Stimulation of Glucose Transport and Glucose Transporter Phosphorylation by Okadaic Acid in Rat Adipocytes*

Okadaic acid, an inhibitor of Type I and IIa protein phosphatases, was recently found to stimulate 2-deoxyglucose uptake in rat adipocytes (Haystead, T. A. J., Sim, A. T. R., Carling, D., Honnor, R. C., Tsukitani, Y., Cohen, P., and Hardie, D. G. (1989) Nature 337, 78-81). In the present experiments the effect of okadaic acid on the phosphorylation and subcellular distribution of the insulin-regulatable glucose transporter (IRGT) was investigated. At maximally effective concentrations, insulin and okadaic acid increased the amount of IRGT in the plasma membrane by 10- and 4-fold, respectively. Thus, the stimulation of glucose transport by okadaic acid was apparently due to an increase in the surface concentration of the IRGT. However, despite its stimulatory actions, okadaic acid partially inhibited the ability of insulin to enhance glucose transport and translocation of the transporter. When cells were incubated with okadaic acid alone or in combination with insulin, phosphorylation of the IRGT in the plasma membrane was increased by approximately 3-fold relative to the intracellular pool of transporters in control cells. Phosphorylation of the IRGT was confined to the presumed cytoplasmic domain at the COOH terminus of the protein. Glucose transporters were dephosphorylated in vitro by Type I or Type IIa protein phosphatases, indicating that inhibition of one or both of these phosphatases could account for the increased phosphorylation produced by okadaic acid. The observation that okadaic acid stimulated translocation of the IRGT implicated a serine/threonine phosphorylation event in triggering movement of the intracellular IRGT-containing vesicles (GTV) to the cell surface. Immunoadsorption of GTV from 32P-labeled adipocytes revealed that the IRGT was the major phosphoprotein in these vesicles. The phosphorylation of at least three other GTV proteins was increased by okadaic acid, and these species would appear to be candidates for regulators of GTV movement to the plasma membrane. It is unlikely that phosphorylation of the IRGT is the signal for translocation because insulin did not increase phosphorylation of the protein. Rather, the inhibitory effect of okadaic acid on insulin-stimulated translocation is consistent with the hypothesis that phosphorylation of the IRGT promotes its internalization.


Okadaic
acid, an inhibitor of Type I and IIa protein phosphatases, was recently found to stimulate 2-deoxyglucose uptake in rat adipocytes (Haystead, T. A. J., Sim, A. T. R., Carling, D., Honnor, R. C., Tsukitani, Y., Cohen, P., and Hardie, D. G. (1989) (GTV) to the cell surface. Immunoadsorption of GTV from 32P-labeled adipocytes revealed that the IRGT was the major phosphoprotein in these vesicles. The phosphorylation of at least three other GTV proteins was increased by okadaic acid, and these species would appear to be candidates for regulators of GTV movement to the plasma membrane.
It is unlikely that phosphorylation of the IRGT is the signal for t,ranslocation because insulin did not increase phosphorylation of the protein.
Rather, the inhibitory effect of okadaic acid on insulin-stimulated translocation is consistent with the hypothesis that phosphorylation of the IRGT promotes its internalization.
Insulin stimulates the transport and metabolism of glucose in muscle and fat cells. The stimulation of transport involves the translocation of transport proteins from an intracellular store to the plasma membrane (l-7). Although the mechanism for this insulin-dependent translocation has not been determined, considerable advances have been made in defining the structure of the glucose transporter itself. cDNAs encoding at least five different transporter species have been cloned (8)(9)(10)(11)(12)(13)(14)(15)(16)(17). The transporter proteins appear to be highly homologous, with overall amino acid sequence identity as high as 76% (17).
