The p38 Mitogen-Activated Protein Kinase Inhibitor SB203580 Reduces Glucose Turnover by the Glucose Transporter-4 of 3T3-L1 Adipocytes in th
http://www.100md.com
内分泌学杂志 2005年第4期
Department of Molecular Cell Biology, Leiden University Medical Center, 2333 AL Leiden, The Netherlands
Address all correspondence and requests for reprints to: Dr. J. Antonie Maassen, Signal Transduction Laboratory, Department of Molecular Cell Biology, Leiden University Medical Center, Wassenaarseweg 72, P.O. Box 9503, 2333 AL Leiden, The Netherlands. E-mail: j.a.maassen@lumc.nl.
Abstract
Insulin induces a profound increase in glucose uptake in 3T3-L1 adipocytes through the activity of the glucose transporter-4 (GLUT4). Apart from GLUT4 translocation toward the plasma membrane, there is also an insulin-induced p38 MAPK-dependent step involved in the regulation of glucose uptake. Consequently, treatment with the p38 MAPK inhibitor SB203580 reduces insulin-induced glucose uptake by approximately 30%. Pretreatment with SB203580 does not alter the apparent Km of GLUT4-mediated glucose uptake but reduces the maximum velocity by approximately 30%. Insulin-induced GLUT4 translocation and exposure of the transporter to the extracellular environment was not altered by pretreatment with SB203580, as evidenced by a lack of effect of the inhibitor on the amount of GLUT4 present in the plasma membrane, as assessed by subcellular fractionation, the amount of GLUT4 that is able to undergo biotinylation on intact adipocytes and the level of extracellular exposure of an ectopically expressed GLUT-green fluorescence protein construct with a hemagglutinin tag in its first extracellular loop. In contrast, labeling of GLUT4 after insulin stimulation by a membrane-impermeable, mannose moiety-containing, photoaffinity-labeling agent [2-N-4(1-azido-2,2,2-trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-propylamine] that binds to the extracellular glucose acceptor domain was markedly reduced by SB203580, although photolabeling with this compound in the absence of insulin was unaffected by SB203580. These data suggest that SB203580 affects glucose turnover by the insulin-responsive GLUT4 transporter in 3T3-L1 adipocytes.
Introduction
ONE OF THE main functions of insulin in the body is maintaining blood glucose homeostasis. To achieve this, insulin induces glucose uptake in responsive tissues such as muscle and adipocytes (1). Stimulated glucose uptake in these cells is mediated primarily by translocation of the insulin-responsive glucose transporter-4 (GLUT4) toward the plasma membrane (PM), involving the contribution of both phosphotidylinositol 3'-kinase and Cbl signaling pathways emanating from the activated insulin receptor (2). After translocation toward the PM, there is a secondary, p38 MAPK-dependent step, leading to enhancement of insulin-induced glucose uptake (3, 4). Consequently, pretreatment of adipocytes with the pharmacological inhibitor SB203580 against the p38 MAPK- and -? isoforms reduces the amount of insulin-induced glucose uptake by roughly 30% (5). This phenomenon may be of importance in a more physiological setting of insulin resistance as well. Previously we reported that in 3T3-L1 adipocytes, treatment with the glucocorticoid dexamethasone induces an insulin-resistant state through up-regulation of the MAPK phosphatases-1 and -4, resulting in p38 MAPK dephosphorylation and concomitantly a reduction in glucose uptake (6).
Compared with the insulin-induced state (Fig. 1A), there are several scenarios that could in theory explain the deleterious effects of SB203580 treatment on insulin-induced glucose uptake: 1) an alteration of the glucose flux (or rate of glucose transfer) through the GLUT4 transporter across the PM (Fig. 1B) (7, 8), 2) a hampering of the transition of the GLUT4 transporter through its occluded (PM inserted, but not fully exposed to the extracellular milieu) state to a fully exposed state (Fig. 1C) (9, 10), or 3) a glucose turnover defect, for example due to an alteration in the oscillation between outward (sugar accepting from the extracellular environment) and inward (sugar release into the cytosol) facing conformations of the GLUT4 transporter (Fig. 1D) (11, 12).
FIG. 1. Theoretical models on the effects of SB203580 on insulin-induced glucose uptake in 3T3-L1 adipocytes. A, Insulin stimulation results in an increase in the exposure of GLUT4 to the extracellular milieu (depicted by open spheres in the PM). As a consequence, 2-DOG uptake is maximally increased (four arrows) and the GLUT4 are accessible to photoaffinity labeling as well as to cell surface biotinylation or antibody binding to an exofacial HA tag. SB203580 reduces insulin-induced glucose uptake (indicated by two arrows). B, If the glucose flux through the transporter is reduced, only an effect on glucose uptake will occur. Photoaffinity label is not transported across the PM and hence is insensitive to alterations in the transfer speed (two arrows, as in A). C, When transition through the occluded state is altered, binding of photoaffinity label is reduced, as is the efficiency of cell surface biotinylation and accessibility of the HA tag (one arrow each). D, When SB203580 induces a turnover defect, a reduction in photoaffinity label is expected. No alterations should be observed in either biotinylation or accessibility of the HA tag, however (two arrows, as in A), because the GLUT4 itself is fully exposed to the extracellular milieu.
As can be readily appreciated from Fig. 1, when challenged with different experimental approaches to analyze GLUT4 translocation and activity, we should be able to discriminate among these models. In this manuscript, we describe such an analysis of the effects of the p38 MAPK inhibitor SB203580 on glucose uptake by the adipocyte.
Materials and Methods
Materials
DMEM was purchased from Invitrogen Life Technologies, Inc. (Gaithersburg, MD); fetal calf serum was obtained from Brunschwig (Amsterdam, The Netherlands; catalog no. A15-043, lot A01127-318); dexamethasone, 1-methyl-3-isobutylxanthine, bovine insulin, and 2-deoxy-D-glucose were obtained from Sigma-Aldrich Corp. (St. Louis, MO); 2-deoxy-D-[14C]glucose and 125I-labeled secondary antibody were purchased from Amersham Biosciences (Little Chalfont, UK). SB203580 was obtained from Promega Corp. (Madison, WI). Monoclonal antibodies against the hemagglutinin (HA) tag were purchased from Abcam (Cambridge, MA). Cy3-conjugated secondary antibodies were purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Goat polyclonal anti-GLUT4 (C-20), rabbit polyclonal against HA tag, and horseradish peroxidate-conjugated donkey antigoat secondary antibody were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Biotin LC hydrazide and sodium meta-periodate were obtained from Perbio (Erembodegem, Belgium).
Cell culture
3T3-L1 fibroblasts were obtained from American Type Culture Collection (Manassas, VA) and differentiated to adipocytes as previously described (13). Mature adipocytes were routinely used 7–14 d after completion of the differentiation process. Only cultures in which more than 95% of the cells displayed adipocyte morphology were used.
Assay of 2-deoxyglucose uptake
3T3-L1 adipocytes, grown in 12-well plates (Costar, Cambridge, MA), were subjected to an assay of 2-deoxy-D-[14C]glucose (0.075 μCi/well) uptake as described previously (14).
Biotinylation
After treatment, cells were washed twice with PBS and subjected to 30-min oxidation by 20 mM sodium-meta-periodate at 4 C in 0.1 M sodium acetate, pH 5.5. Subsequently, cells were washed three times with PBS/45 mM glycerol and twice with 0.1 M 2-N-(morpholino)ethanesulfonic acid (pH 5.0). The biotinylation reaction using 50 mM biotin LC hydrazide was performed in 0.1 M 2-N-(morpholino)ethanesulfonic acid, pH 5.0, for 2 h at 4 C. Three wash steps with PBS/15 mM glycine were used to quench the reaction, and samples were subjected to subcellular fractionation.
Subcellular fractionation
After treatment, cells were washed twice with PBS on ice and scraped in HES buffer [20 mM HEPES (pH 7.4), 1 mM EDTA, and 250 mM sucrose] in the presence of protease inhibitors. Samples were homogenized nine times by three strokes in a Potter homogenizer (Fischer Scientific, Zoetermeer, The Netherlands), after which low density microsomal vesicle (LDM) and PM were isolated by differential centrifugation as described by Simpson et al. (15). Equal amounts of protein (10 μg), as determined using the bicinchoninic acid protein assay reagent (Pierce Chemical Co., Rockford, IL) were subjected to immunoblot analysis.
Photoaffinity labeling
The photoaffinity-labeling reaction using 2-N-4(1-azido-2,2,2-trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-propylamine (ATB-BMPA) was performed as described previously (16, 17). Briefly, samples were treated using a 5-min irradiation time with seven lamps (Southern New England Ultra Violet Co., Branford, CT) at 300 nm (35 watts) in a Rayonet RPR-200 photoincubator in the continued presence of SB203580. After two washes with PBS, cells were lysed in a Nonidet P-40-based buffer [1 mM Na3VO4, 1 mM EGTA, 1 mM EDTA, 50 mM Tris-HCl (pH 7.4), 1% (wt/vol) Nonidet P-40, 0.5% (wt/vol) sodium deoxycholate, 150 mM NaCl, and 5 mM sodium fluoride] in the presence of protease inhibitors (Complete, Roche, Indianapolis, IN), and material was collected using streptavidin bead precipitation.
