Advertisement

Insulin and Contraction Stimulate Exocytosis, but Increased AMP-activated Protein Kinase Activity Resulting from Oxidative Metabolism Stress Slows Endocytosis of GLUT4 in Cardiomyocytes*

  • Jing Yang
    Affiliations
    Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, United Kingdom
    Search for articles by this author
  • Geoffrey D. Holman
    Correspondence
    To whom correspondence should be addressed: Department of Biology and Biochemistry, University of Bath, Claverton Down, Bath BA2 7AY, United Kingdom. Tel.: 44-1225-386874; Fax: 44-1225-386779;
    Affiliations
    Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, United Kingdom
    Search for articles by this author
  • Author Footnotes
    * This work was supported by grants from the Medical Research Council (United Kingdom), Diabetes UK, and the British Heart Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:November 22, 2004DOI:https://doi.org/10.1074/jbc.M410213200
      Stimulations of glucose transport produced by insulin action, contraction, or through a change in cell energy status are mediated by separate signaling pathways. These are the wortmannin-sensitive phosphatidylinositol 3-kinase pathway leading to the intermediate Akt and the wortmannin-insensitive AMP-activated protein kinase (AMPK) pathway. Electrical stimulation of cardiomyocytes produced a rapid, insulin-like, wortmannin-sensitive stimulation of glucose transport activity, but this occurred without extensive activation of Akt. Although AMPK phosphorylation was increased by contraction, this response was not wortmannin-inhibitable and consequently did not correlate with the wortmannin sensitivity of the transport stimulation. Oxidative metabolism stress due to hypoxia or treatment with oligomycin led to increased AMPK activity with a corresponding increase in glucose transport activity. We show here that these separate signaling pathways converge on GLUT4 trafficking at separate steps. The rate of exocytosis of GLUT4 was rapidly stimulated by insulin, but insulin treatment did not alter the endocytosis rate. Like insulin stimulation, electrical stimulation of contraction led to a stimulation of GLUT4 exocytosis without any marked change in endocytosis. By contrast, after oxidative metabolism stress, no stimulation of GLUT4 exocytosis occurred; instead, this treatment led to a reduction in GLUT4 endocytosis.
      Regulation of cellular energy levels and metabolic processes are intricately dependent on the supply of glucose, and there are now known to be multiple mechanisms controlling this supply to specific tissues. Insulin stimulation of glucose transport occurs in response to raised blood glucose or when stimulated storage of glucose, in the form of glycogen and triglyceride, is necessary. A stimulation of glucose uptake is also necessary during metabolic stress such as that which occurs during hypoxia and ischemia. Recent work in this area has identified AMP-activated protein kinase (AMPK)
      The abbreviations used are: AMPK, AMP-activated protein kinase; KRH, Krebs-Ringer-HEPES; AICAR, (5-aminoimidazole-4-carboxamide)-riboside; ZMP, (5-aminoimidazole-4-carboxamide)-riboside-5-phosphate.
      1The abbreviations used are: AMPK, AMP-activated protein kinase; KRH, Krebs-Ringer-HEPES; AICAR, (5-aminoimidazole-4-carboxamide)-riboside; ZMP, (5-aminoimidazole-4-carboxamide)-riboside-5-phosphate.
      as the key regulatory intermediate of these processes. This protein functions as an energy level sensor in organisms ranging from yeast to mammals (
      • Carling D.
      ,
      • Hardie D.G.
      ). Increased AMPK activity is associated with increased glucose transport in cell lines (
      • Fryer L.G.
      • Foufelle F.
      • Barnes K.
      • Baldwin S.A.
      • Woods A.
      • Carling D.
      ,
      • Barnes K.
      • Ingram J.C.
      • Porras O.H.
      • Barros L.F.
      • Hudson E.R.
      • Fryer L.G.
      • Foufelle F.
      • Carling D.
      • Hardie D.G.
      • Baldwin S.A.
      ) and skeletal muscle that have been challenged by a metabolic stress stimulus or more direct activation of the kinase by the AMP analogue, ZMP, which is produced from the nucleoside AICAR (
      • Kurth-Kraczek E.J.
      • Hirshman M.F.
      • Goodyear L.J.
      • Winder W.W.
      ,
      • Koistinen H.A.
      • Galuska D.
      • Chibalin A.V.
      • Yang J.
      • Zierath J.R.
      • Holman G.D.
      • Wallberg-Henriksson H.
      ).
      Contraction of skeletal myocytes is also known to lead to increased levels of glucose transport. Evidence from transgenic mice expressing a dominant inhibitory form of AMPK suggest that although AMPK is involved in regulation of hypoxia-induced glucose transport, it may not be involved in the contraction response (
      • Mu J.
      • Brozinick J.T.
      • Valladares O.
      • Bucan M.
      • Birnbaum M.J.
      ). Similar conclusions have been reached in studies on animal models for obesity-related type 2 diabetes (
      • Barnes B.R.
      • Ryder J.W.
      • Steiler T.L.
      • Fryer L.G.
      • Carling D.
      • Zierath J.R.
      ).
      Electrically stimulated contraction of cardiac myocytes also leads to stimulation of glucose transport. This occurs without a marked stimulation of Akt phosphorylation. However, this effect is unlike that occurring in skeletal muscle because it is inhibited by wortmannin (
      • Till M.
      • Ouwens D.M.
      • Kessler A.
      • Eckel J.
      ). Exercise has been shown to increase heart muscle AMPK activity, and there is an associated increase in GLUT4 translocation (
      • Coven D.L.
      • Hu X.
      • Cong L.
      • Bergeron R.
      • Shulman G.I.
      • Hardie D.G.
      • Young L.H.
      ), but the extent to which this is due to exercise-induced metabolic stress or more directly to increased contractile activity is unclear.
      Although many early signaling events in insulin action, energy status, and contraction-mediated signaling to GLUT4 are now understood, details on how these processes converge on downstream events have not yet been resolved. One approach to the resolution of convergence points is to determine how signaling changes alter the steps involved in the recycling, back and forth, of GLUT4 between its intracellular compartments and the plasma membrane. Signaling could potentially influence any of these multiple steps. The kinetics of GLUT4 trafficking between the plasma membrane and intracellular membrane compartments have been extensively studied in adipocytes and the 3T3-L1 cell line (
      • Satoh S.
      • Nishimura H.
      • Clark A.E.
      • Kozka I.J.
      • Vannucci S.J.
      • Simpson I.A.
      • Quon M.J.
      • Cushman S.W.
      • Holman G.D.
      ,
      • Yang J.
      • Holman G.D.
      ,
      • Yang J.
      • Hodel A.
      • Holman G.D.
      ,
      • Zeigerer A.
      • Lampson M.A.
      • Karylowski O.
      • Sabatini D.D.
      • Adesnik M.
      • Ren M.
      • McGraw T.E.
      ,
      • Czech M.P.
      • Buxton J.M.
      ,
      • Karylowski O.
      • Zeigerer A.
      • Cohen A.
      • McGraw T.E.
      ,
      • Govers R.
      • Coster A.C.
      • James D.E.
      ). However, similar studies have not been previously carried out on native animal skeletal or heart muscle. It is particularly important to investigate GLUT4 trafficking in these myocyte systems because they integrate stimuli that arise from insulin action, from stress stimuli that lead to activation of AMPK, and from contraction-stimulated signaling.
      Any net increase in plasma membrane GLUT4 could be due to an increase in exocytosis or a decrease in endocytosis. Trafficking studies in adipocytes have shown that it is mainly the exocytosis limb of the recycling pathway that is insulin-stimulated (
      • Satoh S.
      • Nishimura H.
      • Clark A.E.
      • Kozka I.J.
      • Vannucci S.J.
      • Simpson I.A.
      • Quon M.J.
