Molecular Aspects of Insulin Receptor Signaling

Aging

Investigators

Linked publications, trials & patents

Abstract

The metabolic and mitogenic actions of insulin are initiated by binding of the hormone to its cell-surface receptor. This results in activation of the intracellular tyrosine kinase domain of the insulin receptor beta-subunit with subsequent phosphorylation of cellular substrates on tyrosine. A number of adaptor and effector proteins have been discovered that interact with the receptor beta-subunit and participate in the mediation of insulin signaling. Our laboratory has recently shown that a thiol-reactive membrane-associated protein, termed TRAP, binds covalently to the cytoplasmic domain of the human insulin receptor when cells are treated with the homobifunctional crosslinking reagent, 1,6- bis-maleimidohexane (Garant et al., 2000). To further our understanding of the biological importance of TRAP in insulin signaling, the protein complex was purified and TRAP was found to be phospholipase C-gamma1 (PLCgamma1) by mass spectrometry analysis. PLCgamma1 is an enzyme that plays a pivotal role in transmembrane signaling by promoting the production of diacylglycerol and inositol 1,4,5-trisphosphate, two second messenger molecules that control calcium mobilization and the activation of protein kinase C. PLCgamma1 was found to be associated with the insulin receptor both in cultured cell lines and in a primary culture of rat hepatocytes, which reflects the potential for physiological significance. In addition to its catalytic domain, PLCgamma1 possesses discrete domains that allow its binding with a number of signaling molecules. The dynamic association between PLCgamma1 and the insulin receptor was found to be dependent on specific domains within both proteins. We have identified some of these motifs by expressing mutant forms of PLCgamma1 and analyzed the pattern of insulin receptor-PLCgamma1 association in intact cells. A reduction in PLCgamma1 expression using siRNA abrogates insulin-dependent regulation of mitogen-activated protein kinase (MAPK), but not that of AKT. Conversely, reconstitution of PLCgamma1 in fibroblasts lacking PLCgamma1 improved MAPK activation by insulin. These results show that PLCgamma1 is a thiol-reactive protein whose association with the insulin receptor could contribute to the activation of MAPK signaling by insulin. Ras-GRP is a member of the family of calcium- and diacylglycerol-dependent guanine nucleotide exchange factors that act on Ras proteins, thereby linking membrane-bound receptors to MAPK activation. Whether such proteins are important for PLCgamma1 signaling in response to insulin remains to be determined. Insulin stimulation often leads to rapid reorganization of the actin cytoskeleton to generate the forces necessary for plasma membrane ruffling formation and a host of other cellular processes, including the redistribution of insulin-regulatable glucose transporter 4 in the cell surface. The stabilization of actin network and its attachment to cellular membranes is orchestrated by actin binding proteins. One group of these proteins, called filamins, has been shown to bind with both actin filaments and a number of macromolecules, notably integrin receptors, small GTPases, and raft-associated caveolin-1. Filamin colocalizes with resident raft proteins, thus providing evidence for the organization and clustering of lipid rafts by the actin cytoskeleton. The aim of this second study was to determine whether filamin A plays a role in activation pathways from the insulin receptor to downstream signaling cascades, leading in turn to the phosphorylation and transcriptional activation of critical genes. We report that human melanoma M2 cells lacking filamin A exhibit normal insulin receptor-mediated tyrosine phosphorylation of IRS-1, Shc proteins, and the proper activation of AKT and MAPK upon stimulation with insulin. In contrast, filamin A-expressing M2A7 cells were unable to elicit insulin-dependent Shc tyrosine phosphorylation, MAPK activation, and insulin-dependent translocation of Shc, SOS1 and MAPK to lipid raft microdomains. Interestingly, filamin A expression did not impair MAPK activation in response to serum or EGF. Coimmunoprecipitation experiments and in vitro binding assays demonstrated that filamin A binds constitutively to the insulin receptor, and that neither insulin nor depolymerization of actin by cytochalasin D affected this interaction. The colocalization of endogenous filamin A with insulin receptor was detected at the surface of the liver-derived HepG2 cells. The interaction between filamin A and the insulin receptor was blocked by overexpression of a carboxy-terminal fragment of filamin A (FLNaCT) in HepG2 cells. FLNaCT had no effect on insulin-dependent tyrosine phosphorylation of insulin receptor and IRS-1, but caused a selective increase in insulin-stimulated phosphorylation of MAPK and activation of the transcription factor, Elk-1. The localization of FLNaCT to membrane ruffles in HepG2 was noted. The fact that filamin A interacts constitutively with the insulin receptor to exert an inhibitory tone along the MAPK activation pathway is novel and potentially very important. With the information from this work, it is possible to construct a model whereby expression of a small fragment of filamin A may translate into potentiation of insulin signaling (e.g., glucose uptake, glycogen synthesis, cell proliferation, cytoskeletal rearrangement, gene expression). The possibility that alterations in expression or posttranslational modification of filamin A and/or ability of filamin A to interact with specific partners are responsible for the development of insulin resistant states will be the subject of future investigations.