Lentivirus-mediated HA tag GLUT4-green fluorescence protein (GFP) transduction in mature 3T3-L1 adipocytes
The GLUT4 construct harboring an HA tag in its first extracellular loop and a C-terminal GFP has been described previously (18, 19). This HA-GLUT4-GFP fusion construct was cloned into a lentiviral backbone under the control of a phosphoglycerate kinase promoter sequence (pRRL-phoshoglycerate kinase). Lentivirus vectors were raised and used to infect mature 3T3-L1 adipocytes, as previously described (20). One week after infection, adipocytes were washed twice with PBS. Adipocytes were fixed using 3.7% formaldehyde in PBS, washed four times with PBS/0.2% BSA, incubated for 30 min in PBS/0.2% BSA, and treated with anti-HA antibodies (1:200 in PBS/0.2% BSA) for 2 h. Plates were washed extensively in PBS/0.2% BSA and incubated with the relevant secondary antibody [Cy3 conjugated (1:100) for immunofluorescence and 125I radiolabeled (0.1 μCi) for quantification] for 1 h. Subsequently, cells were washed six times for 5 min each time with PBS and once with H2O. For immunofluorescence, cells were mounted using Vectashield (Brunschwig) and viewed with a DM-IRBE (Leica, Deerfield, IL). To quantify 125I, cells were lysed overnight in 0.1% sodium dodecyl sulfate/NaOH (0.1 M) and analyzed in a Tri-Carb scintillation counter (Packard, Downers Grove, IL).
Statistical analysis and graph generation
Statistical analysis of the data obtained was performed with an independent-sample t test using SPSS 10.0 (SPSS, Inc., Chicago, IL). Graphs were generated using PRISM 2.01 (GraphPad Software, Inc., San Diego, CA).
Subcellular fractionations subjected to immunoblot analysis were also exposed to a LumiImager (Roche) and quantified using LumiAnalyst software, yielding the amount of signal in Boehringer light units (BLU, an arbitrary unit). Subsequently, data were corrected for protein content and expressed as a relative fraction of GLUT4 residing in either the intracellular LDM fraction or the plasma membrane fraction. Thus, graph data are: LDM [BLU/mg]/(LDM [BLU/mg] + PM [BLU/mg]) and are similar for PM.
With respect to the photoaffinity label, BLU data for the biotin signal were expressed as fold/basal (treated [BLU/mg]/basal [BLU/mg]) and divided by the relative amount of GLUT4 signal in the PM in fold over basal (treated [BLU/mg]/basal [BLU/mg], giving an indication of the amount of photolabel/amount of GLUT4 transporters.
Results
Effects of SB203580 on insulin-induced glucose uptake
The p38 MAPK inhibitor SB203580 was added at several distinct time points, with respect to the time at which the radiolabeled 2-deoxy-D-glucose (2-DOG) was added, in an assay of insulin-induced glucose uptake. These are 15 min before insulin stimulation, concomitant with insulin stimulation, and concomitant with radiolabeled 2-DOG. As shown in Fig. 2A, SB203580 caused a significant reduction of about 30% in insulin-stimulated glucose uptake only when added before the radiolabel. In these experiments, basal levels of 2-DOG uptake were 4.5 ± 0.46 pmol/min·mg protein and were unaltered by similar SB203580 treatments. These observations are in agreement with previous observations (5) and exclude a direct interference of the SB203580 compound with the process of glucose uptake itself.
FIG. 2. A kinetic analysis of the effects of SB203580 on insulin- induced glucose uptake in 3T3-L1 adipocytes. 3T3-L1 adipocytes were stimulated with 100 nM insulin for 15 min and assayed for [14C]2-DOG uptake. A, Adipocytes were treated with 10 μM SB203580 for 15 min before insulin stimulation, concomitant with the insulin stimulation, or concomitant with the addition of radiolabeled 2-DOG or were not treated, as indicated. In these experiments, basal levels of 2-DOG uptake were 4.5 ± 0.46 pmol/min·mg protein and were unaltered by similar SB203580 treatments. The data shown are the mean of three independent experiments, each performed in triplicate, ± SEM. *, P < 0.05 compared with uninhibited, insulin-stimulated cells. B, Insulin-stimulated glucose uptake was measured by employing a range of 2-DOG concentrations, as indicated. Adipocytes were treated with 10 μM SB203580 for 15 min before insulin stimulation. The data shown are the mean of two independent experiment, each performed in triplicate, ± SEM.
Subsequently, the effects of SB203580 on insulin-induced glucose uptake in 3T3-L1 adipocytes were also characterized by performing a kinetic analysis. As shown in Fig. 2B, pretreatment with 10 μM SB203580 reduced the maximum velocity of insulin-induced glucose uptake from 1.0 nmol/min·mg protein to 0.68 nmol. Meanwhile, the insulin-stimulated apparent Km for glucose uptake was 2.2 and 2.0 mM for untreated cells and cells pretreated with SB203580, respectively. These data are similar to those observed in L6 muscle cells stably transfected with an Myc-tagged, GLUT4-expressing construct (21). Inhibition of p38 MAPK activity in an in vitro kinase assay at a concentration of 0.5 μM SB203580 led to an inhibition of about 50% of maximum, in agreement with the 50% inhibitory concentration reported by the manufacturer (Merck & Co., Rahway, NJ) for SB203580 on p38 of 0.6 μM (data not shown).
Effects of SB203580 on insulin-induced GLUT4 translocation
We analyzed the effects of SB203580 on insulin-induced GLUT4 translocation using several different techniques. As shown in Fig. 3 and in support of the literature (5), when employing subcellular fractionation separating intracellular LDM vesicles from the PM by differential centrifugation, SB203580 did not affect insulin-induced GLUT4 translocation toward the PM.
FIG. 3. Effects of SB203580 on insulin-induced GLUT4 translocation. Adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Cells were subjected to subcellular fractionation, and equal amounts of protein (10 μg) were subjected to immunoblot analysis. A, Immunoblots were quantified on a LumiImager and expressed as relative amounts of GLUT4 resident in either the PM or LDM fraction. Data shown are the mean of an experiment, performed in triplicate, ± SEM. Importantly, the amount of GLUT4 observed in the high density microsomal vesicle fraction did not alter significantly in any of the conditions applied. *, P < 0.05 compared with basal levels. B, The top frame shows a representative immunoblot of the subcellular fractionation used to obtain the data shown in A. The bottom frame shows representative immunoblot data from an experiment performed in triplicate when probed with an antibiotin antibody of subcellular fractionation samples performed on 3T3-L1 adipocytes subjected to the biotinylation assay.
Subsequently, we analyzed the exposure of GLUT4 at the surface of the PM. As can be observed from the immunoblot data in Fig. 3B, GLUT4 runs as a smear around 45 kDa rather than as a discrete band due to heavy glycosylation. Treatment of the intact adipocyte with periodate after insulin stimulation causes oxidation of exposed sugar side-chains, allowing subsequent reaction of the resulting sugar aldehydes with the biotinylating reagent. When performing this procedure, no effect of SB203580 on the insulin-induced translocation and exposure to the extracellular milieu was observed (see Fig. 3B). Quantification of these bands resulted in a 2-fold over basal induction by insulin (or insulin with SB203580). However, this quantification should be interpreted with caution, because the procedure enhanced the smearing of the GLUT4 band and concomitantly reduced the signal to noise ratios considerably (see Fig. 3B). Thus, as an alternative to the biotinylation procedure and to obtain a more reliable quantification, 3T3-L1 adipocytes were transduced with a lentivirus vector expressing HA-tagged GLUT4-GFP. The HA tag present in this complex is localized in the first extracellular loop of the GLUT4 transporter (19). Hence, only when fully inserted in the PM will this tag be accessible to anti-HA antibodies supplied in the extracellular milieu. As can be observed in the immunofluorescence data in Fig. 4A, in the basal situation scarcely any extracellular HA tag signal was observed, whereas the GFP signal was readily detectable. Insulin induced a profound translocation of the tagged GLUT4 transporters toward the PM and concomitant extracellular HA tag exposure (see Fig. 4B). Pretreatment with SB203580 had no marked effect on intracellular localization of the tagged GLUT4 transporters in the basal state or on PM exposure after insulin stimulation compared with untreated adipocytes (Fig. 4, C and D). Although lentivirus-mediated infection results typically in 80% infected 3T3-L1 adipocytes, in Fig. 4 we selected less densely transfected areas to visualize the translocation events more clearly. Using a similar technique, but employing a radiolabeled secondary antibody against the HA tag antibody, we were able to quantify the single cell data presented in Fig. 4 in the whole well (containing 50,000 adipocytes). In the basal state, on the average, 1524 ± 8.7 cpm radiolabeled secondary antibody were measured per well. Importantly, mock transfection using a lentivirus expressing only GFP resulted in an average of 974 ± 28 cpm/well without any significant effect of insulin and/or SB203580 treatment. As shown in Fig. 5, insulin induced a 3.1 ± 0.32-fold increase in the amount of HA tag exposure over basal levels, as measured by this approach. Pretreatment of SB203580 did not significantly affect the level of exposure of the HA tag in either the basal (1.1 ± 0.20-fold) or insulin-stimulated (3.4 ± 0.35-fold) state. These data are comparable with those generated using Myc-tagged GLUT4 in stably transfected L6 myotubes (21) and brown adipocytes derived of GLUT4 Myc-expressing mice (22).
