The effect of mosapride (5HT-4 receptor agonist) on insulin sensitivity and GLUT4 translocation

Results After 2 weeks of treatment with mosapride, glucose disposal rates were significantly increased up to those of control (mosapride 5.47 ± 1.72 vs 7.06 ± 2.13, P = 0.004, placebo 5.42 ± 1.85 vs 5.23 ± 1.53 mg kg −1 min −1 ). Fasting plasma glucose (FPG) and insulin levels were decreased. Mosapride increased the contents of GLUT4 in plasma membrane representing the increased recruitment of glucose transporters from intracellular pool. While insulin treatment on human skeletal muscle cell resulted in an increased tyrosine phosphorylation of IRS-1, mosapride did not have any effect. Conclusions Mosapride is effective in decreasing FPG without stimulating insulin secretion in IGT subjects, possibly by inducing GLUT4 translocation in skeletal muscles. Keywords IGT Insulin sensitivity Glucose transport Insulin signaling 1 Introduction Insulin resistance is a key feature of impaired glucose tolerance (IGT) and type 2 diabetes [1] . It is characterized by a diminished ability of insulin-sensitive tissues to take up and metabolize glucose in response to insulin. Skeletal muscle is the primary site of insulin-mediated glucose disposal and contributes significantly to a decreased glucose uptake, as seen in states of insulin resistance [2] . Defects in the early insulin-signaling cascade leading to glucose uptake have been shown to play a key role in the pathogenesis of insulin resistance [3] . In target tissues, such as skeletal muscle, insulin promotes glucose uptake through the translocation of the glucose transporter 4 (GLUT4) from an intracellular vesicular pool to the plasma membrane. Insulin binding to the extracellular α-subunit of its receptor results in the autophosphorylation of tyrosine residues in the receptor β-subunit and activation of tyrosine kinase intrinsic to the β-subunit. This leads to the recruitment and tyrosine phosphorylation of intracellular substrates such as insulin receptor substrates (IRSs) [1–4] . Phosphotyrosines on the IRS proteins bind the p85 regulatory subunit of phosphatidylinositol 3′ (PI3) kinase. PI3-kinase is a heterodimer of a regulatory subunit (p85) and a catalytic subunit (p110), and its activation in response to insulin results primarily through its association with the IRS proteins [4] . Serotonin, also known as 5-hydroxytryptamine, is a neurotransmitter that has been implicated in the regulation of diverse physiological processes, including cellular growth and differentiation [5] , neuronal development [8] , and regulation of blood glucose concentrations [6] . The multiple 5-HT receptor subtypes have been cloned, including 5HT-1, 5HT-2, 5HT-4, 5HT-5, 5HT-6, and 5HT-7 which are members of G-protein-coupled receptors, and 5HT-3, which is a ligand-gated ion channel [7] . Among these subtypes, 5HT-4 subtype is a high affinity receptor located in myenteric plexus of gastrointestinal tract, which accelerates the release of acetylcholines from intestinal cholinergic neurons, contracts the smooth muscle, and thus participates in gastrointestinal motility [7] . However, 5HT-4 receptors have also been detected in skeletal muscles and brain by the RT-PCR analysis [8] . Mosapride citrate is a highly selective 5HT-4 receptor agonist with no affinity for other subtypes of 5-HT receptors except for 5HT-3 receptor, which is weakly antagonized by the metabolite of mosapride. Moreover, it has no affinity for serotonin transporters nor adrenaline α1 , adrenaline α2 or dopamine D2 receptors [8,9] . It is a rapid acting drug, in which the peak plasma concentration ( C max ) is reached 1 h after oral administration, and the plasma levels of the drug reaches a steady state on day 2 after the repeated administration of mosapride three times a day [9] . In terms of adverse effect, unlike cisapride, which acts on serotonin receptors 5HT-1 to -4 and has been associated with fatal arrhythmia, mosapride does not have serious adverse effects except for diarrhea, loose stool, and headache which take place in less than 5% [9,10] . Therefore, mosapride is a widely used prokinetic agent for patients with non-ulcer dyspepsia, diabetic gastroparesis, or reflux esophagitis in Korea [11] . In addition, Ueno et al. have reported that mosapride improves insulin action at skeletal muscles and lowers blood glucose level in patients with type 2 diabetes and that these actions were evident within 10 h after the administration [8,12] , and it has been suggested that their central serotonergic activity mediates the insulin sensitizing action of the mosapride [13] . The aims of this study were to examine the effect of mosapride, 5HT-4 receptor agonist, on insulin sensitivity of subjects with impaired glucose tolerance by euglycemic hyperinsulinemic clamp test and to elucidate the mechanism underlying the insulin sensitizing effect of mosapride using human muscle cell in vitro . 2 Subjects and methods 2.1 Subjects Thirty subjects with impaired glucose tolerance (fasting plasma glucose level of 100–125 mg/dl and/or 2-h plasma glucose level between 141 and 199 mg/dl after standard 75 g oral glucose tolerance test) participated in this study. No subject was taking pharmacological agents that may affect the insulin resistance, such as thiazolidinediones, metformin, and statins or the activity of serotonin and its receptors. None of the subjects had symptoms or signs of gastrointestinal disorders or taking medications related to gastrointestinal motility. Participants were instructed not to alter their body weight or lifestyle habits (eating, drinking, smoking, and exercise) during their participation in the study. It was a single-blind study in which subjects were randomized into two groups after the baseline studies: 20 subjects took mosapride (5 mg t.i.d. daily p.o. Daewoong Pharmacentical, Seoul, Korea), and the remaining 10 subjects took placebo for 2 weeks. Blood samples and clamp test were performed before and after the treatment. Informed consent was obtained from all subjects after explanation of the protocol, and the protocols were approved by the Gangnam Severance Hospital Ethics Committee. 2.2 Anthropometric parameters and biochemical profiles Body weight and height were measured in the morning, without clothing and shoes. BMI was calculated as body weight in kilograms divided by height in meters squared (kg/m 2 ). Serum glucose was measured immediately by an autoanalyzer using the hexokinase method (Roche, Hitachi 747). Serum insulin and c-peptide were determined by an enzyme chemiluminescence immunoassay (ECIA, DPC, Immulite 2000). Serum total cholesterol, HDL-cholesterol and LDL-cholesterol were assessed by the enzymatic methods (Daiichi, Hitachi 747) and serum triglycerides were measured by the enzymatic colorimetric methods (Roche, Hitachi 747). 2.3 Euglycemic hyperinsulinemic clamp test Insulin sensitivity was measured during euglycemic hyperinsulinemic clamp test. After a 10–12 h overnight fast, subjects were admitted to the outpatient clinic at 8:00 A.M. A polyethylene cannula was inserted into an antecubital vein for the infusion of all test substances. A second catheter was inserted retrogradely into an ipsilateral wrist vein on the dorsum of the hand for blood sampling, and the hand was kept in a heated box at 65 °C. Squared priming was performed (0–9 min) with a stepwise decline in the insulin (Humalog, Lilly, U.S.A.) infusion rate every third minute, thereby reducing the insulin infusion rate from 100–80 to 60–40 mU m −2 min −1 . Thereafter, the insulin infusion rate was fixed at 40 mU m −2 min −1 from 9 to 120 min. During the last 30 min of the basal equilibration period, plasma samples were taken at 5–10 min-intervals for the determination of plasma glucose and insulin concentrations. The plasma glucose concentration was measured every 5 min after the start of the insulin infusion, and a variable infusion of 20% glucose was adjusted based on the negative feedback principle to maintain the plasma glucose level at 90 mg/dl with a coefficient of variation <5%. Plasma samples were collected every 15 min from 0 to 90 min and every 5–10 min from 90 to 120 min for the determination of plasma glucose and insulin concentrations. Plasma glucose concentration was maintained constant at euglycemia, using a variable glucose infusion (180 g/l) [14] . Plasma glucose concentration was monitored every 5 min using an automated glucose oxidase method (Glucose Analyzer 2, Beckman Instruments, Fullerton, CA). 