Search

To delineate the signaling pathway leading to glucose transport protein (GLUT4) translocation, we examined the effect of microinjection of the nonhydrolyzable GTP analog, guanosine 5'-O-(3-thiotriphosphate) (GTPgammaS), into 3T3-L1 adipocytes. Thirty minutes after the injection of 5 mM GTPgammaS, 40% of injected cells displayed surface GLUT4 staining indicative of GLUT4 translocation compared with 55% for insulin-treated cells and 10% in control IgG-injected cells. Treatment of the cells with the phosphatidylinositol 3-kinase inhibitor wortmannin or coinjection of GST-p85 SH2 fusion protein had no effect on GTPgammaS-mediated GLUT4 translocation. On the other hand, coinjection of antiphosphotyrosine antibodies (PY20) blocked GTPgammaS-induced GLUT4 translocation by 65%. Furthermore, microinjection of GTPgammaS led to the appearance of tyrosine-phosphorylated proteins around the periphery of the plasma membrane, as observed by immunostaining with PY20. Treatment of the cells with insulin caused a similar phosphotyrosine-staining pattern. Electroporation of GTPgammaS stimulated 2-deoxy-D-glucose transport to 70% of the extent of insulin stimulation. In addition, immunoblotting with phosphotyrosine antibodies after electroporation of GTPgammaS revealed increased tyrosine phosphorylation of several proteins, including 70- to 80-kDa and 120- to 130-kDa species. These results suggest that GTPgammaS acts upon a signaling pathway either downstream of or parallel to activation of phosphatidylinositol 3-kinase and that this pathway involves tyrosine-phosphorylated protein(s).

Insulin's stimulation of glucose transport involves the translocation of vesicles containing the glucose transporter GLUT4 to the plasma membrane. Small GTP-binding proteins have been implicated in the regulation of vesicular traffic. We studied the effects of microinjection of wild-type Rab4 glutathione S-transferase fusion protein (WT Rab4), a GTP-binding defective mutant (Rab4 N121I), a guanosine triphosphatase-defective mutant (Rab4 Q67L), and a Rab4 antibody on insulin-induced GLUT4 translocation in 3T3-L1 adipocytes. Microinjection of Rab4 N121I and Rab4 antibodies had no effect on basal GLUT4 staining, but inhibited insulin-induced GLUT4 translocation by 50% compared with that in control IgG-injected cells. WT Rab4 and Rab4 Q67L microinjection had no effect on either basal or insulin-induced GLUT4 translocation. Premixing and coinjection of the Rab4 antibody with WT Rab4 almost completely abolished its inhibitory effect on insulin-induced GLUT4 translocation. In contrast, microinjection of an antibody directed against the highly conserved region of Rab3 proteins had no effect on insulin-induced GLUT4. These results point to a direct role of Rab4 in insulin-induced GLUT4 translocation, and that this effect is dependent on nucleotide binding to the protein. We also studied the effect of microinjection of the same proteins on insulin-induced actin filament rearrangement (membrane ruffling) in the same cell line. Microinjection of Rab4 N121I and Rab4 antibodies inhibited insulin-induced membrane ruffling by 40%, whereas WT Rab4 or a Rab3 antibody injection had no effect on cytoskeletal rearrangement. In summary, 1) Rab4 is a necessary component of the insulin/GLUT4 translocation signaling pathway; 2) the function of Rab4 in this pathway requires GTP binding; 3) Rab4 also participates in the process of insulin-induced membrane ruffling; and 4) Rab3 proteins do not seem to be involved in these processes.

Rat-1 fibroblasts overexpressing the human insulin receptor undergo rapid actin rearrangement in response to insulin. Breakdown of stress fibers present in quiescent cells is followed by transient membrane ruffling and a return of stress fibers. We investigated the signaling pathways that mediate this insulin-stimulated reorganization of the actin cytoskeleton, which was visualized with rhodamine-phalloidin. Treatment of cells with the phosphatidylinositol 3-kinase (PI3-kinase) inhibitor wortmannin prevented insulin action at the preliminary step of stress fiber breakdown. Cellular microinjection of a polyclonal antibody directed against the p85 subunit of PI3-kinase as well as a purified recombinant p85-SH2 domain protein also inhibited actin reorganization. Transient expression of a constitutively active form of PI3-kinase (p110*) was sufficient to cause both stress fiber breakdown and membrane ruffling in the absence of insulin. Microinjection of a polyclonal anti-Shc antibody or dominant negative N17-Ras protein did not affect actin dynamics, and although constitutively active V12-Ras caused modest cytoskeletal reorganization, this effect was blocked by pretreatment with wortmannin. In summary, activation of PI3-kinase is necessary and sufficient to stimulate actin rearrangement, indicating that PI3-kinase may initiate the only signaling cascade required for insulin to induce cytoskeletal restructuring.