FIG. 4. Immunofluorescence data of 3T3-L1 adipocytes transduced with GFP-GLUT4-HA tag-expressing lentivirus vectors. Transduced adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Adipocytes were fixed and probed with an anti-HA antibody, followed by a Cy3-conjugated secondary antibody (red). The GFP signal of the GLUT4 chimera is in green; yellow indicates an overlap of the GFP-derived GLUT4 signal and the extracellular exposure of the HA tag. Data shown are from cells selected from an experiment, performed in duplicate. For clarity, low density transduced areas (without transduced adipocytes joining each other) were chosen to illustrate the individual cell response to insulin more clearly.
FIG. 5. HA tag quantification data of 3T3-L1 adipocytes transduced with GFP-GLUT4-HA tag-expressing lentivirus vectors. Transduced adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Adipocytes were fixed and probed with an anti-HA antibody, followed by a 125I- radiolabeled secondary antibody. Data shown are depicted as the fold increase over the basal value (mean ± SEM of two experiments, each performed in triplicate). *, P < 0.05 compared with the basal state, the difference between insulin and insulin plus pretreatment with SB203580 was not significant. In the basal state, on the average, 1524 ± 8.7 cpm radiolabeled secondary antibody were measured per well. Each well contains approximately 5.0 x 104 adipocytes.
Effects of SB203580 on insulin-induced photoaffinity labeling with ATB-BMPA
Subsequently, we performed a photoaffinity labeling experiment on insulin-stimulated 3T3-L1 adipocytes pretreated with SB203580. Although it cannot be transported across the PM, the mannose-containing, photoaffinity-labeling compound binds to the exposed exofacial glucose-binding sites of GLUT4 and is irreversibly cross-linked to this site upon exposure to UV light of 300 nm (16). Subsequently, cross-linked GLUT4 transporters can be isolated by performing a streptavidin-bead pull-down employing the biotin moiety present in the photolabeling reagent. As shown in Fig. 6A, pretreatment with SB203580 caused a profound reduction in the amount of photoaffinity labeling of the GLUT4 transporters present in the PM. This observation became particularly evident when the amount of photoaffinity label was expressed relative to the amount of GLUT4 present in the PM (as measured in Fig. 3A), thus giving an indication of the amount of photolabel per GLUT4 transporter as the fold increase over the basal level (see Fig. 6B). These data suggest that upon SB203580 treatment, the insulin-induced increase in accessibility of the exofacial glucose-binding site on GLUT4 for the photoaffinity label is reduced significantly.
FIG. 6. Effects of SB203580 treatment on ATB-BMPA photoaffinity labeling reagent binding to GLUT4. Adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Subsequentially, samples were irradiated for 5 min in the continued presence of SB203580. A, Immunoblot analysis of streptavidin-bead-precipitated material probed with an anti-GLUT4 antibody. Data shown are from a representative immunoblot of an experiment performed in triplicate. B, Data shown as described in A were also subjected to quantification by LumiImager analysis. Data were expressed as described in Materials and Methods using the data obtained in Fig. 3 to yield an indication of the amount of photoaffinity label per the amount of GLUT4 protein as the fold increase over the basal value (mean ± SEM). *, P < 0.05 compared with basal levels; #, P < 0.05 compared with insulin.
Discussion
Insulin-induced glucose uptake in adipocytes is mediated by the insulin-responsive GLUT4 transporter. In response to the activity of insulin-induced phosphotidylinositol 3'-kinase and CAP-Cbl-TC10 signaling pathways, GLUT4-containing storage vesicles rapidly translocate toward the PM (1). 3T3-L1 adipocytes also express GLUT1 transporters, which can undergo translocation by insulin. Theoretically, these transporters could contribute to the observed discrepancy between the 8-fold stimulation of glucose transport by insulin and the only 3- to 4-fold stimulation of GLUT4 translocation. However, we believe that GLUT1 contributes only a small extent to the stimulated glucose transport, based on the following arguments. Application of the inhibitor rottlerin, which specifically inhibits GLUT4-mediated, but not GLUT1-mediated, glucose transport, reduces insulin-stimulated glucose transport to near-basal levels of transport (23). Furthermore, when probing the immunoblots of ATB-BMPA, a photoaffinity labeling experiment with a competent antibody recognizing GLUT1 failed to show any protein (data not shown). This is probably due to the lower levels of insulin-induced GLUT1 translocation compared with GLUT4 and the lower levels of GLUT1 protein present in our fully mature 3T3-L1 adipocytes compared with GLUT4 (20% at best), resulting in the amount of precipitated GLUT1 being below immunoblot detection limits. As a final consideration, we and others have reported on the lack of an effect of SB203580 on GLUT1 translocation and/or the lack of an effect of SB203580 on (strictly GLUT1-mediated) insulin-stimulated glucose uptake in 3T3-L1 fibroblasts (5, 24). Thus, apparently, after GLUT4 membrane insertion, the level of glucose uptake through this transporter is also regulated through the activity of p38 MAPK (3, 4). Consequently, pretreatment of adipocytes with SB203580 (a pharmacological inhibitor of p38 MAPK and -?) or dexamethasone-induced up-regulation of the p38 MAPK phosphatases MKP-1 and -4 in 3T3-L1 adipocytes induces a reduction of approximately 30% in the amount of glucose taken up by the cell (5, 6). However, as outlined in the introduction, the mechanism through which p38 MAPK-mediated enhancement of glucose transport operates on GLUT4 remained poorly defined. To resolve this question, we analyzed the effects of SB203580 treatment on GLUT4-mediated glucose uptake in 3T3-L1 adipocytes in detail.
Importantly, as shown previously for Myc GLUT4 L6 muscle cells (21), in 3T3-L1 adipocytes, treatment with SB203580 does not alter the apparent Km of the system, yet the apparent maximum velocity is markedly attenuated. In other words, the effects of SB203580 on glucose uptake in these cells are consistent with a noncompetitive mechanism. Furthermore, SB203580 did not affect glucose uptake when added simultaneously with the radiolabel, thereby ruling out a direct interference of this pharmacological inhibitor with the actual process of glucose uptake. Several possibilities remain for explaining the effects of SB203580 observed. These are 1) the number of fully deployed GLUT4 transporters in which the glucose-binding site is accessible to external glucose, 2) the speed with which captured glucose is transferred across the PM, and 3) the relative turnover activity of the individual GLUT4 transporters.
The first would be a direct consequence of locking the GLUT4 transporter in a situation in which the glucose transporter is inserted in the PM, but not yet fully accessible to external glucose (an occluded state) (9). This would result in a reduction in photoaffinity labeling by the mannose-containing reagent. However, it should presumably also lead to a reduction of cell surface GLUT4 biotinylation and a reduction of the accessibility of the HA tag in the first extracellular loop of GLUT4 Based upon our observations, interference of SB203580 with the occluded state seems unlikely.
For the second possibility, an alteration of the transfer speed of glucose across the PM (or glucose flux) (4, 5), kinetic analysis demonstrates that the affinity of the GLUT4 transporter for glucose is unaltered by pretreatment SB203580. Yet, a reduction in the amount of photoaffinity labeling of the GLUT4 transporter was observed. Given that the photoaffinity label itself is not transported across the PM (16), an alteration in glucose flux is unlikely to have any effect on the amount of photoaffinity label binding. Thus, these observations also argue against an alteration in the speed of glucose transfer through the GLUT4 transporter. For the third suggestion, an influence of SB203580 on the turnover rate of the GLUT4 transporter, with the transition from the inward facing confirmation to the outward facing confirmation being slowed down, this would result in complete exposure of the GLUT4 transporter to the extracellular environment. The apparent Km of the GLUT4 would not be appreciably affected. However, the number of available glucose-binding sites exposed on the cell surface would be reduced, as observed by the reduced labeling by the photolabeling compound. Thus, in conclusion, a turnover effect on the GLUT4 transporter after SB203580 treatment fits best with the observations presented in this manuscript.
Cytosolic pH and levels of intracellular ATP can modulate the glucose turnover of a homolog glucose transporter, GLUT1, through a process termed intrinsic substrate occlusion (12, 25, 26). Furthermore, the introduction of several point mutations in GLUT1 also affects glucose turnover (27, 28, 29, 30, 31, 32). The predominant explanation for the effects of these mutations entails a hampering of the oscillation of GLUT1 between its outward (glucose-accepting) and inward (glucose-releasing) conformation. Intriguingly, with respect to GLUT4, mutation of Glu329, Glu393, and Arg400 appears to arrest the transporter in an inward facing conformation, inducing a similar effect on efficiency of photoaffinity label binding as we observed (11). The induction of a similar phenomenon by treatment with the p38 MAPK inhibitor SB203580 suggests the possible existence of intracellular insulin-induced signaling pathways and/or components aiding in GLUT4 conformational oscillation present in 3T3-L1 adipocytes, thereby regulating glucose turnover. Elucidation of the signaling components involved should provide exciting new insights for GLUT4-mediated glucose transport and its deregulation in those particular diabetic states in which GLUT4 translocation itself is not affected.
