Title: The Role of Ca2+ in Insulin-stimulated Glucose Transport in 3T3-L1 Cells
Abstract: We have examined the requirement for Ca2+ in the signaling and trafficking pathways involved in insulin-stimulated glucose uptake in 3T3-L1 adipocytes. Chelation of intracellular Ca2+, using 1,2-bis (o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetra (acetoxy- methyl) ester (BAPTA-AM), resulted in >95% inhibition of insulin-stimulated glucose uptake. The calmodulin antagonist, W13, inhibited insulin-stimulated glucose uptake by 60%. Both BAPTA-AM and W13 inhibited Akt phosphorylation by 70–75%. However, analysis of insulin-dose response curves indicated that this inhibition was not sufficient to explain the effects of BAPTA-AM and W13 on glucose uptake. BAPTA-AM inhibited insulin-stimulated translocation of GLUT4 by 50%, as determined by plasma membrane lawn assay and subcellular fractionation. In contrast, the insulin-stimulated appearance of HA-tagged GLUT4 at the cell surface, as measured by surface binding, was blocked by BAPTA-AM. While the ionophores A23187 or ionomycin prevented the inhibition of Akt phosphorylation and GLUT4 translocation by BAPTA-AM, they did not overcome the inhibition of glucose transport. Moreover, glucose uptake of cells pretreated with insulin followed by rapid cooling to 4 °C, to promote cell surface expression of GLUT4 and prevent subsequent endocytosis, was inhibited specifically by BAPTA-AM. This indicates that inhibition of glucose uptake by BAPTA-AM is independent of both trafficking and signal transduction. These data indicate that Ca2+ is involved in at least two different steps of the insulin-dependent recruitment of GLUT4 to the plasma membrane. One involves the translocation step. The second involves the fusion of GLUT4 vesicles with the plasma membrane. These data are consistent with the hypothesis that Ca2+/calmodulin plays a fundamental role in eukaryotic vesicle docking and fusion. Finally, BAPTA-AM may inhibit the activity of the facilitative transporters by binding directly to the transporter itself. We have examined the requirement for Ca2+ in the signaling and trafficking pathways involved in insulin-stimulated glucose uptake in 3T3-L1 adipocytes. Chelation of intracellular Ca2+, using 1,2-bis (o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetra (acetoxy- methyl) ester (BAPTA-AM), resulted in >95% inhibition of insulin-stimulated glucose uptake. The calmodulin antagonist, W13, inhibited insulin-stimulated glucose uptake by 60%. Both BAPTA-AM and W13 inhibited Akt phosphorylation by 70–75%. However, analysis of insulin-dose response curves indicated that this inhibition was not sufficient to explain the effects of BAPTA-AM and W13 on glucose uptake. BAPTA-AM inhibited insulin-stimulated translocation of GLUT4 by 50%, as determined by plasma membrane lawn assay and subcellular fractionation. In contrast, the insulin-stimulated appearance of HA-tagged GLUT4 at the cell surface, as measured by surface binding, was blocked by BAPTA-AM. While the ionophores A23187 or ionomycin prevented the inhibition of Akt phosphorylation and GLUT4 translocation by BAPTA-AM, they did not overcome the inhibition of glucose transport. Moreover, glucose uptake of cells pretreated with insulin followed by rapid cooling to 4 °C, to promote cell surface expression of GLUT4 and prevent subsequent endocytosis, was inhibited specifically by BAPTA-AM. This indicates that inhibition of glucose uptake by BAPTA-AM is independent of both trafficking and signal transduction. These data indicate that Ca2+ is involved in at least two different steps of the insulin-dependent recruitment of GLUT4 to the plasma membrane. One involves the translocation step. The second involves the