Several members of the extensive family of small GTP-binding proteins are regulated by insulin, and have been implicated in insulin action on glucose uptake. These proteins are themselves negatively regulated by a series of specific GAPs (GTPase-activating proteins). Interestingly, there is increasing evidence to suggest that PKB (protein kinase B)-dependent phosphorylation of some GAPs may relieve this negative regulation and so lead to the activation of the target small GTP-binding protein. We review recent evidence that this may be the case, and place specific emphasis on the role of these pathways in insulin-stimulated glucose uptake.
Insulin is the primary hormone that is responsible for the control of glucose homoeostasis. One of the major effects of insulin is to stimulate the translocation of the glucose transporter isoform GLUT4 from an intracellular location to the plasma membrane . This involves a co-ordinated interplay between a number of intracellular processes, including signal transduction, vesicle trafficking and fusion, and actin cytoskeleton rearrangement. Many of these processes involve low-molecular-mass GTP-binding proteins (small GTPases) including members of the Ras, Rad, Rho, Arf and Rab families.
Small GTPases act as molecular switches, cycling between active (GTP-bound) and inactive (GDP-bound) states. The exchange of GDP and GTP is promoted either by GEFs (guanine nucleotide-exchange factors) or by GDIs (guanine nucleotide-dissociation inhibitors) and the transition to the GDP-bound state is promoted by GAPs (GTPase-activating proteins). The role of the various GTPase families in insulin-stimulated GLUT4 translocation has been reviewed previously , and so the purpose of this review is to describe more recent work in this field with specific emphasis placed on the upstream mechanisms by which these proteins are regulated, and then their possible role in insulin-stimulated glucose uptake.
The Rab pathway
Insulin stimulates glucose uptake into muscle and adipose tissue by promoting the rapid translocation of the insulin-responsive isoform of the glucose transporter, GLUT4, from intracellular vesicles to the plasma membrane . In the basal state, the majority of GLUT4 is excluded from the plasma membrane and is in dynamic equilibrium with a number of intracellular compartments, including endosomes and a specialized compartment known as the GSV (GLUT4 storage vesicle) (which is an insulin-sensitive compartment), as well as the endoplasmic reticulum, TGN (trans-Golgi network), Golgi and other compartments (which are most likely to be insulin-insensitive). The primary effect of insulin on GLUT4 translocation to the plasma membrane is to release GLUT4 from the GSV pool, although insulin also increases the translocation of GLUT4 from the recycling endosomal system and decreases the rate of GLUT4 endocytosis [3–5]. Previous studies have implicated both Rab4 and Rab11 in this process. Both have been found to be associated with GLUT4 vesicles by immunoblotting, and both have been shown to redistribute in response to insulin [6,7]. Insulin increases Rab4 GTP loading, and expression of a dominant-negative Rab4 (N121I) decreases insulin-stimulated GLUT4 translocation. Furthermore, a direct interaction of Rab4 with syntaxin 4, which itself has been implicated in insulin-induced GLUT4 translocation, has been reported . However, the role Rab4 plays in GLUT4 trafficking is not clear, and, indeed, a recent study failed to find Rab4 by proteomic analysis of affinity-purified GLUT4 vesicles . In contrast, Rab11 was found to be associated with GLUT4 vesicles in this study  and is thought to be necessary for both the transport of GLUT4 from endosomes to the GSVs and for the insulin-induced translocation to the cell surface .
Recent evidence has implicated a novel Rab GAP, AS160, in regulating GLUT4 translocation in response to insulin [10,11]. This protein has attracted a lot of interest for two reasons: (i) it was identified in a screen for substrates of PKB (protein kinase B or Akt), an enzyme known to play a crucial role in mediating the effect of insulin on glucose transport, and (ii) because the Rab GTPases are known to be key players in many vesicle formation, fusion and trafficking events [10,11].