Acknowledgments
We acknowledge the expert technical advice of Dena Yver and Mary-Jane Zarnowski (Bethesda, MD) on the HAtag quantification. We are also indebted to Dr. G. D. Holman (Bath, UK) for making the photoaffinity-labeling reagent available, and to Dr. Slegt (Leiden, The Netherlands) for technical assistance with the photolabeling procedure. Furthermore, we thank Dr. S. W. Cushman (Bethesda, MD) for making the HA-tagged GLUT4-GFP construct available. Dr. S. W. Cushman is also acknowledged for his constructive involvement in the research described in this manuscript.
References
Saltiel AR, Kahn CR 2001 Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414:799–806
Saltiel AR, Pessin JE 2002 Insulin signaling pathways in time and space. Trends Cell Biol 12:65–71
Somwar R, Perreault M, Kapur S, Taha C, Sweeney G, Ramlal T, Kim DY, Keen J, Cote CH, Klip A, Marette A 2000 Activation of p38 mitogen-activated protein kinase and ? by insulin and contraction in rat skeletal muscle: potential role in the stimulation of glucose transport. Diabetes 49:1794–1800
Somwar R, Koterski S, Sweeney G, Sciott, R, Djuric S, Berg C, Trevillyan J, Scherer PE, Rondinone CM, Klip A 2002 A dominant-negative p38 MAPK mutant and novel selective inhibitors of p38 MAPK reduce insulin-stimulated glucose uptake in 3T3–L1 adipocytes without affecting GLUT4 translocation. J Biol Chem 277:50386–50395
Sweeney G, Somwar R, Ramlal T, Volchuk A, Ueyama A, Klip A 1999 An inhibitor of p38 mitogen-activated protein kinase prevents insulin-stimulated glucose transport but not glucose transporter translocation in 3T3–L1 adipocytes and L6 myotubes. J Biol Chem 274:10071–10078
Bazuine M, Carlotti F, Tafrechi RS, Hoeben RC, Maassen JA 2004 Mitogen-activated protein kinase (MAPK) phosphatase-1 and -4 attenuate p38 MAPK during dexamethasone-induced insulin resistance in 3T3–L1 adipocytes. Mol Endocrinol 18:1697–1707
Michelle FL, Poon V, Klip A 2003 GLUT4 activation: thoughts on possible mechanisms. Acta Physiol Scand 178:287–296
Furtado LM, Somwar R, Sweeney G, Niu W, Klip A 2002 Activation of the glucose transporter GLUT4 by insulin. Biochem Cell Biol 80:569–578
Holman GD, Cushman SW 1994 Subcellular localization and trafficking of the GLUT4 glucose transporter isoform in insulin-responsive cells. Bioessays 16:753–759
Holman GD, Lo-Leggio L, Cushman SW 1994 Insulin-stimulated GLUT4 glucose transporter recycling. A problem in membrane protein subcellular trafficking through multiple pools. J Biol Chem 269:17516–17524
Schurmann A, Doege H, Ohnimus H, Monser V, Buchs A, Joost HG 1997 Role of conserved arginine and glutamate residues on the cytosolic surface of glucose transporters for transporter function. Biochemistry 36:12897–12902
Heard KS, Fidyk N, Carruthers A 2000 ATP-dependent substrate occlusion by the human erythrocyte sugar transporter. Biochemistry 39:3005–3014
Reed BC, Lane MD 1980 Insulin receptor synthesis and turnover in differentiating 3T3–L1 preadipocytes. Proc Natl Acad Sci USA 77:285–289
van den Berghe N, Ouwens DM, Maassen JA, van Mackelenbergh MG, Sips HC, Krans HM 1994 Activation of the Ras/mitogen-activated protein kinase signaling pathway alone is not sufficient to induce glucose uptake in 3T3–L1 adipocytes. Mol Cell Biol 14:2372–2377
Simpson IA, Yver DR, Hissin PJ, Wardzala LJ, Karnieli E, Salans LB, Cushman SW 1983 Insulin-stimulated translocation of glucose transporters in the isolated rat adipose cells: characterization of subcellular fractions. Biochim Biophys Acta 763:393–407
Holman GD, Kozka IJ, Clark AE, Flower CJ, Saltis J, Habberfield AD, Simpson IA, Cushman SW 1990 Cell surface labeling of glucose transporter isoform GLUT4 by bis-mannose photolabel. Correlation with stimulation of glucose transport in rat adipose cells by insulin and phorbol ester. J Biol Chem 265:18172–18179
Satoh S, Nishimura H, Clark AE, Kozka IJ, Vannucci SJ, Simpson IA, Quon MJ, Cushman SW, Holman GD 1993 Use of bismannose photolabel to elucidate insulin-regulated GLUT4 subcellular trafficking kinetics in rat adipose cells. Evidence that exocytosis is a critical site of hormone action. J Biol Chem 268:17820–17829
Dawson K, Aviles-Hernandez A, Cushman SW, Malide D 2001 Insulin-regulated trafficking of dual-labeled glucose transporter 4 in primary rat adipose cells. Biochem Biophys Res Commun 287:445–454
Quon MJ, Guerre-Millo M, Zarnowski MJ, Butte AJ, Em M, Cushman SW, Taylor SI 1994 Tyrosine kinase-deficient mutant human insulin receptors (Met1153–>Ile) overexpressed in transfected rat adipose cells fail to mediate translocation of epitope-tagged GLUT4. Proc Natl Acad Sci USA 91:5587–5591
Carlotti F, Bazuine M, Kekarainen T, Seppen J, Pognonec P, Maassen JA, Hoeben RC 2004 Lentiviral vectors efficiently transduce quiescent mature 3T3–L1 adipocytes. Mol Ther 9:209–217
Somwar R, Kim DY, Sweeney G, Huang C, Niu W, Lador C, Ramlal T, Klip A 2001 GLUT4 translocation precedes the stimulation of glucose uptake by insulin in muscle cells: potential activation of GLUT4 via p38 mitogen-activated protein kinase. Biochem J 359:639–649
Konrad D, Bilan PJ, Nawaz Z, Sweeney G, Niu W, Liu Z, Antonescu CN, Rudich A, Klip A 2002 Need for GLUT4 activation to reach maximum effect of insulin-mediated glucose uptake in brown adipocytes isolated from GLUT4myc expressing mice. Diabetes 51:2719–2726
Bazuine M, Van der Zon GCM, Van de Ven R, Van den Broek PJA, Maassen JA 2004 Rottlerin inhibits multiple steps involved in insulin-induced glucose uptake in 3T3–L1 adipocytes. Biochem Pharmacol 68:105–112
Bazuine M, Ouwens DM, Gomes de Mesquita GDM, Maassen JA 2003 Arsenite stimulated glucose transport in 3T3–L1 adipocytes involves both GLUT4 translocation and p38 MAPK activity. Eur J Biochem 270:3891–3903
Cloherty EK, Levine KB, Graybill C, Carruthers A 2002 Cooperative nucleotide binding to the human erythrocyte sugar transporter. Biochemistry 41:12639–12651
Levine KB, Cloherty EK, Fidyk NJ, Carruthers A 1998 Structural and physiologic determinants of human erythrocyte sugar transport regulation by adenosine triphosphate. Biochemistry 37:12221–12232
Dauterive R, Laroux S, Bunn RC, Chaisson A, Sanson T, Reed BC 1996 C-terminal mutations that alter the turnover number for 3-O-methylglucose transport by GLUT1 and GLUT4. J Biol Chem 271:11414–11421
Katagiri H, Asano T, Ishihara H, Lin JL, Inukai K, Shanahan MF, Tsukuda K, Kikuchi M, Yazaki Y, Oka Y 1993 Role of tryptophan-388 of GLUT1 glucose transporter in glucose-transport activity and photoaffinity-labelling with forskolin. Biochem J 291:861–867
Ishihara H, Asano T, Katagiri H, Lin JL, Tsukuda K, Shibasaki Y, Yazaki Y, Oka Y 1991 The glucose transport activity of GLUT1 is markedly decreased by substitution of a single amino acid with a different charge at residue 415. Biochem Biophys Res Commun 176:922–930
Oka Y, Asano T, Shibasaki Y, Lin JL, Tsukuda K, Katagiri H, Akanuma Y, Takaku F 1990 C-Terminal truncated glucose transporter is locked into an inward-facing form without transport activity. Nature 345:550–553
Mori H, Hashiramoto M, Clark AE, Yang J, Muraoka A, Tamori Y, Kasuga M, Holman GD 1994 Substitution of tyrosine 293 of GLUT1 locks the transporter into an outward facing conformation. J Biol Chem 269:11578–11583
Tamori Y, Hashiramoto M, Clark AE, Mori H, Muraoka A, Kadowaki T, Holman GD, Kasuga M 1994 Substitution at Pro385 of GLUT1 perturbs the glucose transport function by reducing conformational flexibility. J Biol Chem 269:2982–2986(Merlijn Bazuine, Fran?ois)
Address all correspondence and requests for reprints to: Dr. J. Antonie Maassen, Signal Transduction Laboratory, Department of Molecular Cell Biology, Leiden University Medical Center, Wassenaarseweg 72, P.O. Box 9503, 2333 AL Leiden, The Netherlands. E-mail: j.a.maassen@lumc.nl.
