SEARCH

SEARCH BY CITATION

Keywords:

  • GLUT4;
  • insulin;
  • Sec1/Munc18;
  • SNAREs;
  • syntaxin

Abstract

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References

Delivery of the glucose transporter type 4 (GLUT4) from an intracellular location to the cell surface in response to insulin represents a specialized form of membrane traffic, known to be impaired in the disease states of insulin resistance and type 2 diabetes. Like all membrane trafficking events, this translocation of GLUT4 requires members of the SNARE family of proteins. Here, we discuss two SNARE complexes that have been implicated in insulin-regulated GLUT4 traffic: one regulating the final delivery of GLUT4 to the cell surface in response to insulin and the other controlling GLUT4's intracellular trafficking.

inline image


GLUT4 trafficking in fat and muscle cells

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References

Insulin stimulates glucose transport into fat and muscle by regulating delivery of the facilitative glucose transporter type 4 (GLUT4) from an intracellular store to the cell surface (reviewed in (1)). Upon binding its receptor, insulin initiates a signalling cascade that culminates in changes in the trafficking itinerary of GLUT4, releasing it from its intracellular store and delivering it to the cell surface (1). Individuals with insulin resistance and type 2 diabetes exhibit defective insulin-stimulated GLUT4 translocation (1–3) and consequently much effort has gone into defining the trafficking of GLUT4 in adipocytes and muscle cells.

In the absence of insulin, around 95% of cellular GLUT4 localizes to intracellular compartment(s). Insulin stimulates the delivery of GLUT4 from its intracellular store, the precise identity of which remains unclear, to the plasma membrane (1). The resultant 10- to 20-fold increase in cell surface GLUT4 levels accounts for the majority of insulin-stimulated glucose transport observed in these cells. This translocation is achieved by a dramatic increase in the rate constant for exocytosis and a modest inhibition of endocytosis (4). A working model for GLUT4 trafficking in insulin-sensitive cells based on the work from many laboratories is presented in Figure 1.

image

Figure 1. A model for GLUT4 trafficking in insulin-sensitive cells. Intracellular GLUT4 populates two interrelated endosomal cycles. The first is the prototypical endosomal system, trafficking between the plasma membrane and early endosomes (cycle 1). Having entered this pathway from the cell surface, GLUT4 is further sorted into a slowly recycling pathway, operating between recycling endosomes, the TGN and a population of vesicles termed GSVs (cycle 2). According to this model, insulin mobilizes GLUT4 to the cell surface from an intracellular store that moves slowly between the TGN and endosomes in the absence of insulin. Modified from Ref. (1).

Download figure to PowerPoint

Intracellular GLUT4 populates two interrelated endosomal cycles. The first (the prototypical endosomal system) operates between the plasma membrane and early endosomes. This fast trafficking loop serves to effectively internalize GLUT4 from the plasma membrane in the absence of insulin and is dependent on two endocytosis motifs within GLUT4 (5). Once in this cycle, GLUT4 is further sorted into a slowly recycling pathway, operating between recycling endosomes, the trans golgi network (TGN) and a specialized pool of intracellular membranes collectively referred to as GLUT4-storage vesicles (GSVs), an intracellular store that moves slowly between the TGN and endosomes in the absence of insulin. It is from GSVs that GLUT4 is mobilized to the cell surface in response to insulin. Sorting into this cycle depends upon a distinct signal in the extreme C-terminus of GLUT4 (6,7).

GLUT4 is transported between numerous membrane-bound compartments by means of vesicular transport. Membrane traffic in all eukaryotic cells is controlled by the formation of specific SNARE complexes (8). Members of the t-(target) family of SNARE proteins mark specific organelles (9). The formation of complexes between t-SNAREs and their cognate v-SNARE localized to the appropriate donor membrane is sufficient to catalyse bilayer fusion (10) and the SNAREs have been proposed to impart a degree of specificity on membrane traffic (9). SNARE proteins provide the cell with a mechanism by which it regulates membrane traffic, providing an impetus to understand which SNAREs are involved in GLUT4 traffic and how they are controlled.