AS160 phosphorylation has been shown to be increased by insulin on six sites in vivo, five of which conform to the PKB substrate consensus sequence (RXRXX[pS/pT]). Mutants of AS160, which lack these PKB sites, block GLUT4 exocytosis, but not endocytosis, in response to insulin at a step prior to the docking and fusion of GLUT4 vesicles at the plasma membrane [5,12,13]. Furthermore, ablation of AS160 using siRNA (small interfering RNA) led to an increased level of GLUT4 at the plasma membrane in the absence of insulin [9,14]. This would be consistent with the idea that AS160 depletion led to an increased GTP loading of the target Rab and so release of GLUT4 from its intracellular tethering site. Interestingly, phosphorylation of AS160 has been reported to be impaired in skeletal muscle of Type II diabetic subjects, consistent with the known defect in insulin-stimulated PI3K (phosphoinositide 3-kinase) activation in these subjects . Taken together, this strongly implicates AS160 in the mechanism by which insulin promotes GLUT4 translocation to the plasma membrane, although it is unlikely to be the only route of regulation. Other PKB substrates, including the phosphoinositide 3-phosphate 5-kinase, PIKfyve, are almost certainly likely to be additionally involved .
AS160 has been found to interact with the cytosolic tail of IRAP (insulin-responsive aminopeptidase), a known component of GLUT4 vesicles, and, indeed, this may play a part in the mechanism by which it associates with GLUT4 vesicles in the basal state . Insulin has been reported to promote AS160 dissociation from the GLUT4 vesicles, which would be expected to lead to an increase in the GTP-bound conformation of its target Rab(s) [9,14]. Taken together, the data suggest that, in the basal state, AS160 binds to GLUT4 vesicles, negatively regulating its target Rab(s), and that, in response to insulin, it is inactivated by PKB phosphorylation, causing it to dissociate from the GLUT4 vesicles, resulting in the activation of the target Rab(s) that is necessary for the translocation step.
At present, the target Rab for AS160 is not known. It has been shown that the recombinant GAP domain of AS160 is an active GAP for Rabs 2A, 8A, 10 and 14, and that two recent studies have found Rabs 2A, 8A, 10, 11 and 14 present on GLUT4 vesicles [9,17]. Of these, only Rab11 has been implicated previously in GLUT4 trafficking, although it has been suggested that Rab10 may play a role in the formation of the exocyst complex, as it is the mammalian homologue of the GTPase sec4, which has been shown to play that role in yeast [9,18].
The Rho and Rheb pathways
Three other GAPs, TCS2 (tuberous sclerosis complex 2), p122RhoGAP and pp250, have been identified as PKB substrates [19–21]. While none of these proteins has been implicated directly in regulating GLUT4 translocation, an interesting potential theme has emerged concerning how these proteins might be regulated by insulin. TCS2 is a RhebGAP and phosphorylation by PKB has been proposed, but not yet demonstrated directly, to inhibit its GTPase activity. This would allow insulin to increase Rheb GTP binding, and so, by implication, stimulate its activity. It has been proposed that this leads to the activation of mTOR (mammalian target of rapamycin)/Raptor and from there to an increase in translation and cell growth via activation of S6K1 (S6 kinase 1) and 4E-BP1 (eukaryotic initiation factor 4E-binding protein 1) [22,23].
p122RhoGAP, which we have identified recently as a novel PKB substrate and insulin-stimulated phosphoprotein in adipocytes, has been reported to be a RhoA GAP and is the rodent homologue of the tumour-suppressor gene DLC1 (deleted in liver cancer-1). This protein was initially cloned as a PLCδ1 (phospholipase Cδ1)-interacting protein, which increases PLCδ1 activity . PKB phosphorylates p122RhoGAP directly at Ser322in vitro, and insulin stimulates the phosphorylation of this site in both primary adipocytes and CHO.T (transfected Chinese-hamster ovary) cells . In primary adipocytes, insulin-stimulated phosphorylation of p122RhoGAP at Ser322 was absolutely dependent on the activation of PI3K, whereas, in CHO.T cells, the primary determinant of Ser322 phosphorylation was the MAPK (mitogen-activated protein kinase) pathway, with the PI3K pathway playing a more minor role. Insulin-stimulated p122RhoGAP phosphorylation at Ser322 thus occurs via a mechanism that is cell-context-dependent.
Insulin stimulates RhoA GTP binding in several cell types [21,25] and so it is tempting to speculate that the phosphorylation of Ser322 on p122RhoGAP inhibits its GAP activity towards RhoA. The role of the RhoA pathway in insulin-stimulated glucose uptake is controversial, as the C3 toxin from Clostridium botulinum, which inhibits RhoA function, has been reported to mimic , inhibit [27,28] as well as have no effect on  insulin-stimulated glucose uptake and GLUT4 translocation. However, this pathway may be involved in remodelling of the actin cytoskeleton, an event that is known to be important in insulin-stimulated glucose uptake .
