Glucose Transport Activity in Skeletal Muscles from Transgenic Mice Overexpressing GLUTl

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 269, No. 28, Issue of July 15, pp , 1994 Printed in U.S.A. Glucose Transport Activity in Skeletal Muscles from Transgenic Mice Overexpressing GLUTl INCREASED BASAL TRANSPORT IS ASSOCIATED WITH A DEFECTIVE RESPONSE TO DIVERSE STIMULI THAT ACTIVATE GLUT4* (Received for publication, April 7, 1994) Eric A. Gulve$$, Jian-Ming Ren$n, Bess Adkins Marshallll**, Jiaping GaoS, Polly A. HansenS $$, John 0. HolloszyS, and Mike Muecklerll From the Departments of $Medicine and IlCell Biology and Physiology, Washington University School of Medicine, St. Louis. Missouri Glucose transport activity wasexamined in transgenic mice overexpressing the human GLUTl glucose transporter in skeletal muscles. Basal transport activity measured in vitro with the glucose analog 2-deoxy-D-glucose (1 m ~ was ) increased 2-8-fold in four different muscle preparations. Incubation of muscles from control nontransgenic littermates with a maximally effective concentration of insulin or with insulin-like growth factor-1 resulted in glucose transport rates that were 2-3-fold higher than basal. In contrast, insulin did not stimulate glucose transport activity in three different muscle preparations from transgenic animals; insulinlike growth factor-1 was similarly ineffective. Activation of System A amino acid transport activity (measured with the nonmetabolizable analog cw-methylaminoisobutyrate) by insulin was not impaired in muscles from transgenic mice, indicating that the defect does not involve the insulin receptor. In skeletal muscle, glucose transport can be activated by muscle contractions or hypoxia via a pathway separate from that activated by insulin. Incubation of muscles under hypoxic conditions or stimulation of muscles to contract in situ did not in- crease glucose transport activity in muscles from GLUTl-overexpressing mice, in contrast to the stimulatory effects measured in muscles from control animals. These data suggest that increased glucose flux per se into skeletal muscle results in resistance of GLUT4 to activation by insulin and various other stimuli that activate glucose transport by mechanisms distinct from that of insulin. GLUT1-overexpressing mice thus pro- vide a new model system for studying the effects of glucose-induced resistance to activation of glucose transport. * This work was supported in part by National Institutes of Health Grants DK18986 (to J. 0. H.) and DK38495 (to M. M.), the Juvenile Diabetes Foundation International, and the DiabetesResearch and Training Center at Washington University School of Medicine. 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. 0 To whom correspondence should be addressed: GD Searle and Co., MailZone TlF, 800N. Lindbergh Blvd., St. Louis, MO Tel.: ; Fax: Supported by National Institutes of Health Postdoctoral Training Grant AG Current address: Miles Inc., Institute for Metabolic Disorders, 400 Morgan La., West Haven, CT ** SuDDorted bv National Institutes of Health Postdoctoral Training - Grant 6K $$ Supported by National Institutes of Health Postdoctoral Training Grant AG Skeletal muscle is the major site for insulin-mediated wholebody glucose disposal (1, 2) and as such is an important contributor to the maintenance of glucose homeostasis (reviewed in Ref. 3). Impairment of muscle glucose uptake contributes to the insulin resistance characteristic of obesity (3, 4), bed rest (51, and noninsulin-dependent diabetes (3, 4, 6). Skeletal muscle expresses two isoforms of the glucose transporter protein, GLUTl and GLUT4 (for reviews see Refs. 7 and 8). GLUT1, which is constitutively targeted to the plasma mem- brane, is believed to mediate basal glucose transport activity (7, 8). In contrast, in the basal state GLUT4 resides in an inactive intracellular location (7, 8). In response to stimuli such as insulin or contractile activity, GLUT4 is translocated to the skeletal muscle plasma membrane (9-13). We have constructed