Computer modeling based on hydrophobicity of the predicted amino acid sequences indicates that all five species have 12 membrane-spanning domains (8, l7), with intracellular domains located at the beginning, the middle, and the end of the proteins. Skeletal muscle fibers, cardiac myocytes, and adipocytes, the three cell types in which glucose transport is most dramatically stimulated by insulin, have a unique transporter referred to as the insulin-regulated glucose transporter (IRGT) ' (18) or Glut4 (17). In fat cells the concentration of the IRGT in the plasma membrane increases lo-fold or more with insulin (18). This is likely to be the major mechanism by which insulin stimulates glucose transport in these cells although little is known about the biochemical signals that cause movement of transporters to the plasma membrane.
Many of the metabolic effects of insulin are mediated by dephosphorylation of rate-limiting enzymes in various metabolic pathways. For example, the stimulation of glycogen synthesis involves dephosphorylation and activation of glycogen synthase. On the other hand, insulin stimulates the phosphorylation of a number of proteins in fat cells (19,20). Several protein kinases have been shown to be activated when cells are incubated with insulin (21) although the function of these insulin-stimulated phosphorylation reactions has not been established. We found recently that the IRGT was phosphorylated in muscle and fat cells but that its phosphorylation was not increased by insulin (22,23). This was not unexpected since t,he effect. of insulin on stimulating translocation of the IRGT is presumably mediated by elements that regulate the movement of the intracellular vesicles containing the IRGT. In contrast to insulin, isoproterenol and CAMP derivatives stimulated IRGT phosphorylation and inhibited insulin-dependent glucose transport (22). The effects of isoproterenol were not associated with a decrease in the amount of the IRGT in the plasma membrane, suggesting that phosphorylation may regulate the "intrinsic activity" of the transporter.
Okadaic acid is a tumor promoter originally isolated from the sea sponge, Hulichondriu and added to tubes containing SDS (1%) to terminate the phosphatase reaction. The IRGT was immunoprecipitated as described previously (22). The catalytic subunit of Type I protein phosphatase purified from rabbit skeletal muscle (33) was supplied by Dr. Anna DePaoh-Roach (Indiana University School of Medicine). Type IIa catalytic subunit was purified from beef heart (34) and was provided by Dr. Marc Mumby (University of Texas Health Science Center at Dallas). The activities of the respective catalytic subunits were compared by using "P-labeled phosphorylase u as substrate. A unit is defined as the amount of enzyme that released 1 pmol of "P from phosphorylase o in 1 min.
Vesicle Zmmunoadsorption-The epitope of the monoclonal antibodv lF8 is within the cvtonlasmic domain of the IRGT and thus provides accessibility to [he intact intracellular glucose transportercontaining vesicles (GTV). Acrylamide beads (Bio-Rad) were incubated with 1 mg/ml goat anti-mouse IgG (East Acres Biologicals). In experiments involving immunoadsorption of GTV from '*P-labeled cells, beads (200 ~1) were incubated with 1% bovine serum albumin in phosphate-buffered saline for 60 min at 20 "C. Beads were then incubated with GTV-denleted LDM (200 us) from nonlabeled cells for 30 min at 4 'C. Beads were washed three times with Buffer A supplemented with 100 mM NaCl (Buffer B). The suspensions were divided in halves and incubated with either lF8 (50 tig/lOO ~1 of beads) or nonimmune mouse serum (10 ~1) for 60 min at 20 OC. The beads were then washed three times with Buffer B and added to tubes containing 3zP-labeled LDM (10 pg) and bovine serum albumin (0.1%) in Buffer B. After a 2-h incubation at 4 'C, beads were pelleted through a 0.4 M sucrose cushion and washed three times with Buffer B. Beads were subjected to SDS-PAGE following addition of Laemmli sample buffer.
