Insulin and Glucose-dependent Regulation of the Glucose Transport [PDF]

VOl. 264. NO. 11. Issue of Ami1 15. pp. 65874595,1989 Printed in U.S.A.

THEJOURNALOF BIOLOGICAL CHEMISTRY 0 1989 by The American Society for Biochemistry and Molecular Biology. Inc.

Insulin and Glucose-dependent Regulation of the Glucose Transport System in the RatL6 Skeletal Muscle Cell Line* (Received for publication, September 7, 1988)

Patricia S. Walker$, Toolsie Ramlals, Jerald A. Donovan$, Thomas P. Doeringv, Alexander Sandran, Amira KlipQ, and JeffreyE. Pessin$II From the $Department of Physiology and Biophysics and the 7Department ofdnutomy,the University of Iowa, Iowa City, Iowa 52242and the $Department of Cell Biology, The Hospital for Sick Children, Toronto, OntarioM5G 1x8,Canada

Differentiated rat L6 skeletal muscle cell cultures for the maintenance of cellular glucose homeostasis in remaintained inglucose-deficient medium containing 25 sponse to a variety of physiological stimuli. Insulin (for review mM xylose displayed a rapid, reversible, time- and see Ref. 3) and phorbol esters (4, 5) have been demonstrated concentration-dependent 3-5-fold increase in glucose to acutely regulate glucose transport activity in responsive transport activity. Glucose deprivation in the contin- cells via recruitment of preformed glucose transporters from uous presence of insulin (24 h) resulted in an overall an intracellular compartment to the cell surface membrane. 9-10-fold stimulation of glucose transport activity. In Alteration of the intrinsic activity of cell surface glucose contrast, acute (30 min) and chronic (24 h) insulin transporters by 6-adrenergic agonists and antagonists may treatment of L6 cells maintained in high glucose (25 also contribute to theoverall insulin activation process (6-8). mM)-containing medium resulted in a 1.5- and 4-fold Similar to theacute effects of insulin, prolonged administrainduction of glucose transport activity, respectively. tion of insulin in vivo or in vitro has been found to increase Acute glucose deprivation and/or insulin treatment had no significant effect on the total amount of glucose glucose transport activity in fibroblasts (9), adipocytes (1012), and skeletal muscle cells (13-15). In addition, long-term transporter protein, whereas the long-term insulinand glucose-dependent regulation of glucose transport regulation of glucose transport activity has been observed in cultured cells maintained in the absence of glucose, termed activity directly correlated with an increase in the cellular expressionof the glucose transporter protein. starvation or glucose deprivation (16-23), as well as by viral I n situ hybridization of the L6 cells demonstrated a transformation and theoverexpression of oncogenes (23-27). 3-, 4-,and 6-fold increasein glucose transporter Under these conditions, the chronic increase in glucose transport activity has been shown to result from an increase in the mRNA induced by glucose deprivation, insulin, and glucose deprivation plus insulin treatments, respec- number of cell surface glucose transporters based upon cytotively. Similarly, Northern blot analysis of total RNA chalasin B binding and Western blot analysis (10, 18-20,24). isolated from glucose-deprived, insulin, and glucoseSeveral lines of evidence have suggested that various tissues deprived plus insulin-treated cells resulted in a 4-, 3-, may contain more than one form of the glucose transporter and 9-fold induction of glucose transporter mRNA, differing in both abundance and mechanism of physiological respectively. The continuous presence of insulin in the regulation. Immunological data have indicated that insulinmedium, either in the presence or absence of glucose, responsive tissues contain a distinct form of glucose transresulted in a transient alteration of the glucose trans- porter compared to other noninsulin-responsive tissues and porter mRNA. Therelative amount of the glucose several cell types in culture (28, 29). The rat adipose glucose transporter mRNA was maximally increased at 6-12 transporterhas been reported to display different kinetic h which subsequently returned to the basal steadystate level within 48 h. These data demonstrate a role properties compared to other cell types (30,31) and contains for insulin and glucose in the overall regulation of two isoforms based upon [3H]cytochalasin B photoaffinity glucose transporter gene expression which may ac- labeling (32). Further, the rat liver glucose transporter has count for the alterationof glucose transporter activity apparently different cytochalasin B binding properties comof muscle tissue observed in pathophysiological states pared to the human erythrocyte (33, 34). Recently, a cDNA clone encoding a glucose transporter-like protein has been such as type I1 diabetes mellitus. isolated from adult human liver (35) which has 55% sequence identity with the human HepG2 and rat brain glucose transporter cDNA (36,37). The facilitative glucose transport systems are ubiquitous in The molecular basis for the altered expression of cell surface animal cells and are responsible for the movement of glucose glucose transporters by glucose deprivation and chronic inacross the cell surface membrane (1, 2). Regulation of these sulin exposure has notbeen elucidated to date. Several reports carrier proteins can occur at several levels and is necessary have attributed the increased expression of glucose trans* This work was supported by Research Grants HL14388 and DK porter protein by glucose deprivation to specific inhibition of 25295 from the National Institutes of Health andby a research grant glucose transporter degradation (20-22). In contrast, other from the Medical Research Council of Canada. The costs of publica- studies have indicated that an increased glucose transporter tion of this article were defrayed in part by the payment of page biosynthesis was mainly responsible for the increase in glucharges. This article must therefore be hereby marked “advertise- cose transport activity induced by glucose starvation (38-40). ment” in accordance with 18 U.S.C. Section 1734 solely to indicate Recently, we have observed that glucose deprivation of isothis fact. 11 Recipient of Research and Career Development Award DK01822 lated primary rat brain glial cells resulted in an increase of from the National Institutes of Health and towhom correspondence glucose transport activity as well as in theamount of cellular should be addressed. glucose transporter protein and mRNA (41). However, skel6587