Two key questions concerning insulin-regulated trafficking of GLUT4 arise from the model outlined above: ‘How is delivery of GLUT4 from GSVs to the plasma membrane regulated in response to insulin?’ and ‘What controls the effective sorting of GLUT4 into GSVs?’. Here, we consider the role of SNARE proteins in these sorting steps.

Introducing the key t-SNAREs

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References

Two distinct t-SNAREs are important in insulin-regulated GLUT4 trafficking in fat and muscle cells. First, the t-SNARE consisting of syntaxin4 and SNAP-23 (Sx4/S-23) controls fusion of GLUT4-containing vesicles with the plasma membrane, resulting in the final delivery of the transporter to the cell surface in response to insulin (1). Second, the intracellular t-SNARE comprising Sx16 and Sx6 has also been implicated in regulating intracellular GLUT4 traffic (6,11,12), although its precise role remains to be defined. All t-SNAREs contain a member of the Qa family of Sxs (Sx4 and Sx16 are the Qa SNAREs in the examples noted above) (9). These Qa SNAREs bind to a member of the Sec1/Munc 18 (SM) family of proteins (Sx4 binds Munc18c and Sx16 binds mVps45) (13). SM proteins act to control assembly of SNARE complexes and thus modulate the kinetics of bilayer fusion (14).

SNARE Proteins at the Plasma Membrane

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References

In response to insulin, GSVs are delivered the cell surface where they dock and subsequently fuse with the plasma membrane. The identification of members of the VAMP family of v-SNAREs within highly purified GSVs was the first suggestion that translocation involved SNARE proteins (15). Subsequent studies identified VAMP2 as the v-SNARE within GSVs and the Sx4/S-23 binary complex as the plasma membrane t-SNARE (1). A substantial body of experimental work has supported a key role of these SNAREs in the insulin-stimulated delivery of GLUT4 to the plasma membrane in fat and muscle cells. For example, homozygotic disruption of the Sx4 gene results in early embryonic lethality, but heterozygote (Sx4 +/−) knockout mice exhibit impaired glucose tolerance, with a 50% reduction in whole-body glucose uptake, a result attributed to an approximate 50% reduction in insulin-stimulated glucose uptake and GLUT4 translocation in skeletal muscle (16). Similarly, depletion of Sx4 or S-23 using siRNA revealed that these plasma membrane SNAREs are essential for GLUT4–vesicle fusion with the plasma membrane (17).

Which v-SNARE at the plasma membrane?

The identity of the VAMP isoform(s) which mediates GLUT4–vesicle fusion with the plasma membrane is, however, an area of some controversy. Early studies in adipocytes revealed that peptides corresponding to the extended N-terminus of VAMP2 (18) or fusion proteins corresponding to the cytosolic domain of VAMP2 (but not VAMP3) introduced into permeabilized adipocytes inhibited insulin-stimulated GLUT4 translocation to the plasma membrane (19). Similarly, studies using tetanus or Botulinum toxins (which proteolytically cleave VAMP2 and VAMP3) were shown to inhibit insulin-stimulated GLUT4 translocation (20). It is important to note, however, that none of these modalities completely inhibited insulin-stimulated GLUT4 translocation and indeed others reported different results. For example, Hajduch et al. reported that the tetanus toxin-induced cleavage of VAMP2 and VAMP3 in freshly isolated rat adipose cells had no effect on insulin-stimulated glucose transport (21). It should also be noted that translocation of GLUT4 to the plasma membrane in response to non-insulin stimuli (e.g. GTPγS) appeared to involve VAMP3 and not VAMP2 (19), suggesting that multiple pathways for the delivery of GLUT4 to the plasma membrane may operate. Such a contention is also evident in cardiomyocytes, where VAMP2 has been proposed to control insulin-stimulated GLUT4 translocation and VAMP3 in contraction-stimulated GLUT4 translocation (22) (although it should be noted that a role for VAMP3 in contraction-mediated GLUT4 translocation has been questioned, as VAMP3 knockout mice exhibit no impairment in either insulin- or contraction-mediated glucose transport (23,24)). Consistent with this, detailed proteomic characterization of GSVs identified VAMP2, VAMP3 and VAMP8 within GSVs (25), all three of which are known to participate in exocytosis in other cell types and to form an SDS-resistant complex with Sx4/S-23 (26). So, is there an emerging consensus as to which v-SNARE(s) is (are) important?