2.4 Material and primary human cell culture Human muscle cells were provided by Professor Yoon Ghil Park (Muscular Disease Research Center, Gangnam Severance Hospital, Seoul, Korea) and primary cultured as previously described [15] . Cells were grown at 37 °C, in an incubator containing 5% CO 2 incubator. They were then fused for 4 days in DMEM containing 20% fetal bovine serum, 1% penicillin/streptomycin. In culture, after differentiation, most of the human skeletal muscle myoblasts fused into multinucleated myotubes, as shown in Fig. 1 . Immunostaining was performed with an antibody against human sarcomeric actin, which is expressed in differentiated skeletal muscle. 2.5 GLUT4 in human skeletal muscle Cultured human skeletal muscle cells were incubated for 5 h DMEM containing 25 mM glucose in the absence of serum, which results in low and steady basal glucose uptake rates [16] . Then, 10 nM insulin or 100 nM mosapride was added for 1 h at 37 °C. Subcellular fractionation of human skeletal muscle membranes, i.e. plasma, and intracellular membranes were prepared from muscle cells as described previously [16] . Isolated membrane fractions from human skeletal muscle cell were subjected to SDS-PAGE on 8% resolving gels and immunoblotted as previously reported (30 μg protein, 5% silk milking blocking 1 h) [16] . Protein content of each of the isolated membrane fractions was determined using the BCA protein assay kit. Nitrocellulose membranes were probed with antisera against GLUT4 (1:2000, abcam, MA, U.S.A.). Primary antibody detection was performed using either horseradish peroxidase-conjugated anti-rabbit Ig G (1:2000, SAPU, Scotland) or anti-mouse (1:2000, SAPU, Scotland) for 1 h. 2.6 IRS-1 tyrosine phosphorylation Myotubes that were starved overnight and treated with 100 nM mosapride or stimulated with 17 nM insulin for 3, 10, 15, 30, and 60 min. Cells were lysed, IRS-1 were immunoprecipitated with specific antibodies (1:100, Upstate Biotechnology Inc., Lake Placid, NY, U.S.A.), and then separated by SDS-PAGE. After transfer, membrane was first probed with an anti-phosphotyrosine antibody and then stripped and probed with an anti-IRS-1 antibody (Upstate Biotechnology Inc., Lake Placid, NY, U.S.A.). 2.7 Statistical analyses All statistical analyses were performed using SPSS Win 11.0 (Statistical Package for Social Science, SPSS, Chicago, IL, U.S.A.). All data are presented as means ± SD. Pre- to post-therapy values are compared using a paired t -test, with significance reached at P < 0.05. 3 Results 3.1 In vivo study Clinical characteristics of subjects are given in Table 1 . The subjects were matched for age. Basal blood glucose values, other anthropometric parameters and biochemical profiles were comparable between two groups. Mosapride improved glucose utilization (mosapride 5.47 ± 1.72 vs 7.06 ± 2.13, P = 0.004, placebo 5.42 ± 1.85 vs 5.23 ± 1.53 mg kg −1 min −1 ) assessed by euglycemic hyperinsulinemic clamp test. There was a significant decrease in fasting blood glucose level (115.2 ± 21.3 vs 107.11 ± 15.3 mg/dl, P < 0.005) after 2 weeks of treatment of mosapride ( Table 2 ). There were no adverse effects such as diarrhea, loose stools, or headache in either group during the study period. 3.2 GLUT4 expression In order to elucidate the insulin sensitizing action of 5HT-4 receptor agonist, myotubes differentiated from human muscle cells were used ( Fig. 1 ). The protein expression of GLUT4 was examined using specific antibodies directed against this protein in muscle cells treated with insulin and mosapride. Fig. 2 shows the representative immunoblots demonstrating the induction of GLUT4 expression in the plasma membrane and the reduced expression of GLUT4 in intracellular membrane by both insulin and mosapride. 3.3 IRS-1 tyrosine phosphorylation Then, the effect of mosapride on insulin-signaling pathway, particularly on phosphorylation of IRS was examined. IRS-1 is phosphorylated on tyrosine residues by the insulin receptor after insulin stimulation. Fig. 3 demonstrates a representative anti-phosphotyrosine blot of IRS-1 immunoprecipitates induced by insulin and mosapride at a different time course. There was a substantial increase in the phosphorylation of IRS-1 3 min after exposure to insulin, and the rate decreased thereafter and almost vanished by 60 min. Mosapride also increased the phosphorylation of IRS-1 but it was not associated with the effect of insulin. 