 1), and impaired insulin stimulation of IRS-1-associated PI 3-kinase activity (Fig. 2). Impairments in insulin-stimulated PI 3kinase activity in diabetic muscle biopsies have also been reported (29, 31). Some of those same subjects characterized in one of those earlier studies (29) provided tissue for muscle cultures analyzed in the current study. Consideration of the in vivo and in vitro data reveals that the impairment of insulin stimulation of PI 3-kinase was not resolved upon culture of diabetic muscle cells in a normal metabolic milieu, suggesting that the step of activation of PI 3-kinase in diabetic muscle cells may be intrinsically defective and could lead to impairments in insulin activation of GS. A greater sensitivity of diabetic muscle to wortmannin, for insulin stimulation of GS FV, could be attributable to defects in protein expression of the components of the PI 3-kinase signaling system. However, expression of key proteins in this signaling pathway, IRS-1, the p85␣ regulatory subunit, and the p110␤ catalytic subunit, did not differ significantly between nondiabetic and diabetic muscle cells, in agreement with normal expression in muscle strips from diabetic subjects (31). Thus, impairment in PI 3-kinase activation occurs despite normal expression of proteins involved in this process, indicating a reduction in the function of one or more of the components of this system. Possible mechanisms for impaired action could include loss of function mutations or posttranslational modification of these proteins. Indeed, mutations have been identified in p85␣ (48) and IRS-1 (49) but not p110␤ (50). Alternatively, the role of compartmentalization in the organization of insulin signaling has been a topic of recent scrutiny (51), and it may be that defective signaling in diabetic muscle involves mistargeting of key proteins. Impaired insulin stimulation of IRS-1-associated PI 3-kinase could also involve more proximal signaling events, such as insulin receptor kinase activity, a question currently under investigation. We (29) and others (52) have shown that, despite defects in PI 3-kinase activation, at physiological insulin levels, insulin stimulation of Akt phosphorylation is essentially normal in muscle from obese T2D subjects, the same behavior as determined in cultured muscle cells. This second report (52) did find a reduction in insulin activation of Akt activity but only at supraphysiological insulin levels. Full activation of Akt, despite impaired PI 3-kinase, could occur through several mechanisms. One would be that, at least in human skeletal muscle, only a portion of available PI 3-kinase activity is required to fully phosphorylate/activate Akt. Another possibility is that the measurement of total Akt phosphorylation, as performed in the current report, may obscure changes in

a smaller pool of Akt directly linked to GS. The fact that insulin stimulation of GS remains impaired in diabetic muscle cells, even in the presence of intact Akt, indicates that other downstream pathways may play a role in signal transduction to GS. In addition to differences observed in diabetic muscle cells, certain of the current findings in differentiated human myotubes differ, at least in a quantitative manner, from results in other systems. For example, we observed that PD098059 is able to block a significant portion (30 – 40%) of the insulin effect on GS (Fig. 6), indicating that, in differentiated human myotubes, MAPK can significantly influence GS activity. Meanwhile, in other systems, MAPK plays little or no role in the metabolic responses to insulin (12, 20, 22, 24). Interestingly, whereas only a minor portion (⬃20%) of the insulin effect on GS may occur through p70 S6 kinase, in muscle cells from the OND group, the effect of rapamycin was significantly greater (⬃40%) than in LND subjects. Thus, this pathway may be relatively more important for insulin activation of GS in muscle of obese subjects and possibly contribute to the insulin insensitivity of GS in obesity. Additionally, our results indicate that the expression and insulin responsiveness for phosphorylation of p70 S6 kinase and MAPK are essentially normal in diabetic muscle cells. In agreement with our results, expression and insulin-induced phosphorylation of MAPK have been found to be normal in isolated skeletal muscle from T2D patients (31). The impact of type 2 diabetes on p70 S6 kinase has not been previously studied in human muscle, especially in vivo, so the current work is a first report that this signaling enzyme is intrinsically intact in diabetic muscle. Our data are in agreement with a proposed model for signal transduction pathways in which stimulation of PI 3-kinase is an early event in insulin signaling leading to activation of at least three regulatory enzymes and pathways: Akt, p70 S6 kinase, and MAPK. The last two are minor pathways leading to GS stimulation, given that complete inhibition of the enzymes or their activation caused only modest inhibition of synthase activation by insulin. The current studies reveal several features of insulin signaling in T2D skeletal muscle that might not be apparent from studies in other systems. The first is that impairments in PI 3-kinase seen in vivo are not reversed in culture and probably represent an intrinsic defect. Additionally, this signaling defect is not a generalized one, because insulin signaling to p70 S6 kinase, MAPK, and Akt are all intact. Whatever the ultimate mechanisms, it seems that insulin resistance in type 2 diabetes involves multiple insulin signal transduction pathways in addition to defects in final effector systems. Acknowledgments Received February 1, 2001. Accepted May 30, 2001. Address all correspondence and requests for reprints to: R. R. Henry, M.D., Veterans Affairs San Diego Healthcare System (9111G), 3350 La Jolla Village Drive, San Diego, California 92161. E-mail: rrhenry@vapop. ucsd.edu. This work was supported by the Medical Research Service, Department of Veterans Affairs, the Veterans San Diego Healthcare System, a Career Development Award from American Diabetes Association (to S.E.N.), by funds from the Whittier Institute for Diabetes, and by Grant M01-RR-00827 from the General Clinical Research Centers program of the National Center for Research Resources, NIH.