Williams and Pessin used siRNA to deplete VAMP2, VAMP3, VAMP4, VAMP5, VAMP7 and VAMP8 from 3T3-L1 adipocytes (27). Only depletion of VAMP2 was found to result in decreased translocation of GLUT4 to the plasma membrane in response to an acute insulin challenge. These data are consistent with other studies; e.g. we have reported that introduction of the cytosolic domain of VAMP2 (but not VAMP3 or VAMP4) inhibits GLUT4 translocation in permeabilized 3T3-L1 adipocytes (19). Consistent with this, a recent study from Kasuga's group also reported a significant inhibition of insulin-stimulated glucose transport and GLUT4 translocation in 3T3-L1 adipocytes depleted of VAMP2 using siRNA (17).

A different conclusion was reached by the groups of Xu and James using a different experimental system (28). VAMP2 deletion is embryonically lethal, so these groups isolated mouse embryonic fibroblasts (MEFs) from VAMP2 knockout mice and differentiated these into adipocytes. Strikingly, the absence of VAMP2 from these cells was found to have no effect on either the magnitude or the time course of insulin-stimulated GLUT4 translocation (28). These authors report that the simultaneous disruption of VAMP2, VAMP3 and VAMP8 (achieved through a combination of toxin treatment and germline deletion) was required to abolish insulin-stimulated GLUT4 translocation. Importantly, this study reported that the reintroduction of any of these v-SNAREs individually was sufficient to restore close to maximal insulin-stimulated GLUT4 translocation. Such data offer the intriguing hypothesis that there may be plasticity in the v-SNARE requirements for GLUT4 translocation (28).

Although there is an accumulating body of evidence that argues that many SNARE-mediated trafficking steps utilize functionally redundant v-SNARE proteins, it is difficult to reconcile these apparently contradictory findings. One possible explanation may lie with the re-expression of individual v-SNAREs; it is difficult to ensure that the re-expressed v-SNAREs are present in the cell at comparable levels to the endogenous molecules. This is important, as it is well documented that SNARE proteins will form functional complexes with non-physiological partners when expressed at high levels (8). In addition, the use of MEFs from knockout animals may have resulted in compensatory mechanisms that may or may not be physiologically relevant, becoming established.

Nonetheless, these data offer the tantalizing hypothesis that VAMP2, VAMP3 and VAMP8 operate interchangeably for the delivery of GLUT4 to the plasma membrane and may afford the opportunity to impart yet further levels of regulation on this process. This will be discussed further below.

Fusion of GLUT4 vesicle with the plasma membrane

Several recent studies strongly support the notion that, within the context of GLUT4 translocation, the main site of insulin action is at the plasma membrane. Using total internal reflection (TIRF) microscopy of green fluorescent protein (GFP)-labelled GLUT4, Lizunov observed that the GLUT4 vesicles move rapidly along a microtubule network adjacent to the cell surface and that insulin halts this trafficking by tightly tethering vesicles to the plasma membrane, prompting the suggestion that a primary action of insulin is to stimulate the tethering and subsequent fusion of GLUT4 vesicles with the plasma membrane (29). Subsequently, Bai et al. (30) were able to dissect the GLUT4 trafficking events at the plasma membrane into several discrete steps including docking, priming and fusion. Importantly, they went on to measure the transition kinetics between these steps. Detailed quantitative analyses of these processes revealed that insulin accelerates two steps at the cell surface. The first involves vesicle docking and the second prepares these docked GLUT4 vesicles for fusion competence; it is the post-docking step that is the major insulin-regulated step (30).