fusion of GLUT4 vesicles with the plasma membrane. These data are consistent with the hypothesis that Ca2+/calmodulin plays a fundamental role in eukaryotic vesicle docking and fusion. Finally, BAPTA-AM may inhibit the activity of the facilitative transporters by binding directly to the transporter itself. insulin receptor insulin receptor substrate 1,2-bis (o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetra(acetoxymethyl)ester 2-deoxyglucose GLUT4 storage vesicle soluble NSF attachment protein vesicle membrane SNAP receptors target membrane SNAP receptors plasma membrane bovine serum albumin Krebs-Ringer phosphate hemagglutinin Chinese hamster ovary phosphatidylinositol polyacrylamide gel electrophoresis phosphate-buffered saline Insulin stimulates glucose uptake in skeletal muscle and adipose tissue by stimulating the translocation of a facilitative glucose transporter, GLUT4, from an intracellular compartment to the cell surface. In recent years considerable progress has been made in our understanding of the downstream signal transduction pathways that are activated by insulin to mediate the translocation of GLUT4 to the cell surface. Upon insulin binding, the activated insulin receptor (IR)1 tyrosine kinase phosphorylates a number of downstream substrates, most notably the insulin receptor substrate (IRS) family of proteins, including IRS-1 and IRS-2 (1White M.F. Mol. Cell. Biochem. 1998; 182: 3-11Crossref PubMed Scopus (622) Google Scholar). Tyrosyl-phosphorylated IRS-1 and IRS-2 serve as docking stations for SH2 domain-containing proteins such as the class Ia (p85/p110-type) PI 3-kinase (1White M.F. Mol. Cell. Biochem. 1998; 182: 3-11Crossref PubMed Scopus (622) Google Scholar). Activation of PI 3-kinase is essential for insulin-stimulated GLUT4 translocation and glucose uptake (2Shepherd P.R. Withers D.J. Siddle K. Biochem. J. 1998; 333: 471-490Crossref PubMed Scopus (829) Google Scholar) with generation of phosphoinositide 3,4,5-trisphosphate at the plasma membrane (PM) (3Oatey P.B. Venkateswarlu K. Williams A.G. Fletcher L.M. Foulstone E.J. Cullen P.J. Tavare J.M. Biochem. J. 1999; 344: 511-518Crossref PubMed Scopus (100) Google Scholar) serving to recruit and activate pleckstrin homology domain-containing proteins. Recent evidence indicates that the pleckstrin homology domain-containing Ser/Thr kinase Akt (otherwise called protein kinase B) plays a fundamental role in mediating insulin-stimulated GLUT4 translocation (4Kohn A.D. Summers S.A. Birnbaum M.J. Roth R.A. J. Biol. Chem. 1996; 271: 31372-31378Abstract Full Text Full Text PDF PubMed Scopus (1080) Google Scholar, 5Wang Q. Somwar R. Bilan P.J. Liu Z. Jin J. Woodgett J.R. Klip A. Mol. Cell. Biol. 1999; 19: 4008-4018Crossref PubMed Scopus (499) Google Scholar, 6Hill M.M. Clark S.F. Meerloo T. Tucker D. Birnbaum M.J. James D.E. Macaulay S.L. Mol. Cell. Biol. 1999; 19: 7771-7781Crossref PubMed Google Scholar). A direct link between the insulin-signaling cascade and the more distal events associated with GLUT4 trafficking is yet to be identified. The precise nature of the insulin-responsive GLUT4 storage vesicle (GSV) and a detailed molecular description of how insulin promotes translocation of the GSV to the PM remain to be defined. In contrast, the mechanism by which GSVs dock and fuse with the PM is better understood, in part because of the similarity with synaptic vesicle trafficking in neurons. Both of these events involve the pairing of protein complexes in the vesicle compartment (v-SNARES, forvesicle membrane SNAP receptors) with cognate receptor complexes at the target membrane (t-SNARES, fortarget membrane SNAP receptors). Interactions between v-SNARE and t-SNARE proteins, as well as additional accessory proteins, are responsible for formation of the core complex, which is required for membrane docking and fusion (7Mayer A. Curr. Opin. Cell Biol. 1999; 11: 447-452Crossref PubMed Scopus (78) Google Scholar). In adipocytes the core complex is comprised of the v-SNARE, VAMP2, and the t-SNAREs, syntaxin 4 and SNAP23 (8Rea S. James D. Diabetes. 