Sucrose Density Gradient Sedimentution-Adipocytes were incubated in the absence or presence of insulin following labeling with "'P as described above. Homogenates were centrifuged at 48,000 X g for 20 min at 4 OC to remove the plasma membrane, high density microsomes, and mitochondrial/nuclear fractions. The supernatants were concentrated (2 mg/ml, final protein concentration) by using Aquacide (Calbiochem) and then layered gently onto continuous sucrose gradients (12-40%, w/v) prepared in Buffer A. Gradients were centrifuged at 100,000 X g in a Beckman SW 41 rotor for 18 h at 4 'C. Fractions (0.7 ml) were collected at 4 "C by puncturing the bottom of each tube with a needle. The IRGT was immunoprecipitated from each fraction as described above.
Other &futerials-Okadaic acid was generously provided by Dr. Philin Cohen (University of Dundee) and by Dr. M. Gibbs (Pfizer Inc, originally supplied-by Professor Takeshi Yasumoto and Dr. Michio Murata. Tohoku Universitv. Sendai). '*P: and Na? were -, obtained from Du Pant-New England Nuclear. 1z51-labeled protein A was prepared by nonenzymatic iodination of protein A as described previously (35). Highly purified porcine insulin (27 units/mg) was a gift from Lilly.

Stimulation
of 2-Deoxyglucose Uptake by Okadaic Acid-The effect of okadaic acid on the uptake of 2-deoxyglucose by rat adipocytes was measured initially using the experimental design described by Haystead et al. (27) (Fig. 1). Cells were incubated for increasing times in medium containing 0.5 mM 2-deoxy[l-i4C]glucose.
Where indicated, insulin or okadaic acid was added to the medium at the same time as the 2deoxy[l-r4CJglucose ( Fig. 1). Okadaic acid increased uptake after a lag of approximately 5 min, and between 10 and 20 min of incubation, the rate of uptake in the presence of okadaic acid was comparable to the maximum rate of uptake observed by insulin. The present results agree with the earlier findings of Haystead et al. (27) and confirm their observation that okadaic acid increases glucose transport.
From the results in Fig. 1, it was not clear if okadaic acid stimulated glucose transport to the same extent as insulin because the 2-deoxy[I-14C]glucose was added to the cells together with the insulin and okadaic acid. This is because several minutes of incubation are needed for insulin to exert its full effect on transport (36). Therefore, additional experiments were performed to compare the effects of insulin and Rat adipocytes were incubated at 37 'C in low Pi medium for 2 h. Cells were then incubated with insulin (ZK?) (2.5 milliunits/ml) and/or okadaic acid (OKA) (1 pM) for 20 min before 0.1 mM 2-deoxy[l-i4C]glucose was added. The results represent the rate of 2-deoxy[l-i4CJglucose uptake measured after 15 s and are mean values (kS.E.) from five experiments. CON, control.
okadaic acid on stimulating glucose transport. Cells were incubated with the two agents for 20 min, a time sufficient to elicit their respective maximum effects.* The uptake of 2deoxy[l-%]glucose was then measured after a 15-s incubation, which provides an index of the initial rate of sugar uptake. Insulin and okadaic acid stimulated 2-deoxy[l-14C] glucose uptake by approximately 8-and 4-fold, respectively (Fig. 2).
To determine whether the effects of insulin and okadaic acid were additive, cells were incubated with the combination of both agents (Fig. 2). With insulin plus okadaic acid, the rate of 2-deoxyglucose uptake was actually less than that observed with insulin alone (P < 0.05, paired comparison). Similar results were obtained when glucose transport was assessed by measuring the initial rates of 3-0-methylglucose uptake.z Thus, not only was okadaic acid less effective than insulin in stimulating glucose transport, but it also inhibited insulin-stimulated glucose transport. scribed previously (18). Briefly, in the absence of insulin, most of the transporter was found in the LDM fraction with negligible amounts found in the plasma membrane or the mitochondrial/nuclear fractions. In the presence of insulin, transporters in the plasma membrane increased, and those in the LDM decreased. Okadaic acid also stimulated translocation of the IRGT from the LDM to the plasma membrane, but this effect was significantly less than that of insulin. It should be noted that at the concentrations used in these experiments, insulin and okadaic acid produce their respective maximum effects on IRGT translocation.2 When adipocytes were incubated with insulin together with okadaic acid, additivity of the effects was not observed. In fact, okadaic acid impaired the ability of insulin to stimulate translocation (JJ < 0.05, insulin versus insulin plus okadaic acid). The protein recovery in the different membrane fractions was not affected by insulin or okadaic acid,' and the distribution of the Na+-K+-ATPase, a plasma membrane marker, was not significantly different among the treatment groups (Fig. 3). Therefore, it is unlikely that the effects of okadaic acid on opposing insulin-stimulated translocation result from a fractionation artifact.

