Kinetics of GLUT1 and GLUT4 Glucose Transporters Expressed in Xenopus Oocytes*

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 268, No. 12, Issue of April 25, pp , 1993 Printed in U. S. A. Kinetics of GLUT1 and Glucose Transporters Expressed in Xenopus Oocytes* (Received for publication, September 28, 1992) Haruo Nishimura$Q, Federico V. Pallardon(1, Glen A. Seidnerll, Susan Vannucci$**, Ian A. Simpson$, and Morris J. BirnbaumT$$ From the $Experimental Diabetes, Metabolism and Nutrition Section, Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Disease, National Institute of Health, Bethesda, Maryland and the TDepartment of Cellular and Molecular Physiology, Harvard Medical School, Boston, Massachusetts The predominant mechanism by which insulin acti Insulin stimulates glucose transport activity in hormonevates glucose transport in muscle and adipose tissue is responsive tissues primarily by inducing the redistribution of by affecting the redistribution of the facilitated hexose the facilitated hexose carrier isoforms, and, carriers, and, from an intracellular site to the plasma membrane. A quantitative analysis from an intracellular compartment(s) to the plasma memof this process has been hampered by the lack of reli brane (13). is widely distributed whereas is able determinations for kinetic constants catalyzed by expressed exclusively in those tissues in which insulin proeach of these isoforms. In order to obtain such infor duces a marked increase in glucose transport activity, i.e. mation, each transporter was expressed in Xenopus heart, skeletal muscle, and white and brown adipose tissue (4, oocytes by the injection of mrna encoding rat 5). In the latter, is the predominant glucose transor. Equilibrium exchange 30methylglucose porter species (6; for review, see Refs. 7, 8). uptake was measured and the data fitted to a two A persistent, controversial issue in the study of glucose compartment model, yielding K,,, = 26.2 mm and V, = 3.5 nmol/min/cell for and K,,, = 4.3 mm and transport has been the relative contributions of alterations in Vmpr = 0.7 nmol/min/cell for. Measurement of cell surface carrier number and intrinsic catalytic activity to the abundance of cell surface transporters was accom the regulation of substrate flux (8, 9). Several years ago, the plished by two independent protocols: photolabeling lack of correlation between the change in plasma membrane with the impermeant hexose analog 2N4(1azi glucose transporter, as assayed by cytochalasin B binding, 2,2,2trifluoroethyl)benzoyl1,3bis(~mannos4 and hexose uptake appeared to resolved, be first by the cloning yloxy)2propylamine and subcellular fractionation of oocytes. Data obtained by either technique revealed that the ratiof plasma membrane to was about 4; this paralleled the relative maximal velocities for hexose transport, indicating that the turn over numbers for the two isoforms were the same. Moreover, measurement of the concentration of exofacially disposed transporters with 2N4(1azi 2,2,2trifluoroethyl)benzoyllY3bis(~mannos4 yloxy)2propylamine allowed calculation of the turnover number to be about 20,000 min. These data indicate that, at low substrate concentrations, the catalytic efficiency of is significantly greater than GLUT1. Extrapolation to mammalian systems suggests that is responsible for vir tually all of the hexose uptake in insulinresponsive targets, particularly in the presence of hormone. * This work was supported by Grant DK39519 from the National Institutes of Health and a grant from the Juvenile Diabetes Foundation (to M. J. B.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked aduertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 3 Present address: Dept. of Endocrinology and Metabolism, Shizuoka General Hospital, 4271 Kitaando, Shizuoka 420, Japan. 