1997; 46: 1667-1677Crossref PubMed Google Scholar). In view of the similarity in molecular regulation between GLUT4 translocation in adipocytes and synaptic vesicle exocytosis in neurons, it has been suggested that GLUT4 translocation may represent a form of regulated exocytosis. Most regulated exocytic processes share several characteristic features. These include segregation of the cargo to be transported, increased delivery of this cargo to the cell surface in response to secretagogue, and the involvement of Ca2+ in the delivery process. Although numerous studies have examined the role of Ca2+ in insulin-stimulated glucose transport, there remains little consensus concerning its overall role in this process. Investigations in L6 muscle cells, cardiac myocytes, and adipocytes failed to find a clear link between Ca2+ and glucose metabolism (9Klip A. Li G. Logan W.J. Am. J. Physiol. 1984; 247: E297-E304PubMed Google Scholar, 10Klip A. Ramlal T. J. Biol. Chem. 1987; 262: 9141-9146Abstract Full Text PDF PubMed Google Scholar, 11Kelly K.L. Deeney J.T. Corkey B.E. J. Biol. Chem. 1989; 264: 12754-12757Abstract Full Text PDF PubMed Google Scholar, 12Cheung J.Y. Constantine J.M. Bonventre J.V. Am. J. Physiol. 1987; 252: C163-C172Crossref PubMed Google Scholar). In contrast, reduction of intracellular Ca2+ in rat adipocytes markedly inhibited insulin-stimulated glucose transport (13Draznin B. Sussman K. Kao M. Lewis D. Sherman N. J. Biol. Chem. 1987; 262: 14385-14388Abstract Full Text PDF PubMed Google Scholar, 14Pershadsingh H.A. Shade D.L. Delfert D.M. McDonald J.M. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 1025-1029Crossref PubMed Scopus (58) Google Scholar, 15Khil L.Y. Cheon A.J. Chang T.S. Moon C.K. Biochem. Pharmacol. 1997; 54: 97-101Crossref PubMed Scopus (13) Google Scholar). Ca2+ may play a permissive role, or it may actively drive one or more of the steps involved in insulin-stimulated GLUT4 trafficking. In the latter case insulin may increase cytosolic Ca2+ by regulating the activity of a Ca2+channel. In the former case there may be no change in cytosolic Ca2+ with insulin stimulation. Intuitively one might imagine at least two loci where Ca2+ might be involved in mediating the effects of insulin on glucose uptake. Firstly, Ca2+/calmodulin has been implicated in mediating insulin activation of PI 3-kinase and Akt in rat hepatocytes and in 3T3-L1 adipocytes (16Benzeroual K. Pandey S.K. Srivastava A.K. van de Werve G. Haddad P.S. Biochim. Biophys. Acta. 2000; 1495: 14-23Crossref PubMed Scopus (19) Google Scholar, 17Yang C. Watson R.T. Elmendorf J.S. Sacks D.B. Pessin J.E. Mol. Endocrinol. 2000; 14: 317-326Crossref PubMed Scopus (45) Google Scholar). Secondly, several recent studies have reported a key role for Ca2+/calmodulin in the late stages of vesicle docking/fusion (18Peters C. Mayer A. Nature. 1998; 396: 575-580Crossref PubMed Scopus (322) Google Scholar, 19Porat A. Elazar Z. J. Biol. Chem. 2000; 275: 29233-29237Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 20Pryor P.R. Mullock B.M. Bright N.A. Gray S.R. Luzio J.P. J. Cell Biol. 2000; 149: 1053-1062Crossref PubMed Scopus (273) Google Scholar, 21Peters C. Bayer M.J. Buhler S. Andersen J.S. Mann M. Mayer A. Nature. 