6588

Molecular Regulation of Glucose Transporter Expression

figure legends. The harvested cells were lysed in GIT buffer (4 M guanidine isothiocyanate, 2.5 mM NaAc, pH 6.9, 120 mM 2-mercaptoethanol) followed by centrifugation through a 4-ml CsCl cushion (5.7 M csc1, 2.5 mM NaAc, pH 6.0) a t 29,000 rpm in a SW 41 rotor for 21 h at 20 "C. RNA pellets were resuspended in 0.3 M NaAc, pH 6.0, followed by precipitation in 2.5 volumes of 100% ethanol. Northern Blot Analysis-Northern blot analysis was carried out as previously described (41). Briefly, 30 pg of total RNA was denatured by glyoxal, size-fractionated by agarose gel electrophoresis, and transferred to aminobenzyloxymethyl paper. The RNA was visualized and photographed by ultraviolet transillumination to ensure that the total RNA was intact and toallow calculation of molecular weight based on migration of the 28 and 18 S ribosomal subunits. The blot was hybridized to thenick-translated [cr-32P]dATP-labeledfull-length 2.8EXPERIMENTALPROCEDURES kb rat brain glucose transporter cDNA or the 0.7-kb glyceraldehydeMaterials-Tissue culture medium, serum, and reagents were ob- 3-phosphate dehydrogenase control maker cDNA. Hybridization was tained from GIBCO. Cytochalasin B, poly-D-lysine, insulin, and carried out in 50% deionized formamide, 5 X SSPE (150 mM NaCl, RNase A were obtained from Sigma. 2-Deoxy-[3H]glucose, 3-0- 10 mM H2PO4, 1 mM EDTA, pH 7.4), 1 X Denhardt's, 5% dextran methyl-[3H]glucose, [cY-~'P]~ATP, and [cY-~'S]~ATP were purchased sulfate, 0.1 mg/ml denatured sheared salmon sperm DNA, and 1 X from Du Pont-New England Nuclear. lZ6I-Protein A was obtained lo6 cpm/ml 32P-labeledcDNA probe a t 42 "C for 48 h. After being from Amersham Corp. washed twice in 5 X SSPE, 0.1% SDS and three times for 30 min at Cell Culture-The spontaneously fusing L6 skeletal muscle cell 50 "C with 0.1 X SSPE, 0.1% SDS, the blots were exposed to Kodak line was grown and maintained as previously described (4, 43). The XAR-5 film for 24 h. The autoradiographs were quantified by laser cells weregrown in 75-cm2 flasks ina-minimal essential media scanning densitometry to determine the relative amounts of glucose supplemented with 2% fetal bovine serum and 1%(v/v) antimycotic/ transporter mRNA. antibiotic solution (10 pg/ml penicillin, 10 pg/ml streptomycin, 25 Western Blot Analysis-Differentiated L6 cells grown in 150-cm2 pg/ml amphotericin B) at 37 'C. Cultures were maintained in contin- tissue culture dishes were maintained under the appropriate experiuous passages ( ~ 2 0 by ) trypsinization (0.25% trypsin) of semicon- mental conditions as indicated in the individual figure legends. Total fluent nonfused cells. Cellular differentiation was monitored by phase cell membranes were prepared as previously described (41) with slight contrast microscopy and maximal cell fusion (>85%) typically re- modifications. Briefly, the cells were hypotonically lysed in 1 mM quired from 3 to 5 days after reaching confluency. All experimental Tris-HC1, pH 7.4 at 4 "C, followed by mechanical disruption using a procedures were performed on maximally differentiated cells approx- Potter-Elvehjem homogenizer. The cell homogenate was centrifuged imately 8-10 days postconfluence. In all experiments the cells were at 5,000 X g to remove nuclei, mitochondria, and unlysed cells followed incubated in serum-free glucose-free Dulbecco's minimal essential by centrifugation at 100,000 X g for 30 min to obtain the total cell media supplemented with 25 mM glucose (fed) or 25 mM xylose membranes. Membrane protein (50 pg) was subjected to SDS-poly(starved) in the presence or absence of 100 nM insulin as described acrylamide gel electrophoresis on 9% gels (acrylamide:bisacrylamide, in the individual figures. The replacement of mediumglucose for 300.8) and electrophoretically transferred to nitrocellulose memxylose wasperformed to provide for a monosaccharide carbon source branes. The nitrocellulose membranes were incubated overnight a t and asa control for nonspecific starvation effects. Under these culture 4°C in PBS-Tween (137mM NaC1,16 mM Na2HP04,1.5 mM KHzPO4, conditions (24 h), the concentration of medium glucose and insulin 2.6 mM KCl, 1.5 mMMgC12 plus 0.05% Tween 20) and incubated was found to decline to approximately 22 mM and 4 nM, respectively. with a 1:1,000 dilution of the anti-human erythrocyte glucose transWe therefore changed the appropriate tissue culture medium for each porter antibody CYGT-8for 2 h at 23"C. The samples were then cell condition every 24 h during the course of each experimental washed 2 times for 10 min in PBS-Tween, incubated for 1 h with 5 procedure. X IO5 cpm of 1251-labeled protein A followed by extensive rewashing Glucose Transport Assuys-2-Deoxy-~-glucose (2-DG)' end 3-0in PBS-Tween. The membranes were air-dried and subjected to methyl-D-glucose (3-OMG) transport studies were performed on dif- autoradiography. ferentiated L6 cells maintained in 24-well tissue culture plates as I n Situ Hybridization-Differentiated L6 cells were grownon polypreviously described (4, 43). Cells were incubated for various times D-lysine-coated coverslips and maintained for 24 h in Dulbecco's and medium conditions as indicated in the individual figure legends. minimal essential medium containing either25 mM glucose or 25 m M Cell monolayers were washed twice with glucose-free HEPES-buff- xylose in the presence and absence of 100 nM insulin. the coverslips ered saline solution (140 mM NaCl, 20 mM HEPES, 1 mM CaC12, 5 were then washed twice with PBS, pH7.4, fixed for 20 min in freshly mM KC1, and 2.5 mM MgS04, pH7.4). Hexose uptake was determined prepared 4% paraformaldehyde in PBS, washed with PBS, and deby 10-min and 30-s incubations, respectively, with 2-deo~y[~H]glu- hydratedin 70% ethanol a t 4 "C overnight. The coverslips were cose (10 pM,0.2 pCi/nmol) or 3-O-methyl-[3H]glucose(10 pM,0.2 washed 3 times for 5 min each in PBS, 1 time in double distilled pCi/nmol) in HEPES-buffered saline at 23 "C. Hexose uptake was water, and airdried for 10 min. The dried samples were then incubated terminated by three rapid washes with ice-cold PBS (150 mM NaCl, with 0.25 mg/ml Pronase for 10 min a t 23 'C, washed with 2 mg/ml 5 mM NaH2P04, pH 7.4). In the case of 3-OMG uptake, the PBS glycine in PBS, washed twice with PBS, and fixed for 20 min with washes included 1 mM mercuric chloride to inhibit the efflux of 3- 4% formaldehyde in PBS. Following the second fixation, the coverOMG. Under these experimental conditions, it has been previously slips were washed with 2 mg/ml glycine in PBS, twice with 2 X SSC 3- (0.3 M NaCl, 30 mM sodium citrate, pH 8.0), and once with 10 m M established that the time course of both 2-deo~y-[~H]glucose and O-methyl-[3H]glucoseuptake are linear and that hexose transport is triethanolamine, 0.1 mM acetic anhydride in PBS. The samples were the rate-limiting step for monosaccharide uptake (43, 44). The cell- then prehybridized in 50% formamide, 2 X SSC, 10 mM DTT at 50 associated radioactivity was determined by solubilization in 0.05 N "C for 30 min. The pGEM I1 plasmid containing the full-length 2.8NaOH and scintillation counting. Noncarrier-mediaed uptake was kb rat brain glucose transporter cDNA was nick-translated using determined in parallel wells containing 10 J ~ Mcytochalasin B. The [(y-35S]dATP.The fixed cells were hybridized with 3 X lo' cpm/pl protein content/well was determined by the method of Bradford (45), nlck-translated plasmid in 50% deionized formamide, 2 X SSC, 10 and specific uptake was expressed as pmol/min/mg protein from mM DTT, 1mg/ml tRNA, 1 mg/ml sheared sonicated salmon sperm experiments performed in triplicate. DNA at 50 "C for 18 h. Nonspecific background was determined in Isolation of Cellular RNA-Total cellular RNA was isolated as parallel hybridization using the nick-translated [~~-~~S]dATP-labeled previously described (41).Differentiated L6 cells were pooled from pGEM I1 vector DNA. After hybridization was completed the coverfour 100-mm2tissue culture plates previously incubated under the slips were washed as follows: 1)one time in 5 X SSC, 50% formamide, appropriate experimental conditions as indicated in the individual and 10 mM DTT fot3 h at 50 "C; 2) twice in 2X SSC, 50% formamide, and 10 mM DTT for 20 min a t 50 "C; 3) four times in 2 X ssc a t ' The abbreviations used are: 2-DG, 2-deoxy-~-glucose;3-OMG, 3- 23 "C for 1 min; 4) one time in 2 X SSC and 1 mg/ml RNase A for 0-methyl-D-glucose; aGT-8, anti-human erythrocyte glucose trans- 30 min a t 37 'C; 5) one time in 2 X SSC, 50% formamide, and 10 mM porter polyclonal antibody; HEPES, 4-(2-hydroxyethyl)-l-pipera- DTT for 10 min at 50 "C; 6) three times in 0.1 X SSC and 10 mM zineethanesulfonic acid PBS, phosphate-buffered saline; DTT, di- DTT for 20 min at 50 "C; 7) three times in 2 X SSC for 2 min; 8) dehydration in 70% ethanol followed by dehydration in 95% ethanol. thiothreitol; SDS, sodium dodecyl sulfate; kb, kilobase(s).