      • Cushman S.W.
      • Holman G.D.
      ,
      • Yang J.
      • Holman G.D.
      ,
      • Karylowski O.
      • Zeigerer A.
      • Cohen A.
      • McGraw T.E.
      ,
      • Jhun B.H.
      • Rampal A.L.
      • Liu H.
      • Lachaal M.
      • Jung C.Y.
      ). The present study examines whether the stimulatory effect of insulin on GLUT4 translocation in cardiomyocytes is likewise due to stimulation of exocytosis. This is not necessarily the case because relatively small insulin-inhibitory effects on GLUT4 endocytosis have been found in adipocytes (
      • Yang J.
      • Holman G.D.
      ,
      • Czech M.P.
      • Buxton J.M.
      ). However, long-term insulin action on adipocytes leads to a loss of cell surface GLUT4 due to an increased rate of endocytosis (
      • Pryor P.R.
      • Liu S.C.
      • Clark A.E.
      • Yang J.
      • Holman G.D.
      • Tosh D.
      ). In a complex cell type such as the cardiac myocyte, in which multiple signaling pathways may converge, one cannot easily predict the site of insulin action on GLUT4 recycling. Therefore, we have examined here whether the divergent signaling pathways that are activated by the stimuli of insulin action, altered energy status, and electrical stimulation impinge on GLUT4 trafficking at the same or differing sites.

      EXPERIMENTAL PROCEDURES

      Cardiomyocyte Isolation and Stimulation—Cardiomyocytes from adult male Wistar rats (260–280 g) were prepared by collagenase digestion (type II; Worthington Biochemicals) by use of a method described previously (
      • Eckel J.
      • Reinauer H.
      ,
      • Fischer Y.
      • Thomas J.
      • Holman G.D.
      • Rose H.
      • Kammermeier H.
      ) but with the inclusion of 20 mm inosine in the medium. Cell suspensions were adjusted to ≈10% cytocrit in Krebs-Ringer-HEPES (KRH) buffer (6 mm KCl, 1 mm Na2HPO4, 0.2 mm NaHPO4, 1.4 mm MgSO4, 1 mm CaCl2, 128 mm NaCl, and 10 mm HEPES) with 2% fatty acid-free bovine serum albumin (Roche Applied Science). Cell stimulation by insulin was for 2–30 min at 5–30 nm. Stimulation by hypoxia was induced by incubation in KRH buffer that had been pre-gassed with nitrogen for 30 min. The nitrogen atmosphere was maintained throughout subsequent incubations. For electrical stimulation of contraction, 1-ml aliquots of cell suspensions at 37 °C, ≈10% cytocrit, were placed in 19-mm diameter polystyrene dishes. A dish lid was added that had attached electrodes that dipped into the cell suspension. Cells were stimulated for 1–5 min, as indicated in the figure legends, at 100 V with a pulse duration of 1 ms and a frequency of 10 Hz. Contraction was monitored under a microscope.
      Glucose Transport Activity—1-ml aliquots of cardiomyocyte suspension at 10% cytocrit in 2% bovine serum albumin-KRH buffer were maintained at 37 °C and continuously gassed with O2. The transport assay was initiated by the addition of 2-deoxy-d-glucose (100 μm final concentration) containing 0.5 μCi of 2-deoxy-d-[3H]glucose (Amersham Biosciences). Sugar uptake was terminated after 10 min by transferring the cell suspension to microfuge tubes containing 400 μm phloretin in KRH buffer. Background activity was determined by addition of cells to tubes that contained 2-deoxy-d-glucose premixed with phloretin. The samples were immediately centrifuged at 3,500 × g for 1 min. The supernatants were removed, and the cells were washed three times with 1 ml of KRH buffer containing 400 μm phloretin. Cells were lysed with 1 ml of 0.1 m NaOH, and aliquots were taken for determination of radioactivity and protein levels.
      Akt and AMPK Phosphorylation—1-ml aliquots of cardiomyocytes (≈10% cytocrit) were homogenized with 20 complete strokes at 900 revolutions/min in a 5-ml tight-fitting Teflon homogenizer and at 4 °C in 250 μl of HES buffer (20 mm HEPES, pH 7.2, 1 mm EDTA, and 255 mm sucrose) plus protease inhibitors (antipain, pepstatin A, and leupeptin; each at 1 μg/ml), 100 μm 4-(2-aminoethyl)benzenesulfonyl fluoride, and phosphatase inhibitors (10 mm NaF, 1 mm NaMO4, 200 μm Na3VO4, 50 nm cypermethrin, 5 μm dephostain, 100 nm okadaic acid, and 10 pm nuclear inhibitor of protein phosphatase-1; Calbiochem). The homogenized samples were then centrifuged for 30 min at 100,000 rpm. 40-μg samples of supernatant protein were resolved by 8% SDS-PAGE and Western blotted using anti-phospho-AMPK (Thr172), anti-phospho-Akt (Thr308), and anti-phospho-Akt (Ser473) (Cell Signaling). In each case, signals were detected by ECL or advance ECL and quantified using an Optichem detector with associated software (Ultra Violet Products).
      Trafficking of Transferrin Receptors—0.5-ml aliquots of cardiomyocytes under basal conditions or after pre-treatment with 5 μm oligomycin for 60 min at 37 °C were incubated with horseradish peroxidase-conjugated transferrin (2.5 μg/ml; Jackson Laboratories) for an additional 60 min at 4 °C. Cells were then washed to remove non-bound transferrin and maintained at 37 °C for the times indicated in the figure legends. The oligomycin-pre-treated cells were maintained in the presence of oligomycin through all incubations. At the indicated times, cells were rapidly cooled on ice and by adding ice-cold KRH buffer. They were then treated with a solution containing 500 mm NaCl and 200 mm acetic acid (pH 2.8) for 8 min at 4 °C and washed three times in KRH buffer at 4 °C. Backgrounds were determined from samples incubated with competing 2.5 mg/ml holo transferrin. Transferrin uptake was determined by measuring the cell-associated peroxidase activity as described previously (
      • Shepherd P.R.
      • Soos M.A.
      • Siddle K.
      ).
      Trafficking of Photolabeled GLUT4—Cardiomyocytes under basal conditions or after treatment with insulin, oligomycin, or electrically stimulated contraction were prepared as indicated in the figure legends. 1-ml aliquots of cell suspension at 10% cytocrit were transferred to 35-mm-diameter polystyrene dishes cooled to 18 °C to slow transporter recycling. The photolabeling reagent GP15 (
      • Yang J.
      • Hodel A.
      • Holman G.D.
      ,
      • Hashimoto M.
      • Yang J.
      • Holman G.D.
      ) was added for 2 min at a final concentration of 500 μm. Cells were then irradiated at a wavelength of 300–350 nm for 1 min in a Rayonet RPR-100 reactor. After irradiation, cell aliquots for each condition were recombined, transferred to 50-ml tubes, washed once with 50 ml of KRH buffer (pH 6.0) containing 1% bovine serum albumin and 2 mm glucose to remove excess reagent, and then washed twice with 50 ml of KRH buffer (pH 7.4) also containing 1% bovine serum albumin and 2 mm glucose. For internalization experiments, the washed cells were then redivided into 1-ml aliquots at 10% cytocrit and incubated at 37 °C. The maximum internalization signal was determined in each experiment from an insulin-stimulated sample that was returned to the basal state and allowed to internalize the biotinylated GLUT4 for 40 min. At the indicated internalization time points, neutravidin (100 μg/ml; Pierce) was added for 5 min at 37 °C to block those transporters that were still at the cell surface. The cells were then washed three times in 25 ml of ice-cold KRH buffer. For the comparison of internalization rates, the biotin signal that escaped the surface neutravidin treatment was quantified as described below, and data are presented as a percentage of the maximum internalization.