4314


References 1. Daniel PM, Love ER, Pratt OE 1975 Insulin-stimulated entry of glucose into muscle in vivo as a major factor in the regulation of blood glucose. J Physiol 247:273–288 2. Baron AD, Brechtel G, Wallace P, Edelman SV 1988 Rates and tissue sites of non-insulin- and insulin-mediated glucose uptake in humans. Am J Physiol 255:E769 –E774 3. Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA, Shulman RG 1990 Quantitation of muscle glycogen synthesis in normal subjects and subjects with non-insulin-dependent diabetes by 13C nuclear magnetic resonance spectroscopy. N Engl J Med 322:223–228 4. Rodnick K, Piper R, Slot J, James D 1992 Interaction of insulin and exercise on glucose transport in muscle. Diabetes Care 15:1679 –1689 5. Manchester J, Skurat A, Roach P, Hauschka SD, Lawrence Jr LC 1996 Increased glycogen accumulation in transgenic mice overexpressing glycogen synthase in skeletal muscle. Proc Natl Acad Sci USA 93:10707–10711 6. Bogardus C, Lillioja S, Stone K, Mott D 1984 Correlation between muscle glycogen synthase activity and in vivo insulin action in man. J Clin Invest 73:1185–1190 7. Bak J, Pedersen O 1994 Glycogen synthase: characteristics and putative role in insulin insensitivity. Diabetes Ann 8:75–105 8. Thorburn AW, Gumbiner B, Bulacan F, Wallace P, Henry RR 1990 Intracellular glucose oxidation and glycogen synthase activity are reduced in noninsulin-dependent (type II) diabetes independent of impaired glucose uptake. J Clin Invest 85:522–529 9. Henry RR, Ciaraldi TP, Abrams-Carter L, Mudaliar S, Park KS, Nikoulina SE 1996 Glycogen synthase activity is reduced in cultured skeletal muscle cells of non-insulin-dependent diabetes mellitus subjects. Biochemical and molecular mechanisms. J Clin Invest 98:1231–1236 10. Begum N 1995 Stimulation of protein phosphatase-1 activity by insulin in rat adipocytes. Evaluation of the role of mitogen-activated protein kinase pathway. J Biol Chem 270:709 –714 11. Cross DA, Watt PW, Shaw M, et al. 1997 Insulin activates protein kinase B, inhibits glycogen synthase kinase-3 and activates glycogen synthase by rapamycin-insensitive pathways in skeletal muscle and adipose tissue. FEBS Lett 406:211–215 12. Halse R, Rochford JJ, McCormack JG, Vandenheede JR, Hemmings BA, Yeaman SJ 1999 Control of glycogen synthesis in cultures human muscle cells. J Biol Chem 274:776 –780 13. Takada M, Ogawa W, Kitamura T, et al. 1999 Requirement for Akt (protein kinase B) in insulin-induced activation of glycogen synthase and phosphorylation of 4E-BP1 (PHAS-1). J Biol Chem 274:20611–20618 14. Dent P, Lavoinne A, Nakienly S, Caudwell F, Watt P, Cohen P 1990 The molecular mechanism by which insulin stimulates glycogen synthesis in mammalian skeletal muscle. Nature 348:302–308 15. Sutherland C, Cohen P 1994 The alpha-isoform of glycogen synthase kinase-3 from rabbit skeletal muscle is inactivated by p70 S6 kinase or MAP kinaseactivated protein kinase-1 in vitro. FEBS Lett 338:37– 42 16. Sturgill T, Ray L, Erikson E, Maller J 1988 Insulin-stimulated MAP-2 kinase phosphorylates and activates ribosomal protein S6 kinase II. Nature 334:715–718 17. Cohen P 1993 Dissection of the protein phosphorylation cascades involved in insulin and growth factor action. Biochem Soc Trans 21:555–567 18. Yamamoto-Honda R, Tobe K, Kaburagi Y, et al. 1995 Upstream mechanisms of glycogen synthase activation by insulin and insulin growth factor-I. Glycogen synthase activation is antagonized by wortmannin or LY294002 