11 Supported by a North Atlantic Treaty Organization fellowship. Present address: Dept. de Fisiologia, Facultad de Medicina, Universidad de Valencia, Avenue Blasco IhaEz 17, Valencia 46010, Spain. ** Present address: Dept. of Pediatric Neurology, Milton Hershey Medical Center, Pennsylvania State University, Hershey, PA $$ To whom correspondence should be addressed Dept. of Cellular and Molecular Physiology, Harvard Medical School, 25 Shattuck St., Boston, MA of cdnas and preparation of antisera specific to each of the transporter isoforms, and subsequently by the development of impermeant photoaffinity labels of the transporter (46, 10). Nonetheless, the extrapolation of measurements of surface transporter, obtained either by Western blot or photoaffinity labeling, to predict rates of hexose flux requires knowledge of the kinetic parameters of each of the isoforms involved. Such measurements have been difficult to obtain due to the lack of mammalian cells expressing exclusively a single isoform and the inability to definitively establish the number of cell surface transporters. Studies to evaluate the hexose transport kinetics of the different carrier isoforms have included experiments using human erythrocytes, which possess only (for review, see Ref. ll), insulinstimulated adipocytes, in which the major plasma membrane species is (8,12, 13), and Xenopus laeuis oocytes, which have low rates of endogenous uptake but in which, in principle, any cloned transporter can be expressed (1416). Using the latter system, several groups have measured the apparent K, for equilibrium exchange of 30methylDglucose as about 20 mm for and found the K, for either 2 mm or impossible to determine due to low levels of expression (14 16). However, the difficulty in establishing the number of cell surface transporters prevented a reliable determination of the turnover number (V,,./numberof transporters), the other important parameter for estimating intrinsic activity. In the experiments presented below, we determined the apparent K, and Vmax for rat and catalyzed 30 methyldglucose transport activity in Xenopus oocytes by analysis of a twocompartment model and utilized the exofa

2 cia1 affinity photolabel [3H]ATBBMPA' to simultaneously evaluate the number of cell surface glucose transporters. MATERIALS AND METHODS Injection of OocytesStage VI oocytes were prepared from gravid X. laeuis females (Nasco, Fort Atkinson, WI) as described previously (5). For isolation of plasma membrane complexes, the oocytes were dissected manually without exposure to collagenase. The day following harvest, oocytes were injected with 50 nl of either autoclaved water or RNA (0.4 mg/ml) prepared from rat or cloned into psp64t, linearized, and transcribed with SP6 polymerase in the presence of cap analog (5, 17, 18). Oocytes were incubated in modified Barth's solution (19) at 19 "C for 4 days prior to assay. For metabolic labeling, oocytes were incubated in the same solution containing 60 p~ methionine and 1 mci/ml [35S]Translabel (ICN Radiochemicals, Irvine, CA) for 2 days at 19 "C prior to cell fractionation. Preparation of Total MembranesOocytes were homogenized in TES buffer (20 mm Tris, ph 7.4,l mm EDTA, 255 mm sucrose, 1 pm aprotinin, 0.1 WM leupeptin, 1 p~ pepstatin, 50 p~ phenylmethylsulfonyl fluoride) and crude membranes pelleted by centrifugation at 50,000 revolutions/minute in a Beckman rotor. The supernatant contained no glucose transporter detectable by Western blotting (data not shown). Membranes were washed two times by resuspension in TES buffer and centrifugation as above. Plasma membranes were prepared from insulinstimulated adipocytes as described previously (20). Membranes were subjected to SDSPAGE and Western blot analysis with polyclonal antibodies raised against carboxyl termini of and (21). The glucose transporters were visualized using '261labeled protein A and quantitated by cutting the nitrocellulose paper and placing in a ycounter. Preliminary experiments showed the counts/minute to be linearly related to the amount of transporter protein in the range of concentrations utilized. Plasma membrane complexes from oocytes injected with both and mrna were prepared as described by Wall and Pate1 (22), except that only two to three strokes were used in the initial homogenization. Purified membranes were solubilized in 0.4% SDS, 20 mm TrisHC1, ph 7.5, 100 mm NaCl, and 2 mm EDTA, 0.1 volume 20% Triton X100 added, and reacted overnight at 4 "C with antisera specific for the carboxyl termini of or. The immune complexes were adsorbed with protein Asepharose, washed three times with 10 mm sodium phosphate, ph 7.4, 150 mm NaCl, 0.1% Triton X100, and 1 mm EDTA, eluted into transporter loading buffer (23), and visualized by SDSpolyacrylamide electrophoresis and autoradiography. Radioactivity was quantitated using a