2001; 409: 581-588Crossref PubMed Scopus (421) Google Scholar). Thus, Ca2+ could be required both for the proximal signaling events of the insulin cascade and/or in the final stages of docking of GSVs with the plasma membrane. In the present investigation we have re-evaluated the role of Ca2+ in insulin-stimulated glucose transport in 3T3-L1 adipocytes. To do this we have employed the membrane permeable form of the Ca2+-chelating agent BAPTA-AM and the calmodulin antagonist W13. BAPTA-AM and W13 reduced insulin-stimulated glucose uptake by 95 and 60% and Akt phosphorylation by 75 and 70%, respectively. BAPTA-AM reduced GLUT4 translocation to the plasma membrane by 50% as determined by subcellular fractionation analyses. However, by using an antibody binding assay that measured insertion of glucose transporters into the membrane, we observed almost 100% inhibition of insulin-stimulated GLUT4 translocation in response to BAPTA-AM. We also found that treatment with ionophores prevented the inhibition of Akt phosphorylation and GLUT4 translocation by BAPTA-AM. These data indicate that Ca2+/calmodulin is required for the efficient activation of Akt and are consistent with an obligate role for Ca2+ at a late post-docking stage in GLUT4 vesicle fusion. All reagents were from Sigma unless specified otherwise. All tissue culture medium was purchased from Life Technologies Inc., except fetal calf serum, which was obtained from Trace Biosciences (Clayton, Australia). Bovine serum albumin was purchased from ICN (Costa Mesa, CA). Insulin was obtained from Calbiochem. The Ca2+ chelators, BAPTA-AM, BAPTA, and EGTA-AM, and ionophores, A23187 and ionomycin, were also from Calbiochem. The calmodulin antagonist W13 was from Sigma. The polyclonal GLUT4 and IRAP antibodies have been described previously (22James D.E. Strube M. Mueckler M. Nature. 1989; 338: 83-87Crossref PubMed Scopus (667) Google Scholar, 23Shewan A.M. Marsh B.J. Melvin D.R. Martin S. Gould G.W. James D.E. Biochem. J. 2000; 350: 99-107Crossref PubMed Scopus (84) Google Scholar). BCA reagent, used in protein assays, was from Pierce. The anti-phosphotyrosine monoclonal antibody (4G10) was provided by Dr. B. Druker (Oregon Health Sciences University, Portland, OR). Anti-IRS-1 polyclonal antibody was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Anti-p85 and anti-IRS-2 polyclonal antibodies were from Upstate Biotechnology Inc. (Lake Placid, NY). The monoclonal anti-influenza hemagglutinin (HA) epitope antibody was from Babco (Richmond, CA). Antiphospho-Ser473Akt antibody was from New England Biolabs (Beverly, MA). Fluorescein isothiocyanate-conjugated secondary antibodies were from Molecular Probes (Eugene, OR). Peroxidase-coupled secondary antibodies were from Amersham Pharmacia Biotech. 3T3-L1 fibroblasts were cultured and differentiated into adipocytes as described previously (24Piper R. Hess L.J. James D.E. Am. J. Physiol. 1991; 260: C570-C580Crossref PubMed Google Scholar). CHO cells stably overexpressing the IR (CHO.IR cells) were cultured as described previously (25Clark S.F. Martin S. Carozzi A.J. Hill M.M. James D.E. J. Cell Biol. 1998; 140: 1211-1225Crossref PubMed Scopus (159) Google Scholar). In all experiments cells were serum-starved in Krebs-Ringer phosphate (KRP) buffer (12.5 mm HEPES, pH 7.4, 120 mm NaCl, 6 mmKCl, 1.2 mm MgSO4, 1 mmCaCl2, 0.4 mm NaH2PO4, 0.6 mm Na2HPO4) supplemented with 0.2% bovine serum albumin (BSA) for at least 2 h at 37 °C, and all further treatments were performed in the same buffer except where insulin stimulation and 2-DOG uptake were carried out in KRP buffer without Ca2+. In these experiments cells were rinsed in prewarmed KRP buffer as described above, except that it was without CaCl2 and supplemented with 5 mm EGTA, and incubated in this buffer for the duration of insulin stimulation and 2-DOG uptake. Where indicated, cells were incubated with BAPTA-AM (50 µm, made up in Me2SO) for 10 min followed by insulin (1 µm) for 15 min at 37 °C in the continued presence of BAPTA-AM. Incubation of cells in KRP with 0.2% BSA supplemented with 3 mm pyruvate gave comparable results (data not shown). In other experiments, cells were incubated with W13 (70 µm (17Yang C. Watson R.T. Elmendorf J.S. Sacks D.B. Pessin J.E. Mol. Endocrinol. 