a line of transgenic mice overexpressing GLUTl in skeletal muscle (14). These mice possess lower fasting and fed plasma glucose concentrations and dispose of an oral glucose load faster than their control littermates (14). This finding is not due to alterations in pancreatic hormones (14) but rather to an elevated basal glucose transport activity, which results in increased muscle glucose metabolism (15). In the present study we have examined the effects of various stimuli, which normally activate glucose transport activity in skeletal muscle. We find that glucose transport activity cannot be further activated by insulin, insulin-like growth factor-1, hypoxia, or muscle contractions. This defect is specific for certain aspects of carbohydrate metabolism, and may provide a model system for studying the effects of excessive glucose flw into insulin-sensitive tissues. EXPERIMENTALPROCEDURES Muteria~s-2-Deoxy-~-[1,2-~H]glucose was purchased from American Radiolabeled Chemicals. [U-14ClMannitol, a-[l-14clmethylaminoisobutyrate, and ~-[l-~h]mannitol were purchased from DuPont NEN. Porcine insulin (Iletin 11) was from Lilly. Recombinant human insulin-like growth factor-1 (1GF-l)I was from U. S. Biochemical Corp. D-Glucose, bovine serum albumin, and nonradioactive forms of 2-deoxy-~-glucose (2-DG), a-methylaminoisobutyrate (MeAIB), and D-mannitol were obtained from Sigma. Construction of Dansgenic Mice-The construction of transgenic mice overexpressing the human GLUTl glucose transporter was described previously (14). The minigene in this construct contains a kilobase cdna fragment encoding the human GLUTl glucose transporter under the regulation of the 1.2-kilobase rat myosin light chain-2 promoter. Expression of the transgene is restricted to skeletal muscle and does not appreciably affect expression of the GLUT4 isoform (14). Animals used for experiments were littermates resulting from breeding of non-transgenic C57BU6 x SJL F, mice with a single line of GLUTl The abbreviations used are: IGF, insulin-like growth factor; 2-DG, 2-deoxy-~-glucose; MeAIB, a-methylaminoisobutyate; EDL, extensor digitorum longus; KHB, Krebs-Henseleit buffer.

2 Muscle Glucose Dansport in Mice Overexpressing GLUTl transgenic mice (GLUTlb line) carrying one copy of the transgenic locus. Thus litters consisted of a ratio of control and heterozy- CON: Basa1 gous transgenic mice. Deatment of Animals-Animals were housed in a room maintained at 23 "C with a fixed 1ight:dark cycle (lights on from 6 a.m. to 6 p.m.) and given access to Purina chow and water ad libitum. All experiments began in the morning (-10 a.m.). Fed mice were anesthetized by an intraperitoneal injection of sodium pentobarbital (6 mg/100 g of body weight), and intact epitrochlearis, soleus, extensor digitorum longus, and flexor digitorum brevis were excised for incubation. For experiments involving activation of glucose transport by in situ stimulation, the sciatic nerve was isolated in the medial aspect of the upper thigh of anesthetized mice. The nerve was cut, and the distal portion was connected to subminiature electrodes. The knee and ankle joints were fixed at - 100" angles, and one EDL muscle from each mouse was stimulated indirectly via the nerve by a modification of a procedure described previously (16). Briefly, 0.1-ms square wave pulses were de- Sol EDL EPl livered at 100 Hz in 250-ms-long trains at a rate of l/~. Stimulation was performed for four 5-min-long intervals, with 1-min rest periods be- FIG. 1. Basal and maximally insulin-stimulated glucose transtween intervals. Muscles were dissected out immediately after the final port activity in skeletal muscles from transgenic mice and constimulation period. trol littermates. Intact soleus (Sol), EDL, and epitrochlearis (Epi) Muscle Incubation-Muscles were incubated at 35 "C in a Dubnoff muscles were incubated for 30 min at 35 "C in the presence or absence metabolic incubator in 2 ml of pregassed Krebs-Henseleit bicarbonate of 2000 microunits/ml insulin. They were then washed for 10 min at 29 "C, followed by measurement of glucose transport activity with 1 mm buffer supplemented with 0.1% radioimmunoassay grade bovine serum 2-DG as described under "Experimental Procedures." When insulin was albumin, 8 mm glucose, and 32 mm mannitol, with or without insulin or present during the 35 "C incubation it was also included in the wash IGF-1 as specified in the text. Flasks were shaken continuously and transport assay. Values are means f S.E. for 34 (control (CON)) throughout the experiment. The gas concentration in the flasks was and 7 (transgenic(trg)) muscles/group, except for epitrochlearis 95% 0,, 5% CO,, except in experiments involving hypoxia, where the muscles where n = muscles/group. Inset, glucose transport assay gas phase was 95% N,, 5% CO, (17). The incubation time at 35 "C was was assessed in the EDL using 0.2 m~ 2-DG. Values are means f S.E. 30 min except in experiments involving in situ stimulation or hypoxia. for 7 muscles/group. Comparison of insulin to corresponding basal: as- After in situ stimulation, the stimulated and contralateral unstimu- terisk indicates p < 0.001; NS, not significantly different. lated muscles were transferred directly to the wash step (see below). In experiments involving stimulation of glucose transport by hypoxia, all muscles (Le. hypoxic and control oxygenated muscles) were incubated ence of insulin. Uptake of MeAIB into skeletal muscle is linear, with for 45 min. only a minor contribution from nonsaturable processes (22). Muscles After the initial incubation period, all muscles were washed for 10 were processed and transport activity calculated as described above. min at 29 "C in oxygenated medium to remove glucose and to lower the System A transport activity is expressed as nanomoles of MeAIB accutemperature before assay of glucose or amino acid transport activity. mulated per ml of intracellular water per 30 min. The medium (2 ml/flask) consisted of KHB containing bovine serum Measurement of Muscle Glycogen4lycogen was assayed perchlo- in albumin and 40 mm mannitol. The gas phase in the flasks was 95% 0,, ric acid extracts of total muscle homogenates by the amyloglucosidase 5% CO,. If insulin or IGF-1 was present during the "C 35 incubation, it method (24). was also added to the wash step. Statistical Analysis-Data in the text and figures are reported as Determination of Glucose 'IFansport Activity-Glucose transport ac- means S.E. The effects of GLUTl overexpression were assessed two in tivity was measured with the nonmetabolizable glucose analog 2-DG as different ways. First, the ability of muscles from transgenic animals to described previously (18-20). Briefly, muscles were incubated for 20 respond to a given stimulus (insulin, IGF-1, hypoxia, or contractions) min at 29 "C in 1.0 ml of oxygenated KHB supplemented with 1 mm was evaluated. This was assessed for each of group animals (control and 2-deoxy-~-[1,2-~Hlglucose (1.7 mci/mmol) and 39 mm [U-"C]mannitol transgenic) by comparing basal to stimulated conditions by means of a (8.5 pci/mmol). If insulin or IGF-1 was present during the 35 "C incu- Student's t test (paired or unpaired as appropriate). For most experibation, it was also included in the transport assay. The gas phase in the ments, which were performed by pairing muscles from each animal in flasks was 95% 0,, 5% CO,. Uptake of 2-DG into muscles from control the basal versus stimulated state, an error term was calculated for the and transgenic mice under basal and stimulated conditions is linear effect of a given stimulus above basal ("difference"). Therefore the effect over the time period assayed., Furthermore, at stimulated rates of of GLUTl overexpression was also assessed by unpaired t tests comuptake more than 98% of the intracellular 2-DG is phosphorylated. For paring the difference term for control with that for transgenic mice. the experiment involving in situ muscle contractions, muscles were An 01 level of 0.05 was the minimum considered to be statistically processed by homogenizing in 0.3 M perchloric acid at 4 "C. An