      For exocytosis experiments, insulin-stimulated cells were labeled and subsequently washed to remove insulin. They were then maintained at 37 °C for 40 min to obtain maximal internalization. They were then restimulated with insulin or oligomycin or by electrical contraction in the presence of neutravidin. The extent of loss of the internal transporters, as they became quenched by the surface neutravidin, was determined at the indicated time points. To terminate the exocytosis, the cells were washed three times in 25 ml of ice-cold KRH buffer. The data are presented as a percentage of the maximum internalized biotinylated GLUT4 signal present at the beginning of the time course. The fractional change (f) in GLUT4 internalization with time (t) was curvefitted to the equation f = (kex + ken.exp(–t.(ken + kex)))/(ken + kex), and this was used to extract the endocytosis rate constant, ken, as described previously (
      • Satoh S.
      • Nishimura H.
      • Clark A.E.
      • Kozka I.J.
      • Vannucci S.J.
      • Simpson I.A.
      • Quon M.J.
      • Cushman S.W.
      • Holman G.D.
      ). A single exponential function f = exp(–t.kex) was used to calculate the exocytosis rate constant, kex (
      • Yang J.
      • Hodel A.
      • Holman G.D.
      ).
      Determination of Levels of Biotinylated GLUT4—After the termination of trafficking and the washing steps described above, the cardiomyocytes were solubilized in 1% Triton, 1% deoxycholic acid, and 0.2% SDS in phosphate-buffered saline, pH 7.2, with protease inhibitors (antipain, aprotinin, pepstatin, and leupeptin; 1 μg/ml each) and 100 μm 4-(2-amino-ethyl)-benzenesulfonyl fluoride for 50 min. Insoluble material was removed by centrifugation at 20,000 × g for 20 min at 4 °C. The biotinylated transporters from solubilized supernatants were precipitated with 50 μl of 50% slurry of streptavidin-agarose beads (Pierce) overnight at 4 °C with rotation. The streptavidin-agarose beads were then washed three times with 1 ml of phosphate-buffered saline buffer containing 1% (w/v) Thesit (Sigma), three times with 1 ml of phosphate-buffered saline buffer containing 0.1% (w/v) Thesit, and once with 1 ml of phosphate-buffered saline buffer only. Bound biotinylated GLUT4 was eluted in 30 μl of electophoresis sample buffer (62.5 mm Tris-HCl, pH 6.7, 2% (w/v) SDS, and 10% (v/v) glycerol) by heating at 95 °C for 30 min. The eluate was collected, and the elution procedure was repeated. The resulting eluates were pooled, 20 mm dithiothreitol was added, and samples were loaded onto 10% SDS-PAGE gels and transferred to nitrocellulose. GLUT4 was detected with a polyclonal GLUT4 antibody (
      • Holman G.D.
      • Kozka I.J.
      • Clark A.E.
      • Flower C.J.
      • Saltis J.
      • Habberfield A.D.
      • Simpson I.A.
      • Cushman S.W.
      ) and a horseradish peroxidase-linked secondary antibody using advance ECL reagents. The chemiluminescence was quantified using an Optichem detector with associated software (Ultra Violet Products).

      RESULTS

      Activation of Signaling via Akt and AMPK Leads to Contrasting Effects on Glucose Transport Activity—Transport of 2-deoxy-d-glucose in cardiomyocytes was stimulated by both insulin and the oxidative metabolism stress that followed hypoxia or treatment of cardiomyocytes with the mitochondrial inhibitor oligomycin (Fig. 1). The increase in transport activity produced in response to oligomycin was clearly evident within 10 min, but thereafter, there was an additional and gradual increase over a further 50 min (Fig. 1A). The effects of insulin and oligomycin were additive and greater than those of either stimulus alone (Fig. 1B). By contrast, electrical stimulation of cardiomyocyte contraction produced a very rapid stimulation of glucose transport activity. The effect occurred within 1 min and was not increased beyond this level by a longer stimulation period of 5 min (Fig. 1B). In contrast to the oligomycin treatment, the combined effects of insulin and electrically stimulated contraction did not produce a stimulation of transport that was greater than that of either stimulus alone.
      Figure thumbnail gr1
      Fig. 1Stimulation of glucose transport activity in cardiomyocytes. In A, the time course for stimulation of glucose transport by oxidative metabolism stress (induced by treatment with 5 μm oligomycin) was determined. Results are the mean ± S.E. from three experiments. In B, glucose transport activities were determined in cardiomyocytes in the basal state (Ba), after treatment with 5 nm insulin for 30 min (Ins), or after oxidative metabolism stress induced by treatment with 5 μm oligomycin for 60 min (OmO) or hypoxia for 30 min (OmH). The sensitivity to 100 nm wortmannin (W) and the additivity of the insulin and oligomycin (Ins + OmO) responses were determined as indicated. In C, glucose transport activities were determined in cardiomyocytes that had been treated with 5 nm insulin for 30 min (Ins) or by electrical stimulation of contraction for 1 min (Ctr 1m) or 5 min (Ctr 5m). The sensitivity to 100 nm wortmannin (W) and the additivity of the insulin and contraction (Ins + Ctr 5m) responses were determined as indicated. In B and C, data are the mean ± S.E. from three to six experiments, *, p < 0.05 (paired t test compared with insulin).
      The steps that lead to stimulation of transport in response to cell stress and changes in energy status are now known to be mediated by AMPK (
      • Carling D.
      ). The insulin signaling steps that are known to lead to stimulation of wortmannin-inhibitable glucose transport activity are known to occur via phosphatidylinositol 3-kinase and downstream activation of Akt (
      • Czech M.P.
      • Corvera S.
      ). We have therefore examined whether the phosphorylation of AMPK and Akt was altered by oxidative metabolism stress and electrically stimulated contraction of cardiomyocytes. Phospho-AMPK levels were increased 2–3-fold after oxidative metabolism stress induced by hypoxia or after oligomycin treatment. In contrast to the slow increase in stimulated transport activity after oxidative metabolism stress treatment with oligomycin, the increase in phospho-AMPK was similar at 5 and 60 min of stimulation (Fig. 2, A and B). Electrically stimulated contraction produced a comparable response in AMPK levels. Because the glucose transport stimulation produced by contraction was wortmannin-inhibitable, we tested whether the stimulation of AMPK phosphorylation was also wortmannin-sensitive. No inhibition was observed (Fig. 2, A and B). Insulin treatment led to a clear reduction in phospho-AMPK to a level that was only 55% of the basal level.
      Figure thumbnail gr2
      Fig. 2Activation of AMPK in cardiomyocytes. In A, phosphorylations of AMPK at threonine 172 were determined by Western blotting of lysates from cells in the basal state (Ba), after oxidative metabolism stress induced by hypoxia for 30 min (OmH), or after 30 nm insulin treatment for 30 min (Ins). In B, phosphorylations of AMPK were determined by Western blotting of lysates from cells in the basal state (Ba), after treatment with 5 μm oligomycin for 5 min (OmO5) or 60 min (OmO60), after electrically stimulated contraction for 1 min (Ctr1) or 5 min (Ctr5), or after treatment with 30 nm insulin for 30 min (Ins). The sensitivity of the hypoxia and contraction effects to 100 nm wortmannin (W) was also tested as indicated. The data in C are the combined mean ± S.E. from three to five experiments. *, p < 0.05 (paired t test; compared with basal state).
      After insulin stimulation, Akt phosphorylation at threonine 308 and serine 473 was increased by 20-fold and 10-fold, respectively (Fig. 3). No stimulation of either site occurred after the oxidative metabolism stress induced with oligomycin (Fig. 3) or hypoxia.