2000; 14: 317-326Crossref PubMed Scopus (45) Google Scholar), made up in H2O) for 20 min followed by insulin (1 µm) for 15 min at 37 °C in the continued presence of W13. In experiments to investigate the effects of BAPTA-AM on glucose uptake post-insulin treatment, cells were incubated with insulin (1 µm) for 15 min at 37 °C. Following this, cells were rapidly cooled to 4 °C by washing with ice-cold KRP and incubated in the same buffer on ice. The cells were then incubated at 4 °C in the absence or the presence of BAPTA-AM, BAPTA, or EGTA-AM (all at 50 µm) or W13 (70 µm) for the times indicated, such that all cells were incubated at 4 °C for the same duration. In experiments involving the ionophores, A23187 or ionomycin (0.1 µm, made up in MeOH) was added simultaneously to the addition of BAPTA-AM. Treatment of cells with vehicle alone or in combination (at the appropriate final concentrations) was without effect in control experiments (data not shown). Following incubation with the appropriate agents, 3T3-L1 adipocytes were washed twice with ice-cold HES buffer (20 mm HEPES, pH 7.4, 1 mm EDTA, 250 mm sucrose) and homogenized in the same buffer supplemented with phosphatase and protease inhibitors (2 mm sodium orthovanadate, 10 mm sodium fluoride, 1 mmtetrasodium pyrophosphate, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 250 µm phenylmethylsulfonyl fluoride). Subcellular fractions were isolated by differential centrifugation as previously detailed (25Clark S.F. Martin S. Carozzi A.J. Hill M.M. James D.E. J. Cell Biol. 1998; 140: 1211-1225Crossref PubMed Scopus (159) Google Scholar) or by a modified protocol that gave comparable results. The modified protocol differed from that described (25Clark S.F. Martin S. Carozzi A.J. Hill M.M. James D.E. J. Cell Biol. 1998; 140: 1211-1225Crossref PubMed Scopus (159) Google Scholar) in the preparation of the PM fraction. In brief, centrifugation at 2,000 × g for 10 min was performed to remove mitochondria, nuclei, and unbroken cells. The resulting supernatant was then centrifuged at 18,000 × g for 20 min to pellet the crude PM fraction. This pellet was resuspended in HES buffer containing inhibitors and centrifuged again at 2,000 × g for 10 min to remove contaminating material. The supernatant from this was then centrifuged again at 18,000 × g for 20 min to pellet the PM fraction. The high speed pellet (otherwise termed low density microsomal fraction) was prepared from the supernatant from the first 18,000 × g spin as described (25Clark S.F. Martin S. Carozzi A.J. Hill M.M. James D.E. J. Cell Biol. 1998; 140: 1211-1225Crossref PubMed Scopus (159) Google Scholar). The protein content of all samples was determined using BCA reagent. The samples (10 µg) were subjected to SDS-PAGE and transferred to Immobilon-P polyvinylidene difluoride membranes (Millipore Corporation, Bedford, MA), and the membranes were probed with the appropriate primary and horseradish peroxidase-conjugated secondary antibodies. Antibody binding was detected by enhanced chemiluminescence according to the manufacturer's instructions (Supersignal, Pierce). The protein bands were quantified by densitometry (GS-700 Imaging densitometer, Bio-Rad) using nonsaturated exposed x-ray films. 