aliquot of significant. the total homogenate was reserved for assay of glycogen. The remaining homogenate was centrifuged, and aliquots of the supernatant were used RESULTS ANDDISCUSSION for determination of glucose transport activity. For all other experi- Skeletal muscles from GLUTlb transgenic mice carrying a ments muscles were processed by boiling for 10 min in 1 ml of water single copy of the transgenic human GLUTl locus contain sev- (21). Extracts were cooled on ice, vortexed, and centrifuged at 1,000 x g. Aliquots of muscle supernatant and incubation medium were eralfold higher levels of GLUTl as assessed by Western blot counted for radioactivity, and the extracellular and intracellular 2-DG analysis (14, 15). Overexpression of GLUTl results in an inaccumulation was calculated. Quantification of glucose transport activ- crease in basal glucose transport activity (Ref. 15 and Fig. 1) ity by boiling muscles yields values that are not significantly different relative to that measured in muscles from nontransgenic litfrom those obtained by acid extraction (20). Glucose transport activity termates. Our previous findings (15) characterized muscles (epis expressed as micromoles of 2-DG accumulated per ml of intracellular itrochlearis and EDL) enriched in Type 2b fibers (25, 26). The water per 20 min. Determination of System A Amino Acid Dansport Activity-Activity present results extend those findings to the soleus (Fig. l), of the System A neutral amino acid transporter was assessed using the which is comprised of primarily Type 1 fibers (25), and to the nonmetabolizable amino acid analog MeAIB, which is specific for this flexor digitorum brevis, composed mainly of m e 2a fibers (27). transporter in muscle (22) and other tissues. Uptake was assayed as described previously (23), except that the incubation period was 30 min, and radioactivity was increased to 0.18 pci/ml (['4C]MeAIB) and 0.9 pci/ml ([3Hlmannitol). Briefly, muscles were incubated at 29 "C in 1.0 ml of KHB supplemented with 0.1 mm MeAIB, in the absence or pres- ' P. A. Hansen, E. A. Gulve, and J. 0. Holloszy, unpublished results mm 2-DG I In the flexor, basal glucose transport activity averaged (n = 8) and (n = 5) pmol of 2-DG/ml/20 min in muscles from control and transgenic mice, respectively (means 2 S.E.;p < 0.001). In different experiments transport activity in muscles from transgenic animals was 2-3-fold greater in the soleus and flexor digitorurn brevis and 4-8-fold higher in the epitrochlearis and EDL. The finding of elevated glucose trans-

3 18368 Muscle Glucose Dunsport in Mice Overexpressing GLUTl -1.5 Control Transgenic FIG. 3. Effect of insulin on system A amino acid transport activity in EDL muscles from transgenic mice and control littermates. Paired muscles were incubated for 30 min at 35 C in the presence or absence of 2000 microunits/ml insulin. They were then washed for 10 min at 29 C, followed by measurement of system A amino acid transport activity with 0.1 mm MeAIB as described under Experimental Procedures. When insulin was present during the 35 C incubation it was also included in the wash and transport assay. Values are means? S.E. for 8 muscles/group. The change in transport activity relative to.1.5 J the unstimulated muscle (Difference) has been calculated. Comparison Control Transgenic of insulin to corresponding basal: asterisk indicatesp < 0.001, Compari- FIG. 2. Effect of IGF-1 and hypoxia on glucose transport activ- son of transgenic difference to control difference: NS, not significantly ity in EDL muscles from transgenic mice and control litter- different. mates. a, muscles (7-8/group) were incubated for 30 min at 35 C in the presence or absence of25 nm IGF-1. b, muscles (5/group) were incubated for 45 min at 35 C in either oxygenated or hypoxic medium. All muscles In order to determine whether the muscles from animals were then washed for 10 min at 29 C in oxygenated medium, followed overexpressing GLUTl exhibit a generalized resistance to inby measurement of glucose transport activity with 