      J. Yang and G. D. Holman, unpublished data.
      Electrical contraction did not lead to an increase in Akt phosphorylation at threonine 308 at all (Fig. 3B) but did lead to a very slight (∼1.4-fold) increase in phosphorylation at serine 473 (Fig. 3C). The slight increase in Akt phosphorylation at serine 473 was reversed by wortmannin treatment (Fig. 3C). However, these effects were very small when compared with the insulin response in the same experiments, and the effects did not reach statistical significance over the number of experiments in which the response was studied. Akt is likely to be present in excess of that needed to activate processes leading to GLUT4 translocation. It may be that a slight activation of Akt is all that is required to account for the contraction-mediated increase in transport. These results are consistent with those reported previously by Till et al. (
      • Till M.
      • Ouwens D.M.
      • Kessler A.
      • Eckel J.
      ). They found that although the glucose transport increase that follows electrical stimulation of cardiomyocytes was wortmannin-sensitive, there was no increase in Akt phosphorylation as determined using the anti-phosphothreonine 308 antibody. However this study did not examine phosphorylation of serine 473.
      Figure thumbnail gr3
      Fig. 3Activation of Akt in cardiomyocytes. In A, phosphorylations of Akt at threonine 308 (top panel) and serine 473 (bottom panel) were determined by Western blotting of lysates from cells in the basal state (Ba), after insulin treatment (Ins) for 2 min or 30 min, after oxidative metabolism stress induced by treatment with 5 μm oligomycin (Omo) for 5 or 60 min, or after electrically stimulated contraction (Ctr) for 1 and 5 min. The sensitivity of the contraction effect to 100 nm wortmannin (Ctr 5min + W) was also tested. The Western blotting data using the anti-phosphothreonine 308 antibody (B) and the anti-phosphoserine 473 antibody (C) have been quantified. The data are the mean ± S.E. from four experiments. *, p < 0.05 (analysis of variance; compared with basal state).
      GLUT4 Exocytosis in Cardiomyocytes—The GLUT4 tagging and trafficking kinetic approach that we have used for cardiomyocytes involves use of the biotinylated photolabel GP15. This reagent has three parts. A glucose moiety is present that gives the label high affinity and specificity for GLUT4. A diazirine photolabel group allows covalent tagging of the GLUT4 after UV irradiation. Third, a biotin moiety is attached to the rest of the ligand via a very long spacer. This long spacer is important for binding to avidin in intact cells and in the absence of detergent (
      • Hashimoto M.
      • Yang J.
      • Holman G.D.
      ). The avidin only binds to the transporters at the surface. GLUT4 inside the cell escapes this quenching and can be subsequently isolated in a precipitation step using immobilized streptavidin. The streptavidin-precipitated material can then be analyzed by Western blotting for GLUT4 (
      • Yang J.
      • Hodel A.
      • Holman G.D.
      ).
      To measure exocytosis in cardiomyocytes, the tagged GLUT4 was first internalized. Avidin was then added to the cells to capture any GLUT4 that was re-translocated back to the plasma membrane upon stimulation. The GLUT4 signal decreased more rapidly in insulin-treated cells than in basal cells (Fig. 4A). Curve-fitting (Fig. 4B) to a simple exponential function then gave the exocytosis rate constant. The exocytosis rate constant of GLUT4 was 3-fold higher in insulin-stimulated cells compared with basal cells (Fig. 5B). These data provide the first evidence that insulin stimulates exocytosis in native animal myocytes. Although this has been inferred from comparisons with adipocytes and from examination of myocytes by microscopy, a direct measurement of the insulin-stimulated exocytosis has not been made previously. As noted previously for the rat adipocyte system (
      • Yang J.
      • Hodel A.
      • Holman G.D.
      ), there appeared to be a tendency for the exocytosis curves to depart from a single exponential time course. This may be because of GLUT4 distribution between two intracellular pools (
      • Govers R.
      • Coster A.C.
      • James D.E.
      ); however, because of the limited number of data points that can be obtained from the cardiomyocytes, we could not clearly resolve whether this was the case.
      Figure thumbnail gr4
      Fig. 4Insulin stimulates exocytosis of biotin-tagged GLUT4 in cardiomyocytes. Cardiomyocyte GLUT4 was tagged by labeling with the biotinylated photolabel GP15 as described under “Experimental Procedures.” The tagged GLUT4 was allowed to internalize for 40 min. Extravidin was then added to the cells to quench the GLUT4 that returned to the surface after stimulation. The remaining GLUT4 was precipitated using immobilized streptavidin and then resolved by SDS-PAGE and Western blotting. In A, the rate of decrease in Western blotting signal was monitored in cells in the basal state (Ba) or in cells that were stimulated by the addition of 30 nm insulin (Ins). In B, the Western blotting data were quantified, and curve-fitting to a single exponential decay function was carried out. Data are the mean ± S.E. from seven experiments.
      Figure thumbnail gr5
      Fig. 5GLUT4 exocytosis in cardiomyocytes is stimulated by electrically induced contraction but not by oxidative metabolism stress. Cardiomyocyte GLUT4 was tagged by labeling with the biotinylated photolabel GP15. The tagged GLUT4 was allowed to internalize for 40 min. Extravidin was then added to the cells to quench the GLUT4 that returned to the surface after stimulation. The remaining GLUT4 was precipitated using immobilized streptavidin and then resolved by SDS-PAGE and Western blotting. In A, the rate of decrease in internal GP15-labeled GLUT4 was calculated, and curve-fitting to a single exponential decay function was carried out. The calculated exocytosis rate constants were determined for cells in the basal state (Ba) or in cells that were stimulated by electrical contraction (Ctr) for 5 min after the internalization step. Oxidative metabolism-stressed cells were treated with 10 μm oligomycin (OmO) before and throughout the internalization period and throughout the following exocytosis assay. In B, exocytosis rate constants are compared in cells maintained in the basal state (Ba) or after insulin (Ins), oxidative metabolism stress (OmO), or contraction (Ctr) treatments. Data are the mean ± S.E. from four to seven experiments. *, p < 0.05 (unpaired t test; compared with basal state).
      Electrical stimulation of contraction also significantly increased the rate of GLUT4 exocytosis (Fig. 5A). The measurements were made over only 20 min in the contracting cells because the effects of contraction reversed after this time. To examine the effects of oligomycin on GLUT4 exocytosis, cells were treated before assay as described in the insulin plus oligomycin reversal experiments (see “Internalization of GLUT4 in Cardiomyocytes”). This pre-treatment with oligomycin was considered necessary because the onset of oligomycin action was slow in comparison with the time course of the exocytosis assay. By contrast to the stimulatory effects of insulin and contraction on GLUT4 exocytosis, oligomycin treatment did not produce an increase in GLUT4 exocytosis. The rate constant calculated (Fig. 5B) was the same as that in basal cells.
      Internalization of GLUT4 in Cardiomyocytes—Because oligomycin treatment did not increase exocytosis, we examined whether it might alter the internalization of GLUT4 from the cell surface. We first measured the changes in glucose transport activity that followed the removal of an insulin stimulus. The insulin reversal condition was achieved by using a pH 6.0 buffer wash step to remove insulin from its receptor (
      • Calderhead D.M.
      • Kitagawa K.
      • Tanner L.I.
      • Holman G.D.
      • Lienhard G.E.
      ). After insulin removal, glucose transport activity declined to the basal level over 40 min. The rate of reversal of transport activity was reduced by ∼50% when cells were treated with insulin and oligomycin before insulin removal (Fig. 6A). This is consistent with a recent study on reversal of insulin action in the presence of rotenone (
      • Becker C.
      • Sevilla L.
      • Tomas E.
      • Palacin M.
      • Zorzano A.
      • Fischer Y.
      ). A similar reduction in the reversal rate occurred when cardiomyocytes were maintained under hypoxic conditions (Fig. 6A).