Molecular Dynamics phosphorimager equipped with ImageQuant software. Photoaffinity Labeling of Glucose TransportersPhotoaffinity labeling of glucose transporters with [3H]ATBBMPA was performed according the method of Holman et al. (6) with modifications. Intact oocytes or adipocyte plasma membranes were incubated in 150 pl of modified Barth's medium with [3H]ATBBMPA (500 pci) in a 200 pl fluorescence quartz cuvette at room temperature. The cuvette was irradiated six times for 30 s using a Rayonet photochemical reactor equipped with 300nm lamps. Cells or membranes were solubilized in a detergent buffer (1% SDS, 1% Triton X100, 0.4% deoxycholate, 66 mm EDTA, 50 mm TrisHC1, ph 7.4) for 1 h at 4 "C, centrifuged at 10,000 X g for 30 min., and the supernatant was diluted 10fold with 1% Triton X100, 66 mm EDTA, 50 mm TrisHC1, such that the final concentration of SDS was 0.1%. The extract was incubated with a or aglut1coated protein ASepharose CL4B for 12 h at 4 "C. Immunoprecipitations were performed three successive times for each sample and the pellets pooled, resulting in adsorption of >90% of the transporter. The beads were washed, the immune complexes eluted with sample buffer and subjected to SDSPAGE, and quantitated by scintillation counting as described (6). Measurement of 2Deoxyglucose Transport ActivityGroups of 10 oocytes were incubated in 2 ml of modified Barth's medium containing 0.1 mm 2deoxyglucose and 1 pci of [3H]2deoxyglucose for 30 min at 20 'C. Hexose uptake was linear for at least 30 min, as reported previously (5). The reaction was terminated by washing the oocytes three times with icecold phosphatebuffered saline, quickly aspirating the medium, and dispensing the oocytes to glass scintillation vials. Kinetics of and in Xenopus Oocytes % sodium dodecyl sulfate (0.5 ml) was added to each vial, which were incubated at room temperature with shaking for at least 2 h prior to liquid scintillation spectroscopy. Measurement of 30Methylglucose Equilibrium Exchunge Groups of 2530 RNA or waterinjected oocytes were incubated for approximately 18 h at 19 "c in varying concentrations ( mm) of 30methylDglucose (30MG). Cells were divided into subgroups of three to four oocytes in 0.5 mlof modified Barth's solution containing 2 pci of [3H]3OmethylDglucose and the appropriate equilibrium concentration of 30MG. "Zero time" uptake was determined by adding Dglucose (200 mm, final concentration) prior to radioactive 30MG. Uptake at 22 "Cwas terminated by quickly washing three times with 8 ml of icecold phosphatebuffered saline mm phloretin. Individual oocytes were solubilized and counted as described above. Preliminary experiments of equilibrium exchange 30MG uptake indicated that, even after substantial periods of time, the intracellular space had not completely equilibrated with extracellular hexose. As illustrated in Fig. L4, oocytes previously injected with mrna or water had accumulated at 2 h only about half the [3Hl30MG taken up at 20 h. This was interpreted as indicative of the presence a second, more slowly filled, intracellular compartment. Therefore, data were analyzed by a twocompartment model, as follows. When two intracellular compartments are present in series, the concentration of glucose at time t can be expressed as the following: Xl(t) = Xoeklt (Es. 1) where Xl(t) is the concentration of [3H]30MG in the extracellular medium, and Xz(t) and XJt) are the concentrations in the first and second compartment in the cell. When each compartment has a The abbreviations used are: ATBBMPA, 2N4(1azi2,2,2tri ing (A) or (B) and 4 days later assayed for transport fl~0r0ethyl)benz0yll,3bis(~mannos4yloxy)2propy~a~in~; 30 of 30MG. Oocytes were incubated for 18 h in nonradioactive 5 mm MG, 30methylDglucose; PAGE, polyacrylamide gel electrophore 30MG, tracer [3H]30MG was added, and accumulation of radiosis; dpm, disintegrations/minute. activity measured at the times indicated. lime (min) FIG. 1. Time course of 30methylDglucose uptake into oocytes. Oocytes were injected with water or 20 ng of mrna encod

