If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported in part by National Institutes of Health Grant DK33651, by the Veterans Administration Medical Research Service, by a mentor-based fellowship grant from the American Diabetes Association (to T. H.), and by National Institutes of Health Research Fellowship DK09415 (to A. J. M.). The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. § Both authors contributed equally to this work. ¶ Participant in the Biomedical Sciences Doctoral Program. Lucille P. Markey predoctoral fellow. ∥ Participant in the Whittier Diabetes Program.
Insulin stimulation of 3T3-L1 adipocytes causes rapid translocation of actin and the GLUT4 glucose transporter to the plasma membrane. Both processes depend on the activity of phosphatidylinositol 3-kinase. Using single cell microinjection, we have transiently expressed a constitutively activated mutant of phosphatidylinositol 3-kinase, p110*, in 3T3-L1 adipocytes. Fluorescent detection of GLUT4 protein and actin within these cells demonstrates that expression of p110* is sufficient to cause translocation of GLUT4 to the plasma membrane and the formation of actin membrane ruffles. These effects are inhibited by wortmannin in the p110*-expressing cells, indicating that the phosphatidylinositol 3-kinase activity of the protein is required. Overexpression of an identical protein containing a point mutation in the kinase domain, p110*Δkin, was incapable of mediating either action, confirming that neither the microinjection process nor a nonspecific effect of the protein was responsible for the observed effects. These data suggest that although insulin is capable of inducing numerous signaling pathways, the isolated activation of phosphatidylinositol 3-kinase can initiate the signaling cascade leading to both actin rearrangement and GLUT4 translocation in the absence of insulin stimulation.
Insulin regulates plasma glucose levels primarily through stimulation of glucose uptake into target tissues and suppression of hepatic glucose production (
). Despite the importance of this biological effect, the signaling mechanisms leading to insulin-stimulated GLUT4 translocation remain poorly defined. Recent work has demonstrated that this event is dependent on insulin-stimulated phosphatidylinositol 3-kinase (PI3K)
Recent reports demonstrate that a constitutively active mutant of PI3K (p110*), which is composed of an amino-terminal fusion of the p110 catalytic subunit with the iSH2 activation domain of p85, is capable of stimulating PI3K-dependent gene expression in the absence of added growth factors (
). Using single cell microinjection of an expression vector for p110* and fluorescence microscopy, we have studied the direct role of PI3K activation on actin rearrangement and GLUT4 translocation in 3T3-L1 adipocytes.
Porcine insulin was kindly provided by Lilly. Rabbit polyclonal GLUT4 antibody (F349) and construction of the CMV expression vectors for p110* and p110*Δkin was as described previously (
). Sheep IgG, AMCA-conjugated anti-sheep, and fluorescein isothiocyanate-conjugated anti-rabbit antibodies were from Jackson Immunoresearch Laboratories, Inc. Tetramethylrhodamine-conjugated phalloidin, wortmannin, and all other reagents were purchased from Sigma.
Cell Culture and Microinjection
3T3-L1 cells were maintained, differentiated into adipocytes, and reseeded onto glass coverslips as described previously (
). Prior to microinjection, cells were rendered quiescent by serum deprivation for 18-24 h. Individual nuclei of 3T3-L1 adipocytes were then injected with either CMV-p110* or CMV-p110*Δkin (0.2 mg/ml) in microinjection buffer (5 mM sodium phosphate, pH 7.2, and 100 mM KCl) mixed with preimmune sheep IgG (10 mg/ml) to allow the detection of injected cells. Five to six hours following injection, cells were fixed with 3.7% formaldehyde for 10 min. For wortmannin treatment of CMV-p110*-injected cells, 1 µM wortmannin was added to the media 1 h prior to fixation. Control cells (basal and insulin-stimulated) were mock injected with microinjection buffer containing sheep IgG alone and stimulated with 100 ng/ml insulin 15 min prior to formaldehyde fixation.
Immunostaining and Fluorescence Microscopy
GLUT4 Protein Staining
Formaldehyde-fixed cells were washed in PBS for 10 min and then permeabilized and blocked with 0.1% Triton X-100 and 1% fetal calf serum in PBS for 10 min. Cells were then incubated overnight at 4°C with polyclonal rabbit anti-GLUT4 antibody (F349, 1 µg/ml), which was diluted in PBS with 1% fetal calf serum. After washing with PBS for 10 min, cells were incubated with fluorescein-conjugated donkey anti-rabbit antibody (1:100) to detect F349 and AMCA-conjugated anti-sheep antibody (1:100) to detect injected cells.
Cells were washed and permeabilized as above and then incubated in PBS with rhodamine-phalloidin (0.5 µg/ml) to visualize the location of polymerized actin (membrane ruffles) and AMCA-conjugated donkey anti-sheep antibody to detect injected cells.