      Figure thumbnail gr6
      Fig. 6Internalization of GLUT4 in cardiomyocytes is reduced by oxidative metabolism stress treatments. Cardiomyocytes were treated with 5 nm insulin for 30 min (Ins to Ba) or subjected to oxidative metabolism stress induced by 5 μm oligomycin treatment for 60 min plus insulin for 30 min (Ins + OmO to OmO) or hypoxia plus insulin for 30 min (Ins + OmH to OmH). Cells were then washed in a pH 6.0 buffer to remove insulin and allow reversal to the basal state. In A, the return to the basal state was determined from the decrease in 2-deoxy-glucose transport activity. In B, cell surface biotin-tagged GLUT4 was quenched by avidin addition at the indicated times, and the remaining internalized biotin-tagged GLUT4 that escaped the quenching was precipitated, resolved by SDS-PAGE, and Western blotted for GLUT4. Curve-fitting was carried out to determine the endocytosis rate constants (C). Data are the mean ± S.E. from three experiments (transport) and four experiments (GP15 internalization). *, p < 0.05 (unpaired t test; compared with Ins to Ba).
      To determine whether the reductions in transport could be directly attributable to internalization of GLUT4, we carried out GP15 internalization experiments. When insulin was removed from cells that had been labeled in the insulin-stimulated state, tagged GLUT4 escaped avidin addition to the cells with a time course that paralleled the decrease in transport activity (Fig. 6B). It is mainly the endocytosis rate constant that determines the shape of the internalization curve because the recycling or exocytosis rate is lowered by cutting off insulin signaling. Because of limitations in the number of data points that could be obtained, only endocytosis rate constants have been extracted from the curve-fitting (Fig. 6C). Analysis of the endocytosis rate constant values derived from both the transport and the labeling data revealed that the endocytosis process was slowed by ∼50% after oligomycin treatment.
      Steady-state Recycling of GLUT4 in Cardiomyocytes—Studies using a benzophenone-substituted C4-linked bis-mannose compound were the first to show that under steady-state conditions of insulin treatment, GLUT4 is continuously recycled. Some of the transporters labeled at the plasma membrane of insulin-stimulated cells were isolated in the light microsome fraction (
      • Holman G.D.
      • Karim A.R.
      • Karim B.
      ). Subsequent steady-state trafficking studies, in which GLUT4 was tagged using a benzophenone-substituted C3-linked bis-glucose (
      • Jhun B.H.
      • Rampal A.L.
      • Liu H.
      • Lachaal M.
      • Jung C.Y.
      ) and a diazirine-substituted C4-linked bis-mannose (
      • Satoh S.
      • Nishimura H.
      • Clark A.E.
      • Kozka I.J.
      • Vannucci S.J.
      • Simpson I.A.
      • Quon M.J.
      • Cushman S.W.
      • Holman G.D.
      ,
      • Yang J.
      • Holman G.D.
      ) and trypsin treatment (
      • Czech M.P.
      • Buxton J.M.
      ), revealed that insulin action is primarily to stimulate the exocytosis limb of the trafficking pathway. More recent studies using a hemagglutinin epitope-tagged GLUT4 have confirmed that GLUT4 is continually recycled at the steady state and that the tagged GLUT4 takes up antibody under conditions in which the level of glucose transport stimulation at steady state is constant (
      • Karylowski O.
      • Zeigerer A.
      • Cohen A.
      • McGraw T.E.
      ,
      • Govers R.
      • Coster A.C.
      • James D.E.
      ). These more recent studies have added considerably to our understanding of the movement and distribution of GLUT4 between its intracellular compartments. Because the steady-state approach reveals important aspects of GLUT4 recycling, we have used this kinetic approach in the cardiomyocyte system.
      In cardiomyocytes, as in the other systems described above, GLUT4 continuously recycles in the basal and insulin-stimulated states, and this is detected as an increasing GP15 signal that escapes the cell surface avidin quenching over time (Fig. 7A). Because the GP15 labeling is lower in the basal state, the error bars on the time course are larger. In insulin-stimulated cells, an equilibrium distribution of GLUT4 was reached in which approximately half of the labeled transporters escaped the cell surface quenching by avidin and remained detectable. In the basal state, a higher proportion of transporters escaped the quenching, but the initial rate constant for the loss from the surface was indistinguishable from that obtained in insulin-stimulated cells (Fig. 7B). After electrically stimulated contraction, there was also continuous GLUT4 recycling, and, like the insulin samples, the proportion lost from the surface was less than that occurring in the basal state. By contrast to this, oligomycin treatment resulted in a very marked reduction in steady-state recycling. This is consistent with the insulin reversal experiment (Fig. 7). The initial rate constants derived from curve-fitting (Fig. 7B) suggested that endocytosis was reduced to only 25% of the level occurring in either the basal state or the insulin- or contraction-stimulated state.
      Figure thumbnail gr7
      Fig. 7Oxidative metabolism stress treatment of cardiomyocytes slows the steady-state recycling of GLUT4 but not transferrin receptors. In A, cardiomyocytes were maintained in the basal state (Ba) or stimulated by treatment with 5 nm insulin (Ins) for 30 min, oxidative metabolism stress induced by 5 μm oligomycin treatment for 60 min (OmO), or electrically stimulated contraction for 5 min (Ctr). Cells were then cell surface-labeled with GP15. The internalization of the biotin-tagged GLUT4 was allowed to proceed, and then avidin was added to the cells at the indicated times to quench any tagged GLUT4 that remained at the cell surface. The remaining internalized biotin-tagged GLUT4 that escaped the quenching was precipitated, resolved by SDS-PAGE, and Western blotted for GLUT4. The GLUT4 levels were quantified, and then curve-fitting was carried out to determine the endocytosis rate constants (B). In C, the internalization of transferrin was determined in cardiomyocytes that were either maintained in the basal state (Ba) or treated with oligomycin (OmO). Data are the mean ± S.E. from three to five experiments in each case. *, p < 0.05 (unpaired t test; compared with basal state).
      To determine the extent to which the effects of oxidative metabolism stress after both hypoxia and oligomycin treatments were specific to glucose transport and GLUT4, we measured transferrin internalization. The rate of transferrin internalization was very rapid in cardiomyocytes, and transferrin internalization occurred at a rate that was ≈5 times faster than GLUT4 internalization. This rapid internalization rate is comparable with that reported previously for 3T3-L1 cells (
      • Tanner L.I.
      • Lienhard G.E.
      ). The difference in internalization rates of GLUT4 and transferrin receptors suggests some separation in the processing at the level of clathrin vesicle formation. No significant difference in the internalization rate constant was evident when comparing cells maintained with and without oligomycin treatment (Fig. 7C). The apparent endocytosis rate constants were 0.59 ± 0.18 and 0.70 ± 0.08 min–1 for oligomycin-treated and untreated cells, respectively (n = 3).

      DISCUSSION

      Stimulation of Glucose Transport in Cardiomyocytes—Heart cells are continually active, and although they can utilize a wide range of metabolic fuels, glucose supply to active heart is critical (
      • Abel E.D.
      ). Although hearts are continually contracting, increased body activity and exercise will increase the rate of heart contraction, and consequently, there will be increased demands for glucose. Insulin regulation of this activity is essential because persons with type 2 diabetes have increased risk of heart failure, and cardiovascular disease is the leading cause of death in diabetic patients (
      • Taegtmeyer H.
      • McNulty P.
      • Young M.E.
      ). Impairments in glucose supply and utilization occur in heart from type 2 diabetic patients (
      • Scheuermann-Freestone M.
      • Madsen P.L.
      • Manners D.
      • Blamire A.M.
      • Buckingham R.E.
      • Styles P.
      • Radda G.K.
      • Neubauer S.
      • Clarke K.