Abstract
Insulin induces a profound increase in glucose uptake in 3T3-L1 adipocytes through the activity of the glucose transporter-4 (GLUT4). Apart from GLUT4 translocation toward the plasma membrane, there is also an insulin-induced p38 MAPK-dependent step involved in the regulation of glucose uptake. Consequently, treatment with the p38 MAPK inhibitor SB203580 reduces insulin-induced glucose uptake by approximately 30%. Pretreatment with SB203580 does not alter the apparent Km of GLUT4-mediated glucose uptake but reduces the maximum velocity by approximately 30%. Insulin-induced GLUT4 translocation and exposure of the transporter to the extracellular environment was not altered by pretreatment with SB203580, as evidenced by a lack of effect of the inhibitor on the amount of GLUT4 present in the plasma membrane, as assessed by subcellular fractionation, the amount of GLUT4 that is able to undergo biotinylation on intact adipocytes and the level of extracellular exposure of an ectopically expressed GLUT-green fluorescence protein construct with a hemagglutinin tag in its first extracellular loop. In contrast, labeling of GLUT4 after insulin stimulation by a membrane-impermeable, mannose moiety-containing, photoaffinity-labeling agent [2-N-4(1-azido-2,2,2-trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-propylamine] that binds to the extracellular glucose acceptor domain was markedly reduced by SB203580, although photolabeling with this compound in the absence of insulin was unaffected by SB203580. These data suggest that SB203580 affects glucose turnover by the insulin-responsive GLUT4 transporter in 3T3-L1 adipocytes.
Introduction
ONE OF THE main functions of insulin in the body is maintaining blood glucose homeostasis. To achieve this, insulin induces glucose uptake in responsive tissues such as muscle and adipocytes (1). Stimulated glucose uptake in these cells is mediated primarily by translocation of the insulin-responsive glucose transporter-4 (GLUT4) toward the plasma membrane (PM), involving the contribution of both phosphotidylinositol 3'-kinase and Cbl signaling pathways emanating from the activated insulin receptor (2). After translocation toward the PM, there is a secondary, p38 MAPK-dependent step, leading to enhancement of insulin-induced glucose uptake (3, 4). Consequently, pretreatment of adipocytes with the pharmacological inhibitor SB203580 against the p38 MAPK- and -? isoforms reduces the amount of insulin-induced glucose uptake by roughly 30% (5). This phenomenon may be of importance in a more physiological setting of insulin resistance as well. Previously we reported that in 3T3-L1 adipocytes, treatment with the glucocorticoid dexamethasone induces an insulin-resistant state through up-regulation of the MAPK phosphatases-1 and -4, resulting in p38 MAPK dephosphorylation and concomitantly a reduction in glucose uptake (6).
Compared with the insulin-induced state (Fig. 1A), there are several scenarios that could in theory explain the deleterious effects of SB203580 treatment on insulin-induced glucose uptake: 1) an alteration of the glucose flux (or rate of glucose transfer) through the GLUT4 transporter across the PM (Fig. 1B) (7, 8), 2) a hampering of the transition of the GLUT4 transporter through its occluded (PM inserted, but not fully exposed to the extracellular milieu) state to a fully exposed state (Fig. 1C) (9, 10), or 3) a glucose turnover defect, for example due to an alteration in the oscillation between outward (sugar accepting from the extracellular environment) and inward (sugar release into the cytosol) facing conformations of the GLUT4 transporter (Fig. 1D) (11, 12).
FIG. 1. Theoretical models on the effects of SB203580 on insulin-induced glucose uptake in 3T3-L1 adipocytes. A, Insulin stimulation results in an increase in the exposure of GLUT4 to the extracellular milieu (depicted by open spheres in the PM). As a consequence, 2-DOG uptake is maximally increased (four arrows) and the GLUT4 are accessible to photoaffinity labeling as well as to cell surface biotinylation or antibody binding to an exofacial HA tag. SB203580 reduces insulin-induced glucose uptake (indicated by two arrows). B, If the glucose flux through the transporter is reduced, only an effect on glucose uptake will occur. Photoaffinity label is not transported across the PM and hence is insensitive to alterations in the transfer speed (two arrows, as in A). C, When transition through the occluded state is altered, binding of photoaffinity label is reduced, as is the efficiency of cell surface biotinylation and accessibility of the HA tag (one arrow each). D, When SB203580 induces a turnover defect, a reduction in photoaffinity label is expected. No alterations should be observed in either biotinylation or accessibility of the HA tag, however (two arrows, as in A), because the GLUT4 itself is fully exposed to the extracellular milieu.
As can be readily appreciated from Fig. 1, when challenged with different experimental approaches to analyze GLUT4 translocation and activity, we should be able to discriminate among these models. In this manuscript, we describe such an analysis of the effects of the p38 MAPK inhibitor SB203580 on glucose uptake by the adipocyte.
Materials and Methods
Materials
DMEM was purchased from Invitrogen Life Technologies, Inc. (Gaithersburg, MD); fetal calf serum was obtained from Brunschwig (Amsterdam, The Netherlands; catalog no. A15-043, lot A01127-318); dexamethasone, 1-methyl-3-isobutylxanthine, bovine insulin, and 2-deoxy-D-glucose were obtained from Sigma-Aldrich Corp. (St. Louis, MO); 2-deoxy-D-[14C]glucose and 125I-labeled secondary antibody were purchased from Amersham Biosciences (Little Chalfont, UK). SB203580 was obtained from Promega Corp. (Madison, WI). Monoclonal antibodies against the hemagglutinin (HA) tag were purchased from Abcam (Cambridge, MA). Cy3-conjugated secondary antibodies were purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Goat polyclonal anti-GLUT4 (C-20), rabbit polyclonal against HA tag, and horseradish peroxidate-conjugated donkey antigoat secondary antibody were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Biotin LC hydrazide and sodium meta-periodate were obtained from Perbio (Erembodegem, Belgium).
Cell culture
3T3-L1 fibroblasts were obtained from American Type Culture Collection (Manassas, VA) and differentiated to adipocytes as previously described (13). Mature adipocytes were routinely used 7–14 d after completion of the differentiation process. Only cultures in which more than 95% of the cells displayed adipocyte morphology were used.
Assay of 2-deoxyglucose uptake
3T3-L1 adipocytes, grown in 12-well plates (Costar, Cambridge, MA), were subjected to an assay of 2-deoxy-D-[14C]glucose (0.075 μCi/well) uptake as described previously (14).
Biotinylation
After treatment, cells were washed twice with PBS and subjected to 30-min oxidation by 20 mM sodium-meta-periodate at 4 C in 0.1 M sodium acetate, pH 5.5. Subsequently, cells were washed three times with PBS/45 mM glycerol and twice with 0.1 M 2-N-(morpholino)ethanesulfonic acid (pH 5.0). The biotinylation reaction using 50 mM biotin LC hydrazide was performed in 0.1 M 2-N-(morpholino)ethanesulfonic acid, pH 5.0, for 2 h at 4 C. Three wash steps with PBS/15 mM glycine were used to quench the reaction, and samples were subjected to subcellular fractionation.
Subcellular fractionation
After treatment, cells were washed twice with PBS on ice and scraped in HES buffer [20 mM HEPES (pH 7.4), 1 mM EDTA, and 250 mM sucrose] in the presence of protease inhibitors. Samples were homogenized nine times by three strokes in a Potter homogenizer (Fischer Scientific, Zoetermeer, The Netherlands), after which low density microsomal vesicle (LDM) and PM were isolated by differential centrifugation as described by Simpson et al. (15). Equal amounts of protein (10 μg), as determined using the bicinchoninic acid protein assay reagent (Pierce Chemical Co., Rockford, IL) were subjected to immunoblot analysis.
Photoaffinity labeling
The photoaffinity-labeling reaction using 2-N-4(1-azido-2,2,2-trifluoroethyl)benzoyl-1,3-bis(D-mannose-4-yloxy)-2-propylamine (ATB-BMPA) was performed as described previously (16, 17). Briefly, samples were treated using a 5-min irradiation time with seven lamps (Southern New England Ultra Violet Co., Branford, CT) at 300 nm (35 watts) in a Rayonet RPR-200 photoincubator in the continued presence of SB203580. After two washes with PBS, cells were lysed in a Nonidet P-40-based buffer [1 mM Na3VO4, 1 mM EGTA, 1 mM EDTA, 50 mM Tris-HCl (pH 7.4), 1% (wt/vol) Nonidet P-40, 0.5% (wt/vol) sodium deoxycholate, 150 mM NaCl, and 5 mM sodium fluoride] in the presence of protease inhibitors (Complete, Roche, Indianapolis, IN), and material was collected using streptavidin bead precipitation.