Numerous studies over the past 10 years or so have established the half-time of GLUT4 delivery to the plasma membrane in response to insulin to be approximately 5 min, with peak levels of GLUT4 being evident 10–15 min after insulin exposure (31). Further insight into this event has been gained from analysis of the effect of insulin on the post-fusion distribution of GLUT4 in the plasma membrane of rat adipose cells. Stenkula et al. (32) observed that insulin regulates not only the exocytosis of GLUT4 vesicles, but also the dispersal of GLUT4 within the plasma membrane. Using a combination of TIRF microscopy and wide-field imaging, this group showed that the distribution of GLUT4 in the plasma membrane is not homogeneous, but rather exists either in discrete GLUT4 clusters or diffusely dispersed throughout the plasma membrane. In the basal state, the amounts of GLUT4 clustered and dispersed were similar. However, upon insulin stimulation the dispersed signal increased approximately fourfold (32). This study revealed that in the basal state, GLUT4 vesicle fusion with the plasma membrane was not accompanied by a dispersal of GLUT4; rather the GLUT4 remains clustered at the fusion site from where it is rapidly re-internalized. In response to insulin, there is a burst of exocytosis that is coupled to the dispersal of GLUT4 from the fusion site into the plasma membrane (Figure 2). Kinetic analyses of these events suggest that this dispersal is stimulated approximately 60-fold by insulin (32). These latter data can be interpreted to invoke distinct mechanisms of exocytosis of GLUT4 in the basal and insulin-stimulated states. Intriguingly, the insulin-stimulated increase in the rate of GLUT4–vesicle fusion with the plasma membrane was found to be transient: after a peak at approximately 2–3 min after exposure to insulin, the fusion frequency declines to levels only slightly higher than those observed in basal cells. This implies that, after the initial ‘burst’ of GLUT4 exocytosis, it is the increased dispersal of GLUT4 from the fusion site which accounts for the observed increase in cell surface GLUT4 levels at 10–15 min (32).

image

Figure 2. Insulin stimulates the docking and fusion and subsequent post-fusion dispersal of GLUT4 in the plasma membrane. GLUT4 (red lozenge)-containing vesicles (1) traffic to the plasma membrane where they dock (2) and fuse (3). Insulin promotes the docking event via PI3K/Akt/AS160 (see text for details). In the absence of insulin, the GLUT4 molecules remain clustered and are re-internalized from the same site (4). However, in the presence of insulin, GLUT4 is dispersed away from the site of fusion (dashed line) (5). This insulin-stimulated dispersal of GLUT4 is accelerated approximately 60-fold by insulin (32) and is the major insulin-regulated step (30).

Download figure to PowerPoint

What is the role of the v-SNARE at the cell surface: more than just fusion?

Sx4/S-23/V2 constitutes a SNARE complex that is capable of mediating liposome fusion in vitro(33). Hence, one clear function of this complex (and the VAMP isoform in this context) is to drive membrane fusion. Whether this is the sole function of the VAMP's has recently begun to be addressed. In the study by Zhao et al. (28) introduced above, GLUT4 vesicle-docking events were completely abrogated in adipocytes depleted of VAMP2, VAMP3 and VAMP8, suggesting that v-SNAREs are indispensable for the stabilized docking of GLUT4 vesicles with the plasma membrane. Whether each of these VAMP isoforms could rescue this effect or whether it is restricted to a single VAMP isoform was not tested. Further to this finding, Kawaguchi et al. used an siRNA approach to deplete VAMP2 from adipocytes and found association of GLUT4 vesicles with the cell surface, as assessed by a plasma membrane lawn assay, to be unaffected. However, insulin-stimulated glucose transport into these cells was reduced and thus this study concluded that although VAMP2 depletion did not affect docking of GLUT4 vesicles, their subsequent fusion was inhibited (17). It is, however, worth noting that another study reported a diminution of GLUT4 vesicle association with the cell surface upon VAMP2 depletion (27), adding to the controversy regarding both the identity of the key VAMP and the role(s) they play.

How is the plasma membrane SNARE complex regulated?

As introduced above, several studies have suggested that a key facet of insulin action occurs after the arrival of GLUT4 vesicles at the plasma membrane, suggesting that a step after the initial, stabilized docking of these vesicles at the cell surface is under acute regulation by insulin. Moreover, using an elegant cell-free system which recapitulates the final fusion step of GLUT4 vesicles with purified plasma membranes, Koumarov et al. (34) showed that the fusion of GLUT4 vesicles with plasma membranes is not a constitutive event, but rather is activated approximately eightfold by insulin. Such data posit that there is an insulin-dependent ‘switch’ to fusion competency after docking of GSVs at the plasma membrane. So what might this activation of fusion represent?