3 Kinetics of and in Xenopus Oocytes 8516 volume V2and V3,net substrate uptake into thecell Si(t) is, Si(t) = V2XAt)+ V3X3(t) (Eq. 4) Thus, totalhexose uptake is expressed as a twoexponential equation, S i ( t ) = aeklf + be&+ c (Es.5) in which a, b, and c are constants. Because Si(0) = 0, the above equation can be transformed to, Si(t)= ml(l e+) + m2(1 e&) (Eq. 6) in which ml and m2 are constants. The initial velocity, u(o), is the derivative of Si(t ) a t time ) = mlkl + mzh 2ooo (Eq. 7) loo0 (Eq. 8) 0 For each concentrationof 30MG, uptakea t five to eight time points was measured, and the radioactivity associated with waterinjected cells subtracted. Typically, waterinjected oocytes accumulated less than 10% the 30MG of transporterexpressing calls, although a t the earliest time points this approached 50% (Fig. 1). Values were curvefitted to Equation 6 using the program MLAB (Civilized Software, Inc., Bethesda, MD) with the MarquardtLevenberg iterative curve fitting algorithm. Next,ml, kl, m2,and k2 were determined, and initial velocity was calculated using Equation 8. CalculationsStatistical significance was tested with one way analysis of variance followed by Duncan's multiple range test and unpaired Student's t test, as appropriate, and differences were accepted as significant at thep < 0.05 level Slim Number FIG. 3. Photoaffinity labeling of glucose transporters inoocytes. 10 oocytes were injected with water (Control, A), CLUTl or mrna (0)and exofacially photolabeled with mrna (0). 13H]ATBBMPA.The cells were solubilized and immunoprecipitated with antisera directed against (GLUT],Control) or (). A gel profile is shown which is representative of a t least three independentexperiments. The molecular size standards are indicated in kda. were not detected by Western blot of t o t a l membranes of waterinjected cells (Fig. 2). Quantitation of Glucose Transporters in OocytesTo determeasurement of Surface Transporters by Labeling with mine the quantity of glucose transporters expressed in oocytes ATBBMPAFig. 3 shows a typical profile of immunoprecipinjected with in uitrosynthesizedrna, crude membranes itated transporter from oocytes injected with water, CLUTl were preparedfrom10 oocytes, yielding about 200 pg of mrna, ormrnaand labeled with ['HI ATBprotein/cell. Membranes were subjected to SDSPAGE and BMPA. Waterinjected oocytes contained little or no radiowesternblotusingantisera specific for the and activity in immunoprecipitates using antisera directed against transporter isoforms. As a standard, plasma memeither transporter isoform. Oocytes injected with mrna enbranes from insulinstimulated rat adipocytes were included coding or and exofacially labeled contained on the same immunoblot. This membrane fraction has been 1404 f 83 dpm/cell ( n = 3) and dpm/cell ( n = 3) shown to contain22 pmol of glucose transporter/mg of protein associated witheachtransporter, respectively. Inparallel as determined by cytochalasin B bindingand a ratio of experiments, plasma membranes from insulinstimulated fat to of 1O:l as assayed by labeling with ATB cells, which contain 2pmol of and 20 pmol of BMPA (6). A representative immunoblot is shown in Fig. 2. /mg of protein (see above), were photolabeled in the There was no difference in the total level of expression of same concentrationof [3H]ATBBMPAwith a geometry ideneach of the transporters: 1.8 f 0.4 pmol of /cell and tical to that used for intact oocytes. Data from two independ2.4 f 0.6 pmol of /cell ( p > 0.05, n = 3). and ent experimentsrevealed that the incorporation of photolabel into adipocyte transporters were, respectively 563, 532 dpm/ 0.05 pmol of GLUT1/25 pg of protein and 5897, 5634 dpm/ 0.5 pmol of /25 pg of protein. These results confirm kda that both transporters are labeled with identical efficiencies 200 of 1.1 X 10' dpm ATBBMPA/pmol transporter. Measurement of Glucose Transporters in Plasma Membrane loo ComplexesManually defolliculated oocytes were injected 68with both and mrna and incubated for 4 43days at 19 "C,during the last 2 days in the presence of [3sS] amino acids. Plasma membrane complexes were isolated, solubilized, and divided into 3 equal aliquots for immunoprecip28 itation with nonimmune sera, aglut1, or a. Results of a typical experiment are shown in Fig. 4 A, in which immunoprecipitates from whole cell homogenates are also included. Immunoprecipitation of membranes from GLUT1FIG. 2. Expression of glucose transporters in Xenopus oo or expressing oocytes with antisera directed against cytes. Oocytes were injected with water (Control)or 20 ng of in oitro the other carrier isoform or nonimmune serum consistently synthesized or mrna. Total membranes were preyielded no detectable radioactive species (data not shown). pared and subjected to Western blot using antisera specific for or. As a standard, plasma membranes purified from The results from six independent experiments are summainsulinstimulated rat adipocytes (Fat, 25 pg) was also included. rized in Fig. 4B; 7.6% of total cellular and 2.0% of cofractionated with plasma membrane complexes. In Molecular mass markers are indicated in kda. RESULTS