Stimulation of Protein Phosphovlution by Okadaic Acid-
Okadaic acid stimulated the phosphorylation of proteins in all subcellular fractions except the mitochondrial/nuclear fraction. Phosphorylation of most of the major phosphoproteins in the plasma membrane, LDM, and soluble fractions was increased by okadaic acid (Fig. 5). These species appeared to include proteins whose phosphorylation was also increased by insulin. An example is the iVr = 130,000 species, which is probably ATP-citrate lyase (37), found in the LDM and soluble fractions. However, okadaic acid also increased the phosphorylation of proteins not significantly phosphorylated in response to insulin, such as the JVfr = 90,000 species present in the soluble fraction.
To determine whether the IRGT was phosphorylated in response to okadaic acid, the protein was immunoprecipitated from subcellular fractions of ,=P-labeled adipocytes (Fig. 6). Samples of the immunoprecipitate were then subjected to SDS-PAGE, and proteins were transferred to nitrocellulose. The amount of '*P-labeled transporter was determined by autoradiography ( Fig. 6@. Okadaic acid increased phosphorylation of the IRGT in both the LDM and plasma membrane fractions. However, because okadaic promotes translocation of the transporter, it was necessary to measure the amount of transporter in the fractions in order to assess the magnitude of the effect of okadaic acid on increasing phosphorylation. Therefore, the nitrocellulose sheets were incubated with R820 and '*Y-labeled protein A to enable estimation of the relative amounts of IRGT present (Fig. 6A). Values for specific activity in arbitrary units could then be determined by dividing the amount of '*P by the amount of 1251. To allow comparison among experiments, the specific activities in the different fractions were expressed relative to that in the LDM from control cells (22).
A potential problem with expressing specific activity relative to the control LDM is that the specific activity of the IRGT in the precursor pool from which the transporters move to the plasma membrane may differ from that of the whole LDM population. However, we found no evidence that the "*P-labeled IRGT was found in a subset of the GTV within the LDM when sucrose gradient analyses were performed on LDM from control and insulin-treated cells. The '*P-labeled IRGT was found in gradient fractions identical to those containing the bulk of the intracellular IRGT (Fig. 7). Furthermore, the density of the GTV isolated from control cells was indistinguishable from those of insulin-treated cells.
In agreement with our previous results (22,23), the specific activity of '*P-labeled IRGT was significantly higher in the plasma membranes of control cells than in the membranes of insulin-treated cells (Fig. 8@. It seems likely that translocation per se accounts at least in part for the decrease in specific activity of the plasma membrane transporters produced by insulin. The transporters that are translocated from the LDM to the plasma membrane in response to insulin have a lower specific activity than those transporters already in the plasma membrane. Consequently, the specific activity of the plasma membrane fraction of IRGT would be expected to resemble more closely that of these newly inserted transporters because of their relative abundance. For this reason, we believe it is most appropriate to make all comparisons of "*P-labeled IRGT relative to the specific activity of the $'P-labeled transporters in the donor compartment (i.e. the control LDM fraction) of control cells. It is possible that changes in =Plabeled transporters may occur in the plasma membrane were subjected to SDS-PAGE, and the proteins were transferred to nitrocellulose sheets. The sheets were exposed to film for estimation of '*P-labeled transporter (A) and then blotted with R820 (B). Autoradiograms, which are shown as inserts, were scanned for optical density. The amounts of '*P-labeled IRGT or the immunoreactive IRGT relative t.o the radioactivity detected in the respective peak fractions from control (0) and insulin-treated (0) cells are shown. fraction in an insulin-dependent manner. However, such changes should still be evident when the specific activities are expressed relative to that of the transporters in the control LDM.