      ) and animal models of type 2 diabetes (
      • Desrois M.
      • Sidell R.J.
      • Gauguier D.
      • King L.M.
      • Radda G.K.
      • Clarke K.
      ). Studies of the multiple modes of regulation of glucose transport and how they are related to one another in heart are therefore of mechanistic importance and possible clinical relevance. We have examined here three modes of regulation that include insulin stimulation, metabolic stress-related AMPK stimulation, and stimulated contraction.
      Elevations in glucose transport activity occur in response to hypoxia, inhibition of mitochondrial ATP generation, and activation of AMPK activity after treatment with AICAR, which generates the AMP mimic ZMP in skeletal muscle. The biguanine drugs metformin and phenformin (
      • Fryer L.G.
      • Parbu-Patel A.
      • Carling D.
      ,
      • Zhou G.
      • Myers R.
      • Li Y.
      • Chen Y.
      • Shen X.
      • Fenyk-Melody J.
      • Wu M.
      • Ventre J.
      • Doebber T.
      • Fujii N.
      • Musi N.
      • Hirshman M.F.
      • Goodyear L.J.
      • Moller D.E.
      ) also lead to AMPK activation. We have used hypoxia and oligomycin to induce oxidative metabolism stress in the cardiomyocytes because this approach has been used previously to activate AMPK in cardiomyocytes (
      • Marsin A.S.
      • Bertrand L.
      • Rider M.H.
      • Deprez J.
      • Beauloye C.
      • Van den Vincent M.F.
      • Berghe G.
      • Carling D.
      • Hue L.
      ). The alternative pharmacological reagent AICAR that is widely used for stimulating skeletal muscle AMPK was unsuitable for activation of AMPK in cardiomyocytes because these cells lack the adenylate kinase necessary for conversion of the AICAR to the AMP mimic, ZMP (
      • Marsin A.S.
      • Bertrand L.
      • Rider M.H.
      • Deprez J.
      • Beauloye C.
      • Van den Vincent M.F.
      • Berghe G.
      • Carling D.
      • Hue L.
      ). Although AICAR treatment (for up to 60 min) does not lead to an increase in cardiomyocyte glucose transport activity, 2 this reagent has been used successfully to activate AMPK and glucose transport in isolated heart papillary muscle strips (
      • Russell R.R.
      • II I
      • Bergeron R.
      • Shulman G.I.
      • Young L.H.
      ,
      • Li J.
      • Hu X.
      • Selvakumar P.
      • Russell R.R.
      • II I
      • Cushman S.W.
      • Holman G.D.
      • Young L.H.
      ).
      Aspects of the changes in glucose transport activity that occur in response to hypoxia and oligomycin treatment indicate that the signaling from oxidative metabolism stress may impinge on a different site than that which occurs downstream of an insulin stimulus or a stimulus from electrically stimulated contraction. The response to combined treatments with insulin and oligomycin results in a greater stimulation of glucose transport activity than that seen with either stimulus alone. In addition, the rate of onset of the increased transport activity that follows oligomycin treatment is much slower than that which follows either an insulin- or contraction-induced stimulation. By contrast, electrically stimulated contraction does not produce an additive response when combined with insulin treatment. It is also very rapid in its speed of onset.
      Signaling to Glucose Transport in Cardiomyocytes—The differing modes of stimulation of transport activity are reflected in differing signaling changes. Whereas insulin markedly stimulates Akt phosphorylation, oxidative metabolism stress does not. We cannot rule out a very slight, wortmannin-sensitive stimulation of Akt phosphorylation at serine 473 in response to contraction. The changes in AMPK activity also differ between the three regulatory modes studied. Whereas insulin lowers AMPK activity, both oxidative metabolism stress and contraction lead to increases in AMPK activity. In some respects, contraction produces a similar response to insulin (for example, rapid action on glucose transport and the wortmannin sensitivity of the response), but in other respects (including the elevation of AMPK activity), it produces a response that is similar to oxidative metabolism stress. It may be that the duration of the rise in AMPK activity produced by contraction is insufficient to produce the types of changes in GLUT4 trafficking (see the text below) that are produced by the changes in long-term oxidative metabolism stress or long-term exercise.
      The signaling steps that lead to increased glucose transport in contracting muscle are variable between skeletal muscle fiber types. The levels of activation of these steps are also dependent on the duration and intensity of the signal leading to contraction (
      • Sakamoto K.
      • Goodyear L.J.
      ,
      • Sakamoto K.
      • Aschenbach W.G.
      • Hirshman M.F.
      • Goodyear L.J.
      ). In slow twitch oxidative muscle fibers, such as those that occur in soleus muscle, contraction stimulates glucose transport via a pathway that is not mediated by Akt and not inhibited by the phosphatidylinositol 3-kinase inhibitor wortmannin (
      • Lund S.
      • Holman G.D.
      • Schmitz O.
      • Pedersen O.
      ,
      • Lund S.
      • Pryor P.R.
      • Østergaard S.
      • Schmitz O.
      • Pedersen O.
      • Holman G.D.
      ,
      • Brozinick J.T.
      • Birnbaum M.J.
      ). Furthermore, in this muscle type, the response to contraction and insulin are additive (
      • Lund S.
      • Holman G.D.
      • Schmitz O.
      • Pedersen O.
      ,
      • Lund S.
      • Pryor P.R.
      • Østergaard S.
      • Schmitz O.
      • Pedersen O.
      • Holman G.D.
      ,
      • Brozinick J.T.
      • Birnbaum M.J.
      ). In fast twitch glycolytic fibers such as those found in extensor digitorum longus muscle, contraction leads to significant stimulation of Akt (
      • Sakamoto K.
      • Goodyear L.J.
      ,
      • Sakamoto K.
      • Aschenbach W.G.
      • Hirshman M.F.
      • Goodyear L.J.
      ,
      • Sakamoto K.
      • Hirshman M.F.
      • Aschenbach W.G.
      • Goodyear L.J.
      ). In this muscle type, contraction responses on glucose transport are inhibited by wortmannin (
      • Sakamoto K.
      • Aschenbach W.G.
      • Hirshman M.F.
      • Goodyear L.J.
      ,
      • Sakamoto K.
      • Hirshman M.F.
      • Aschenbach W.G.
      • Goodyear L.J.
      ). Therefore, cardiomyocytes (although very dependent on oxidative metabolism) (
      • Abel E.D.
      ), may more closely resemble fast twitch than slow twitch skeletal muscle in the signaling responses that follow a contraction stimulus.
      Several additional pathways whereby electrically induced contraction can lead to stimulated glucose transport have been proposed, and these include activation of calcium-dependent kinases and stress-activated kinases (
      • Sakamoto K.
      • Goodyear L.J.
      ,
      • Wright D.C.
      • Hucker K.A.
      • Holloszy J.O.
      • Han D.H.
      ). The signaling issues are complex, and other kinases and intermediates would have to be studied to further resolve the different signaling modes of action. Our focus has instead been on resolution of the end points of the signaling and the points of convergence of these signaling processes on the kinetics of GLUT4 trafficking.
      Regulation of GLUT4 Trafficking by Insulin—We have determined GLUT4 trafficking rates by measuring exocytosis and internalization of biotin-tagged GLUT4. Use of the biotin-photolabel tagging approach to study GLUT4 trafficking has advantages over other methods for measuring this trafficking because it can be used in native animal myocytes and without the need to isolate cell membrane fractions. Indeed, we have been able to show for the first time that insulin increases the rate of exocytosis of GLUT4 in cardiomyocytes, and we have been able to compare this with other stimulatory processes that impinge on this pathway. However, there a number of limitations of the technique as used here. These are mainly related to the numbers of cells that can be used in the assays and therefore the number of data points that can be obtained for the time courses. There is evidence, from studies on adipocyte trafficking, that GLUT4 is moved between multiple intracellular pools (
      • Karylowski O.