eta1 muscle is the major tissue responsible for insulin-stimulated glucose uptake in vivo (13, 42). In spite of numerous reports on the acute effects of insulin on skeletal muscle, the effects of chronic insulin and glucose deprivation have not been examined in detail. In this study, we have characterized the acute and chronic effect of insulin on glucose transport activity, protein, and mRNA in both the control and glucosedeprived rat L6 skeletal muscle cell line. These data demonstrate a potential role for insulin and glucose in the overall regulation of muscle cell glucose transporter gene expression.

Molecular Regulation

of Glucose Transporter Expression

Slides were then coated with autoradiographic emulsion and stored at -20 ‘C for 2-4 weeks after which the coverslips were developed and stained with hematoxylin plus eosin. The slides were then mounted and thegrains counted under light microscopy. RESULTS

We have previously observed that glucose starvation of primary rat brain glial cultures results in a time- and concentration-dependent increase in glucose transport activity, protein, andmRNA (41). In order to fully examine this regulatory property of the glucose transporter in an insulin-responsive cell type, the differentiated rat L6 skeletal muscle cells were maintained for 24 h invarious amounts of glucose plus xylose such that thetotal initialmedium monosaccharide concentration was 25 mM. In a concentration-dependent manner, glucose deprivation maximally stimulated 2-DG transport activity approximately 5-fold compared to cells maintained for 24 h in the presence of 25 mM glucose (Fig. lA). Half-maximal stimulation was found to occur a t 3.5 mM glucose whereas maximal stimulation required the complete absence of exogenously added glucose. This starvation-induced increase in glucose transport activity was also found to be readily reversible (Fig. 1B).Readdition of 25 mM glucose to the glucosedeprived L6 cells resulted inthereturnto basal glucose transport activity within 2-4 h. We next determined the effect of D-glucose deprivation in conjunction with insulin treatment on glucose transport activity. The differentiated L6 cells were maintained for 24 h in medium containing either 25 mM glucose or 25 mM xylose in

40

i

B,

6589

the absence or continuous presence of 100 nM insulin (Fig. 2). Incubation of the L6 cells with 25 mM glucose plus 100 nM insulin for 24 h resulted in a 3-fold increase in 2-DG (Fig. 2 A ) and an approximate 2-fold increase in 3-OMG (Fig. 223) transport activities compared to control cells maintained in the absence of insulin. Glucose deprivation for 24 h produced a 6-fold increase in 2-DG (Fig. 2 A ) and a &fold increase in 3-OMG transport activitycompared to thecontrol cells maintained in 25 mM glucose. However, incubation of L6 cells in the absence of glucose but in the continuous presence of insulin for 24 h resulted in a 13-fold increase in 2-DG (Fig. 2 A ) and a 14-fold increase in 3-OMG (Fig. 2B) transport activities. These data demonstrate that under these experimental conditions 2-DG transport is an accurate measure of glucose transport activity and thatglucose deprivation in the continuous presence of insulin markedly enhances glucose transport activity. The time dependence of glucose transporter activity regulation by glucose deprivation and chronic insulin treatments were determined in Fig. 3. The differentiated L6 cells were maintained in medium containing 25 mM glucose or 25 mM xylose either in the presence or absence of 100 nM insulin for the indicated times. Cells maintained in the presence of both glucose and insulin resulted in a complex time-dependent alteration of glucose transport activity relative to control cultures maintained in the continuous presence of glucose alone (Fig. 3A). The initial phase of insulin stimulation was relatively rapid with the maximal increase in glucose transport activity observed at 2 h, followed bya decline toward the basal glucose transport activity level within 4 h. Subsequently, a second phase of insulin-stimulated glucose transport activity occurred at 4 h which gradually increased over the next 24-h

I

QlC

Xyl

refed

(2h) (4h) FIG. 1. Glucose concentration dependence and reversibility glucose glucose of the starvation-stimulated increase in glucose transport fed starved activity. A , differentiated L6 cells were incubated with the indicated concentrations of glucose plus the appropriateconcentrations of FIG. 2. Effect of insulin and glucose deprivation on 2-DG xylose to equal 25 mM monosaccharide for 24 h. The initial rate of and 3-OMG uptake in differential L6 cell cultures. Differencytochalasin B-sensitive 2-DG uptake was determined as described tiated L6 cells were incubated in medium containing 25 mM glucose under “ExperimentalProcedures.” B, the differentiated L6 cells were or 25 mM xylose in the presence (hatched bars) or absence (open incubated for 24 h with 25 mM glucose (glc) or 25 mM xylose (xyl). bars) of 100 nM insulin for 24 h. Cytochalasin B-sensitive uptake of The xylose-incubated cells were then medium changed into 25 mM 2-DG (A) or 3-OMG ( B )was determined as described under “Experglucose (refed) for 2 and 4 h. Cytochalasin B-sensitive 2-DG uptake imental Procedures.” Each point represents the mean f S.E. for was then determined as described under “Experimental Procedures.” triplicate determinations. This is a representative experiment which Each point represents the mean & S.E. of six independent determi- was independently performed four times for 2-DG uptake and two nations each preformed in triplicate. times for 3-OMG uptake. (24h)