1 mm 2-DG as de- sulin action, we measured System A amino acid transport acscribed under Experimental Procedures. When IGF-1 was present tivity in muscles from control and transgenic animals. The during the 35 C incubation it was also included in the wash and transport assay. Values are means? S.E. for muscles paired with or without System A transporter is the primary insulin-regulatable neustimulation by IGF-1 or hypoxia. The effect of each stimulus relative to tral amino acid transport system in skeletal muscle (28). Sysbasal conditions (Difference) is also shown. Comparison of IGF-1 or tem A activity, measured with the nonmetabolizable analog hypoxia to corresponding basal: asterisk indicates p < 0,001; NS, not MeAIB, was 2-fold higher in insulin-treated muscles from nonsignificantly different. Comparison of transgenic difference to control transgenic mice relative to that measured in contralateral difference: dagger indicates p < muscles incubated without insulin (Fig. 3). Basal MeAIB transport activity in muscles from transgenic animals has been con- port was lower in muscles from the transgenic animals relative firmed with the glucose analog 3-O-methyl-~-glucose (15). to their nontransgenic littermates; however, System A trans- Fig. 1 also demonstrates that muscles from transgenic ani- port activity was stimulated to the same extent as in the nonmals do not increase transport activity further when incubated transgenic controls. This finding indicates that the inability of with a maximally effective concentration of insulin (2000 insulin to stimulate glucose transport activity is not due to a microunits/ml), in contrast to muscles from nontransgenic lit- defect in upstream insulin signaling events. termates. This pattern was seen consistently in three different Insulin and IGF-1 signaling share in common the fact that muscle types in multiple experiments. Activity of the enzyme their actions are mediated by tyrosine kinase moieties in their hexokinase is slightly (15%) greater in muscles from mice over- respective receptors (29). Furthermore, effects of insulin and expressing GLUTl (15). To ensure that hexokinase activity was IGF-1 on skeletal muscle glucose transport activity are not not limiting for 2-DG phosphorylation in insulin-treated additive, suggesting a shared intracellular mechanism (30,311. muscles from transgenic mice, we assayed transport activity in another experiment after lowering the 2-DG concentration by 80%. Transport rates were reduced to -20% of those measured at 1 mm 2-DG, yet insulin was still ineffective in promoting an increase in transport activity in muscles from transgenic animals (Fig. 1, inset). In most experiments, paired muscles from a given mouse were incubated with and without insulin. Nontransgenic animals consistently showed a higher rate of trans- port in the insulin-treated muscles, relative to basal conditions, whereas this was rarely the case in muscles from transgenic mice. Recent experiments involving euglycernic hyperinsulinemic clamps have demonstrated that insulin action is impaired in vivo as well.3 We then examined whether the inability to further increase glucose transport activity in muscles from transgenic mice is restricted to insulin-mediated transport. Treatment of muscles with 25 nm IGF-1 more than doubled glucose transport activity in EDL muscles from control mice (Fig. 2a). In contrast, IGF-1 was ineffective in muscles from mice overexpressing GLUT1. These results demonstrate that the defective response to insulin cannot be attributed solely to impaired insulin receptor function. B. A. Marshall and M. Mueckler, manuscript in preparation. Control Transgenic In skeletal muscle, glucose transport can be activated by a pathway which is distinct from that stimulated by insulin. The effects of muscle contractions (18, 32-34) or hypoxia (17) are additive to the effects of insulin and appear to be mediated by increases in intracellular calcium concentration (17,35). When incubated under hypoxic conditions, muscles from control animals