2-Deoxy-[3H]glucose uptake was measured as described previously (26Clark S.F. Molero J.C. James D.E. J. Biol. Chem. 2000; 275: 3819-3826Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). In brief, 3T3-L1 adipocytes in 12-well plates were incubated in the absence or the presence of compounds, as indicated, in 500 µl of KRP with 0.2% BSA. The assay was initiated by adding 50 µl of 1 mm2-deoxy-[3H]glucose (20 µCi/mmol)/KRP and terminated after 1–2 min by washing cells rapidly three times with ice-cold PBS. The cells were solubilized in 1% Triton X-100, and 3H was quantitated by scintillation counting (Packard 1900CA liquid scintillation analyzer, Packard Instrument Co.). Glucose uptake was measured in duplicate in all treatments. Nonspecific uptake of 2-deoxy-[3H]glucose was determined by the addition of cytochalasin B (50 µm) to the appropriate controls prior to the commencement of assays. Measurement of the transport of the nonmetabolizable glucose analogue 3-O-[methyl-3H]glucose was performed essentially as described above except that the assay was terminated after 45 s by washing cells three times in ice-cold PBS containing phloretin (100 µm). The PM lawn assay was performed essentially as described (27Robinson L.J. Pang S. Harris D.S. Heuser J. James D.E. J. Cell Biol. 1992; 117: 1181-1196Crossref PubMed Scopus (254) Google Scholar). In brief, 3T3-L1 adipocytes grown on glass coverslips were treated as indicated and then sonicated using a probe sonicator (Kontes Co., Vineland, NJ) to generate a lawn of PM fragments attached to the coverslip. The coverslips were then incubated in GLUT4-specific antiserum, followed by incubation with fluorescein isothiocyanate-conjugated secondary antibody. Coverslips were washed with PBS, mounted onto glass microscope slides, and viewed using a 63×/1.4 Zeiss oil immersion objective on a Zeiss Axiovert fluorescence microscope equipped with a Bio-Rad MRC-600 laser confocal imaging system. Duplicate coverslips were prepared for each condition, and six random images of PM lawn were collected from each. The images were quantified using NIH 1.62 software. An exofacial HA epitope-tagged GLUT4 construct containing a single HA epitope in the first exofacial loop between transmembrane domains 1 and 2 (kindly provided by Dr. Michael Quon, National Institutes of Health, Bethesda, MD) was inserted into the retroviral expression vector pBabepuro and used to generate 3T3-L1 adipocytes stably expressing HA-GLUT4 as described (23Shewan A.M. Marsh B.J. Melvin D.R. Martin S. Gould G.W. James D.E. Biochem. J. 2000; 350: 99-107Crossref PubMed Scopus (84) Google Scholar). Following treatment of 3T3-L1 adipocytes stably expressing HA-GLUT4 cells were rinsed once in ice-cold PBS and then fixed in 2% paraformaldehyde with PBS for 15 min. Excess fixative was neutralized with 0.15 m glycine with PBS and blocked using 1% BSA with PBS for 30 min. The coverslips were incubated in anti-HA antibody (16B12) in 1% BSA with PBS for 1 h followed by a fluorescein isothiocyanate-conjugated secondary antibody in 1% BSA with PBS for 30 min. The coverslips were washed with PBS, mounted onto glass microscope slides, and viewed as described above. 3T3-L1 adipocytes expressing HA-GLUT4 were grown in 24-well plates. Following the appropriate treatment cells were washed twice in PBS and fixed in 2% paraformaldehyde with PBS for five min. The cells were blocked in 2.5% normal swine serum with PBS for 30 min and then incubated in anti-HA antibody (16B12 at 1:500) in 1% normal swine serum with PBS or 1% normal swine serum with PBS alone as a control for 60 min. After washing in 0.1% BSA with PBS (3 times for 10 min each time), the cells were incubated with anti-mouse horseradish peroxidase conjugate (1:5000) in 2.5% normal rat serum with PBS for 30 min. The cells were then washed in PBS (3 × 10 min) followed by incubation ino-phenylenediamine