Okadaic acid increased the phosphorylation of IRGT in the LDM fraction by only 60%. In contrast, in the presence of okadaic acid alone or okadaic acid plus insulin, there was a 3-4-fold increase in the phosphorylation of transporters in the plasma membrane fraction (Fig. 8EJ). Assuming that the specific activity of [32P]phosphate in the transporter was equal to that of intracellular [Y-~*P]ATP and by using measurements of glucose-inhibitable cytochalasin B binding sites as an estimate of transporter number, we have calculated that there is approximately 0.2 mol of phosphate per mol of IRGT in the LDM of control cells incubated under these conditions (22). Although this value is likely to be lower than the actual stoichiometry (22), it indicates that there is at least 0.6-0.8 mol of phosphate per mol of plasma membrane IRGT after okadaic acid treatment.
Essentially all of the %'P in transporters from control or insulin-treated cells is recovered in a single CNBr fragment, denoted . This fragment appears to encompass the COOH terminus of the IRGT because it binds to the COOHterminal antibody, R820 (12). As shown in Fig. 9, CB-T was the only "*P-labeled CNBr fragment evident in transporter immunoprecipitated from okadaic acid-treated cells. Thus, even after inhibiting protein phosphatases with okadaic acid, phosphorylation of the IRGT was restricted to a relatively small region in the COOH-terminal intracellular domain.

Phosphoproteins
Associated with Intracellular Glucose

Transporter
Vesicles-The GTV appear to be correctly oriented following homogenization because they can be isolated quantitatively using antibodies directed against cytoplasmic domains of the glucose transporter (38). Using this technique, GTV from "P-labeled cells were isolated, and the major phosphoprotein constituents of the vesicles were identified (Fig. 10). An Mr = 45,000 protein that we have identified as the IRGT is the major '*P-labeled protein in the GTV isolated from control cells. At least three other phosphoproteins (denoted PGTV-1 to PGTV-3 in Fig. 10) were identified which underwent increased phosphorylation in response to okadaic acid. These proteins were not evident in the supernatant fractions from the vesicle immunoprecipitation or when immunoadsorption was performed using a nonimmune IgG (Fig.  10). The phosphorylation of an Mr = 32,000 protein in the immunoprecipitates was increased markedly after insulin and okadaic acid. However, this species is probably not associated with the GTV because it was recovered using nonimmune IgG. "P-labeled adipocytes were incubated for 20 min in the absence of insulin (CON or in the presence of insulin (INS'), okadaic acid (OKA), or okadaic acid plus insulin. Following homogenization of cells, the LDM were isolated by differential centrifugation, and samples were incubated at 4 'C for 2 h with phosphatase inhibitors and acrylamide beads that had been conjugated with either lF8 (+) or nonimmune mouse serum (- Therefore, the decrease in "P is because of phosphatase activity rather than a decrease in the number of transporters immunoprecipitated.

Dephosphorylation
The results are expressed as the percentage of the '*P content of the transporters before incubation at 30 'C.
tivity is related to the decrease in the amount of "P-labeled transporter.
Type I and Type IIa phosphatases dephosphorylated the transporter at comparable rates (Fig. 11). With 0.5 milliunit/ml of the enzyme, dephosphorylation appeared to reach a plateau when approximately 40% of the '*P was removed. However, increasing the phosphatase concentration resulted in a more rapid and complete dephosphorylation of the transporter.
Therefore, all of the sites in the transporter occupied by "P could be dephosphorylated by the Type IIa enzyme.

