Neomycin prevents the wortmannin inhibition of insulin-stimulated Glut4 translocation and glucose transport in 3T3-L1 adipocytes.

Insulin stimulates the movement of the facilitative glucose transporter glucose transporter-4 (Glut4) from an intracellular compartment to the plasma membrane in adipocytes and muscle cells, resulting in an increased rate of glucose uptake. Insulin-stimulated Glut4 translocation and glucose transport are abolished by wortmannin, a specific inhibitor of phosphatidylinositol 3'-kinase (PI3K). Here, we demonstrate that neomycin, a drug that masks the cellular substrate of PI3K, phosphatidylinositol 4,5-bisphosphate (PIP), prevents wortmannin inhibition of insulin-stimulated (2)Glut4 translocation and glucose transport without activating protein kinase B, a downstream effector of PI3K. These results suggest that PIP(2) may have an important regulatory function in insulin-stimulated Glut4 translocation and glucose transport.

Cell Culture-3T3-L1 fibroblasts were grown in DMEM containing 10% calf serum at 37°C in a humidified atmosphere of 10% CO 2 . For differentiation into adipocytes, fibroblasts were grown to confluence, and the medium was replaced with DMEM containing 10% fetal calf serum, 250 nM dexamethasone, 110 g/ml isobutylmethylxanthinine, and 172 nM insulin for 2 days. Medium was then replaced with DMEM containing 10% fetal calf serum and 172 nM insulin for a further 2 days. Adipocytes were used between 10 and 12 days following differentiation.
2-Deoxy-D-glucose Transport Assays in Cultured 3T3-L1 Adipocytes-Differentiated adipocytes were incubated in serum-free medium for 2 h in the presence or absence of 10 mM neomycin. For neomycin-treated cells, neomycin was maintained at a concentration of 10 mM throughout the experiment. The serum-starved cells were washed twice in 2 ml/well of Krebs-Ringer-Hepes buffer (25 mM Hepes, pH 7.4, 118 mM NaCl, 5 mM NaHCO 3 , 4.7 mM KCl, 1.2 mM KH 2 PO 4, 1.2 mM MgSO 4 ⅐7H 2 O, 2.5 mM CaCl 2 ). The cells were then incubated for 30 min at 37°C in the presence or absence 1 M insulin; wortmannin-treated cells were incubated with 100 nM wortmannin for a total of 40 min, including 10 min prior to the addition of insulin. [ 3 H]2-Deoxyglucose (at a final concentration of 10 M, 1 Ci/ml) was added to each well for 4 min, and the cells were quickly washed in ice-cold PBS and solubilized in Triton X-100, and [ 3 H]2-deoxy-D-glucose uptake was measured by scintillation counting. Glut-specific glucose uptake was measured by subtracting values for [ 3 H]2-deoxy-D-glucose uptake in the presence of 10 M cytochalasin B.
Immunofluorescence Analysis of Plasma Membrane Lawns-3T3-L1 adipocytes were cultured on collagen-coated glass coverslips. Cells were serum-starved in the presence or absence of 10 mM neomycin for 2 h, followed by stimulation with 1 M insulin (in the presence or absence of 100 nM wortmannin as indicated in the figure legends; wortmannin was added to the well 10 min prior to insulin treatment) for 20 min. Plasma membrane lawns were prepared as described in Ref. 26. Cell were transferred onto ice and washed three times with ice-cold PBS and fixed in 4% paraformaldehyde. Cells were washed three times with PBS, followed by three washes in PBS, 5% goat serum over a time period of 15 min. Following this, cells were incubated with primary antibody diluted in PBS, 1% goat serum for 1 h. After this time, cells were washed three times with PBS, 1% goat serum and incubated with fluorescein isothiocyanate-conjugated secondary antibodies diluted in PBS, 1% goat serum for 1 h. Finally, cells were washed with five changes of PBS over a 15-min period and mounted on glass slides and visualized using a Zeiss Pascal on an Axiovert microscope using X40 1.3-numerical aperture plan apochromat objective (Carl Zeiss Ltd., Welwyn Garden City, UK). Images were quantified using Image J from the National Institutes of Health.
Protein Kinase B Activity Assays-3T3-L1 adipocytes, treated as outlined in glucose transport methods, were briefly washed with icecold PBS on ice and solubilized in lysis buffer. Lysates were centrifuged at 14,000 ϫ g for 10 min and insoluble material removed. PKB immunoprecipitation and kinase assay were performed as described previously (27).