dihydrochloride reagent made up according to the manufacturer's instructions (Sigma) for 30 min in the dark. Finally, the A of the supernatant was read at 450 nm. The data were normalized to the insulin response detected in the same experiment and are expressed as percentages of the average insulin effect observed. The significance of various treatments was determined using the Student's ttest. For reasons of clarity, statistical significance, or lack thereof, between parameters is detailed only in situations pertinent to the discussion. It has been suggested that extracellular Ca2+ may play a role in insulin-stimulated glucose uptake in adipocytes (13Draznin B. Sussman K. Kao M. Lewis D. Sherman N. J. Biol. Chem. 1987; 262: 14385-14388Abstract Full Text PDF PubMed Google Scholar). Consistent with this we observed a 30% decrease in insulin-stimulated 2-DOG uptake when cells were incubated in Ca2+ free buffer supplemented with 5 mm EGTA (Fig. 1A). To further investigate the role of Ca2+ in this process we performed experiments using the membrane-permeable form of BAPTA, namely BAPTA-AM, which is freely taken up into cells where it is hydrolyzed by cytosolic esterases and trapped intracellularly as the active chelator BAPTA. This reagent exchanges Ca2+ more than 100 times faster than other agents such as EGTA, because of the faster rates of association and dissociation. Pretreatment of 3T3-L1 adipocytes with BAPTA-AM for 10 min resulted in a dose-dependent inhibition of insulin-stimulated 2-DOG uptake with an IC50 of 15 µm (Fig. 1 B). In all further experiments we used a BAPTA-AM concentration of 50 µm, at which we observed almost complete (>95%) inhibition of insulin-stimulated 2-DOG uptake (Fig. 1 C). BAPTA-AM also caused significant inhibition of basal 2-DOG uptake (Fig. 1 C). Identical results were obtained when cells were treated with BAPTA-AM in Ca2+-free buffer (basal, 10% ± 2; BAPTA-AM, 6% ± 1; insulin, 100%; insulin + BAPTA-AM, 9% ± 2; n = 4). Moreover, the nonesterified form of BAPTA or EGTA-AM had no significant effect on basal (control, 10% ± 2; BAPTA, 11% ± 3; EGTA-AM, 12% ± 2; n = 3) or insulin-stimulated 2-DOG uptake (control, 100%; BAPTA, 102% ± 5; EGTA-AM, 97% ± 3; n = 3). The inhibitory effect of BAPTA-AM did not involve an effect on intracellular ATP levels or hexokinase activity because it also inhibited insulin-stimulated transport of the nonmetabolizable glucose analogue 3-O-methylglucose (Fig. 1 D). We next examined the effects of BAPTA-AM on GLUT4 translocation to the cell surface using the PM lawn assay (Fig. 2). The morphology of the plasma membrane fragments was unaffected by BAPTA-AM. Insulin increased the level of GLUT4 at the PM by 4–5-fold. Pretreatment with BAPTA-AM inhibited insulin-stimulated GLUT4 levels in the PM lawns by ∼ 50%. It has previously been reported that BAPTA-AM has no effect on GLUT4 translocation using this assay (28Chen D. Elmendorf J.S. Olson A.L. Li X. Earp H.S. Pessin J.E. J. Biol. Chem. 1997; 272: 27401-27410Abstract Full Text Full Text PDF PubMed Scopus (139) Google Scholar). However, quantitation of GLUT4 in PM lawns was not performed in that study, in which case it is conceivable that a 50% inhibition may have been overlooked. To further investigate the apparent inhibition of insulin-stimulated GLUT4 translocation by BAPTA-AM, we examined its effects on insulin-stimulated GLUT4 translocation by subcellular fractionation using differential centrifugation and immunoblotting (Fig.3). In the absence of insulin very little GLUT4 was detected in the PM fraction obtained from basal cells. Consistent with the PM lawn data, insulin treatment resulted in a 5-fold increase in GLUT4 levels within the PM fraction (Fig. 3). Although BAPTA-AM alone had no