Slides were analyzed on a Zeiss Axiophot fluorescence microscope. Cells that displayed AMCA-stained nuclei (coinjected sheep IgG) were scored for the presence of plasma membrane-associated GLUT4 staining or actin membrane ruffles. The percentage of injected cells displaying each phenotype is represented as GLUT4 translocation or ruffling index, respectively.
RESULTS AND DISCUSSION
Insulin stimulates translocation of the GLUT4 glucose transporter from an endosomal compartment within the cell to the plasma membrane of differentiated 3T3-L1 adipocytes (
) have reported evidence indicating that activation of PI3K is a necessary step in this insulin bioeffect. To determine whether activation of PI3K is sufficient for GLUT4 translocation, an expression vector (
) for a constitutively active form of PI3K (p110*) driven by the cytomegalovirus promoter (CMV-p110*) was microinjected along with preimmune sheep IgG into the nucleus of differentiated, serum-starved 3T3-L1 adipocytes in the absence of insulin. Five to six hours following injection, cells were formaldehyde-fixed and stained immunofluorescently for both coinjected sheep IgG and GLUT4 protein. Injected cells were identified by the presence of nuclear staining for sheep IgG (Fig. 1C). Immunofluorescent staining of the same cells for GLUT4 demonstrates that introduction of the p110* expression vector leads to recruitment of GLUT4 to the cell surface (Fig. 1D).
Quantitation of cells staining positive for GLUT4 protein at the plasma membrane showed that insulin induces a ∼5-fold stimulation of GLUT4 translocation, while cells injected with CMV-p110* exhibit nearly 75% of this response in the absence of insulin (Fig. 2). Treatment of CMV-p110*-injected cells with wortmannin reduced this effect, indicating that the ability of CMV-p110* to induce GLUT4 translocation depends on stimulation of PI3K catalytic activity. Furthermore, microinjection of an identical CMV-driven expression vector with a point mutation in the ATP-binding site of the kinase domain (CMV-p110*Δkin) (
) had no effect on GLUT4 translocation (Fig. 2), confirming that this response was not a nonspecific result of protein overexpression or the nuclear microinjection process.
Insulin also stimulates the rapid localization of polymerized actin to the cell surface in 3T3-L1 adipocytes (Fig. 3, A and B). This effect becomes maximal approximately 15 min following administration of insulin, which correlates closely with the early time course of GLUT4 translocation (
). In the current experiments, we investigated whether PI3K was also sufficient to mediate this process by using the CMV-p110* microinjection approach. Similar to the findings for GLUT4 translocation, cells that were nuclear injected with the CMV-p110* (Fig. 3C) displayed surface localization of polymerized actin in membrane ruffling structures (Fig. 3D). Likewise, this activity of the p110* was inhibited by co-treatment with wortmannin, and expression of the p110*Δkin had no detectable effect (Fig. 4). These data demonstrate that PI3K activation is capable of inducing both actin rearrangement and GLUT4 translocation in the absence of insulin stimulation.
Insulin-stimulated membrane ruffling and GLUT4 translocation are temporally correlated (Fig. 1, Fig. 2, Fig. 3) and also follow a similar early signaling pathway that is PI3K-dependent and Ras-independent (
), so it seems possible that insulin-stimulated PI3K activation could cause GLUT4 translocation via a Rac-dependent mechanism. On the other hand, recent evidence suggests that the Rac protein is not required or sufficient to induce glucose uptake into cells (
). However, translocation of GLUT4 to the plasma membrane may not be sufficient to cause glucose transport as several reports indicate that GLUT4 translocation can occur independently of glucose uptake induction (
). Therefore, Rac may be involved in GLUT4 translocation without being sufficient to induce glucose uptake in the absence of insulin. It is possible that additional insulin-stimulated events are required to induce maximal glucose transport, independent of and in addition to, GLUT4 translocation.
We demonstrate that actin rearrangement and GLUT4 translocation are both induced by constitutive activation of PI3K. However, it is important to recognize that the constitutively active p110* is likely incapable of p85-directed localization within the cell due to intramolecular association with the p85 iSH2 domain at its amino terminus. Recent studies indicate that PI3K is localized to GLUT4 vesicles following insulin stimulation (
). It is possible that targeting of PI3K to different intracellular compartments may be important for the two distinct responses (membrane ruffling and GLUT4 vesicle translocation), and this hypothesis remains to be tested. The ability of insulin to induce both actin rearrangement and GLUT4 translocation appears to be dependent solely on induction of a signaling cascade by PI3K, since transient expression of constitutively active PI3K is sufficient to mediate these responses in the absence of insulin.
We thank Jay Nelson and Peter Vollenweider for assistance in the culturing of the 3T3-L1 adipocyte cell line and David Rose for many helpful discussions and critical comments on the experiments performed.
Ellenberg and Rifkin's Diabetes Mellitus. 4th Ed. Elsevier Science Publishing Co., Inc.,
New York1994: 121