4 Discussion The course of type 2 diabetes is slow and metabolic abnormalities that lead to hyperglycemia are established long before the development of overt diabetes [17,18] . This state, in which abnormal glucose metabolism exists with the supranormal range of glucose level below the diagnostic criteria for diabetes, is referred to as pre-diabetes [19] . Insulin resistance is a primary pathophysiology of IGT as well as type 2 diabetes. Skeletal muscle is the major organ for insulin-stimulated glucose disposal [20] , and glucose transport is rate limiting for such disposal [2] . Impaired activation of glucose transport in muscle is a major contributor to insulin resistance in type 2 diabetes and IGT [2] . Defects in the insulin-signaling cascade leading to GLUT4 translocation and glucose uptake play an important role in the pathogenesis of insulin resistance in skeletal muscles [2,21] . Most of the data have been gathered in patients with full-blown type 2 diabetes. However, little is known regarding signaling defects in prediabetic stages such as IGT. The mechanisms underlying defective insulin-stimulated glucose transport in IGT and type 2 diabetes are uncertain. Except for morbid obesity, insulin-sensitive GLUT4 glucose transporter levels in skeletal muscle are not altered [22] and further studies have suggested that there may be defects in insulin signaling and translocation of glucose transporters to the plasma membrane. Previous work has shown that mosapride treatment has a glucose lowering effect with a simultaneous decrease in circulating insulin level in patients with type 2 diabetes [8] . They also have demonstrated that 5HT-4 receptors are expressed in muscle although at a lesser degree than in brain and intestine but higher than in liver and fat tissue by RT-PCR analysis [8] . In the present group of IGT patients after 2 weeks of mosapride treatment, the simultaneous fall in blood glucose and insulin concentration suggest an overall improvement in insulin action. This was confirmed by the increased glucose utilization in the euglycemic hyperinsulinemic clamp test. Such a glucose lowering effect of mosapride without an increase in insulin secretion suggests skeletal muscle to be the potential site of action for mosapride [8] . To test this hypothesis, we performed SDS-PAGE and immunoblotting to determine changes in translocation of GLUT4 from internal membrane to plasma membrane before and after mosapride treatment. To gain some insights into whether components of the insulin-signaling pathway participate in 5HT-4 receptor signaling, we investigated whether 5HT-4 receptor stimulation modulate the phosphorylation status of IRS-1. IRS-1 tyrosine phosphorylation showed a tendency to be lower in the mosapride treated cell. All together, these results suggest that 5HT-4 receptor agonist causes a rapid stimulation in glucose transport that occurs as a result of the increased recruitment of glucose transporters from an intracellular pool to the cell surface, and post-receptor signaling events other than IRS-1 phosphorylation is responsible for eliciting this stimulation. Recent studies suggest that insulin stimulates glucose transport through insulin receptor-mediated tyrosine phosphorylation of IRS-1 or other intermediates that activate PI3K, which, through increases in PI-3,4,5-(PO 4 ) 3 (PIP 3 ), activate downstream effectors protein kinase B (PKB/Akt) [23,24] and atypical protein kinase Cs (aPKCs) ζ and λ/ι [25,26] . Although defects in IRS-1-dependent PI3-kinase activation by insulin in muscle of type 2 diabetic human subjects have been reported [27–29] , information on downstream activators of glucose transport is controversial or lacking. Thus, defective PKB activation was seen during incubation of muscle strips of nonobese type 2 diabetic humans [28] , whereas PKB activation was undiminished in muscle biopsies taken during clamp studies in obese type 2 diabetic humans [29] . Further, it is currently unknown whether aPKC activation is defective in