Molecular Regulation of Glucose Transporter Expression

6590

TIME

3h 48h

6 h24h

12h

23 KGLUCOSE INSULIN

++--+”+

++--+-+

++-- ++” - + - + -+”+ ”+-+ ++“

FIG.4. Time course of glucose transporter protein induction by insulin treatment and glucose deprivation. Differentiated L6 cells were maintained in the presence of 25 mM glucose or 25 mM xylose with or without 100 nM insulin for the times indicated. Total cellular membranes were prepared and subjected to Western blot analysis using the aGT-8anti-human erythrocyteglucose transporter antibody as described under “ExperimentalProcedures.” Each time point is an independent experiment performed one time at 3 h, three times a t 6,12, and 24 h, and twice a t 48 h. -

0 024

6 18

12

the effect on glucose transport activity (Fig. 3B), glucose deprivation in the presence of insulin resulted in a dramatic FIG.3. Time course of glucose transport activity stimulated by insulin and glucose deprivation. A, differentiated L6 cells were increase in the amount of immunoreactiveglucose transporter maintained for the indicated times in medium containing 25 mM protein (Fig. 4). Althoughthe increase in glucose transporter glucose (0,glc) or 25 mM glucose plus 100 nM insulin (0,glc, ins). protein by insulin treatment was initially detected by 6-12 h, Cytochalasin B-sensitive 2-DG uptake was determined as described it alsooccurred significantly more slowly than the initial under “Experimental Procedures.” B , the L6 cells were maintained rapid stimulation (2 h) of glucose transport activity (Fig. 3B). for the indicated times in medium containing 25 mM xylose (0,xyl) However, the time course of insulin-stimulated glucose transor 25 mM xylose plus 100 nM insulin (0, xyl, ins). Each point represents the mean f S.E.of six independent determinations each porter protein increase was temporally related to the gradual second phase increase of glucose transport activity (Fig. 3B). performed in triplicate. It should be noted that each time point of the Western blot analysis presented in Fig. 4 was an individual experiment period. The net effect was on overall 4-fold stimulation of such that comparisons can be made only with respectto the glucose transport activity by 24 h of continuous insulin excontrol (glucose-fed)cell membranes within a given time posure (Fig. 3A). Glucose deprivation of the L6cells, maintained in the point. Similar results were also obtained when the total L6 presence of 25 mM xylose, resulted in a rapid %fold increase cellular membranes were prepared by extraction with an SDS in glucose transport activity (Fig. 3 B ) . Half-maximal stimu- lysis buffer (data not shown). The effect of chronic insulin treatment and glucose deprilation of glucose transport activity occurred by 15 min and vation on the regulation of glucose transporter mRNA in the maximal stimulation by 2 h. In contrast to insulin treatment, differentiated L6 cells was next determined by Northern blot glucose deprivation resulted in only a relatively small increanalysis. Total RNA isolated from the differentiated L6 cells mental increase in the amount of glucose transport activity maintained in the absence of glucose (25 mM xylose) had a 3expressedover the next 24 h (Fig. 3 B ) . However,glucose fold increase in the steady-state level of the 2.8-kb glucose deprivation in the continuous presence of insulin resulted in transporter mRNA (Fig. 6,A and C) when comparedto RNA a rapid (2-h) 5-fold increase in glucose transport activity isolated from control cultures maintained in the presence of which also gradually increased to greater than 10-fold over 25 mM glucose (time 0). Maximal induction was observed by the next 24 h (Fig. 3B). These data suggest that theregulation 24 h with a half-maximal increase in the glucose transporter of glucose transport activity by chronic insulin exposure and mRNA at 6-12 h. To ensure that theeffect of glucose depriglucose deprivation arise by different cellular processes and vation was specific for the glucose transporter mRNA, the that these alterations in glucose transport activity occur in a blots were also probed with the cDNA for glyceraldehyde-3synergistic fashion. phosphate dehydrogenase. The expression of the 1.5-kb glycTo compare the relative number of membrane-associated eraldehyde-3-phosphatedehydrogenase mRNA did not signifglucose transporters with glucose transport activity, Western icantly change in either the control (Fig. 5, B and C) or blot analysis was performed onthe isolated L6 cellmembranes glucose-deprived (Fig. B6,and C) L6 cell cultures. In addition, (Fig. 4). Total particulate membranes were prepared from the L6 cells maintained in the presence of 25 mM glucose over differentiated L6 cells maintained in medium containing 25 a 48-h time period had nosignificant alteration in the steadymM glucose or 25 mM xylose either in the presence or absence state levels of the glucose transporter mRNA (Fig. 5, A and of 100 nM insulin for the indicated times. The aGT-8 identi- 0 . fied a closely spaced and diffuse protein doublet with M , = Incubation of the L6 cells in the continuous presence of 43,000 and 46,000. Glucose deprivation for 48 h resulted in a 100 nM insulin, either in the presence (Fig. 7, A and C) or time-dependent increase in the relative amount of immuno- absence (Fig. 8, A and C) of glucose, resulted in a transient reactive glucosetransporter protein compared to control (25 alteration of the glucose transporter mRNA. In this experimM glucose) cells. The glucose deprivation-induced increase ment, incubation of L6 cells inthe continuous presence of 25 in glucose transporter protein (Fig. 4) occurred significantly mM glucose plus 100 nM insulin resulted in an approximate later (24 h) than the starvation-induced increase (2 h) in 3-fold induction in the glucose transporter mRNA expression glucose transport activity (Fig. 3B). Cells maintained in 25 by 6 h (Fig. 7). In threeindependent experiments, the insulinmM glucose plus 100 nM insulin also resulted in a time- dependent increase in the expression of the glucose transdependent increase in glucose transporter protein. Similar to porter mRNA was somewhat variable, ranging from 3- to 5Time (hours)

6591

Molecular Regulation of Glucose Transporter Expression A)

A)

5.1Kb-c

1

2.0Kb.I

2 . O K h

1IME.h

5.1Kb.I

o

TIME,h

I 3 6 12 2448

5.1Kb”

5.1Kb-c

2.OKb”

2.OKb”

4

o I

3 6 12 2448

-

C>

.b m c

al

Y

c - 2 0) >

.-

Y

0

0)

