Phosphatidylinositol 3-Kinase Suppresses Glucose-Stimulated Insulin Secretion by Affecting Post-Cytosolic [Ca 2 ] Elevation Signals Kazuhiro Eto, 1 Tokuyuki Yamashita, 1 Yoshiharu Tsubamoto, 1 Yasuo Terauchi, 1 Kenzo Hirose, 2 Naoto Kubota, 1 Shigeo Yamashita, 1 Junko Taka, 1 Shinobu Satoh, 3 Hisahiko Sekihara, 3 Kazuyuki Tobe, 1 Masamitsu Iino, 2 Mitsuhiko Noda, 1 Satoshi Kimura, 1 and Takashi Kadowaki 1 The role of phosphatidylinositol (PI) 3-kinase in the regulation of pancreatic -cell function was investi- gated. PI 3-kinase activity in p85 regulatory subunit– deficient (p85 / ) islets was decreased to 20% of that in wild-type controls. Insulin content and mass of rough endoplasmic reticula were decreased in -cells from p85 / mice with increased insulin sensitivity. However, p85 / -cells exhibited a marked increase in the insulin secretory response to higher concentra- tions of glucose. When PI 3-kinase in wild-type islets was suppressed by wortmannin or LY294002, the secre- tion was also substantially potentiated. Wortmannin’s potentiating effect was not due to augmentation in glucose metabolism or cytosolic [Ca 2 ] elevation. Re- sults of p85 / islets and wortmannin-treated wild- type islets stimulated with diazoxide and KCl showed that inhibition of PI 3-kinase activity exerted its effect on secretion, at least in part, distal to a cytosolic [Ca 2 ] elevation. These results suggest that PI 3-kinase activ- ity normally plays a crucial role in the suppression of glucose-stimulated insulin secretion. Diabetes 51: 87–97, 2002 D efects in insulin secretion from pancreatic -cells and insulin resistance in the target tis- sues are two major elements that cause type 2 diabetes (1,2). However, diabetes develops only when adequate secretion of the hormone, in proportion to insulin resistance, fails to occur (3– 6). This suggests that preservation of glucose-stimulated insulin secretion is crucial for maintaining glucose homeostasis in the pres- ence of insulin resistance. To trigger insulin secretion, glucose metabolism and subsequent mitochondrial ATP generation are required (7,8). An increase in the cellular ATP-to-ADP ratio closes ATP-sensitive K + (K ATP ) chan- nels, which depolarizes the plasma membrane and evokes Ca 2+ influx into the cytosol. This finally induces exocyto- sis of insulin from secretory vesicles (9,10). Moreover, increased glucose metabolism augments insulin secretion under conditions in which cytosolic [Ca 2+ ] is fixed at higher levels, where glucose metabolism amplifies efficacy of Ca 2+ on exocytosis of insulin granules (11–13). The mediators of this amplifying pathway may include ATP (14,15), guanosine triphosphate (16), long-chain acyl-CoAs (17), and glutamate (18). Phosphatidylinositol (PI) 3-kinase is a key element in transducing various actions of insulin on downstream effectors (19,20) and has been implicated in controlling vesicle trafficking and secretory function in many cell types (21–27). Pancreatic -ells also contain PI 3-kinase activity (28 –30). Its activity is important for stimulating the transcription of the preproinsulin gene and liver-type pyruvate kinase gene (31,32). Insulin is hypothesized to enhance its own gene transcription in an autocrine manner through insulin receptor and PI 3-kinase pathway. How- ever, the role of this enzyme in -cell secretory function remains largely unknown. Involvement of PI 3-kinase in glucose-stimulated insulin secretion has been studied by some investigators using the pharmacological PI 3-kinase inhibitor wortmannin. Thus, 100 nmol/l wortmannin had no effect on insulin secretion induced by basal concentra- tion of glucose in HIT-T15 cells (33), and 50 –100 nmol/l wortmannin potentiated glucose-stimulated insulin secre- tion at higher glucose ranges in MIN6 cells (29); the same concentrations of wortmannin seemed to show no effect on secretion induced by glucose plus carbamyl choline in rat islets or -TC3 cells (30). Very recently, two articles reported wortmannin’s potentiating effect on glucose-stim- ulated insulin secretion at 8 –16.7 mmol/l glucose in rat islets (34,35). However, the mechanism of the potentiation has been elusive and a few experiments only suggested that inositol phosphate accumulation or protein kinase C activity in islets was not correlated with the phenomenon (35). To investigate the role of PI 3-kinase in -cell function, we attempted to analyze islets from class I PI 3-kinase p85 regulatory subunit– deficient (p85 -/- ) mice, a -cell model with genetically decreased PI 3-kinase activity. The mice showed increased systemic insulin sensitivity attrib- utable to upregulation of alternative PI 3-kinase regulatory subunits (36). PI 3-kinase activity in p85 -/- islets was From the 1 Department of Metabolic Diseases, Graduate School of Medicine, University of Tokyo, Tokyo, Japan; the 2 Department of Pharmacology, Grad- uate School of Medicine, University of Tokyo, Tokyo, Japan; and the 3 Third Department of Internal Medicine, Yokohama City University, Yokohama, Japan. Address correspondence and reprint requests to Takashi Kadowaki, Depart- ment of Metabolic Diseases, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail: kadowaki- 3im@h.u-tokyo.ac.jp. Received for publication 5 March 2001 and accepted in revised form 28 September 2001. ER, endoplasmic reticulum; IRS-1, insulin receptor substrate-1; K ATP , ATP- sensitive K + channel; KRB, Krebs-Ringer bicarbonate; MAP, mitogen-activated protein; PDE, phosphodiesterase; PI, phosphatidylinositol. DIABETES, VOL. 51, JANUARY 2002 87