muscles of type 2 diabetic subjects. Concerning IGT, decreased incremental but normal absolute levels of IRS-1-dependent PI3K and no significant reduction in PKB phosphorylation were seen in human muscle in a euglycemic hyperinsulinemic clamp test [30] , suggesting that other factors may contribute more directly to a defective insulin-stimulated glucose disposal. On the other hand, defects in glucose transport and IRS-2-dependent PI3K and aPKC activation were observed in cultured myocytes obtained from obese/impaired glucose tolerant humans [31] . Further study should focus on activation of PI3K (phophoinositide 3-kinase) with treatment of mosapride and possibility that 5HT-4 receptor signaling may converge at some point downstream of PI3K. There are several limitations to this study. First of all, this was not a double-blinded study, where study investigator was aware of the type of medication each patient took. However, since the anthropometric measurements and hyperinsulinemic euglycemic clamp studies were conducted by a technician unaware of the type of medication each patient took, it would not have affected the study results significantly. Secondly, the in vivo and in vitro studies were not conducted on the same subjects. Also, although the participants were told not to change their lifestyle habits including eating, drinking, smoking, and physical activity and maintain their body weight throughout the study, we did not monitor the calorie intake, which could have affected the metabolic parameters and insulin resistance. Moreover, in in vitro study, the expressions of GLUT4 and IRS-1 were not quantified. It would have provided more concrete data. Lastly, it would have been interesting to see the effect of insulin and mosapride on GLUT4 expression and IRS-1 phosphorylation in different glycemic conditions. In summary, these results suggest that 5HT-4 receptor agonist, mosapride, is effective in decreasing plasma glucose concentrations without stimulating insulin secretion in IGT patients by recruiting glucose transporters from an intracellular pool to the cell surface. And, IRS-1 does not seem to be a downstream target for 5HT-4 receptor. Thus, we can speculate that since 5HT-4 receptors are expressed fairly abundantly in brain and the central serotonergic activity has been shown to correlated with insulin sensitivity [13] , mosapride may improve the insulin sensitivity by increasing serotonergic activity. Signaling molecules other than IRS-1 like PI3K, aPKCs or downstream activators might be responsible. Further studies are warranted to elucidate how mosapride stimulates GLUT4 translocation. Conflict of interest statement There are no conflicts of interest. References [1] R.A. De Fronzo Pathogenesis of type 2 (non-insulin dependent) diabetes mellitus: a balanced overview Diabetologia 35 1992 389 497 [2] G.I. Shulman Cellular mechanisms of insulin resistance J. Clin. Invest. 106 2000 171 176 [3] A. Guma J.R. Zierath H. Wallberg-Henriksson A. Klip Insulin induces translocation of GLUT-4 glucose transporters in human skeletal muscle Am. J. Physiol. 268 1995 E613 E622 [4] M.F. White The insulin signaling system and the IRS proteins Diabetologia 40 Suppl. 2 1997 S2 S17 [5] L. Tecott S. Shtrom D. Julius Expression of a serotonin-gated ion channel in embryonic neural and nonneural tissues Mol. Cell. Neurosci. 6 1995 43 55 [6] Y. Fischer J. Thomas J. Kamp E. Jüngling H. Rose Carpéné 5-Hydroxytryptamine stimulates glucose transport in cardiomyocytes via a monoamine oxidase-dependent reaction J. Biol. Chem. 311 1995 575 583 [7] K. Taniyama N. Makimoto A. Furuichi Y. Sakurai-Yamashita Y. Nagase M. Kaibara Functions of peripheral 5-hydroxytryptamin receptors, especially 5-hydroxytryptamine4 receptor, in gastrointestinal motility J. Gastroenterol. 