E

C

12

24

36

48

lime (hours) FIG. 5. Time course of the amount of glucose transporter mRNA in control L6 cell cultures. Differentiated L6 cells were maintained in the presence of 25 mM glucose for the indicated times, and the total cellular RNA was isolated as described under “Experimental Procedures.” A , Northern blot analysisof total cellular RNA (30 pg/lane) was probed against the nick-translated [a-”PIdATPcDNA. B, labeled2.8-kbfull-lengthratbrainglucosetransporter rehybridization of the Northern blot used in A with the nick-translated 0.7-kb cDNA for glyceraldehyde-3-phosphate dehydrogenase. C, quantitative laser scanning densitometry of the glucose transporter mRNA Northern blot presented in panel A (0)and of the glyceraldehyde-3-phosphate dehydrogenasemRNA Northern blot presented in panel B (0).This is a representative experiment independently performed four times. fold at 6-12 h (data not shown). The L6 cell cultures maintained inthe absence of glucose (25 mM xylose) supplemented with 100 nM insulin resulted in a 9-fold induction of glucose transporter mRNA by 12 h and ranged from 7- to 9-fold in three separate experiments (Fig. 8). The insulin-dependent increase of the glucose transporter mRNA in both the 25 mM glucose-maintained and -deprived cultures was maximal by 6-12 h but returned to the basal level within 48 h of continuous insulin exposure. Insulin treatment did not significantly alter the expression of the 1.5-kb glyceraldehyde-3-phosphate dehydrogenase mRNA in either the glucose-maintained (Fig. 7, B and C) or the glucose-deprived (Fig.8, B and C) L6 cells, although glyceraldehyde-3-phosphate dehydrogenase expression has beenshown to be stimulated by insulin in 3T3 adipocytes and H35 hepatoma cells (46). To determine the cellular localization of the glucose transporter mRNA, in situ hybridization was performed on the glucose-deprived and insulin-treated differentiated L6 cell cultures (Fig. 9). Differentiated L6 cells were incubated for 24 h in medium containing 25 mM glucose or 25 mM xylose in eitherthe presence or absence of 100 nM insulin. The cells were then hybridized against the full-length nick-translated [~x-~~S]dATP-labeled glucose transporter cDNA. Under all the experimental conditions employed, the glucose transporter mRNA was found to be distributed over the cytoplasm, nuclear, and perinuclear regions of the multinucleated L6 myotubes (Fig. 9). Consistent with the Northern blot data (Figs. 5-8), autoradiographic analysis demonstrated that glucose deprivation for 24 h resulted in a%fold increase in the specific

12

24

36

48

lime (hours) FIG. 6. Time course of the amount of glucose transporter mRNA by glucose deprivation. Differentiated L6 cellswere maintained in the absence of glucose (25 mM xylose) for the indicated times, and the total cellular RNA was isolated as described under A, Northern blotanalysis of total cellular “Experimental Procedures.” RNA (30 pgllane) was probed against the nick-translated [cx-~*P] dATP-labeled 2.8-kbfull-length rat brainglucose transportercDNA. B, rehybridization of the Northern blot used in A withthe nicktranslated 0.7-kb cDNA for glyceraldehyde-3-phosphate dehydrogenase. C, quantitative laser scanningdensitometry of the glucose transporter mRNA Northern blot presented in A (0)and of the glyceraldehyde-3-phosphate dehydrogenasemRNA Northern blot presented in B (0).This is a representative experiment independently performed four times. hybridization to the L6 cells (Figs. 9C and 10) compared to the control glucose-maintained cells (Figs. 9A and 10). Similarly, cells maintained in 100 nM insulin in either the presence (Fig. 9B) or absence (Fig. 9D) of glucose resulted in a 4- and 6-fold increase in the glucose transporter-specific mRNA grains compared to thecontrol cells (Fig. lo), respectively. In all cases, nonspecific hybridization to control plasmid DNA was found to be constant. DISCUSSION

Skeletal muscle is the major tissue responsible for the maintenance of glucose homeostasis in vivo(13,42). The rat L6 myoblasts are a rapidly growing cell line that displays many of the phenotypic characteristics of skeletal muscle and can form multinucleated muscle fibers with a concomitant cessation of DNA synthesis (47). In thisstudy, we have taken advantage of the L6 muscle cell lines as an in vitro model system in order to examine the insulin- and glucose-dependent regulation of this muscle cell glucosetransporter. Glucose deprivation and acute insulin treatment of the differentiated L6 cells resulted in a 4-6-fold increase of glucose transport activity as determined by the initial rate of 2DG uptake (Figs. 1-3). The starvation-induced increase in glucose uptake was completely reversiblewithin 2-4 h following the readdition of glucose (Fig. 1).These observations are consistent with the previously reported increases in glucose transport activity by glucose starvation in muscle cells (16, 19) as well as in several other cell culture systems (16-22,39, 40). The starvation-induced increase in glucose transport

6592

Molecular Regulation of Glucose Transporter Expression 1

*)

5.1Kb”

..

-

..

,

,;.

.

2 . O K e

TIME,h

TIME,h

0 1 3 6 122448

5.1Kb” 2.OKb-L

0 1 3 6 12 2448

4

.b v)

K 0)

Y

c - 2 0) >

.CI

0 0

t Y

C

24

12

36

48

Time (hours)

0‘ 0 24

12

36

48

Time (hours)

FIG. 8. Time course of the amount of glucose transporter FIG. 7. Time courseof the amount of cellular glucose transporter mRNA by continuous insulin treatment. Differentiated mRNA in the differentiated L6 cells by the combination of L6 cells were maintained in medium containing 25 mM glucose plus glucose deprivation and continuous insulin treatment. The L6 100 nM insulin for the indicated times and total cellular RNA was cells were maintained in glucose-deficient medium (25 mM xylose) isolated as described under “Experimental Procedures.” A, Northern supplementedwith 100 nM insulin for the indicated times.Total blot analysis of total cellular RNA (30 pg/lane) was probed against cellular RNA was isolated as described under “Experimental Procethe nick-translated [a-32P]dATP-labeled2.8-kb full-length rat brain dures.” A, Northern blot analysis of total cellular RNA (30 pgllane) was probed against the nick-translated [c~-’~P]dATP-labeled2.8-kb glucose transporter cDNA. B, rehybridization of the Northern blot used in A with the nick-translated 0.7-kb cDNA for glyceraldehyde- full-length rat brainglucose transporter cDNA. B, rehybridization of 3-phosphate dehydrogenase. C, quantitative laser scanning densitom- the Northern blot used in A with the nick-translated 0.7-kb cDNA etry of the glucose transporter mRNA Northern blot presented inA for glyceraldehyde-3-phosphate dehydrogenase. C, quantitative laser scanning densitometry of the glucose transporter mRNA Northern (0)and of the glyceraldehyde-3-phosphatedehydrogenase mRNA Northern blot presented in B (0).This isa representative experiment blot presented in A (0)and of the glyceraldehyde-3-phosphate deB (0).This is a hydrogenase mRNA Northern blot presented in independently performed four times. representative experiment independentlyperformed four times.