Neomycin Restores the Insulin-stimulated Glucose Transport
Inhibited by the PI3K Inhibitor Wortmannin-To analyze the possible role of PIP 2 in insulin-stimulated glucose transport, we examined the effect of the aminoglycoside, neomycin, on glucose transport into 3T3-L1 adipocytes. This agent was chosen as it binds with high specificity to PIP 2 (24,25) and has been used previously in several studies examining PIP 2 function (28 -30). The results in Fig. 1A show that 10 mM neomycin had no significant effect on the basal level of glucose transport compared with vehicle. Incubation of adipocytes in 100 nM insulin gave rise to a characteristic 10-fold increase in glucose transport (Fig. 1A). Neomycin treatment caused a small (24 Ϯ 10%) but significant inhibition of insulin-stimulated glucose transport (p ϭ 0.05). As expected, insulin-stimulated glucose transport was completely abolished by the PI3K inhibitor, wortmannin (Fig. 1A). However, remarkably, this inhibition of glucose transport by wortmannin was restored by 47 Ϯ 13%, p Ͻ 0.005 in the presence of neomycin (Fig. 1A). Neomycin prevented wortmannin inhibition of insulin-stimulated glucose transport in a dose-dependent manner (Fig. 1B). These results suggest that an important aspect of insulin-stimulated PI3K activation is the subsequent loss or masking of PIP 2 .
To rule out any interfering effects of neomycin on insulin signaling or on the ability of wortmannin to inhibit PI3K, the phosphorylation and activity of the PI3K effector, PKB, was assessed. As shown in Fig. 2A and quantified in Fig. 2B, in both the presence and absence of neomycin, insulin stimulation results in the significant phosphorylation of serine (S) at postion 473 in PKB when compared with basal conditions (p Ͻ 0.0001, n ϭ 4). The magnitude of this response was not different between neomycin-treated and untreated cells. The insulindependent PKB phosphorylation seen in both the presence and absence of neomycin is reduced to the same extent by the PI3K inhibitor wortmannin (84% reduction in the presence and 87% reduction in the absence of neomycin, p Ͻ 0.001, n ϭ 4). Thus, neomycin has no effect on insulin stimulation of PKB serine 473 phosphorylation, and furthermore, the inhibition of the insulin-dependent PKB serine 473 phosphorylation seen in the presence of wortmannin is unaffected by neomycin treatment. However, such phosphorylation data does not consider effects on PKB activity.
As analysis of PKB phosphorylation is not a direct measure of activity, the ability of immunoprecipitated total cellular PKB to phosphorylate a synthetic peptide substrate (KRPRAATF),

Effects of Neomycin on Insulin Action 20568
under the same treatment conditions used previously, was measured. As shown in Fig. 2C, insulin stimulated a marked rise in PKB activity; the magnitude of this response was similar in neomycin-treated cells (Fig. 2C). As expected, wortmannin treatment caused a marked inhibition of insulin-stimulated PKB activity (54 Ϯ 3%). In agreement with the results of the PKB phosphorylation data (Fig. 2, A and B), neomycin did not prevent the wortmannin-induced inhibition of insulin-stimulated PKB activity (Fig. 2C). Thus, the reversal of wortmannininhibited insulin-stimulated glucose transport is independent of any effects of neomycin on PKB activation.
Neomycin Prevents Wortmannin Inhibition of Insulindependent Glut4 Translocation to the Plasma Membrane-The results presented in Fig. 1, A and B, clearly demonstrate that neomycin prevents the inhibition of insulin-stimulated glucose transport elicited by wortmannin. As the increased rate of glucose transport promoted by insulin is largely a result of translocation of Glut4 from an intracellular storage compartment to the plasma membrane, we examined the effect of neomycin treatment on Glut4 insertion into the plasma membrane. For this, plasma membrane sheets derived from 3T3-L1 adipocytes, prepared by sonication of cells attached to coverslips, were stained with Glut4 antibody as described under "Materials and Methods." Fig. 3A shows that plasma membrane sheets prepared from unstimulated cells have only very low levels of Glut4 staining in both the presence and absence of neomycin. Insulin elicited a marked increase in plasma membrane Glut4 content, and interestingly this was further enhanced by neomycin (53 Ϯ 4% increase, p ϭ 0.05, n ϭ 3). As expected, wortmannin treatment resulted in a marked decrease in the insulin-dependent increase in Glut4 levels at the plasma membrane. However, consistent with the glucose transport data, the presence of neomycin prevented wortmannin inhibition of the insulin-dependent increase in plasma membrane Glut4 content. Quantification of the fluorescence intensity from 30 fields of view per condition is shown in Fig. 3B. DISCUSSION It is widely accepted that the lipid kinase PI3K plays an essential role in insulin stimulated glucose transport (4,6,22,31). This is borne out by the complete inhibition of insulinstimulated glucose transport by the fungal metabolite and specific inhibitor of PI3K, wortmannin (6). Intensive research into the downstream effectors of PI3K activity involved in insulin signaling have resulted in the identification of molecules such as protein kinase B and atypical PKCs that bind PIP 3 (13,32). However, the exact requirement for these kinases in insulinstimulated glucose transport is unresolved. Indeed, although PI3K activity is necessary for insulin-stimulated Glut4 trans-FIG. 2. Wortmannin inhibition of insulin-dependent PKB phosphorylation and activity are unaffected by neomycin treatment. 3T3-L1 adipocytes were incubated for 2 h before and maintained throughout the experiment in the presence and absence of 10 mM neomycin. Cells were mock-treated, stimulated for 10 min with 1 M insulin, or treated with 100 nM wortmannin prior to insulin stimulation. A, adipocytes were lysed and equal amounts of protein (10 g/lane) were resolved by SDS-PAGE followed by transfer to nitrocellulose for immunoblotting analysis using antibodies specific for PKB and serine 473-phosphorylated PKB. Shown is a representative blot. B, PKB and phosphoserine 473 PKB levels for each condition were quantified. Data shown are the mean Ϯ S.E. from three separate experiments. C, lysates were prepared and assayed for PKB activity as described under "Materials and Methods." Results are expressed as the mean Ϯ S.E. of three separate experiments performed in triplicate.

