The roles of the alpha- and beta-isoforms of phosphatidylinositol (PI) 3'-kinase p85 regulatory subunit were studied with isoform-specific antisera in three model systems in which the insulin receptor mediates rapid phosphorylation of insulin receptor substrate-1 (IRS-1). Insulin receptor signaling stimulated the association of IRS-1 with p85 alpha protein, and p85 alpha-associated PI 3-kinase activity in 3T3-L1 adipocytes, and in transfected Chinese hamster ovary cells (CHO-T) and COS-1 cells expressing high levels of human insulin receptors. While not detectable in 3T3-L1 adipocytes, the p85 beta isoform was also found to associate with IRS-1 in response to insulin receptor activation in COS-1 and CHO-T cells. However, selective immunoprecipitation of p85 beta from unstimulated COS-1 or CHO-T cell lysates was accompanied by higher levels of PI 3-kinase activity than that associated with p85 alpha. Remarkably, the large stimulation of PI 3-kinase activity associated with p85 alpha (7.8 +/- 2.0-fold, n = 6) in insulin-treated CHO-T cells was not observed in p85 beta immunoprecipitates (1.8 +/- 0.6-fold, n = 6), and in COS-1 cells p85 beta-associated PI 3-kinase activity was completely insensitive to stimulation by the insulin receptor. These data suggest the novel hypothesis that binding of p85 beta to IRS-1 complexes in COS-1 and CHO-T cells does not mediate marked activation of PI 3-kinase activity as does p85 alpha.

Insulin is thought to exert its effects on cellular function through the phosphorylation or dephosphorylation of specific regulatory substrates. We have analyzed the effects of okadaic acid, a potent inhibitor of type 1 and 2A protein phosphatases, on the ability of insulin to stimulate glucose transport in rat adipocytes. Insulin and okadaic acid caused a 20-25- and a 3-6-fold increase, respectively, in the rate of 2-deoxyglucose accumulation by adipose cells. When added to cells previously treated with okadaic acid, insulin failed to stimulate 2-deoxyglucose accumulation beyond the levels observed with okadaic acid alone. Treatment of cells with okadaic acid did not inhibit the effect of insulin to stimulate tyrosine autophosphorylation of its receptor. These results indicate that okadaic acid potently inhibits the effects of insulin to stimulate glucose uptake and/or utilization at a step after receptor activation. To clarify the mechanism of inhibition by okadaic acid, the intrinsic activity of the plasma membrane glucose transporters was analyzed by measuring the rate of uptake of 3-O-methylglucose by adipose cells, and the concentration of adipocyte/skeletal muscle isoform of the glucose transporter (GLUT-4) in plasma membranes isolated from these cells. Insulin caused a 15-20-fold stimulation of 3-O-methylglucose uptake and a 2-3-fold increase in the levels of GLUT-4 detected by immunoblotting of isolated plasma membranes; okadaic acid caused a 2-fold increase in 3-O-methylglucose uptake, and a 1.5-fold increase in plasma membrane GLUT-4. Pretreatment of cells with okadaic acid blocked the effect of insulin to stimulate 3-O-methylglucose uptake and to increase the plasma membrane concentration of GLUT-4 beyond the levels observed with okadaic acid alone. These results indicate that the effect of okadaic acid to inhibit the effect of insulin on glucose uptake is exerted at a step prior to the recruitment of glucose transporters to the cell surface, and suggest that a phosphatase activity may be critical for this process.

Insulin action leads to the rapid stimulation of a cytosolic Kemptide (Leu-Arg-Arg-Ala-Ser-Leu-Gly) kinase (KIK) that has been recently purified to near homogeneity (Klarlund, J. K., Bradford, A. P., Milla, M. G., and Czech, M. P. (1990) J. Biol. Chem. 265, 227-234). To examine its activation mechanism, purified KIK was treated with purified protein phosphatases. The catalytic subunit of phosphatase 2A inhibited the activity of control KIK by about 50% and abolished the 5-fold elevation in KIK activity due to insulin action. The catalytic subunit of phosphatase 1 with equivalent activity based on dephosphorylation of 32P-labeled phosphorylase alpha had no effect on either control or insulin-stimulated KIK activity. The deactivation of insulin-stimulated KIK by phosphatase 2A was time- and concentration-dependent and was blocked by phosphatase inhibitors. The purified native complexes of phosphatase 2A, phosphatase 2A1, and phosphatase 2A2 similarly deactivated KIK. Analyis of control or insulin-stimulated KIK with two antiphosphotyrosine antibodies by immunoblotting and immunoprecipitation failed to detect the presence of phosphotyrosine in the kinase. These results indicate that KIK is activated by phosphorylation as part of a kinase cascade emanating from insulin receptor stimulation.

The use of adult rat cardiomyocytes to model cardiac glycogen metabolism was investigated by monitoring the response of glycogen phosphorylase and glycogen synthase to epinephrine and insulin treatment. Cardiomyocytes derived from normal rats respond to epinephrine in the range of 1 X 10(-7) to 5.5 X 10(-6) M epinephrine with an increase in the percent of phosphorylase in the AMP-independent form from 11.5 to 24.8%. In the same cells, insulin in the range of 10(-9) to 10(-7) M increased the glucose 6-phosphate independent form of glycogen synthase from 30.5 to 40.5%. Cells derived from alloxan-diabetic hearts exhibit a hypersensitive phosphorylase activation and a refractile synthase inactivation in response to epinephrine treatment. This pattern is similar to that recorded using perfused heart preparations. The data presented suggests that adult rat cardiomyocytes represent a valid model of glycogen metabolism in both the normal and alloxan-diabetic rat.