but not by rapamycin or by inhibiting p21ras. J Biol Chem 270:2729 –2734 19. Standaert ML, Bandyopadhyay G, Farese RV 1995 Studies with wortmannin suggest a role for phosphatidylinositol 3-kinase in the activation of glycogen synthase and mitogen-activated protein kinase by insulin in rat adipocytes: comparison of insulin and protein kinase C modulators. Biochem Biophys Res Commun 209:1082–1088 20. Moule K, Edgell N, Welsh G, et al. 1995 Multiple signaling pathways involved in stimulation of fatty acid and glycogen synthesis by insulin in rat epididimal fat cells. Biochem J 311:595– 601 21. Shepherd PR, Nave´ BT, Siddle K 1995 Insulin stimulation of glycogen synthesis and glycogen synthase activity is blocked by wortmannin and rapamycin in 3T3–L1 adipocytes: evidence for the involvement of phosphoinositide 3-kinase and p70 ribosomal protein-S6 kinase. Biochem J 305:25–28 22. Hurel SJ, Rochford JJ, Borthwick AC, et al. 1996 Insulin action in cultured human myoblasts: contribution of different signalling pathways to regulation of glycogen synthesis. Biochem J 320:871– 877 23. Ueki K, Yamamoto-Honda R, Kaburagi Y, et al. 1998 Potential role of protein kinase B in insulin-induced glucose transport, glycogen synthesis, and protein synthesis. J Biol Chem 273:5315–5322 24. Azpiazu I, Saltiel AR, DePaoli-Roach AA, Lawrence JC 1996 Regulation of both glycogen synthase and PHAS-I by insulin in rat skeletal muscle involves mitogen-activated protein kinase-independent and rapamycin-sensitive pathways. J Biol Chem 271:5033–5039 25. Lin T-A, Lawrence J 1994 Activation of ribosomal protein S6 kinases does not increase glycogen synthesis or glucose transport in rat adipocytes. J Biol Chem 269:21255–21269


26. Chang PY, Le Marchand-Brustel Y, Cheatham LA, Moller DE 1995 Insulin stimulation of mitogen-activated protein kinase, p90rsk, and p70 S6 kinase in skeletal muscle of normal and insulin-resistant mice. Implications for the regulation of glycogen synthase. J Biol Chem 270:29928 –29935 27. Wells AM, Sutcliffe IC, Johnson AB, Taylor R 1993 Abnormal activation of glycogen synthesis in fibroblasts from NIDDM subjects. Evidence for an abnormality specific to glucose metabolism. Diabetes 42:583–589 28. Henry RR, Abrams L, Nikoulina S, Ciaraldi TP 1995 Insulin action and glucose metabolism in nondiabetic control and NIDDM subjects. Comparison using human skeletal muscle cell cultures. Diabetes 44:936 –946 29. Kim YB, Nikoulina SE, Ciaraldi TP, Henry RR, Kahn BB 1999 Normal insulin-dependent activation of Akt/protein kinase B, with diminished activation of phosphoinositide 3-kinase, in muscle in type 2 diabetes. J Clin Invest 104:733–741 30. Bjornholm M, Kawano Y, Lehtihet M, Zierath JR 1997 Insulin receptor substrate-1 phosphorylation and phosphatidylinositol 3-kinase activity in skeletal muscle from NIDDM subjects after in vivo insulin stimulation. Diabetes 46: 524 –527 31. Krook A, Bjornholm M, Galuska D, et al. 2000 Characterisation of signal transduction and glucose transport in skeletal muscle from type 2 diabetic patients. Diabetes 49:284 –291 32. Ciaraldi TP, Abrams L, Nikoulina S, Mudaliar S, Henry RR 1995 Glucose transport in cultured human skeletal muscle cells. Regulation by insulin and glucose in nondiabetic and non-insulin-dependent diabetes mellitus subjects. J Clin Invest 96:2820 –2827 33. Thorburn AW, Gumbiner B, Bulacan F, Brechtel G, Henry RR 1991 Multiple defects in muscle glycogen synthase activity contribute to reduced glycogen synthesis in non-insulin dependent diabetes mellitus. J Clin Invest 87:489 – 495 34. 