showed the expected increase in glucose transport activity, whereas those from transgenic mice did not respond with a further increase in glucose transport (Fig. 2b). The mechanisms that trigger calcium release in hypoxic conditions are not well understood. Because muscles from GLUTl transgenic animals contain extensive glycogen depots (15), we considered the possibility that hypoxia might not be as effective a stimulus for increasing intracellular calcium (and thus for stimulating glu- cose transport activity) as it is in muscles from control animals. To circumvent this potential problem we stimulated EDL muscles to contract in situ, thereby raising intracellular calcium directly by release from the sarcoplasmic reticulum. The stimulation period was doubled from that previously shown to result in a maximally effective stimulus for increasing skeletal muscle glucose transport activity (16). In muscles from transgenic mice glucose transport activity was actually lower after contractions, in contrast to the situation in the controls (Fig. 4a). The reason for the reduction in transport activity is not

4 -1.5 Muscle Glucose Dansport in Mice Overexpressing GLUTl Basal Contractions 13 * Difference Transgenic Control = 4 63 Difference % 75 Transgenic Control FIG. 4. Effect of in situ muscle contractions on glucose trans- flux results in a postbinding defect (36, 37) characterized by port activity and glycogen concentration in EDL muscles from impaired insulin-stimulated glucose transporter translocation transgenic mice and control littermates. The sciatic nerve to one limb was isolated, and EDL muscles were indirectly activated to con- (37). Unfortunately translocation of glucose transporters is tract in situ with a tetanic stimulation protocol as described under very difficult to measure in skeletal muscle by conventional Experimental Procedures. Muscles from the stimulated and contralat- subcellular fractionation techniques because of extensive crosseral unstimulated leg were then dissected out and washed 29 at C for contamination of the plasma membrane fraction with intracel- 10 min before measurement of glucose transport activity with 1 mm 2-DG. A portion of each muscle extract was used for measurement of lular membranes (13, 41). Both of our laboratories have at- glycogen concentration. a, glucose transport activity. b, muscle glycogen concentration. Values are means t S.E. for 7 (control) and 8 (transgenic) pairs of muscles. The change in glucose transport activity or glycogen relative to the unstimulated muscle (Difference) has been calculated. Comparison of contractions to corresponding basal: asterisk indicates p < 0.001, Comparison of transgenic difference to control difference: dagger indicates p < apparent, but the important finding from this experiment is that contractile activity did not increase glucose transport activity. In situ contractions stimulated glycogen breakdown (60-70% depletion of glycogen) to a similar extent in muscles from the control and transgenic mice (Fig. 4b), suggesting that the inability of muscles from transgenic mice to further increase glucose transport cannot be explained by an insufficient contractile stimulus. Furthermore, the impaired response to contractions in transgenic muscle is specific for activation of glu- cose transport and does not extend to activation of glycogenolysis. Studies performed in vivo and in vitro have suggested that increased flux of glucose into insulin-sensitive tissues results in the induction of insulin resistance (36-38). Our findings strongly suggest that increased glucose flux into muscle by itself can induce insulin resistance at the level of glucose transport activity. Previous studies in adipocytes and skeletal muscle have not examined the effects of excessive glucose flux on glucose transport activated by stimuli other than insulin, i.e. IGF-1, hypoxia, or muscle contractions. Thus these earlier studies assumed that excessive glucose flux induced glucose transport impairment that is specific for insulin. Our findings are