Action of insulin on the surface morphology of hepatocytes: role of phosphatidylinositol 3-kinase in insulin-induced shape change of microvilli.

In previous studies we have shown that the insulin-responding glucose transporter isoform of 3T3-L1 adipocytes, GluT4, is almost completely located on microvilli. Furthermore, insulin caused the integration of these microvilli into the plasma membrane, suggesting that insulin-induced stimulation of glucose uptake may be due to the destruction of the cytoskeletal diffusion barrier formed by the actin filament bundle of the microvillar shaft regions [Lange et al. (1990) FEBS Lett. 261, 459-463; Lange et al. (1990) FEBS Lett. 276, 39-41]. Similar shape changes in microvilli were observed when the transport rates of adipocytes were modulated by glucose feeding or starvation. Here we demonstrate that the action of insulin on the surface morphology of hepatocytes is identical to that on 3T3L1 adipocytes; small and narrow microvilli on the surface of unstimulated hepatocytes were rapidly shortened and dilated on top of large domed surface areas. The aspect and mechanism of this effect are closely related to "membrane ruffling" induced by insulin and other growth factors. Pretreatment of hepatocytes with the PI 3-kinase inhibitor wortmannin (100 nM), which completely prevents transport stimulation by insulin in adipocytes and other cell types, also inhibited insulin-induced shape changes in microvilli on the hepatocyte surface. In contrast, vasopressin-induced microvillar shape changes in hepatocytes [Lange et al. (1997) Exp. Cell Res. 234, 486-497] were insensitive to wortmannin pretreatment. These findings indicate that PI 3-kinase products are necessary for stimulation of submembrane microfilament dynamics and that cytoskeletal reorganization is critically involved in insulin stimulation of transport processes. The mechanism of the insulin-induced cytoskeletal reorganization can be explained on the basis of the recent finding of Lu et al. [Biochemistry 35(1996) 14027-14034] that PI 3-kinase products exhibit much higher affinity for the profilin-actin complex than the primary products, PIP and PIP2. Thus, activated PI 3-kinase may direct a flux of profilin-actin complexes to the membrane locations of activated insulin receptors, where, due to the release of actin monomers after binding of profilactin to PI(3,4)P2 and PI(3,4,5)P3, massive actin polymerization is initiated. As a consequence, PI 3-kinase activation initiates a vectorial reorganization of the cellular actin system to membrane sites neighboring activated insulin receptors, giving rise to local membrane stress as visualized by extensive surface deformations and shortening of microvilli. In addition, extensive high-affinity binding of F-actin-barbed endcapping proteins enhances the cytoplasmic concentration of rapidly polymerizing filament ends. Consequently, the actin monomer concentration is lowered and the (cytoplasmic) pointed ends of the microvillar shaft bundle depolymerize and become shorter. The observations presented strengthen the previously postulated diffusion-barrier concept of glucose- and ion-uptake regulation and provide a mechanistic basis for explaining the action of insulin and other growth factors on transport processes across the plasma membrane.