DISCUSSION
The present results indicate that okadaic acid stimulates translocation of glucose transport proteins from an intracellular store to the plasma membrane (Fig. 3). In this respect the action of okadaic acid resembles that of insulin, and it seems reasonable to conclude that the increase in plasma membrane transporters produced by okadaic acid is primarily responsible for the observed increase in glucose transport. Our results also indicate that okadaic acid increases the phosphorylation of the IRGT (Fig. 6), suggesting that the transporter is dephosphorylated in cells by either Type I or Type IIa protein phosphatases.
Further support for this hypothesis is provided by the findings that the transporter could be dephosphorylated in vitro by the catalytic subunits of either of these phosphatases (Fig. 11). Several differences were noted between the effects of okadaic acid and insulin on glucose transport.
Unlike okadaic acid, insulin did not stimulate phosphorylation of the IRGT (Fig. 8). This suggests that phosphorylation of the transporter per se is not required for its translocation to the cell surface. With okadaic acid, but not with insulin, there was a pronounced lag before increased transport was observed (Fig. 1). This lag might be because of the delay in diffusion of okadaic acid into the cell (27). Alternatively, the hysteresis is suggestive of an indirect action of okadaic acid on stimulating transport, and the possibility has not been eliminated that transport activation is mediated by a metabolite that requires time to accumulate. Free fatty acids are possible candidates because their release is stimulated by okadaic acid, and recent findings suggest that they stimulate glucose transport and translocation of the IRGT (39). However, the effect of okadaic acid on lipolysis is considerably less than that of isoproterenol (27). This argues against a role of fatty acids in mediating the effect of okadaic acid because isoproterenol plus adenosine deaminase markedly stimulates lipolysis but does not stimulate glucose transport in rat adipocytes (40). A further difference between the actions of insulin and okadaic acid was that okadaic acid was less effective in increasing both the uptake of 2-deoxyglucose and the number of transporters in the plasma membrane (Figs. 2 and 4). This difference was observed using concentrations of okadaic acid and insulin sufficient to produce their respective maximum effects. Finally, okadaic acid partially inhibited insulin's ability to stimulate transport and transporter translocation.
Because of these differences, it seems clear that insulin and okadaic acid act by separate mechanisms although they may still share certain steps in the pathways of activation.
The ability of insulin and okadaic acid to stimulate translocation of the IRGT raises the possibility that both agents stimulate the phosphorylation of the same regulatory protein responsible for triggering movement of vesicles to the cell surface. A number of proteins whose phosphorylation was increased by both insulin and okadaic acid were evident in the cell extract (Fig. 5). However, the effects of okadaic acid on protein phosphorylation were far more widespread than those of insulin.
Thus, establishing a link between phosphorylation of one of the extract phosphoproteins and augmented glucose transport will be difficult. Therefore, as an initial step in identifying phosphoproteins involved in regulating glucose transport, it seemed reasonable to focus on those associated with the GTV. The three phosphoproteins that were identified in the vesicles (Fig. 10) are attractive candidates for regulators of movement of the GTV to the plasma membrane.
One interpretation of the differences between the extent of glucose transport activation by insulin and okadaic acid is that okadaic acid has at least two actions, an insulin-like action to stimulate translocation of the IRGT and a second action that opposes the first. The second might be at the level of the insulin receptor since there is evidence that increased serine/threonine phosphorylation of the receptor inhibits its activity (41). However, because insulin does not stimulate IRGT phosphorylation, it is tempting to propose that it is the action of okadaic acid on increasing transporter phosphorylation which is involved in opposing the effect of insulin.