35 2000 575 582 [8] N. Ueno A. Inui A. Asakawa F. Takao Y. Komatsu K. Kotani Mosapride, a HT-4 receptor agonist, improves insulin sensitivity and glycemic control in patients with type 2 diabetes mellitus Diabetologia 45 2002 727 792 [9] M.P. Curran D.M. Robinson Mosapride in gastrointestinal disorders Drugs 68 2008 981 991 [10] F. Potet T. Bouyssou D. Escande Gastrointestinal prokinetic drugs have a different affinity for human cardiac ether-a-gogo K + channel J. Pharmacol. Exp. Ther. 289 2001 1007 1012 [11] Y. Mine T. Yoshikawa S. Oku R. Nagai N. Yoshida K. Hosoki Comparison of effect of mosapride citrate and existing 5-HT4 receptor agonist on gastrointestinal motility in vivo and in vitro J. Pharmacol. Exp. Ther. 283 1997 1000 1008 [12] N. Ueno Y. Satoh Rapid effect of mosapride citrate, 5-HT4 receptor agonist, on fasting blood glucose in Type 2 diabetes patients J. Diabetes Complications 23 2009 255 256 [13] J. Horaceck M. Kuzmiakova C. Hoschl M. Andel R. Bahbohn The relationship between central serotonergic activity and insulin sensitivity in healthy volunteers Psychoneuroendocrinology 24 1999 785 797 [14] R.A. Fronzo J.D. Tobin R. Andres Glucose clamp technique: a method for quantifying insulin secretion and resistance Am. J. Physiol. 237 1979 E214 E223 [15] R.R. Henry L. Abrams S. Nikoulina T.P. Ciaraldi Insulin action and glucose metabolism in nondiabetic control and NIDDM subjects: comparison using human skeletal muscle cell cultures Diabetes 44 1995 936 946 [16] P.J. Bilan Y. Mitsumoto T. Ramlal A. Klip Acute and long-term effects of insulin-like growth factor I on glucose transporters in muscle cells. Translocation and biosynthesis FEBS Lett. 298 1992 285 290 [17] G. Gulli E. Ferrannini M. Stern S. Haffner R.A. DeFronzo The metabolic profile of NIDDM is fully established in glucose-tolerant offspring of two Mexican-American NIDDM parents Diabetes 41 1992 1575 1586 [18] J.F. Gautier C. Wilson C. Weyer D. Mott W.C. Knowler M. Cavaghan Low acute insulin secretory responses in adult offspring of people with early onset type 2 diabetes Diabetes 50 2001 1828 1833 [19] F. Vendrame P.A. Gottlieb Prediabetes: prediction and prevention trials Endocrinol. Metab. Clin. North Am. 33 2004 75 92 [20] R.A. DeFronzo The triumvirate: β-cell, muscle, liver Diabetes 37 1988 667 687 [21] W.T Garvey L. Maianu J.H. Zhu G. Brechtel-Hook P. Wallace A.D. Baron Evidence for defects in the trafficking and translocation of GLUT4 glucose transporters in skeletal muscle as a cause of human insulin resistance J. Clin. Invest. 101 1998 2377 2386 [22] B.B. Kahn Facilitative glucose transporters: regulatory mechanisms and dysregulation in diabetes J. Clin. Invest. 89 1992 1367 1374 [23] M.M. Hill S.F. Clark D.F. Tucker M.J. Birnbaum D.E. James S.L. Macaulay A role for protein kinase β/Akt 2 in insulin-stimulated Glut4 translocation in adipocytes Mol. Cell. Biol. 19 1999 7771 7781 [24] A.D. Kohn S.A. Summers M.J. Birnbaum R.A. Roth 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 1996 31372 31378 [25] G. Bandyopadhyay M.L. Standaert L. Zhao B. Yu A. Avignon L. Galloway Activation of protein kinase C (α,β,ζ) by insulin in 3T3/L1 cells: transfection studies suggest a role for PKC-ζ in glucose transport J. Biol. Chem. 272 1997 2551 2558 [26] G. Bandyopadhyay M.L. Standaert U. Kikkawa Y. Ono J. Moscat R.V. Farese Effects of transiently expressed atypical (ζ,λ), conventional (α,β) and novel (θ,ι) protein kinase C isoforms on insulin-stimulated translocation of opitope-tagged Glut4 glucose transporters in rat adipocytes: specific interchangeable effects of protein kinases C-ζ and C-λ J. Biol. Chem. 337 1999 461 470 [27] M. Bjornholm Y. Kawano M. Lehtihet J.R. Zierath Insulin receptor substrate-1 phosphorylation and phosphatidylinositol 3-kinase activity in skeletal muscle from NIDDM subjects after in vivo insulin stimulation Diabetes 46 1997 524 527 [28] A. Krook R.A. Roth X.J. Jiang J.R. Zierath H. Wallberg-Henriksson Insulin-stimulated Akt kinase activity is reduced in skeletal muscle from NIDDM subjects Diabetes 47 1998 1281 1286 [29] Y.B. Kim S.E. Nikoulina R.P. Ciaraldi R.R. Henry B.B. Kahn 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 1999 733 741 [30] H. Storgaard X.M. Song C.B. Jensen S. Madsbad M. Björnholm A. Vaag Insulin signal transduction in skeletal muscle from glucose-intolerant relatives with type 2 diabetes Diabetes 50 2001 2770 2778 [31] P. Vollenweider B. Menard P. Nicod Insulin resistance, defective insulin receptor substrate 2-associated phosphatidylinositol-3′ kinase activation, and impaired atypical protein kinasde C (ζ/λ) activation in myotubes from obese patients with impaired glucose tolerance Diabetes 51 2002 1052 1059