novel in that they demonstrate that elevated glucose flux does not result in impairment solely of insulin-mediated glucose transport. The defect in our model system is most likely generalized resistance of GLUT4 activation to all known stimuli. It seems likely that this defect resides in the machinery linking GLUT4 to activation by diverse stimuli. The in vivo environment of these GLUTl-overexpressing mice differs from the conditions seen in non-insulin-dependent diabetes, which is characterized by hyperglycemia and often by hyperinsulinemia (3). In contrast, the GLUTl transgenic mice are characterized by hypoglycemia in the face of unaltered levels of circulating insulin and glucagon. The advantage of this model is that it makes possible examination of the molecular basis of insulin resistance due solely to elevated glucose flux, without the confounding effects of hyperglycemia and hyperinsulinemia, which may independently contribute to insulin resistance (3, 39). The mechanism for the defective response to activators of glucose transport in muscles overexpressing GLUTl is not yet known. The defective response to hypoxia or muscle contractions, which activate glucose transport by a pathway distinct from insulin, indicates that the primary defect resulting from elevated glucose flux does not involve impairment of the insulin receptor and its tyrosine kinase signaling. Resistance to activation of transport may be due to impaired translocation of the GLUT4 isoform (in the face of unaltered total GLUT4 levels). Translocation of GLUT4 normally occurs in response to stimuli such as insulin, hypoxia, or muscle contractions (9-13, 17). It is known that increased glucose flux can result in impaired insulin action without a change in total GLUT4 levels (37, 40). Studies in adipocytes have demonstrated that elevated glucose tempted to measure translocation in mouse muscle without success (i.e. inconsistent insulin-stimulated translocation even in muscles from control animals). To the best of our knowledge, GLUT4 translocation in mouse muscle has not been measured by laboratories under any conditions. Current protocols for rat skeletal muscle require large amounts of starting material; furthermore, the relative increases ( % in different laboratories; Refs and 17) in plasma membrane transporter number induced by insulin are considerably less than the relative increases ( %; Refs. 9, 10, 17, 18, and 32-34) in glucose transport activity. In summary, GLUTl transgenic mice provide a novel model system for studying the contribution of excessive glucose flux to the development of impaired glucose transport activity. Elevated glucose flux apparently results in a postreceptor defect in distal elements of the signaling pathway, which does not affect other pathways activated by insulin, such as amino acid transport. Thus in this model, muscle resistance to insulin and other stimuli that activate GLUT4 is dissociated from hyperglycemia or perturbations in pancreatic hormone levels that can secondarily alter glucose transport. Acknowledgments-We thank Dan Johnson, Xiang-Jing Wang, Connie Skillington, Guofeng Zhou, and May Chen for their excellent technical assistance. REFERENCES 1. DeFronzo, R. A., Jacot, E., Jequier, E., Maeder, E., Wahren, J., and Felber, J. P. (1981) Diabetes 30, Kraegen, E. W., James, D. E., Jenkins, A. B., and Chisholm, D. J. (1985) Am. J. Physiol. 248, E353-E DeFronzo, R. A. (1988) Diabetes 37, Dohm, G. L., Tapscott, E. B., Pones, W. J., Dabbs, D. J., Flickinger, E. G., Meelheim, D., Fushiki, T., Atkinson, S. M., Elton, C. W., and Caro, J. F, (1988) J. Clin. Inuest. 82, Mikines, K. J., Richter, E. A,, Dela, F., and Galbo, H. (1991) J. Appl. Physiol. 70, Rothman, D. L., Shulman, R. G., and Shulman, G. I. (1992) J. Clin. Inuest. 89, Mueckler, M. (1990) Diabetes 39, Burant, C. F., Sivitz, W. I., Fukumoto, H., Kayano, T., Nagamatsu, S., Seino, S., Pessin, J. E., and Bell, G. I. (1991) Recent Prog. Horm. Res. 47, Hip, A., Ramlal, T., Bilan, P. J., Cartee, G. D., Gulve, E. A., and Holloszy, J. 0.