4 A Total PMC 1 oocyte 30oocytes Kinetics of and in Xenopus Oocytes 8517 kda A m r 15ooo loo00 5ooo n lime (min) mm mm B l2 c FIG. 4. Immunoprecipitation of metabolically labeled gluco88 transporter in oocyte plaema membrane complexes. Panel A, oocytes were injected with both and mrna and incubated with a mixture of [R5S]amino acids. Either intact oocytes (Total) or plasma membrane complexes (PMC) were solubilized and immunoprecipitated with antibody specific for or. Immune complexes were eluted, subjected to SDSPAGE, and visualized by fluorography. For total extract, transporter immunoprecipitated from a single oocyte was loaded in each lane, whereas 30 oocytes provide the material for each lane of immunoprecipitate from plasma membrane complexes. The migration of molecular weight markera is indicated. Panel B, the data from six independent experiments such as that shown in panel A are summarized as percent of total cellular transporter present in the plasma membrane complex fraction. Data represent the mean f S.E. spite of significant variability among experiments in the recovery of glucose transport protein in the plasma membrane complexes, the ratio of / in this fraction was remarkably consistent: 3.8 f (n = 6). 2Deoxyglucose UptakeUptake of 0.1 mm 2deoxyglucose was measured as 118 * 18 and 81 & 13 pmol/30 min/cell for GLUT1 and injected oocytes, respectively. Water injected cells accumulated less than 5 pmol/30 min/cell). These values were unaltered by insulin treatment for 30 min (data not shown). Kinetics of 30Methylglucose TransportAs described above, the primary data from 30MG uptake was best described by a twocompartment model, in which there is a rapidly filled compartment which equilibrates more slowly with a second intracellular compartment. Data from a representative 30MG equilibrium exchange transport assay are B m lime (min) \ I 0 0 \ 0 GLUT v / [SI 0 I FIG. 5. Equilibrium exchange 30 C transport into oocytes expressing GLUT1 or CLUTI. Oocytes injected with water or mrna encoding either (panel A) or (panel R) were incubated for 4 days at 19 C, with the indicated concentration of 30 C present during the last 18 h. Uptake was initiated by the addition of tracer 13H]30MG and the assay terminated at the times indicated. The accumulation of ( HJ30MG in waterinjected oocytes has been subtracted. Data are the mean f S.E. of three to four oocytes from an experiment representative of at least three independent experiments. Panel C, WoolfHofstee plot of the data from panepb A and B. Initial velocities were calculated as described under Materials and Methods. shown in Fig. 5. Oocytes injected with mrna encoding either or were incubated to equilibrium with nonradioactive 30MG and the assay initiated by the addition of radioactive tracer. Concentrations of hexose tested ranged from 0.3 to 30 mm for and 1 to 100 mm for CLUT1. By using Equations 6 and 8, initial velocities were determined and plotted against substrate concentrations according to

5 8518 Kinetics of and in Xenopus Oocytes Woolf, as advocated by Hofstee (24) (Fig. 5C). From a series of experiments of this kind, the kinetic constants derived were K,,, = 26.2 f 4.9 and V,, = 3491 k 448 pmol/min/cell (n = 3) for and K,,, = 4.3 f 0.6 mm and V,,, = 666 f 187 pmol/min/cell (n = 3) for expressing oocytes (Table I). tain absolute plasma membrane abundance with the aim of deriving turnover numbers. The first approach utilized was exofacial labeling with the impermeant glucose analog, ATB BMPA. This reagent was developed by Holman and his colleagues (6,2527) and has proven useful for measuring the number of cell surface and transporters in mammalian cells. There now exists much evidence that ATB DISCUSSION BMPA binds to both transporter isoforms with the same The primary goals of these experiments were to