Hydrolysis of glycosyl-phosphatidylinositol in response to insulin is reduced in cells bearing kinase-deficient insulin receptors.

A glycosyl-phosphatidylinositol (GPI) has been previously identified that serves as a precursor of the polar head group that mimics and may mediate some of the intracellular actions of insulin. Since many of the biological activities of insulin may depend upon the activity of the insulin receptor kinase, we evaluated the requirement for this activity in insulin-dependent GPI hydrolysis. For the analysis we used stably transfected CHO cell lines, expressing either the wild-type human insulin receptor or a mutant receptor that lacks tyrosine kinase activity (Chou et al., 1987) and a stably transfected CHO cell line, expressing the wild-type human insulin-like growth factor I (IGF-1) receptor (Steele-Perkins et al., 1988). A GPI was identified in both types of transfected cells and in both sets of parental cells by metabolic labeling with [3H]glucosamine or [3H]galactose. The isolated glycolipid was sensitive to hydrolysis by phospholipase C and to deamination by nitrous acid. Insulin induced a time- and dose-dependent hydrolysis of the GPI in the parental line and in the transfected cell types. Cells bearing normal human receptors hydrolyzed up to 70% of their radiolabeled GPI within 2 min of the addition of 0.1 nM insulin, whereas parental cells and cells expressing the mutant receptor hydrolyzed only 20-30% in response to 100 nM insulin. IGF-1 (5-50 nM) had little effect on GPI hydrolysis in these cells as well as in CHO cells expressing the human IGF-1 receptor. It is concluded that insulin-dependent GPI hydrolysis is mediated by the normal but not by a kinase-deficient insulin receptor.