How does insulin-signalling impact the trafficking machinery at the plasma membrane?

Important insight into the interface of GLUT4 trafficking and insulin signalling came with the identification of a protein, termed AS160 (also known as TBC1D4), which was identified as a target of the insulin-activated kinase Akt (35) and the closely related TBC1D1 (36). These proteins contain GTPase-activating (GAP) domains for Rab proteins (35), small GTPases known to regulate SNARE-mediated membrane fusion (8). AS160 is phosphorylated by Akt in response to insulin on multiple sites and mutation of these sites to alanines impairs insulin-stimulated GLUT4 translocation (25). These and other data led to a model which proposed that insulin-stimulated phosphorylation of AS160 inactivates the Rab GAP function, an event required for GLUT4 translocation. Hence, AS160 is active as a Rab GAP which serves to maintain GLUT4 within the GSVs. Consistent with this, knockdown of AS160 using short-hairpin RNA (shRNA) increases plasma membrane GLUT4 levels in an insulin-independent manner (37). These, and related, studies clearly point to AS160 as being an important point of intersection of signalling and trafficking. Recent studies on the behaviour of GLUT4 vesicles using TIRF microscopy revealed that the Akt/AS160 signalling axis regulates the initial docking stage of GLUT4 vesicles arriving at the plasma membrane and has little effect on the subsequent fusion event as wortmannin treatment (to inhibit PI3Kinase and thus prevent activation of Akt) or expression of the alanine-mutant of AS160 both significantly decreased the docking frequency of GLUT4 vesicles at the plasma membrane in both basal and insulin-stimulated adipocytes (30).

Role of Munc18c

Although the studies on AS160 discussed above indicate that insulin regulates the docking of GLUT4 vesicles with the cell surface, it is clear that a major locus of insulin action occurs after this (32,34). What might this post-docking regulatory node represent? All SNARE complexes are subject to strict regulation to insure that the trafficking steps which they catalyse take place within the correct spatial and temporal coordinates. Key regulators of Sx (t-SNARE) function are members of the SM family of proteins (14). Like the SNAREs, the SM family is highly conserved through evolution and understanding their precise role in membrane fusion represents an important question in cell biology. Some confusion regarding the role of these proteins exists, as they appear to bind t-SNAREs via multiple distinct modes of binding, depending on the studied t-SNARE/SM pair (13,14). However, it has recently been proposed that these different modes of binding may play a role at distinct stages of the SNARE cycle (14). There is, however, uniform agreement that these proteins play essential roles in the fusion steps catalysed by all SNAREs. The SM protein that binds to Sx4 is Munc18c, and there is a wealth of evidence that implicates this SM protein in the control of insulin-stimulated GLUT4 translocation to the plasma membrane.

Homozygous depletion of Munc18c is embryonically lethal, but heterozygous knockout mice exhibit decreased insulin sensitivity and a >50% reduction in skeletal muscle insulin-stimulated GLUT4 translocation, strongly supporting the notion that Munc18c is a key regulator of insulin-stimulated GLUT4 translocation (38). Using adipocytes derived from MEFs from Munc18c −/− mice, Kanda et al. (39) showed that GLUT4 translocation is enhanced by the absence of Munc18. Such data suggest that Munc18c inhibits insulin-stimulated externalization of GLUT4 and argue that the disruption of the interaction between Sx4 and Munc18c in adipocytes might result in enhancement of insulin-stimulated GLUT4 translocation. Consistent with this, overexpression of Munc18c inhibits insulin-stimulated GLUT4 translocation in 3T3-L1 adipocytes (40) and Munc18c inhibits fusion of liposomes mediated by Sx4/S-23 and VAMP2 in vitro(33), supporting the notion of Munc18c-inhibiting fusion of GLUT4 vesicles with the plasma membrane.