determine affinity and photolabels with virtually 100% efficiency (6, 25, the apparent K,,, and V,, for equilibrium uptake of 30MG 27). Thus, the major theoretical drawback of this reagent is in Xenopus oocytes injected with mrna encoding the that it might not recognize transporters which reside on the or glucose transporters and under the same condi cell surface but are catalytically silent. tions measure the number of carriers on the cell surface. This In this study, we have adapted the ATBBMPAlabeling would permit, for the first time in regard to, a realistic technique for Xenopus oocytes. The primary data can be assessment of turnover number in cells expressing exclusively one transporter. Establishing apparent affinities was the initial task and largely served to confirm previously published converted into numbers of transporters in two ways. In the first, insulintreated rat adipocyte plasma membranes, which contain a measurable quantity of and, served work (14, 15). Nonetheless, we found that the presence of a as standards and were photolabeled in precisely the same second, slowly equilibrating compartment for distribution of manner as intact oocytes (6). Since about 1.1 X lo4 dpm ATBhexose confounded a simple linear transformation of the data, BMPA was incorporated into a picomole of either transporter which were instead analyzed by fitting to the equations described under Materials and Methods. The precise biochemical nature of this apparent intracellular space remains unclear, but might well represent the yolk platelets, which ocin adipocyte membranes, the amount of and on the surface of the oocyte can be calculated as 128 f 13 and 40 f 8 fmol/cell, respectively (Table I). Alternatively, one can take advantage of the fact that essentially all bound ATBcupy about 50% of the total volume of the oocyte (22). BMPA is covalently linked to transporter upon exposure to Utilizing such a twocompartment model eliminated the requirement for determining intracellular space and provided a relatively good fit for the data. Values obtained for the K,,, ultraviolet light to derive a value for total binding sites, Bmax (27). By using the dissociation constant, Kd, for ATBBMPA which is known to be about 150 pm (27), the concentration of were 26.2 k 4.9 and 4.3 k 0.6mM for and, free ligand, F, which equals 333 pm, deriving the bound ligand, respectively (Table I). The apparent affinity of for 3 B, from the specific activity, 10 Ci/mmol, and the Michaelis 0MG compares favorably with values previously reported, equation, whether determined in human erythrocytes or Xenopus oocytes (Table 11). There has been one prior study which in the K,,, for expressed in oocytes was found to be 1.8 mm, or about half of that reported here (14). Measurements of one derives the total cell surface binding, B,,,, to be 103 f 13 s affinity for hexose in mammalian systems have fmol/cell for and 32 f 7 fmol/cell for, in relied on the assumption that this isoform catalyzes the major portion of 30MG uptake in insulinstimulated adipocytes. The good correlation between estimates in adipocytes and the current values (Table 11), as well as calculations described good agreement with the above values. As illustrated in Fig. 2, the total cellular transporter can be estimated from quantitative Western blot to be the same for and and the fraction expressed on the cell surface 7.2 and 1.8%, below based on the presently reported affinities and turnover respectively (Table I). numbers of each of the isoforms, provide direct evidence that A completely independent method for determining the ratio is catalytically dominant in hormonetreated cells. fat of cell surface to was to coinject mrna Measuring cell surface transporter has proven a difficult encoding both carrier isoforms into the same population of task in past experiments. Because of the problems inherent to all current methodologies, we elected to utilize two indeoocytes and prepare a plasma membraneenriched fraction. Wall and Pate1 (22) have described a method for the purifipendent techniques. It should be emphasized that the critical cation of plasma membrane complexes almost completely parameter required to calculate the relative intrinsic activities of the two transporter isoforms is the ratio of their cell surface expression. Nonetheless, we have attempted to