activity was concentration-dependentwith half-maximal stimulation occurring between 2 and 5 mM glucose (Fig. 1). This glucose concentration-dependent inhibition of glucose transport activity by extracellular glucose is also consistent with data previously reported in other cell types (13, 16). To confirm that the initial rate of 2-DG uptake was an accurate reflection of glucose transport activity, 3-OMG uptake was determinedinparallel (Fig. 2). Chronicinsulin treatment and/or glucose deprivation resulted in a similar increase of 3-OMG uptake as observed for 2-DG transport measurements. However, it should be notedthat insulin stimulation of 3-OMG uptake in cells maintained in high glucosecontaining medium was somewhat less than thatobserved by 2-DG uptake. This apparent difference in the insulin stimulation of glucose transport activity has been previously attributed to the presence of both high and low affinity glucose transporters with relative preferences for either 2-DG or 3OMG, respectively (19,48). The time-dependent effect of continuous insulin exposure on the L6 cells maintained either in 25 mM glucose or 25 mM xylose revealed a complex regulatory pattern of glucose transport activity (Fig. 3). The initial phase of the insulin stimulation of glucose transport activity was relatively rapid, reaching a maximal 3-fold stimulation by 2 h. During the next2 h, the glucose transport activity was observed to return nearly to the basal state. However, over the remaining 24-h period, the amount of glucose transport activity steadily increased. This complex response of glucose transport activity during the continuous exposure of the L6 cells to insulin cannot be

accounted for by the desensitization of the L6 insulin receptors per se, since a t long time points insulin continued to promote a further elevation in glucose transport activity. In contrast, thetime dependence of glucose deprivation primarily resulted in the initial rapid phase of glucose transport activation (maximal stimulation by 2 h) without a significant incremental increase during the following 24-h time course examined (Fig. 3B). In addition, the stimulation of glucose transport by chronic insulin treatment and glucose deprivation was found to be significantly greater than additive (Figs. 2 and 3). Thus, these data indicate that theacute and chronic insulin-dependent regulation of the glucose transporter occurs via a t least two distinct cellular mechanisms. It is possible that 100 nM insulin could function via activation of the insulin-like growth factor 1 receptor which is relatively abundant in this cell line (49). However, previous studies have demonstrated that thisconcentration of insulin is specific for insulin receptor-mediated biological responsiveness in the rat L6 skeletal muscle cells (49, 50). Further, consistent with these reports we have observed that 100 nM insulin was only able to displace approximately 15% of specifically bound tracer (0.5 nM) 1251-insulin-likegrowth factor 1.2 The stimulation of glucose transport activity by acute insulin treatment has been previously reported in L6 cells (4, 43, 50) as well as in other insulin-sensitive cell systems such

* P. S. Walker, T. Ramlal, J. A. Donovan, T. P. Doering, A. Sandra, A. Klip, and J. E. Pessin, unpublished results.

Molecular Regulation of Gluc:ose Transporter Expression

glc (244

glc

6593

+

ins (24h)

xyl (24h)

+

x y ~ ins

(24h)

FIG. 10. Morphometric analysis of the in situ hybridization from the glucose-starved and insulin-treated differentiated L6 cells. The differentiated L6 cells were maintained in medium containing 25 mM glucose (glc), 25 mM glucose plus 100 nM insulin (glc + ins), 25 mM xylose (xyl),or 25 mM xylose plus 100 nM insulin (xyl + ins). The open bars represent the percent increase in the total number of grains counted over the control arbitrarily set to 100%. The hatched bars represent the percent of total grains due to nonspecific hybridization as described under “Experimental Procedures.” Each bar represents the mean f S.E. of the mean from three independently performed experiments. The total number of cells counted under each condition was 300.

FIG. 9. Localization of glucose transporter mRNA by in situ hybridization. Differentiated L6 cells were grown, fixed, and hybridized against the nick-translated [a-35S]dATP-labeled2.8-kb rat brain glucose transporter cDNA as described under “Experimental Procedures.” A , control cultures were maintained for 24 h in 25 mM glucose; B, cultures maintainedfor 24 h in medium containing 25 mM glucose plus 100 nM insulin; C, cultures maintained for 24 h in 25 mM xylose-containing medium; D, cultures maintained for 24 h in 25 mM xylose plus 100 nM insulin-containing medium. In each case a prominent multinucleated myotube is presented. The magnification was X 2000. N , nuclei; arrows are directed toward typical silver grains. The autoradiographic exposure was for 15 days.

as adipocytes and muscle tissue (for reviews see Refs. 3, 13, and 23). It has been well established that themajor action of acute insulin treatment in these insulin-responsivetarget cells is to induce a rapid translocation (recruitment) of preformed glucose transporters to the cell surface membrane (3). In the differentiated L6 cells, acute insulin treatment was observed to stimulate glucose transport activity in the glucose-fed control cells (Figs. 2 and 3). However, similar to thefindings in other cells (51, 52), acute insulin treatment of the L6 cells maintained for 24 h in the absence of glucose (25 mM xylose) failed to stimulate glucose uptake above that stimulated by glucose deprivation alone.3 These data in the differentiated rat L6 skeletal musclecell line suggest that both glucose deprivation and acute insulin treatment increase glucose transport activity by inducing a rapid translocation of preformed intracellular glucose transporters to the cell surface. However, they do not directly distinguish between recruitment of an intracellular glucose transporter pool and the possibility of intrinsic activation of pre-existing cell surface

3T.Ramlal, P. S. Walker, J. E. Pessin, and A. Klip, unpublished results.

transporter molecules. Clarification of the precise cellular mechanism bywhichglucose deprivation activates glucose transport activity will require further studies to directly determine the subcellular distribution of the L6 glucose transporter protein under glucose deprivation conditions. Consistent with this interpretation, Western blot analysis using aGT-8 revealed that both acute glucose deprivation and insulin treatments had no significant effect onthe netamount of total membrane-associated glucose transporters. In contrast, long term insulin treatment increased the total membrane-associated glucose transporter both in the absence and presence of glucosein the culture medium. This couldbe accounted for by either an increase in glucose transporter biosynthesis and/or a decrease in glucose transporter degradation. Both protein synthesis-dependent and -independent increases in glucose transport activity in muscle cells following chronic (5 h) insulin treatment havebeenpreviously reported (14). It is important to recognize that cellsmay contain multiple immunogenic forms of the glucose transporter (28, 29) and that this particular antibody is directed against the same glucose transporter speciespredicted by HepG2 and ratbrain glucose transporter cDNA used in Figs. 5-8 (36, 37). In addition, it should also be noted that the amount of total cellular glucose transporter protein may not necessarily correlate with glucose transport activity since the latter is a measurement of functional glucose transporter residing in the cell surface membrane. Interestingly, glucose deprivation itself was also observed to stimulate an increase in the total amount of membraneassociated glucose transporter, although to a lesser extent and over a longer time period than cells maintained in the presence of insulin (Fig. 4).Further, glucose deprivation also increased the expression of the glucose transporter mRNA (Fig. 6) but to a lesser extent and over a longer period of than in the presence of insulin (Figs. 7 and 8). We speculate that this cellular response was to replenish the intracellular glucose transporter poolswhichwould not be detected byglucose transport activity measurements in the intact cells. In this regard, several studies have indicated that the glucose starvation-induced increase in glucose transport activity can occur by either an increase in glucose transporter biosynthesis, which is RNA- and/or protein synthesis-dependent (38-401, or via an inhibition of glucosetransporter protein degradation,