significant effect on GLUT4 translocation, it caused a significant reduction (∼ 50%) in insulin-stimulated GLUT4 translocation. Moreover, insulin-stimulated translocation of the insulin responsive aminopeptidase (IRAP), which shows virtually identical trafficking properties to GLUT4 (29Garza L.A. Birnbaum M.J. J. Biol. Chem. 2000; 275: 2560-2567Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar), was similarly inhibited by BAPTA-AM (Fig. 3 A). The above data suggest that although BAPTA-AM caused almost quantitative inhibition of insulin-stimulated glucose uptake, this reagent inhibited GLUT4 translocation by only 50%. This discrepancy is unlikely to reflect a technical limitation of our ability to quantify GLUT4 translocation because we observed quantitatively similar results using two different fractionation techniques. In view of recent findings implicating a role for Ca2+ at a post-docking step in vesicle transport (7Mayer A. Curr. Opin. Cell Biol. 1999; 11: 447-452Crossref PubMed Scopus (78) Google Scholar), we reasoned that in the presence of BAPTA-AM, GSVs may dock at the PM but be blocked in their ability to fuse with the cell surface. Such docked vesicles may remain attached to the PM during preparation of PM fractions by the lawn technique or by subcellular fractionation, but because they have not fused with the cell surface they may not contribute to glucose entry into the cell. To examine this possibility we developed a surface binding assay utilizing 3T3-L1 adipocytes expressing an exofacial tagged HA-GLUT4 construct (23Shewan A.M. Marsh B.J. Melvin D.R. Martin S. Gould G.W. James D.E. Biochem. J. 2000; 350: 99-107Crossref PubMed Scopus (84) Google Scholar). This assay will only detect GLUT4 if it has inserted into the cell surface lipid bilayer, thus providing an estimate of vesicle docking and fusion. In the absence of insulin, we observed no detectable labeling of the cell surface using the anti-HA antibody in cells expressing HA-GLUT4 (Fig.4A). In insulin-treated cells we observed a marked increase in surface labeling of most cells in the culture (Fig. 4 A). Pretreatment with BAPTA-AM resulted in complete inhibition of insulin-stimulated HA-GLUT4 translocation to the cell surface (Fig. 4 A). To obtain more quantitative data we performed further experiments where the cell surface expression of HA-GLUT4 was quantified using a colorimetric assay (Fig. 4 B) (30Wang Q. Khayat Z. Kishi K. Ebina Y. Klip A. FEBS Lett. 1998; 427: 193-197Crossref PubMed Scopus (184) Google Scholar). Consistent with the immunofluorescence data, pretreatment with BAPTA-AM resulted in an almost total inhibition of insulin-stimulated cell surface expression of HA-GLUT4. These data implicate a role for Ca2+ in the fusion of GSVs with the plasma membrane and potentially resolve the discrepancy between the effects of BAPTA-AM on glucose transport and GLUT4 translocation as measured by the PM lawn technique or differential centrifugation (Figs. Figure 1, Figure 2, Figure 3). The above data implicate a role for Ca2+ at two different stages of the GLUT4 translocation process. Firstly, BAPTA-AM was shown to inhibit the translocation of GLUT4 to the cell surface by ∼50%. Secondly, BAPTA-AM blocked the apparent insertion of GLUT4 into the plasma membrane. Collectively, these two effects may account for the almost complete inhibition of insulin-stimulated glucose transport by BAPTA-AM. The first effect of BAPTA-AM may involve a limitation in the availability of docking or fusion sites at the plasma membrane or an additional inhibitory effect of BAPTA-AM at a step between insulin binding to its receptor and that of GLUT4 docking at the PM. Pretreatment with BAPTA-AM had no significant effect on insulin-stimulated t