ascerdevoid of intracellular membranes, as determined by electron microscopy. Using this procedure, we measured the plasma membrane and to be 7.6 and 2.0% of total TABLE I Summary of kinetics of hexose transport activity in Xenopus oocytes expressing rat or 2Deoxyglucose Surface 30MG transport Total Surfaceftotal Turnover no? uptake K, V,, (ATBBMPA) (Western) ATBBMpAb PMC prnolf30 rninfcell m.w fmolfcell pmolfrninfcell prnolfcell % min X I@ 118 f f & f & f % & & f f f f & f f 0.4 NS p < 0.05 p < 0.05 p < 0.01 NS p < 0.01 p < 0.05 NS Values calculated on the basis of ATBBMPA labeling of intact oocytes, using adipocyte plasma membranes as standards. * Values are the number of surface transporters as determined by ATB labeling divided by total cellular transporter as ascertained by Western blot of crude membranes, using adipocyte plasma membranes as standards. Values represent the fraction of surface transporters, as determined by the immunoprecipitation of metabolically labeled transporters from plasma membrane complexes and from total oocyte homogenate. Calculated using the number of surface transporters determined by ATBBMPA labeling. Comparison of values for and. NS, not significant.

6 Kinetics of GLUT1 and in Xenopus Oocytes 8519 TABLE I1 Summary of kinetic constants for or 30MG equilibrium exchange transport Cell type Temperature Ref. E TN E TN mm X 1 fy min" mm XI@ min" "C Xenopus oocytes" NDb Xenopus oocytes 20 1 ND ND Xenopus oocytes 17 ND ND 16 ND 18 Xenopus oocytes study This Human erythrocytes Reviewed in 11 adipocytes Rat 5.0' ND 37 Reviewed in 8 Rat ND 5.6' 3T3Ll adipocytes 23d 7.2' 7.2d 7.Y Data are derived from experiments in which mammalian transporters were expressed in oocytes. * Not determined. These values are derived from experiments using insulinstimulated adipocytes and assume that all measurable transport is catalyzed by GLUTI. See "Discussion" for validity of this assumption. Values calculated from the displacement of ATBBMPA. e See Ref. 27 for an explanation of the derivation of these turnover numbers. cellular transporter, respectively (Fig. 4). The striking correlation between the estimate of surface transporters as ascertained by ATBBMPA photolabeling and subcellular fractionation provides strong support for the reliability of these measurements. Moreover, these data argue against the existence of a significant pool of plasma membrane transporters which are not labeled by ATBBMPA in oocytes. Since the ratio of surface / as determined in either of these series of experiments approximates the ratio of the V,, values, the turnover numbers of the two isoforms are essentially the same (Table I). It is possible to apply the kinetic constants obtained from oocyte experiments to measurements of and in adipocytes. The velocity of transport into an adipocyte expressing both and is, (Eq. 10) in which the subscripts, and 4, indicate the constants pertaining to and, respectively. Since, V, = TN[GT] (Eq. 11) in which TN is the turnover number and [GT] the number of cell surface transporters, the relative uptake in the presence and absence of insulin, v+l/vl, can be expressed as, Km1 + [SI. + ~. K + [SI Moreover, since, as discussed above, TNl = TN,, under the usual assay conditions, i.e. [SI << K,, Using the values for K, in Table I, (Eq. 13) one can use previously reported values for obtained by ATB BMPA photolabeling of 3T3Ll adipocytes in the presence or absence of insulin to derive as 12.2 (10) or 8.3 (27, 28), and compare these values to hormonestimulated hexose uptake of about 10fold as measured by equilibrium exchange 30 MG transport (27, 29) and 1020fold as assayed by zerotrans 2deoxyglucose uptake (10, 28, 3032). In isolated rat adipocytes, the ATBBMPA binding data of Holman et al. (6) can be applied to Equation 14 to yield a predicted insulin stimulation of 19fold, in which catalyzes virtually all hexose flux in both the presence and absence of hormone. Should the basal ratio of plasma membrane to, which is difficult to measure by photoaffinity labeling, be higher than 1:2 (6), the predicted stimulation by insulin would also be greater. In addition, these data would explain the decrease in apparent K, for hexose uptake which some investigators have found to accompany insulin stimulation (29, 3335). In any case, the translocation of and appears to