6594

Molecular Regulation of Glucose Transporter Expression

independent of RNA and protein synthesis (20-22). We have recently reported that in primary cultures of rat brain glial cells, glucose deprivation results in a coordinate increase in the expression of glucose transport activity, protein, and mRNA (41). These apparent divergent resultsmost likely reflect cell-specific differential regulation of multiple glucose transporter gene products (28,29,35). Northern blot analysis demonstrated an appropriate timedependent increase in the glucose transporter mRNA (Figs. 5-8), preceding the increase in glucose transporter protein (Fig. 4). Surprisingly, the total cellular RNA isolated from both glucose-starved and glucose-fed L6 cells maintained in the presence of insulin revealed a transient induction of glucose transporter mRNA. Continuous insulin treatment initially (6-12 h) increased the amount of glucose transporter mRNA, which then progressively decreased to thebasal level by 48 h. This transient insulin response of the glucose transporter mRNA is similar to that observed for other growth factor-regulated genes such as myc andfos (53,54). The return of the glucose transporter mRNA to thebasal level within 48 h of continuous insulinexposure could potentially resultfrom an insulin-induced down-regulation of the insulin receptor. However, this appears unlikely since the insulin stimulation of glucose transport activity (Fig. 3) and glucose transporter protein (Fig. 4) did not decline over the same time period. Nevertheless, this type of cellular response at themRNA level suggests that once the glucose transporter is synthesized, it is a relatively stable protein such that continuous biosynthesis is not necessary to maintainelevated levels subsequent to the initial increase. Further, these data would predict that at significantly longer time points that those examined in this study, the amountof glucose transporter protein would eventually return to the basal state due to the decrease in the amount of expressed glucose transporter mRNA. In addition to the Northernblot analysis, the induction of glucose transporter mRNA by both insulin and glucose deprivation was confirmed by in situ hybridization (Figs. 9 and 10). Recent evidence has suggested that the glucose transporter is storedinintracellular vesicles (55,56) in close proximity to the plasma membrane (28, 57). Several studies have demonstrated that cytoplasmic mRNA such as those for actin, tubulin, and vimentin are translated while associated with the cytoskeleton and remain in a polarized position (5860). I n situ hybridization was therefore used to determine if glucose transporter biosynthesis was co-localized to a particular membrane-targeted organelle. However, under these experimental conditions microscopic analysis indicated a predominantly cytoplasmic and perinuclear distribution of the glucose transporter mRNA with no evidence for specific localization in asubsarcolemmal position. In futurestudies, this methodology will be useful to furtherexamine the nutritional and insulin-dependent regulation of glucose transporter mRNA expression in specific cell types in vivo. After submission of this paper, cDNA clones for the human fetal skeletal muscle (61) and ratliver (62) glucose transporters were identified inaddition to the previously reported HepG2 (36), rat brain (37), and human liver (35) glucose transporter cDNA clones. Expression of these different glucose transporter mRNAs apparently occurs with a distinct but overlapping tissue distribution in uivo. The relationship between these other glucose transporter subtypes with the insulin- and glucose-dependent regulation of rat brainglucose transporter expression is presently unknown. In summary, these data demonstrate that the rat L6 skeletal musclecell line containsa glucose- and insulin-sensitive glucose transporter which is regulated in ahighly coordinated

and complex fashion. The cellular response tothe acute actions of glucose deprivation and insulin treatment is characterized by a rapid increase in glucose transport activity which occurs without any alteration in glucose transporter protein and mRNA. In contrast, the long term effects of insulin on glucose transport activity correlate with a timedependent increase in both the glucose transporter protein and mRNA. Further studies willbe necessary in order to determine the molecular events responsible for the transient insulin responsiveness of glucose transporter mRNA expression and to elucidate a role for the long term decrease in the glucose transporter mRNA in the peripheral resistance of muscle tissue to insulin action. Acknowledgments-We wish to thank Drs. Morris Birnbaum and Ora Rosen for providing us with the full-length rat brain glucose transporter cDNA and Dr. Michael Weber for the use of aGT-8. We also wish to thank Mark Meier for his assistance in the culturing of the L6 cells and Dr. Brian Van Ness for helpful discussions during the course of these studies. Note Added in Proof-Recently, Hiraki et al. (63) demonstrated that serum stimulation of confluent NIH3T3 fibroblasts also results in a transient induction of the rat brain glucose transporter mRNA with a similar time course to that observed for the c-10s protooncogene.

1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.

REFERENCES Wheeler, T. J., and Hinkle, P. C. (1985) Annu. Rev. Physwl. 47, 503-517 Plagemann, P. G.W., and Richey, D. (1974) Biochim. Biophys. Acta 3 4 4 , 263-305 Simpson, I. A., and Cushman, S. W. (1986) Annu. Rev. Biochem. 55,1059-1089 Klip, A., and Ramlal, T. (1987) Biochem. J. 2 4 2 , 131-136 Muhlbacher, C., Karnieli, E., Schaff, P., Mushack, J., Rattenhuber, E., and Haring, H. U.(1988) Biochem. J. 249,865-870 Joost, H. G., Weber, T. M., Cushman, S. W., and Simpson, I. A. (1987) J. Biol. Chem. 2 6 2 , 11261-11267 Kuroda, M., Honnor, R. C., Cushman, S. W., Londos, C., and Simpson, I. A. (1987) J. Biol. Chem. 2 6 2 , 245-253 Baly, D. L., and Horuk, R. (1987) J. BioZ. Chem. 262,21-24 Rosen, 0.M., Smith, C. J., Fung, C., and Rubin, C. S. (1978) J. Biol. Chem. 253,7579-7583 Kahn, B. B., Horton, E. S., and Cushman, S. W. (1987) J. Clin. Invest. 79,853-858 Ciaraldi, T. P.,and Olefsky, J. M. (1983) Endocrinology 1 1 3 , 1739-1745 Wardzala, L. J., Hirshman, M., Pofcher, E., Horton, E. D., Mead, P. M., Cushman, S. W., and Horton, E. S. (1985) J. Clin. Invest. 76,460-469 Wallberg-Henriksson, H. (1987) Acta Physwl. Scand. 5 6 4 , 1-80 Young, D. A., Uhl, J. J., Cartee, G. D., and Holloszy, J. 0.(1986) J. Biol. Chem. 261,16049-16053 Bequinot, F., Kahn, C. R., Moses, A. C., and Smith, R. J. (1986) Endocrinology 118,446-455 Sasson, S., and Cerasi, E. (1986) J. Bwl. Chem. 261, 1682716833 Yamada, K., Tillotson, L.G., and Isselbacher, K. J. (1983) J. Biol. Chem. 258,9786-9792 Pessin, J. E., Tillotson, L. G., Yamada, K., Gitomer, W., CarterSu, C., Mora, R., Isselbacher, K. J., and Czech, M. P. (1982) Proc. Natl. Acad. Sci. U. S. A. 7 9 , 2286-2209 D'Amore, T., and Lo, T. C. Y. (1986) J. Cell. Physwl. 127, 95105 Haspel, H. C., Wilk, E. W., Birnbaum, M. J., Cushman, S. W., and Rosen, 0.M. (1986) J . Bid. Chem. 261,6778-6789 Lemmon, S. K., Sens, D.A., and Buse, M. G. (1985) J. Cell. Physiol. 125,456-464 van Putten, J. P. M., and Krans, H. M. J. (1985) J . Bwl. Chem. 2 6 0 , 7996-8001 Pessin, J. E., and Czech, M. P. (1985) in The Enzymes ofBiologica1 Membranes (Martonosi, A. N., ed) Vol. 2, PP.497-522, Plenum Publishing Corp., New York Salter, D. W., Baldwin, S. A., Lienhard, G. E., and Weber, M. J. "