      • Zeigerer A.
      • Cohen A.
      • McGraw T.E.
      ,
      • Govers R.
      • Coster A.C.
      • James D.E.
      ,
      • Holman G.D.
      • Leggio L.L.
      • Cushman S.W.
      ). On leaving the plasma membrane, it first enters the endosome system and then is sorted into a specialized GLUT4 compartment. In cardiomyocytes, there is an additional complexity because GLUT4 could be recruited from T-tubule locations in addition to its perinuclear storage reservoir (
      • Becker C.
      • Sevilla L.
      • Tomas E.
      • Palacin M.
      • Zorzano A.
      • Fischer Y.
      ). The presence of multiple pools needs to be considered in analysis of whether insulin alters the endocytosis of GLUT4. If there is a local circuit between plasma membrane and endosome GLUT4, then some GLUT4 may return from the endosomes to the plasma membrane instead of being trapped in a second intracellular compartment. Therefore, internalization is a net flux process, and it is difficult to resolve, with the limited numbers of data points on the time courses, whether insulin reduces plasma membrane loss or increases the return of GLUT4 from endosomes; both processes would appear to slow net internalization and affect the proportion of GLUT4 at the surface. Our data suggest that insulin does not markedly reduce endocytosis in cardiomyocytes, but we cannot rule out a small net reduction in these steps.
      Regulation of GLUT4 Trafficking by Contraction—Electrically stimulated contraction leads to an increase in GLUT4 exocytosis without any clear reduction in GLUT4 endocytosis. The increased exocytosis response produced by contraction of cardiomyocytes occurs with only a very small elevation of Akt phosphorylation. However, Akt has been clearly shown to be a key intermediate in insulin stimulation of GLUT4 exocytosis (
      • Jiang Z.Y.
      • Zhou Q.L.
      • Coleman K.A.
      • Chouinard M.
      • Boese Q.
      • Czech M.P.
      ). During contraction, some bypassing of this step may occur. Direct stimulation of cytoskeletal proteins may result from electrically stimulated contraction, and these changes may facilitate GLUT4 exocytosis. Cytoskeletal proteins are known to be permissive in insulin-stimulated GLUT4 translocation (
      • Tong P.
      • Khayat Z.A.
      • Huang C.
      • Patel N.
      • Ueyama A.
      • Klip A.
      ,
      • Chunqiu H.J.
      • Pessin J.E.
      ,
      • Bose A.
      • Robida S.
      • Furcinitti P.S.
      • Chawla A.
      • Fogarty K.
      • Corvera S.
      • Czech M.P.
      ). Although AMPK phosphorylation increases after contraction, this increase is not inhibited by wortmannin and does not correspond with the trafficking changes. It may be that the stimulus is of insufficient duration to produce a detectable block in GLUT4 endocytosis.
      Regulation of GLUT4 Trafficking by Oxidative Metabolism Stress—The trafficking data collectively indicate that oxidative metabolism stress (induced by hypoxia or oligomycin treatment) leads to a lowering of the GLUT4 endocytosis rate without any marked acceleration of the exocytosis rate. By contrast, transferrin internalization is very rapid in cardiomyocytes, in both the presence and absence of oxidative metabolism stress. Therefore, some degree of specificity occurs in the slowing of GLUT4 internalization. The level of reduction in GLUT4 endocytosis is greater in cells that are treated with only oligomycin alone (Fig. 7) compared with cells treated with insulin and oligomycin with a subsequent insulin removal step (Fig. 6). This may occur because of the observed reduction in AMPK signaling in insulin-treated cells. Alternatively, there may be a more complex trafficking in operation, in which insulin and oxidative metabolism stress recruit from and return GLUT4 to separate pools (
      • Becker C.
      • Sevilla L.
      • Tomas E.
      • Palacin M.
      • Zorzano A.
      • Fischer Y.
      ). If this is the case, then only the return of the GLUT4 to the pool associated with oxidative metabolism stress would be expected to be inhibited by oligomycin. The return to the insulin-recruited pool would not be expected to be affected because insulin removal would allow these transporters to be returned at an uninhibited rate. The combined effect would be the observed greater inhibition of internalization produced by oligomycin treatment compared with oligomycin plus insulin treatment.
      The increase in AMPK phosphorylation that occurs in response to oligomycin treatment is the same at 5 and 60 min, but it takes a longer time for the transport stimulation to reach a maximum. These changes are consistent with the observed slow GLUT4 exocytosis rate. When there is no stimulation of exocytosis, a sustained AMPK stimulus may be needed to continuously block endocytosis and thereby capture GLUT4 at the surface. Continuous contraction during exercise might also be expected to block endocytosis if the AMPK signal were sustained.
      In contrast to the insulin-mediated and slight reductions in endocytosis of GLUT4 that have been noted above, prolonged insulin action leads to an acceleration of endocytosis (
      • Pryor P.R.
      • Liu S.C.
      • Clark A.E.
      • Yang J.
      • Holman G.D.
      • Tosh D.
      ). This acceleration is reversed by long-term metformin treatment, which produces an inhibitory effect on endocytosis (
      • Pryor P.R.
      • Liu S.C.
      • Clark A.E.
      • Yang J.
      • Holman G.D.
      • Tosh D.
      ). The inhibition of endocytosis that follows metformin treatment is consistent with the known stimulatory effect of metformin on AMPK activity (
      • Fryer L.G.
      • Parbu-Patel A.
      • Carling D.
      ,
      • Zhou G.
      • Myers R.
      • Li Y.
      • Chen Y.
      • Shen X.
      • Fenyk-Melody J.
      • Wu M.
      • Ventre J.
      • Doebber T.
      • Fujii N.
      • Musi N.
      • Hirshman M.F.
      • Goodyear L.J.
      • Moller D.E.
      ) and our present study, which indicates that activation of AMPK through oxidative metabolism stress leads to inhibition of GLUT4 endocytosis.
      The kinetic approach to studying GLUT4 trafficking described here may ultimately be applicable to human skeletal muscle of persons with and without type 2 diabetes. In heart and skeletal muscle of diabetic persons, it will also be important to further understand how the interactions among insulin, oxidative metabolism stress, and contraction/exercise pathways occur and consequently lead to glucose transport stimulation. In heart, because GLUT4 recruitment can be increased by either contraction-stimulated exocytosis or oxidative metabolism stress-mediated reduction in endocytosis, there are potentially multiple means by which insulin resistance can be bypassed.

      References

        • Carling D.
        Trends Biochem. Sci. 2004; 29: 18-24
        • Hardie D.G.
        Endocrinology. 2003; 144: 5179-5183
        • Fryer L.G.
        • Foufelle F.
        • Barnes K.
        • Baldwin S.A.
        • Woods A.
        • Carling D.
        Biochem. J. 2002; 363: 167-174
        • Barnes K.
        • Ingram J.C.
        • Porras O.H.
        • Barros L.F.
        • Hudson E.R.
        • Fryer L.G.
        • Foufelle F.
        • Carling D.
        • Hardie D.G.
        • Baldwin S.A.
        J. Cell Sci. 2002; 115: 2433-2442
        • Kurth-Kraczek E.J.
        • Hirshman M.F.
        • Goodyear L.J.
        • Winder W.W.
        Diabetes. 1999; 48: 1667-1671
        • Koistinen H.A.
        • Galuska D.
        • Chibalin A.V.
        • Yang J.
        • Zierath J.R.
        • Holman G.D.
        • Wallberg-Henriksson H.
        Diabetes. 2003; 52: 1066-1072
        • Mu J.
        • Brozinick J.T.
        • Valladares O.
        • Bucan M.
        • Birnbaum M.J.