FIG. 3. Effect of neomycin on plasma membrane Glut4 levels.
3T3-L1 adipocytes were incubated for 2 h before and maintained throughout the experiment in the presence and absence of 10 mM neomycin. Cells were mock-treated, stimulated for 30 min with 1 M insulin, or treated with 100 nM wortmannin for 10 min prior to insulin stimulation. Plasma membranes lawns were prepared as described under "Materials and Methods" and subjected to indirect immunofluorescence using a polyclonal Glut4 antibody. A, representative fields for each condition are shown. B, averaged data from three separate experiments Ϯ S.E. (ϳ indicates p ϭ 0.05; * indicates p Ͻ 0.01).

Effects of Neomycin on Insulin Action 20569
location, the role of PIP 3 in this process is not clear. For example, overexpression of the pleckstrin homology domains of GRP-1 or PKB that bind specifically to PIP 3 do not prevent insulin-dependent Glut4 translocation and insertion into the plasma membrane (31,33). Furthermore, expression of a dominant-negative mutant of the lipid phosphatase PTEN that increased PIP 3 levels had no effect on either basal or insulinstimulated glucose transport (34).
In this study, we find that treatment of cells with the aminoglycoside, neomycin, prevents the wortmannin inhibition of insulin-stimulated glucose transport. Neomycin has this effect without influencing the phosphorylation state or activity of PKB, a downstream effector of PI3K, demonstrating that the effects of neomycin are not due to interference with wortmannin action. In addition to the effects on glucose transport, neomycin also prevented wortmannin inhibition of insulinstimulated Glut4 translocation. Interestingly, assessment of Glut4 translocation reveals that the insulin-dependent plasma membrane Glut4 content is increased in the presence of neomycin compared with insulin alone. This increase in insulindependent Glut4 content in the presence of neomycin is in contrast with the modest inhibition of insulin-stimulated glucose transport under the same conditions. This difference is also observed between the insulin-stimulated glucose transport rates and Glut4 translocation data from wortmannin-treated cells in the presence of neomycin. Such a difference may indicate that masking of PIP 2 with neomycin as well as stimulating Glut4 translocation also has a downstream effect on the activity of the glucose transporter. Nevertheless, the data are consistent in that neomycin prevents the wortmannin-induced inhibition of both insulin-stimulated glucose transport and Glut4 translocation.
It is interesting to speculate as to the mechanism by which neomycin elicits these effects on Glut4 translocation and glucose transport. Neomycin is characterized by an affinity to bind phosphoinositides in particular, PIP 2 (24,25). Binding of neomycin to PIP 2 results in the masking of this lipid (25,29). Taken together with our observations, it may be reasoned that the dynamic turnover of PIP 2 that takes place on activation of PI3K may be an important signal in stimulating glucose transport. It is important to spell out what dynamic turnover means; in this context we use it to describe a rapid change in the real time depletion and resynthesis of PIP 2 brought on by agonist stimulation. The idea that stimulus-dependent dynamic turnover of PIP 2 can affect exocytosis has recently been described for insulin secretion by MIN6 cells (35) and human growth hormone release in PC12 cells (36). In both these systems the involvement of ADP-ribosylation factor-6 in the maintenance of PIP 2 levels was important in exocytosis. More directly, other agonists like endothelin-1 and thrombin can elicit an increase in glucose transporter translocation to the plasma membrane and glucose transport, although not to the same extent as insulin (37)(38)(39). Both endothelin-1 and thrombin receptor agonism result in dramatic decreases in PIP 2 (40,41); in the case of endothelin-1 this is believed to involve PLC activation and subsequent cleavage of PIP 2 to yield IP 3 and diacylglycerol (40). The ability of neomycin to bind PIP 2 may mimic the depletion of PIP 2 , which together with other insulin-dependent signals may be sufficient to stimulate Glut4 translocation and glucose transport. Finally, this study suggests that insulin-stimulated Glut4 translocation can proceed in the absence of PKB activation.