24.

Molecular Regulation of Gluccose Transporter Expression (1982) Proc. Natl. Acad. Sci. U. S.A. 79,1540-1544 25. Flier, J. S., Mueckler, M. M., Usher, P., and Lodish, H. F. (1987) Science 235,1492-1495 26. Birnbaum, M. J., Haspel, H. C., and Rosen, 0.M. (1987) Science 235,1495-1498 27. White, M., and Weber, M. J. (1988) Mol. Cell. Biol. 8 , 138-144 28. Wang, C. (1987) J. Bwl. Chem. 262,15689-15695 29. James, D. E., Brown, R., Navarro, J., and Pilch, P. F. (1988) Nature 333,183-185 30. Taylor, L. P., and Holman, G. D. (1981) Biochim. Biophys. Acta 642.78-86 31. Holman, G. D., and Rees, W. D. (1982) Biochim. Biophys. Acta 685,78-86 32. Horuk, R., Matthei, S., Olefsky, J. M., Baly, D., Cushman, S. W., and Simpson, I. A. (1986) J. Biol. Chem. 261,1823-1828 33. Alexrod, J. D., and Pilch, P. F. (1983) Biochemistry 2 2 , 22222227 34. Ciaraldi, T. P., Horuk, R., and Matthaei, S. (1986) Biochem. J. 240,115-123 35. Fukumoto, H., Seino, S., Imura, H., Seino, Y., Eddy, R. L., Fukushima, Y., Byers, M.G., Shows, T. B., and Bell, G.I. (1988) Proc. Natl. Acad. Sci. U. S. A. 8 5 , 5434-5438 36. Mueckler, M., Caruso, C., Baldwin, S. A., Panico, M., Blench, I., Morris, H. R., Allard, W. J., Lienhard, G. E., and Lodish, H. F. (1985) Science 229,941-945 37. Birnbaum, M. J., Haspel, H. C., and Rosen, 0.M. (1986) Proc. Natl. Acad. Sci. U. S. A. 8 3 , 5784-5788 38. Shawver, L. K., Olson, S. A., White, M.K., and Weber, M. J. (1987) Mol. Cell. Biol. 7 , 2112-2118 39. Paris, S., Pouyssegur, J., and Ailhaud, G. (1980) Biochim. Biophys. Acta602,644-652 40. Christopher, C. W., Colby, W. W., and Ullrey, D. (1976) J. Cell. Physiol. 89,683-692 41. Walker, P. S., Donovan, J. A., Van Ness, B. G., Fellows, R. E., and Pessin, J. E. (1988) J . Biol. Chem. 2 6 3 , 15594-15601 42. James, D. E.,Burleigh, K. M., and Kraegen, E. W. (1985) Diabetes 34,1049-1054

6595

43. Klip, A., Li, G., and Logan, W. J. (1984) Am. J. Physiol. 247, E291-296 44. Klip, A., Li, G., and Logan, W. J. (1982) Biochim. Biophys. Acta 687,265-280 45. Bradford, M. M. (1976) Anal. Biochem. 7 2 , 248-254 46. Alexander, M., Curtis, G., Avruch, J., and Goodman, H.M. (1985) J. Bwl. Chem. 260,1197&11985 47. Yaffe, D. (1968) Proc. Natl. Acad. Sci. U. S. A. 6 1 , 477-483 48. Christopher, C.W., Kohlbacher, M. S., and Amos, H. (1976) Biochem. J. 158,439-450 49. Bequinot, F., Kahn, C. R. Moses, A. C., and Smith, R. J. (1985) J.Bwl. Chem. 260,15892-15898 50. Bequinot, F., Kahn, C. R., Moses, A. C., and Smith, R. J. (1986) Endocrinology 1 8 , 446-455 51. Germinario, R. J., Ozaki, S., andKalant, N. (1984)Arch. Biochem. Biophys. 234,559-566 52. Cigolini, M., Cavallo, E., Zancanaro, C., Micciolo, R., Benati, D., and Bosello, 0.(1985) Acta Diabetol. Lat. 2 2 , 351-355 53. Kelly, K., Cochran, B. H., Stiles, C. D., and Leder, P. (1983) Cell 35,603-610 54. Taub, R., Roy, A,,Dieter, R., and Koontz, J. (1987) J. Biol. Chem. 262,10893-10897 55. Biber, J. W., and Lienhard, G. E. (1986) J. Biol. Chem. 2 6 1 , 16180-16184 56. James, D. E., Lederman, L., and Pilch, P. F. (1987) J. Biol. Chem. 262,11817-11824 57. Haspel, H. C., Rosenfeld, M. G., and Rosen, 0.M. (1988) J . Biol. Chem. 263,398-403 58. Fulton, A. B., Wan, K. M., and Penman, S. (1980) Cell 2 0 , 849857 59. Bonneau, A. M., Darveau, A., and Sonenberg, N. (1985) J. Cell Biol. 100,1209-1218 60. Lawerence, J . B., and Singer, R. H. (1986) Cell 45,407-415 61. Kayano, T., Fukumoto, H., Eddy, R. L., Fan, Y.-S., Byers, M. G., Shows, T. B., and Bell, G. I. (1988) J. Biol. Chem. 263,1524515248 62. Thorens, B., Sarkar, H.K., Kaback, H. R., and Lodish, H. F. (1988) Cell 55,281-290 63. Hiraki, Y., Rosen, 0. M., and Birnbaum, M. J. (1988) J. Biol. Chem. 263,13655-13662 ~I