The third GAP identified as a PKB substrate is pp250, a protein of unknown function, but which contains a predicted C-terminal GAP domain for Rheb and Rap . It will be intriguing not only to confirm the identity of the small GTP-binding protein that is a target for pp250 and whether insulin inhibits its GAP activation via PKB-dependent phosphorylation, but also to explore any possible role this protein plays in insulin-stimulated glucose uptake.
The TC10 pathway
While it is well established that insulin stimulates GLUT4 translocation via a signalling pathway involving the class 1a PI3Ks, the hormone has also been reported to stimulate GLUT4 translocation via a PI3K-independent pathway. This pathway, which is localized to lipid rafts via the binding of CAP (Cbl-associated protein) to the integral membrane protein flotillin, involves the recruitment of the adaptor proteins APS (adapter protein with a pleckstrin homology and an Src homology 2 domain), Cbl and CrkII to the activated insulin receptor. This leads to the recruitment of the GEF, C3G, which stimulates the GDP–GTP exchange on TC10 and thus its activation. TC10 is a member of the Rho family of GTPases and its activation is thought to be involved in the remodelling of the actin cytoskeleton as well as the induction of the formation of phosphatidylinositol 3-phosphate, both of which have been reported to be necessary for insulin-stimulated GLUT4 translocation [31–33].
Two downstream effectors of TC10 have been identified, both of which have been proposed to play a role in the translocation of GLUT4. The first, CIP4/2 (Cdc42-interacting protein 4/2), interacts with TC10 in a GTP-dependent manner, and is an adaptor protein that may have a role in the regulation of actin dynamics . The second, TCGAP (TC10/Cdc42 GAP), is a multidomain GAP that interacts with both TC10 and cdc42, itself a Rho GTPase family member that has been reported to mediate insulin signalling to GLUT4 translocation [35,36]. TC10 has been reported to interact with Exo70, one of the components of the exocyst complex that is required for the targeting and docking of vesicles at specific plasma membrane fusion sites and which has been shown to be required for the targeting of GLUT4 to the plasma membrane in response to insulin . This suggests that the TC10–exocyst complex may be involved in the targeting, docking and fusion of GLUT4 to specific sites on the plasma membrane. However, recently, the importance of this pathway has been questioned, since GLUT4 translocation and glucose uptake were not inhibited either in 3T3-L1 adipocytes when CAP, c-Cbl, Cbl-b or CrkII was selectively ablated using siRNA or in adipocytes from mice in which Cbl or APS was knocked out [38–41]. Furthermore, there is evidence that a functional TC10-dependent pathway is not necessary for actin remodelling or GLUT4 translocation in muscle cells .
Two interesting themes have emerged in recent years concerning insulin action. First, small GTPases play a crucial role in many of insulin's physiologically relevant effects, including the regulation of protein kinase cascades, gene expression and cellular metabolism (e.g. Ras and Rheb), protein trafficking (e.g. Rabs) and cytoskeletal dynamics (e.g. Rho-family members). The second theme concerns the mechanism by which some of these proteins may be regulated, namely their inhibition via PKB-directed phosphorylation of the regulatory GAP (1). This and the biological roles of the small GTPases are certain to be areas of fertile future interest in the quest to delineate insulin action.
Regulation of GAP function by PKB-dependent phosphorylation
Insulin, Calcium and the Control of Mammalian Metabolism: Focused Meeting to honour the retirement of Professor Dick Denton FRS, held at Willis Hall, University of Bristol, U.K., 22–23 September 2005. Organized and edited by A. Halestrap, G. Rutter and J.M. Tavaré (Bristol, U.K.).
adapter protein with a pleckstrin homology and an Src homology 2 domain
transfected Chinese-hamster ovary
guanine nucleotide-exchange factor
glucose transporter 4
GLUT4 storage vesicle
protein kinase B
small interfering RNA
tuberous sclerosis complex 2
This work was supported by grants from the Medical Research Council (to J.M.T.) and British Heart Foundation (to I.H.).