5 ~ ~~~~ Muscle Glucose Dunsport in (1990) Biochem. Biophys. Res. Commun. 172, Hirshman, M. E, Goodyear, L. J., Wardzala, L. J., Horton, E. D., and Horton, E. S. (1990) J. Biol. Chem. 265, Douen, A. G., Ramlal, T., Rastogi, S., Bilan, P. J., Cartee, G. D., Vranic, M., Holloszy, J. O., and Klip, A. (1990) J. Biol. Chem. 265, Goodyear, L. J., Hirshman, M. F., and Horton, E. S. (1991)Am. J. Physiol. 261, E795-E Rodnick, K. J., Slot, J. W., Studelska, D. R., Hanpeter, D. E., Robinson, L. J., Geuze, H. J., and James, D. E. (1992) J. Biol. Chem. 267, Marshall, B. A,, Ren, J.-M., Johnson, D. W., Gibbs, E. M., Lillquist, J. S., Soeller, W. C., Holloszy, J. 0.. and Mueckler, M. (1993) J. Biol. Chem. 268, Ren, J.-M., Marshall, B. A,, Gulve, E. A,, Gao, G., Johnson, D. W., Holloszy, J. O., and Mueckler, M. (1993) J. Biol. Chem. 268, Gao, J., Gulve, E.A., and Holloszy, J. 0. (1994)Am. J. Physiol. 266, E186-El Cartee, G. D., Douen, A. G., Ramlal, T., Klip, A., and Holloszy, J. 0. (1991) J. Appl. Physiol. 70, Henriksen, E. J., Bourey, R. E., Rodnick, K. J., Koranyi, L., Permutt, M. A,, and Holloszy, J. 0. (1990) Am. J. Physiol. 259, E593-E Gulve, E. A,, Henriksen, E. J., Rodnick, K. J., Youn, J. H., and Holloszy, J. 0. (1993) Am. J. Physiol. 264, E319-E Hansen, P. A,, Gulve, E. G., and Holloszy, J. 0. (1994) J. Appl. Physiol. 76, Burant, C. F., Lemmon, S. K., Treutelaar, M. K., and Buse, M. G. (1984)Am. J. Physiol. 247, E657-E Maroni, B. J., Karapanos, G., and Mitch, W. E. (1986) Am. J. Physiol. 251, F74-F Gulve, E. A,, Cartee, G.D., Youn, J. H., and Holloszy, J. 0. (1991) Am. J. Physiol. 260, C88-C Passoneau, J. V., and Lauderdale, V. R. (1974) Anal. Biochem. 60, Mice Overexpressing GLUT1 25. Ariano, M. A., Armstrong, R. B., and Edgerton, V. R. (1973) J. Histochem. Cytochem. 21, Nesher, R., Karl, I. E., Kaiser, K. K., and Kipnis, D. M. (1980) Am. J. Physiol. 239, E454-E Carlsen, R. C., Larson, D. B., and Walsh, D. A. (1985) Can. J. Physiol. Pharmacol. 63, Riggs, T. R., and McKirahan, K J. (1973) J. Biol. Chem. 248, Yarden, Y., and Ullrich, A. (1988) Annu. Rev. Biochem. 67, Poggi, C., LeMarchand-Brustel, Y., Zapf, J., Froesch, E. R., and Freychet, P. (1979) Endocrinology 105, Sowell, M. O., Robinson, K. A,, and Buse, M. G. (1988)Am. J. Physiol. 255, E159-E Nesher, R., Karl, I. E., and Kipnis, D. M. (1985) Am. J. Physiol. 249, (3226- c232 "" 33. Zorzano, A,, Balon, T. W., Goodman, M. N., and Ruderman, N. B. (1986) Am. J. Phvsiol E21-E Dola;, P. L., Tapscott, E. B., Dorton, P. J., and Dohm, G. L. (1993) Biochem. J. 289, Holloszy, J. O., Constable, S. H., and Young, D. A. (1986) Diabetes Metab. Reu. 1,40% Mandarino, L., Baker, B., Rizza, R., Genest, J., and Gerich, J. (1984) Diabetologia 27, Garvey, W. T., Olefsky, J. M., Matthaei, S., and Marshall, S. (1987) J. Biol. Chem. 262, Hager, S. R., Jochen, A. L., and Kalkhoff, R. K. (1991) Am. J. Physiol. 260, E353-E Garvey, W. T., Olefsky, J. M., and Marshall, S. (1986) Diabetes 35, Hager, S. R., Pastorek, D., Jochen, A. L., and Meier, D. (1991) Biochem. Biophys. Res. Commun. 181, Watanabe, T., Smith, M. M.. Robinson. E W.. and Kono. T. (1984)J. Biol. Chem. 259,