How may a ‘negative’ regulation of Sx4/S-23 by Munc18c be alleviated upon insulin treatment? Some tantalizing possibilities are worthy of consideration. First, Munc18c is tyrosine phosphorylated upon insulin-stimulation of 3T3-L1 adipocytes (41). In the β-cell line MIN6, elevated glucose results in the tyrosine phosphorylation of Munc18c and the dissociation of the Munc18c/Sx4 complex (42). This suggests that the tyrosine phosphorylation of Munc18c acts to dissociate the Munc18c/Sx4 complex, thus facilitating fusion of insulin-containing granules at the surface of β-cells. Interestingly, upon dissociation of Munc18c from Sx4 under these conditions, Sx4 associates with another potential regulatory molecule, Doc2b (43,44). Doc2b is a positive regulator of Sx4 function, suggesting that, at least in β-cells, the switch in binding of Munc18c to Doc2b on Sx4 is mechanistically coupled to an increase in fusion of vesicles at the plasma membrane. It is interesting to note that Doc2b is required for insulin-stimulated glucose transport and GLUT4 trafficking in adipocytes (45), suggesting that this paradigm is worthy of further study.

Stimulation of 3T3-L1 adipocytes with platelet-derived growth factor (PDGF) also promotes tyrosine phosphorylation of Munc18c concomitant with dissociation of the Sx-4/Munc18c complex (46). Expression of a mutant form of Munc18c (Y521A) was found to inhibit PDGF-stimulated GLUT4 translocation, suggesting that the tyrosine phosphorylation of Munc18c may impact the trafficking machinery involved in GLUT4 mobilization. However, the effect of the Y521A mutant on insulin-stimulated GLUT4 trafficking remains unclear, as overexpression of both wild-type or mutant forms of Munc18c both inhibited GLUT4 translocation (46). It will be important to determine whether the tyrosine phosphorylation of Munc18c is itself sufficient to result in the dissociation of Munc18c from Sx-4 (or whether other proteins are involved, such as Doc2b (45)) and the consequences of abrogation of Munc18c phosphorylation on insulin action in more amenable experimental systems.

In addition, it is well established that Munc18c can be phosphorylated on Ser/Thr residues by kinases such as Protein Kinase C (47) or Cyclin-dependent kinase-5 (48). How/whether these post-translational events are linked to insulin action remain to be clarified.

SNAREs Involved in Intracellular GLUT4 Sorting

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References

Sorting of GLUT4 into its insulin-sensitive compartment(s) has received comparatively less attention than events regulating GLUT4 insertion into the plasma membrane. As alluded to in Figure 1, GLUT4 is sorted from recycling endosomes into a slow, futile cycle (cycle 2), which culminates into the sorting of GLUT4 into an insulin-responsive compartment, termed GSVs (1). Clearly, if cells are unable to correctly sort GLUT4 into GSVs, then it is probably that insulin-stimulated GLUT4 translocation will be impaired. This is highlighted in a recent study that defines a role for a clathrin isoform in the sorting of GLUT4 into GSVs (49). Clathrin forms a coat on intracellular vesicles to facilitate selective transport of proteins during trafficking (50). Humans express two clathrin heavy chain isoforms, one of which, CHC22, is highly expressed in skeletal muscle and exhibits a high degree of colocalization with GLUT4 (49). Adaptor proteins have been shown to link clathrin to membranes and vesicle cargo (50). Interestingly, CHC22 co-immunoprecipitates both VAMP2 and the adaptor protein GGA2 (Golgi-associated, gamma adaptin ear-containing, Arf binding protein-2; also implicated in targeting GLUT4 into GSVs (51)) (49). Strikingly, depletion of CHC22 in cultured myotubes results in a loss of the GSV compartment (49). It is important to note that in the physiological context, adipocytes and muscle are required to continuously deliver GLUT4 to GSVs.

Sx6 and Sx16 and the intracellular trafficking of GLUT4

The first demonstration of a requirement for SNARE proteins came from work revealing a role for the intracellular Sx6 in GLUT4 sorting. Overexpression of a mutant form of Sx6 lacking the transmembrane anchor (and thus acting as a ‘dominant negative inhibitor’ of endogenous Sx6) delays re-internalization of GLUT4 from the cell surface upon insulin removal (11). Although the locus of action of Sx6 remains to be defined, studies in other systems suggest that this SNARE acts at the TGN, consistent with a role for Sx6 in sorting into GSVs. This conclusion is further supported by data showing that the recycling of the insulin-responsive aminopeptidase (IRAP; a GSV resident protein) from the cell surface back to GSVs requires Sx6 (52).