Lentivirus-mediated HA tag GLUT4-green fluorescence protein (GFP) transduction in mature 3T3-L1 adipocytes
The GLUT4 construct harboring an HA tag in its first extracellular loop and a C-terminal GFP has been described previously (18, 19). This HA-GLUT4-GFP fusion construct was cloned into a lentiviral backbone under the control of a phosphoglycerate kinase promoter sequence (pRRL-phoshoglycerate kinase). Lentivirus vectors were raised and used to infect mature 3T3-L1 adipocytes, as previously described (20). One week after infection, adipocytes were washed twice with PBS. Adipocytes were fixed using 3.7% formaldehyde in PBS, washed four times with PBS/0.2% BSA, incubated for 30 min in PBS/0.2% BSA, and treated with anti-HA antibodies (1:200 in PBS/0.2% BSA) for 2 h. Plates were washed extensively in PBS/0.2% BSA and incubated with the relevant secondary antibody [Cy3 conjugated (1:100) for immunofluorescence and 125I radiolabeled (0.1 μCi) for quantification] for 1 h. Subsequently, cells were washed six times for 5 min each time with PBS and once with H2O. For immunofluorescence, cells were mounted using Vectashield (Brunschwig) and viewed with a DM-IRBE (Leica, Deerfield, IL). To quantify 125I, cells were lysed overnight in 0.1% sodium dodecyl sulfate/NaOH (0.1 M) and analyzed in a Tri-Carb scintillation counter (Packard, Downers Grove, IL).
Statistical analysis and graph generation
Statistical analysis of the data obtained was performed with an independent-sample t test using SPSS 10.0 (SPSS, Inc., Chicago, IL). Graphs were generated using PRISM 2.01 (GraphPad Software, Inc., San Diego, CA).
Subcellular fractionations subjected to immunoblot analysis were also exposed to a LumiImager (Roche) and quantified using LumiAnalyst software, yielding the amount of signal in Boehringer light units (BLU, an arbitrary unit). Subsequently, data were corrected for protein content and expressed as a relative fraction of GLUT4 residing in either the intracellular LDM fraction or the plasma membrane fraction. Thus, graph data are: LDM [BLU/mg]/(LDM [BLU/mg] + PM [BLU/mg]) and are similar for PM.
With respect to the photoaffinity label, BLU data for the biotin signal were expressed as fold/basal (treated [BLU/mg]/basal [BLU/mg]) and divided by the relative amount of GLUT4 signal in the PM in fold over basal (treated [BLU/mg]/basal [BLU/mg], giving an indication of the amount of photolabel/amount of GLUT4 transporters.
Results
Effects of SB203580 on insulin-induced glucose uptake
The p38 MAPK inhibitor SB203580 was added at several distinct time points, with respect to the time at which the radiolabeled 2-deoxy-D-glucose (2-DOG) was added, in an assay of insulin-induced glucose uptake. These are 15 min before insulin stimulation, concomitant with insulin stimulation, and concomitant with radiolabeled 2-DOG. As shown in Fig. 2A, SB203580 caused a significant reduction of about 30% in insulin-stimulated glucose uptake only when added before the radiolabel. In these experiments, basal levels of 2-DOG uptake were 4.5 ± 0.46 pmol/min·mg protein and were unaltered by similar SB203580 treatments. These observations are in agreement with previous observations (5) and exclude a direct interference of the SB203580 compound with the process of glucose uptake itself.
FIG. 2. A kinetic analysis of the effects of SB203580 on insulin- induced glucose uptake in 3T3-L1 adipocytes. 3T3-L1 adipocytes were stimulated with 100 nM insulin for 15 min and assayed for [14C]2-DOG uptake. A, Adipocytes were treated with 10 μM SB203580 for 15 min before insulin stimulation, concomitant with the insulin stimulation, or concomitant with the addition of radiolabeled 2-DOG or were not treated, as indicated. In these experiments, basal levels of 2-DOG uptake were 4.5 ± 0.46 pmol/min·mg protein and were unaltered by similar SB203580 treatments. The data shown are the mean of three independent experiments, each performed in triplicate, ± SEM. *, P < 0.05 compared with uninhibited, insulin-stimulated cells. B, Insulin-stimulated glucose uptake was measured by employing a range of 2-DOG concentrations, as indicated. Adipocytes were treated with 10 μM SB203580 for 15 min before insulin stimulation. The data shown are the mean of two independent experiment, each performed in triplicate, ± SEM.
Subsequently, the effects of SB203580 on insulin-induced glucose uptake in 3T3-L1 adipocytes were also characterized by performing a kinetic analysis. As shown in Fig. 2B, pretreatment with 10 μM SB203580 reduced the maximum velocity of insulin-induced glucose uptake from 1.0 nmol/min·mg protein to 0.68 nmol. Meanwhile, the insulin-stimulated apparent Km for glucose uptake was 2.2 and 2.0 mM for untreated cells and cells pretreated with SB203580, respectively. These data are similar to those observed in L6 muscle cells stably transfected with an Myc-tagged, GLUT4-expressing construct (21). Inhibition of p38 MAPK activity in an in vitro kinase assay at a concentration of 0.5 μM SB203580 led to an inhibition of about 50% of maximum, in agreement with the 50% inhibitory concentration reported by the manufacturer (Merck & Co., Rahway, NJ) for SB203580 on p38 of 0.6 μM (data not shown).
Effects of SB203580 on insulin-induced GLUT4 translocation
We analyzed the effects of SB203580 on insulin-induced GLUT4 translocation using several different techniques. As shown in Fig. 3 and in support of the literature (5), when employing subcellular fractionation separating intracellular LDM vesicles from the PM by differential centrifugation, SB203580 did not affect insulin-induced GLUT4 translocation toward the PM.
FIG. 3. Effects of SB203580 on insulin-induced GLUT4 translocation. Adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Cells were subjected to subcellular fractionation, and equal amounts of protein (10 μg) were subjected to immunoblot analysis. A, Immunoblots were quantified on a LumiImager and expressed as relative amounts of GLUT4 resident in either the PM or LDM fraction. Data shown are the mean of an experiment, performed in triplicate, ± SEM. Importantly, the amount of GLUT4 observed in the high density microsomal vesicle fraction did not alter significantly in any of the conditions applied. *, P < 0.05 compared with basal levels. B, The top frame shows a representative immunoblot of the subcellular fractionation used to obtain the data shown in A. The bottom frame shows representative immunoblot data from an experiment performed in triplicate when probed with an antibiotin antibody of subcellular fractionation samples performed on 3T3-L1 adipocytes subjected to the biotinylation assay.
Subsequently, we analyzed the exposure of GLUT4 at the surface of the PM. As can be observed from the immunoblot data in Fig. 3B, GLUT4 runs as a smear around 45 kDa rather than as a discrete band due to heavy glycosylation. Treatment of the intact adipocyte with periodate after insulin stimulation causes oxidation of exposed sugar side-chains, allowing subsequent reaction of the resulting sugar aldehydes with the biotinylating reagent. When performing this procedure, no effect of SB203580 on the insulin-induced translocation and exposure to the extracellular milieu was observed (see Fig. 3B). Quantification of these bands resulted in a 2-fold over basal induction by insulin (or insulin with SB203580). However, this quantification should be interpreted with caution, because the procedure enhanced the smearing of the GLUT4 band and concomitantly reduced the signal to noise ratios considerably (see Fig. 3B). Thus, as an alternative to the biotinylation procedure and to obtain a more reliable quantification, 3T3-L1 adipocytes were transduced with a lentivirus vector expressing HA-tagged GLUT4-GFP. The HA tag present in this complex is localized in the first extracellular loop of the GLUT4 transporter (19). Hence, only when fully inserted in the PM will this tag be accessible to anti-HA antibodies supplied in the extracellular milieu. As can be observed in the immunofluorescence data in Fig. 4A, in the basal situation scarcely any extracellular HA tag signal was observed, whereas the GFP signal was readily detectable. Insulin induced a profound translocation of the tagged GLUT4 transporters toward the PM and concomitant extracellular HA tag exposure (see Fig. 4B). Pretreatment with SB203580 had no marked effect on intracellular localization of the tagged GLUT4 transporters in the basal state or on PM exposure after insulin stimulation compared with untreated adipocytes (Fig. 4, C and D). Although lentivirus-mediated infection results typically in 80% infected 3T3-L1 adipocytes, in Fig. 4 we selected less densely transfected areas to visualize the translocation events more clearly. Using a similar technique, but employing a radiolabeled secondary antibody against the HA tag antibody, we were able to quantify the single cell data presented in Fig. 4 in the whole well (containing 50,000 adipocytes). In the basal state, on the average, 1524 ± 8.7 cpm radiolabeled secondary antibody were measured per well. Importantly, mock transfection using a lentivirus expressing only GFP resulted in an average of 974 ± 28 cpm/well without any significant effect of insulin and/or SB203580 treatment. As shown in Fig. 5, insulin induced a 3.1 ± 0.32-fold increase in the amount of HA tag exposure over basal levels, as measured by this approach. Pretreatment of SB203580 did not significantly affect the level of exposure of the HA tag in either the basal (1.1 ± 0.20-fold) or insulin-stimulated (3.4 ± 0.35-fold) state. These data are comparable with those generated using Myc-tagged GLUT4 in stably transfected L6 myotubes (21) and brown adipocytes derived of GLUT4 Myc-expressing mice (22).