1997 Report of the expert committee on the diagnosis and classification of diabetes mellitus Diabetes Care 20:1183–1197 35. Thomas G, Hall MN 1997 TOR signaling and control of cell growth. Curr Opin Cell Biol 9:782–787 36. Nikoulina SE, Ciaraldi TP, Abrams-Carter L, Mudaliar S, Park KS, Henry RR 1997 Regulation of glycogen synthase activity in cultured skeletal muscle cells from subjects with type II diabetes: role of chronic hyperinsulinemia and hyperglycemia. Diabetes 46:1017–1024 37. Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248 –254 38. Kelly KL, Ruderman NB 1993 Insulin-stimulated phosphatidylinositol 3kinase. Association with a 185-kDa tyrosine-phosphorylated protein (IRS-1) and localization in a low density membrane vesicle. J Biol Chem 268:4391– 4398 39. Burnette WN 1981 “Western blotting”: electrophoretic transfer of proteins from sodium dodecyl sulfate–polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal Biochem 112:195–203 40. Burgering B, Coffer P 1995 Protein kinase B (c-Akt) in phosphatydilinositol3-OH kinase signal transduction. Nature 376:599 – 602 41. Shepherd PR, Nave´ BT, Siddle K 1995 Insulin activates glycogen synthase by a novel PI 3-kinase/p70s6k dependent pathway in 3T3–L1 adipocytes. Biochem Soc Trans 23:202S 42. Shepherd P, Nave B, Rincon J, et al. 1997 Involvement of phosphoinositide 3-kinase in insulin stimulation of MAP-kinase and phosphorylation of protein kinase-B in human skeletal muscle: implications for glucose metabolism. Diabetologia 1172–1177 43. Klippel A, Reinhard C, Kavanaugh W, Apell G, Escobedo M, Williams L 1996 Membrane localization of phosphatidylinositol 3-kinase is sufficient to activate multiple signal-transducing kinase pathways. Mol Cell Biol 16:4117– 4127 44. Lawrence Jr JC, Roach PJ 1967 New insights into the role and mechanism of glycogen synthase activation by insulin. Diabetes 46:541–547 45. Tsakiridis T, McDowell H, Walker T, et al. 1995 Multiple roles of Phosphatidylinositol 3-kinase in regulation of glucose transport, amino acid transport, and glucose transporters in L6 skeletal muscle cells. Endocrinology 136:4315– 4322 46. Kohn A, Summers S, Birnbaum M, Roth R 1996 Expression of a constitutively active Akt Ser/Thr kinase in 3T3–L1 adipocytes stimulates glucose uptake and glucose transporter 4 translocation. J Biol Chem 271:31372–31378 47. Moxham CM, Tabrizchi A, Davis RJ, Malbon CC 1996 Jun N-terminal kinase mediates activation of skeletal muscle glycogen synthase by insulin in vivo. J Biol Chem 271:30765–30773 48. Hansen T, Andersen CB, Echwald SM, et al. 1997 Identification of a common amino acid polymorphism in the p85␣ regulatory subunit of phosphatidylinositol 3-kinase. Diabetes 46:494 –501 49. Almind K, Bjorbaek C, Vestergaard H, Hansen T, Echwald S, Pedersen O 1993 Amino acid polymorphisms of insulin receptor substrate-1 in non-insulindependent diabetes mellitus. Lancet 342:828 – 832 50. Kossila M, Sinkovic M, Karkkainen P, et al. 2000 Gene encoding the catalytic subunit p110␤ of human phosphatidylinositol 3-kinase. Diabetes 49:1740 –1743 51. Mastick C, Brady M, Printen J, Ribon V, Saltiel A 1998 Spatial determinants of specificity in insulin action. Mol Cell Biochem 182:65–71 52. Krook A, Roth RA, Jiang XJ, Zierath JR, Wallberg-Henriksson H 1998 Insulin-stimulated Akt kinase activity is reduced in skeletal muscle from NIDDM subjects. Diabetes 47:1281–1286