Two further studies examining the role of Sx16 (which forms a t-SNARE complex with Sx6) in GLUT4 traffic have strengthened this hypothesis. Shewan et al. (6) found that Sx16 exhibits a high degree of colocalization with GLUT4 and further showed that GLUT4 recycles through a subdomain of the TGN enriched in Sx16 and Sx6. Moreover, depletion of Sx16 (or overexpression of a dominant negative Sx16 mutant) in adipocytes results in reduced insulin-stimulated glucose transport and a reduction in cellular GLUT4 levels (12). Such data suggest that Sx16 acts to facilitate trafficking of GLUT4 into GSVs and that disruption of this pathway results in mistargeting of GLUT4.

VAMP isoforms and intracellular GLUT4 traffic

A recent siRNA screen on the role of VAMP isoforms of GLUT4 trafficking in 3T3-L1 adipocytes revealed not only a role for VAMP2 in GLUT4 traffic to the cell surface, but also roles for VAMP4 and VAMP8 in different facets of GLUT4 intracellular traffic (27). Knockdown of VAMP4 redirects a pool of GLUT4 to the plasma membrane in the absence of insulin. VAMP4 is known to control traffic between endosomes and the TGN and to play a role in the retrograde transport of immature secretory granules to the TGN, prompting the suggestion that VAMP4 controls retrieval of missorted GLUT4 into the GSVs from endosomes or the TGN (27). GLUT4 undergoes a continuous cycle of exocytosis and endocytosis from the plasma membrane, even in the absence of insulin. Depletion of VAMP8 results in elevated levels of GLUT4 at the plasma membrane in the absence of insulin, a phenotype explained by a dramatic reduction in the rate of GLUT4 endocytosis. Such data suggest that VAMP8 is involved in the intracellular sorting of GLUT4 after its retrieval from the cell surface, ‘en route’ back to the GSVs (27).

A somewhat different view was reached by analysis of the role of VAMP isoforms in cultured cardiomyocytes, where VAMP7 appears to control intracellular GLUT4 sequestration (53). This may reflect the fact that these cells do not express significant levels of VAMP8. Further studies aimed at clarifying the role of VAMP4, VAMP7 and VAMP8, and determining the consequence of ablation of these genes in transgenic animals is clearly warranted.

Intracellular SM proteins

The data described above support an important role for Sx16 in GLUT4 sorting, but how may this t-SNARE be regulated? Like the situation described above for Sx4, it is clear that Sx16 is regulated by a member of the SM protein family (54,55). mVps45 is a poorly studied protein, with little being known of its biological function other than its binding to Sx16 (55). We have found that the depletion of mVps45 from 3T3-L1 adipocytes results in a significant diminution of insulin-stimulated glucose transport and GLUT4 translocation (Roccisana et al., manuscript in preparation). These data further implicate the Sx6/Sx16 t-SNARE and its regulatory protein mVps45 in the intracellular sorting of GLUT4 in insulin-sensitive cells. Further work is ongoing in our laboratory to define the locus of action of Sx16 and mVps45 and to determine how this SM/Sx complex is regulated in response to insulin.

The studies described above establish SNARE proteins as key mediators of GLUT4 traffic in insulin-sensitive cells. Defining the precise trafficking step that each SNARE regulates, as well as the mechanism(s) by which such regulation is achieved will allow us to better understand the trafficking itinerary of GLUT4 in both the absence and presence of insulin.

Acknowledgments

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References

We thank Dr Ian Salt for critical reading of the manuscript. Work in the authors' laboratories is supported by Diabetes UK, the Biotechnology and Biological Sciences Research Council, the Wellcome Trust and the Diabetes Research and Wellness Foundation. N. J. B. is a Prize Fellow of the Lister Institute of Preventive Medicine.

References

  1. Top of page
  2. Abstract
  3. GLUT4 trafficking in fat and muscle cells
  4. Introducing the key t-SNAREs
  5. SNARE Proteins at the Plasma Membrane
  6. SNAREs Involved in Intracellular GLUT4 Sorting
  7. Acknowledgments
  8. References