FIG. 4. Immunofluorescence data of 3T3-L1 adipocytes transduced with GFP-GLUT4-HA tag-expressing lentivirus vectors. Transduced adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Adipocytes were fixed and probed with an anti-HA antibody, followed by a Cy3-conjugated secondary antibody (red). The GFP signal of the GLUT4 chimera is in green; yellow indicates an overlap of the GFP-derived GLUT4 signal and the extracellular exposure of the HA tag. Data shown are from cells selected from an experiment, performed in duplicate. For clarity, low density transduced areas (without transduced adipocytes joining each other) were chosen to illustrate the individual cell response to insulin more clearly.
FIG. 5. HA tag quantification data of 3T3-L1 adipocytes transduced with GFP-GLUT4-HA tag-expressing lentivirus vectors. Transduced adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Adipocytes were fixed and probed with an anti-HA antibody, followed by a 125I- radiolabeled secondary antibody. Data shown are depicted as the fold increase over the basal value (mean ± SEM of two experiments, each performed in triplicate). *, P < 0.05 compared with the basal state, the difference between insulin and insulin plus pretreatment with SB203580 was not significant. In the basal state, on the average, 1524 ± 8.7 cpm radiolabeled secondary antibody were measured per well. Each well contains approximately 5.0 x 104 adipocytes.
Effects of SB203580 on insulin-induced photoaffinity labeling with ATB-BMPA
Subsequently, we performed a photoaffinity labeling experiment on insulin-stimulated 3T3-L1 adipocytes pretreated with SB203580. Although it cannot be transported across the PM, the mannose-containing, photoaffinity-labeling compound binds to the exposed exofacial glucose-binding sites of GLUT4 and is irreversibly cross-linked to this site upon exposure to UV light of 300 nm (16). Subsequently, cross-linked GLUT4 transporters can be isolated by performing a streptavidin-bead pull-down employing the biotin moiety present in the photolabeling reagent. As shown in Fig. 6A, pretreatment with SB203580 caused a profound reduction in the amount of photoaffinity labeling of the GLUT4 transporters present in the PM. This observation became particularly evident when the amount of photoaffinity label was expressed relative to the amount of GLUT4 present in the PM (as measured in Fig. 3A), thus giving an indication of the amount of photolabel per GLUT4 transporter as the fold increase over the basal level (see Fig. 6B). These data suggest that upon SB203580 treatment, the insulin-induced increase in accessibility of the exofacial glucose-binding site on GLUT4 for the photoaffinity label is reduced significantly.
FIG. 6. Effects of SB203580 treatment on ATB-BMPA photoaffinity labeling reagent binding to GLUT4. Adipocytes were pretreated with 10 μM SB203580 for 15 min and subsequently stimulated with 100 nM insulin for 15 min, as indicated. Subsequentially, samples were irradiated for 5 min in the continued presence of SB203580. A, Immunoblot analysis of streptavidin-bead-precipitated material probed with an anti-GLUT4 antibody. Data shown are from a representative immunoblot of an experiment performed in triplicate. B, Data shown as described in A were also subjected to quantification by LumiImager analysis. Data were expressed as described in Materials and Methods using the data obtained in Fig. 3 to yield an indication of the amount of photoaffinity label per the amount of GLUT4 protein as the fold increase over the basal value (mean ± SEM). *, P < 0.05 compared with basal levels; #, P < 0.05 compared with insulin.
Discussion
Insulin-induced glucose uptake in adipocytes is mediated by the insulin-responsive GLUT4 transporter. In response to the activity of insulin-induced phosphotidylinositol 3'-kinase and CAP-Cbl-TC10 signaling pathways, GLUT4-containing storage vesicles rapidly translocate toward the PM (1). 3T3-L1 adipocytes also express GLUT1 transporters, which can undergo translocation by insulin. Theoretically, these transporters could contribute to the observed discrepancy between the 8-fold stimulation of glucose transport by insulin and the only 3- to 4-fold stimulation of GLUT4 translocation. However, we believe that GLUT1 contributes only a small extent to the stimulated glucose transport, based on the following arguments. Application of the inhibitor rottlerin, which specifically inhibits GLUT4-mediated, but not GLUT1-mediated, glucose transport, reduces insulin-stimulated glucose transport to near-basal levels of transport (23). Furthermore, when probing the immunoblots of ATB-BMPA, a photoaffinity labeling experiment with a competent antibody recognizing GLUT1 failed to show any protein (data not shown). This is probably due to the lower levels of insulin-induced GLUT1 translocation compared with GLUT4 and the lower levels of GLUT1 protein present in our fully mature 3T3-L1 adipocytes compared with GLUT4 (20% at best), resulting in the amount of precipitated GLUT1 being below immunoblot detection limits. As a final consideration, we and others have reported on the lack of an effect of SB203580 on GLUT1 translocation and/or the lack of an effect of SB203580 on (strictly GLUT1-mediated) insulin-stimulated glucose uptake in 3T3-L1 fibroblasts (5, 24). Thus, apparently, after GLUT4 membrane insertion, the level of glucose uptake through this transporter is also regulated through the activity of p38 MAPK (3, 4). Consequently, pretreatment of adipocytes with SB203580 (a pharmacological inhibitor of p38 MAPK and -?) or dexamethasone-induced up-regulation of the p38 MAPK phosphatases MKP-1 and -4 in 3T3-L1 adipocytes induces a reduction of approximately 30% in the amount of glucose taken up by the cell (5, 6). However, as outlined in the introduction, the mechanism through which p38 MAPK-mediated enhancement of glucose transport operates on GLUT4 remained poorly defined. To resolve this question, we analyzed the effects of SB203580 treatment on GLUT4-mediated glucose uptake in 3T3-L1 adipocytes in detail.
Importantly, as shown previously for Myc GLUT4 L6 muscle cells (21), in 3T3-L1 adipocytes, treatment with SB203580 does not alter the apparent Km of the system, yet the apparent maximum velocity is markedly attenuated. In other words, the effects of SB203580 on glucose uptake in these cells are consistent with a noncompetitive mechanism. Furthermore, SB203580 did not affect glucose uptake when added simultaneously with the radiolabel, thereby ruling out a direct interference of this pharmacological inhibitor with the actual process of glucose uptake. Several possibilities remain for explaining the effects of SB203580 observed. These are 1) the number of fully deployed GLUT4 transporters in which the glucose-binding site is accessible to external glucose, 2) the speed with which captured glucose is transferred across the PM, and 3) the relative turnover activity of the individual GLUT4 transporters.
The first would be a direct consequence of locking the GLUT4 transporter in a situation in which the glucose transporter is inserted in the PM, but not yet fully accessible to external glucose (an occluded state) (9). This would result in a reduction in photoaffinity labeling by the mannose-containing reagent. However, it should presumably also lead to a reduction of cell surface GLUT4 biotinylation and a reduction of the accessibility of the HA tag in the first extracellular loop of GLUT4 Based upon our observations, interference of SB203580 with the occluded state seems unlikely.
For the second possibility, an alteration of the transfer speed of glucose across the PM (or glucose flux) (4, 5), kinetic analysis demonstrates that the affinity of the GLUT4 transporter for glucose is unaltered by pretreatment SB203580. Yet, a reduction in the amount of photoaffinity labeling of the GLUT4 transporter was observed. Given that the photoaffinity label itself is not transported across the PM (16), an alteration in glucose flux is unlikely to have any effect on the amount of photoaffinity label binding. Thus, these observations also argue against an alteration in the speed of glucose transfer through the GLUT4 transporter. For the third suggestion, an influence of SB203580 on the turnover rate of the GLUT4 transporter, with the transition from the inward facing confirmation to the outward facing confirmation being slowed down, this would result in complete exposure of the GLUT4 transporter to the extracellular environment. The apparent Km of the GLUT4 would not be appreciably affected. However, the number of available glucose-binding sites exposed on the cell surface would be reduced, as observed by the reduced labeling by the photolabeling compound. Thus, in conclusion, a turnover effect on the GLUT4 transporter after SB203580 treatment fits best with the observations presented in this manuscript.
Cytosolic pH and levels of intracellular ATP can modulate the glucose turnover of a homolog glucose transporter, GLUT1, through a process termed intrinsic substrate occlusion (12, 25, 26). Furthermore, the introduction of several point mutations in GLUT1 also affects glucose turnover (27, 28, 29, 30, 31, 32). The predominant explanation for the effects of these mutations entails a hampering of the oscillation of GLUT1 between its outward (glucose-accepting) and inward (glucose-releasing) conformation. Intriguingly, with respect to GLUT4, mutation of Glu329, Glu393, and Arg400 appears to arrest the transporter in an inward facing conformation, inducing a similar effect on efficiency of photoaffinity label binding as we observed (11). The induction of a similar phenomenon by treatment with the p38 MAPK inhibitor SB203580 suggests the possible existence of intracellular insulin-induced signaling pathways and/or components aiding in GLUT4 conformational oscillation present in 3T3-L1 adipocytes, thereby regulating glucose turnover. Elucidation of the signaling components involved should provide exciting new insights for GLUT4-mediated glucose transport and its deregulation in those particular diabetic states in which GLUT4 translocation itself is not affected.