        Mol. Cell. 2001; 7: 1085-1094
        • Barnes B.R.
        • Ryder J.W.
        • Steiler T.L.
        • Fryer L.G.
        • Carling D.
        • Zierath J.R.
        Diabetes. 2002; 51: 2703-2708
        • Till M.
        • Ouwens D.M.
        • Kessler A.
        • Eckel J.
        Biochem. J. 2000; 346 (Pt 3): 841-847
        • Coven D.L.
        • Hu X.
        • Cong L.
        • Bergeron R.
        • Shulman G.I.
        • Hardie D.G.
        • Young L.H.
        Am. J. Physiol. Endocrinol. Metab. 2003; 285: E629-E636
        • Satoh S.
        • Nishimura H.
        • Clark A.E.
        • Kozka I.J.
        • Vannucci S.J.
        • Simpson I.A.
        • Quon M.J.
        • Cushman S.W.
        • Holman G.D.
        J. Biol. Chem. 1993; 268: 17820-17829
        • Yang J.
        • Holman G.D.
        J. Biol. Chem. 1993; 268: 4600-4603
        • Yang J.
        • Hodel A.
        • Holman G.D.
        J. Biol. Chem. 2002; 277: 6559-6566
        • Zeigerer A.
        • Lampson M.A.
        • Karylowski O.
        • Sabatini D.D.
        • Adesnik M.
        • Ren M.
        • McGraw T.E.
        Mol. Biol. Cell. 2002; 13: 2421-2435
        • Czech M.P.
        • Buxton J.M.
        J. Biol. Chem. 1993; 268: 9187-9190
        • Karylowski O.
        • Zeigerer A.
        • Cohen A.
        • McGraw T.E.
        Mol. Biol. Cell. 2004; 15: 870-882
        • Govers R.
        • Coster A.C.
        • James D.E.
        Mol. Cell. Biol. 2004; 24: 6456-6466
        • Jhun B.H.
        • Rampal A.L.
        • Liu H.
        • Lachaal M.
        • Jung C.Y.
        J. Biol. Chem. 1992; 267: 17710-17715
        • Pryor P.R.
        • Liu S.C.
        • Clark A.E.
        • Yang J.
        • Holman G.D.
        • Tosh D.
        Biochem. J. 2000; 348: 83-91
        • Eckel J.
        • Reinauer H.
        Biochim. Biophys. Acta. 1980; 629: 510-521
        • Fischer Y.
        • Thomas J.
        • Holman G.D.
        • Rose H.
        • Kammermeier H.
        Am. J. Physiol. 1996; 270: C1204-C1210
        • Shepherd P.R.
        • Soos M.A.
        • Siddle K.
        Biochem. Biophys. Res. Commun. 1995; 211: 535-539
        • Hashimoto M.
        • Yang J.
        • Holman G.D.
        Chembiochem. 2001; 2: 52-59
        • Holman G.D.
        • Kozka I.J.
        • Clark A.E.
        • Flower C.J.
        • Saltis J.
        • Habberfield A.D.
        • Simpson I.A.
        • Cushman S.W.
        J. Biol. Chem. 1990; 265: 18172-18179
        • Czech M.P.
        • Corvera S.
        J. Biol. Chem. 1999; 274: 1865-1868
        • Calderhead D.M.
        • Kitagawa K.
        • Tanner L.I.
        • Holman G.D.
        • Lienhard G.E.
        J. Biol. Chem. 1990; 265: 13800-13808
        • Becker C.
        • Sevilla L.
        • Tomas E.
        • Palacin M.
        • Zorzano A.
        • Fischer Y.
        Endocrinology. 2001; 142: 5267-5276
        • Holman G.D.
        • Karim A.R.
        • Karim B.
        Biochim. Biophys. Acta. 1988; 946: 75-84
        • Tanner L.I.
        • Lienhard G.E.
        J. Biol. Chem. 1987; 262: 8975-8980
        • Abel E.D.
        Front. Biosci. 2004; 9: 201-215
        • Taegtmeyer H.
        • McNulty P.
        • Young M.E.
        Circulation. 2002; 105: 1727-1733
        • Scheuermann-Freestone M.
        • Madsen P.L.
        • Manners D.
        • Blamire A.M.
        • Buckingham R.E.
        • Styles P.
        • Radda G.K.
        • Neubauer S.
        • Clarke K.
        Circulation. 2003; 107: 3040-3046
        • Desrois M.
        • Sidell R.J.
        • Gauguier D.
        • King L.M.
        • Radda G.K.
        • Clarke K.
        Cardiovasc. Res. 2004; 61: 288-296
        • Fryer L.G.
        • Parbu-Patel A.
        • Carling D.
        J. Biol. Chem. 2002; 277: 25226-25232
        • Zhou G.
        • Myers R.
        • Li Y.
        • Chen Y.
        • Shen X.
        • Fenyk-Melody J.
        • Wu M.
        • Ventre J.
        • Doebber T.
        • Fujii N.
        • Musi N.
        • Hirshman M.F.
        • Goodyear L.J.
        • Moller D.E.
        J. Clin. Investig. 2001; 108: 1167-1174
        • Marsin A.S.
        • Bertrand L.
        • Rider M.H.
        • Deprez J.
        • Beauloye C.
        • Van den Vincent M.F.
        • Berghe G.
        • Carling D.
        • Hue L.
        Curr. Biol. 2000; 10: 1247-1255
        • Russell R.R.
        • II I
        • Bergeron R.
        • Shulman G.I.
        • Young L.H.
        Am. J. Physiol. 1999; 277: H643-H649
        • Li J.
        • Hu X.
        • Selvakumar P.
        • Russell R.R.
        • II I
        • Cushman S.W.
        • Holman G.D.
        • Young L.H.
        Am. J. Physiol. Endocrinol. Metab. 2004; 287: E834-E841
        • Sakamoto K.
        • Goodyear L.J.
        J. Appl. Physiol. 2002; 93: 369-383
        • Sakamoto K.
        • Aschenbach W.G.
        • Hirshman M.F.
        • Goodyear L.J.
        Am. J. Physiol. Endocrinol. Metab. 2003; 285: E1081-E1088
        • Lund S.
        • Holman G.D.
        • Schmitz O.
        • Pedersen O.
        Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5817-5821
        • Lund S.
        • Pryor P.R.
        • Østergaard S.
        • Schmitz O.
        • Pedersen O.
        • Holman G.D.
        FEBS Lett. 1998; 425: 472-474
        • Brozinick J.T.
        • Birnbaum M.J.
        J. Biol. Chem. 1998; 273: 14679-14682
        • Sakamoto K.
        • Hirshman M.F.
        • Aschenbach W.G.
        • Goodyear L.J.
        J. Biol. Chem. 2002; 277: 11910-11917
        • Wright D.C.
        • Hucker K.A.
        • Holloszy J.O.
        • Han D.H.
        Diabetes. 2004; 53: 330-335
        • Holman G.D.
        • Leggio L.L.
        • Cushman S.W.
        J. Biol. Chem. 1994; 269: 17516-17524
        • Jiang Z.Y.
        • Zhou Q.L.
        • Coleman K.A.
        • Chouinard M.
        • Boese Q.
        • Czech M.P.
        Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 7569-7574
        • Tong P.
        • Khayat Z.A.
        • Huang C.
        • Patel N.
        • Ueyama A.
        • Klip A.
        J. Clin. Investig. 2001; 108: 371-381
        • Chunqiu H.J.
        • Pessin J.E.
        Mol. Biol. Cell. 2003; 14: 3578-3591
        • Bose A.
        • Robida S.
        • Furcinitti P.S.
        • Chawla A.
        • Fogarty K.
        • Corvera S.
        • Czech M.P.
        Mol. Cell. Biol. 2004; 24: 5447-5458