The role of phosphorylation in modulating transporter function has not been defined. However, there is evidence that the phosphorylation state of the IRGT is altered under various experimental conditions. We have shown previously that phosphorylation of the IRGT is increased by jY-adrenergic agonists and CAMP derivatives (22, 23). Incubation of adipocytes with the tumor promoter, phorbol 12-myristate 13acetate, also stimulates phosphorylation of the IRGT. ' The results with isoproterenol were of particular interest because /3-adrenergic receptor stimulation inhibits insulin-stimulated glucose transport without decreasing the numbers of transporters in the plasma membrane. Based on these observations we proposed that phosphorylation of the transporter by CAMP-dependent protein kinase decreased the ability of the IRGT to transport glucose. Presumably this is not the case with okadaic acid, whose inhibition of insulin-stimulated glucose transport appears to involve inhibition of transporter translocation.
The actual site of phosphorylation increased in response to okadaic acid has not been determined although it is located in the same CNBr fragment as the isoproterenolstimulated phosphorylation site(s). It should be noted that there are several serine and threonine residues that are potential sites of phosphorylation in this fragment, and it seems possible that the IRGT tail contains multiple phosphorylation sites that have different functional roles. Phosphorylation has been proposed to regulate endocytosis of a number of membrane proteins, including the receptors for epidermal growth factor (42), insulin (43), and insulin-like growth factor II (IGF-II) (44). The IGF-II receptor is of particular interest because like the glucose transporter, its cell surface concentration is increased following incubation of cells with insulin (45). Another similarity between the proteins is that the phosphorylation state of the IGF-II receptor in the plasma membrane from control cells is significantly higher than that in the plasma membrane of insulin-treated cells (44). It has been proposed that the phosphorylated IGF-II receptor recycles more rapidly than the nonphosphorylated receptor and that insulin causes dephosphorylation of the receptor at the cell surface (44).
We suggest that phosphorylation of the IRGT triggers its internalization.
In order for this hypothesis to accommodate the effects of insulin and okadaic acid on glucose transport, it is necessary to assume that the IRGT recycles between the intracellular compartment and the cell surface. This is likely because the glucose transporter repopulates the intracellular compartment following insulin removal (46). The difference between the maximum effects of insulin and okadaic acid would be consistent with an action of phosphorylation to increase the rate of IRGT internalization since okadaic acid increased, and insulin decreased, the phosphorylation state of transporters in the plasma membrane. Furthermore, it might be significant that transporters in the plasma membrane of control cells are more highly phosphorylated than those in the LDM. It is an intriguing possibility that the high basal phosphorylation is a mechanism for maintaining a low number of the IRGT proteins in plasma membrane in the absence of insulin. It may be noteworthy that the 32P-labeled IRGT in the basal and insulin-treated plasma membrane fraction was not recovered in a Triton X-lOO-insoluble fraction of the plasma membrane. ' This result is in contrast to previous findings with the IGF-II receptor (47). However, the IRGT may not cluster in coated pits to the same extent as the IGF-II receptor.
Further characterization of the phosphorylation sites within the IRGT and establishing their functional roles may yield important information about insulin action. A working hypothesis is that insulin has two actions in regulating the increased expression of the IRGT at the cell surface. One is to increase movement of the intracellular vesicles to the plasma membrane. The second is to promote retention of the transporter in the plasma membrane. The latter effect might involve maintaining the IRGT in a relatively dephosphorylated state, an action that does not occur with okadaic acid and possibly other agents such as phorbol 12-myristate 13acetate. The more potent stimulatory effect of okadaic acid on phosphorylation of transporters in the plasma membrane fraction compared with those in the LDM suggests that phosphorylation/dephosphorylation occurs either within or close to the plasma membrane.
In summary, our results lend further support to the hypothesis that translocation is the major mechanism for acutely regulating glucose transport in rat adipocytes. We have demonstrated that the IRGT is a target for cellular protein kinases and protein phosphatases although the physiological role of transporter phosphorylation is yet to be determined. Our findings are consistent with the hypothesis that increased serinelthreonine phosphorylation of a protein, possibly associated with the GTV, stimulates movement of transporters to the plasma membrane whereas phosphorylation of the glucose transporter itself accelerates its internalization.