Acknowledgments
We acknowledge the expert technical advice of Dena Yver and Mary-Jane Zarnowski (Bethesda, MD) on the HAtag quantification. We are also indebted to Dr. G. D. Holman (Bath, UK) for making the photoaffinity-labeling reagent available, and to Dr. Slegt (Leiden, The Netherlands) for technical assistance with the photolabeling procedure. Furthermore, we thank Dr. S. W. Cushman (Bethesda, MD) for making the HA-tagged GLUT4-GFP construct available. Dr. S. W. Cushman is also acknowledged for his constructive involvement in the research described in this manuscript.
References
Saltiel AR, Kahn CR 2001 Insulin signalling and the regulation of glucose and lipid metabolism. Nature 414:799–806
Saltiel AR, Pessin JE 2002 Insulin signaling pathways in time and space. Trends Cell Biol 12:65–71
Somwar R, Perreault M, Kapur S, Taha C, Sweeney G, Ramlal T, Kim DY, Keen J, Cote CH, Klip A, Marette A 2000 Activation of p38 mitogen-activated protein kinase and ? by insulin and contraction in rat skeletal muscle: potential role in the stimulation of glucose transport. Diabetes 49:1794–1800
Somwar R, Koterski S, Sweeney G, Sciott, R, Djuric S, Berg C, Trevillyan J, Scherer PE, Rondinone CM, Klip A 2002 A dominant-negative p38 MAPK mutant and novel selective inhibitors of p38 MAPK reduce insulin-stimulated glucose uptake in 3T3–L1 adipocytes without affecting GLUT4 translocation. J Biol Chem 277:50386–50395
Sweeney G, Somwar R, Ramlal T, Volchuk A, Ueyama A, Klip A 1999 An inhibitor of p38 mitogen-activated protein kinase prevents insulin-stimulated glucose transport but not glucose transporter translocation in 3T3–L1 adipocytes and L6 myotubes. J Biol Chem 274:10071–10078
Bazuine M, Carlotti F, Tafrechi RS, Hoeben RC, Maassen JA 2004 Mitogen-activated protein kinase (MAPK) phosphatase-1 and -4 attenuate p38 MAPK during dexamethasone-induced insulin resistance in 3T3–L1 adipocytes. Mol Endocrinol 18:1697–1707
Michelle FL, Poon V, Klip A 2003 GLUT4 activation: thoughts on possible mechanisms. Acta Physiol Scand 178:287–296
Furtado LM, Somwar R, Sweeney G, Niu W, Klip A 2002 Activation of the glucose transporter GLUT4 by insulin. Biochem Cell Biol 80:569–578
Holman GD, Cushman SW 1994 Subcellular localization and trafficking of the GLUT4 glucose transporter isoform in insulin-responsive cells. Bioessays 16:753–759
Holman GD, Lo-Leggio L, Cushman SW 1994 Insulin-stimulated GLUT4 glucose transporter recycling. A problem in membrane protein subcellular trafficking through multiple pools. J Biol Chem 269:17516–17524
Schurmann A, Doege H, Ohnimus H, Monser V, Buchs A, Joost HG 1997 Role of conserved arginine and glutamate residues on the cytosolic surface of glucose transporters for transporter function. Biochemistry 36:12897–12902
Heard KS, Fidyk N, Carruthers A 2000 ATP-dependent substrate occlusion by the human erythrocyte sugar transporter. Biochemistry 39:3005–3014
Reed BC, Lane MD 1980 Insulin receptor synthesis and turnover in differentiating 3T3–L1 preadipocytes. Proc Natl Acad Sci USA 77:285–289
van den Berghe N, Ouwens DM, Maassen JA, van Mackelenbergh MG, Sips HC, Krans HM 1994 Activation of the Ras/mitogen-activated protein kinase signaling pathway alone is not sufficient to induce glucose uptake in 3T3–L1 adipocytes. Mol Cell Biol 14:2372–2377
Simpson IA, Yver DR, Hissin PJ, Wardzala LJ, Karnieli E, Salans LB, Cushman SW 1983 Insulin-stimulated translocation of glucose transporters in the isolated rat adipose cells: characterization of subcellular fractions. Biochim Biophys Acta 763:393–407
Holman GD, Kozka IJ, Clark AE, Flower CJ, Saltis J, Habberfield AD, Simpson IA, Cushman SW 1990 Cell surface labeling of glucose transporter isoform GLUT4 by bis-mannose photolabel. Correlation with stimulation of glucose transport in rat adipose cells by insulin and phorbol ester. J Biol Chem 265:18172–18179
Satoh S, Nishimura H, Clark AE, Kozka IJ, Vannucci SJ, Simpson IA, Quon MJ, Cushman SW, Holman GD 1993 Use of bismannose photolabel to elucidate insulin-regulated GLUT4 subcellular trafficking kinetics in rat adipose cells. Evidence that exocytosis is a critical site of hormone action. J Biol Chem 268:17820–17829
Dawson K, Aviles-Hernandez A, Cushman SW, Malide D 2001 Insulin-regulated trafficking of dual-labeled glucose transporter 4 in primary rat adipose cells. Biochem Biophys Res Commun 287:445–454
Quon MJ, Guerre-Millo M, Zarnowski MJ, Butte AJ, Em M, Cushman SW, Taylor SI 1994 Tyrosine kinase-deficient mutant human insulin receptors (Met1153–>Ile) overexpressed in transfected rat adipose cells fail to mediate translocation of epitope-tagged GLUT4. Proc Natl Acad Sci USA 91:5587–5591
Carlotti F, Bazuine M, Kekarainen T, Seppen J, Pognonec P, Maassen JA, Hoeben RC 2004 Lentiviral vectors efficiently transduce quiescent mature 3T3–L1 adipocytes. Mol Ther 9:209–217
Somwar R, Kim DY, Sweeney G, Huang C, Niu W, Lador C, Ramlal T, Klip A 2001 GLUT4 translocation precedes the stimulation of glucose uptake by insulin in muscle cells: potential activation of GLUT4 via p38 mitogen-activated protein kinase. Biochem J 359:639–649
Konrad D, Bilan PJ, Nawaz Z, Sweeney G, Niu W, Liu Z, Antonescu CN, Rudich A, Klip A 2002 Need for GLUT4 activation to reach maximum effect of insulin-mediated glucose uptake in brown adipocytes isolated from GLUT4myc expressing mice. Diabetes 51:2719–2726
Bazuine M, Van der Zon GCM, Van de Ven R, Van den Broek PJA, Maassen JA 2004 Rottlerin inhibits multiple steps involved in insulin-induced glucose uptake in 3T3–L1 adipocytes. Biochem Pharmacol 68:105–112
Bazuine M, Ouwens DM, Gomes de Mesquita GDM, Maassen JA 2003 Arsenite stimulated glucose transport in 3T3–L1 adipocytes involves both GLUT4 translocation and p38 MAPK activity. Eur J Biochem 270:3891–3903
Cloherty EK, Levine KB, Graybill C, Carruthers A 2002 Cooperative nucleotide binding to the human erythrocyte sugar transporter. Biochemistry 41:12639–12651
Levine KB, Cloherty EK, Fidyk NJ, Carruthers A 1998 Structural and physiologic determinants of human erythrocyte sugar transport regulation by adenosine triphosphate. Biochemistry 37:12221–12232
Dauterive R, Laroux S, Bunn RC, Chaisson A, Sanson T, Reed BC 1996 C-terminal mutations that alter the turnover number for 3-O-methylglucose transport by GLUT1 and GLUT4. J Biol Chem 271:11414–11421
Katagiri H, Asano T, Ishihara H, Lin JL, Inukai K, Shanahan MF, Tsukuda K, Kikuchi M, Yazaki Y, Oka Y 1993 Role of tryptophan-388 of GLUT1 glucose transporter in glucose-transport activity and photoaffinity-labelling with forskolin. Biochem J 291:861–867
Ishihara H, Asano T, Katagiri H, Lin JL, Tsukuda K, Shibasaki Y, Yazaki Y, Oka Y 1991 The glucose transport activity of GLUT1 is markedly decreased by substitution of a single amino acid with a different charge at residue 415. Biochem Biophys Res Commun 176:922–930
Oka Y, Asano T, Shibasaki Y, Lin JL, Tsukuda K, Katagiri H, Akanuma Y, Takaku F 1990 C-Terminal truncated glucose transporter is locked into an inward-facing form without transport activity. Nature 345:550–553
Mori H, Hashiramoto M, Clark AE, Yang J, Muraoka A, Tamori Y, Kasuga M, Holman GD 1994 Substitution of tyrosine 293 of GLUT1 locks the transporter into an outward facing conformation. J Biol Chem 269:11578–11583
Tamori Y, Hashiramoto M, Clark AE, Mori H, Muraoka A, Kadowaki T, Holman GD, Kasuga M 1994 Substitution at Pro385 of GLUT1 perturbs the glucose transport function by reducing conformational flexibility. J Biol Chem 269:2982–2986(Merlijn Bazuine, Fran?ois)