explain most or all of the insulinstimulated hexose uptake into 3T3Ll or rat adipocytes. Since the present data predict the insulin response to be at least 50% of that observed experimentally, if a translocationindependent mechanism does exist, it is likely to contribute modestly to the overall hormonedependent increase in flux. Moreover, the present studies cannot address the issue of the mechanism of potential alterations in hexose uptake unaccompanied by redistributions in transporter. A number of perturbations, including inhibition of protein synthesis and exposure to adrenergic agents, modulate both transport activity and ATB BMPA binding without concomitant alterations in immunologically detectable transporter in plasma membraneenriched fractions (29, 32, 36). In the present study, the correlation between cell surface transporter as determined by Western blot and photolabeling argues strongly against the existence of latent plasma membraneassociated or in mrnainjected oocytes and suggests that the kinetic constants determined represent "basal" values intrinsic to the carriers. Perhaps Xenopus oocytes will ultimately provide an experimental system in which to investigate the process by which transport activity is suppressed in mammalian cells (37). Several assumptions have been made in the previous argument. Since little is known about glycosylation of membrane proteins in X. oocytes it is difficult to ascertain whether alternative posttranslational processing might affect the kinetics. More generally, the most critical assumption is that data obtained from transport measurements at 22 "C in Xenopus oocytes can be applied to mammalian systems. The strongest support for such an assumption is the correlation

7 ~~ ~~~ 8520 Kinetics of and in Xenopus Oocytes between kinetic constants obtained in this study and those reported previously, particularly for for which there is substantial published data derived from experiments in human erythrocytes (Table 11). The good agreement between the turnover number reported herein, 27,000 min', and that derived from measurements in human erythrocytes emphasizes the need to accurately assess plasma membrane transporters in oocytes, in which a relatively modest percentage are present at the cell surface. It is unclear why a prior study obtained a turnover number for significantly higher than that found here, although it is possible that there was incomplete recovery of plasma membrane transporter (14). The kinetic constants for, K, = 4.3 mm and TN = 17,000 min", represent the first direct measurements in a system expressing exclusively the single isoform. Previous estimates of apparent affinity are based on the assumption that adipocytes express exclusively on their cell surface, which is certainly not the case in the nonhormonestimulated state (6). Nonetheless, the reported K, has been in the range of 5 mm, in agreement with the present study (Table 11). An accurate measure of the turnover number has been more difficult to comeby. Palfreyman et al. (27) attempted to estimate turnover numbers for the glucose transporter isoforms in 3T3Ll cells by calculating fractional occupancy of and at different substrate concentrations by displacement of ATBBMPA and fitting these data to an equation describing 30MG transport velocity. They obtained turnover numbers slightly greater than those reported herein but were in agreement in the important aspect that there is no difference between and (Table 11). The current experiments, therefore, offer direct support for the proposal that is intrinsically more active than GLUT1; this is solely on the basis of a lower K,, with both isoforms displaying similar turnover numbers. AcknowledgmentsWe thank the Hofmann LaRoche Co. for providing the anti antibody and Drs. G. Holman for helpful discussions and for providing ATBBMPA and JihI. Yeh for critical reading of the manuscript. REFERENCES 1. Cushman, S. W., and Wardzala, L. J. (1980) J. Biol. Chem. 255, Suzuki, K., and Kono, T. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, James, D. E., Brown, R., Navarro, J., and Pilch, P. F. (1988) Nature 333, James, D. E., Strube, M., and Mueckler, M. (1989) Nature 338, Birnbaum, M. J. (1989) Cell 57, Holman, G. D., Kozka, I. J., Clark, A. E., Flower, C. J., Saltis, J., Habberfield, A. D., Simpson, I. A,, and Cushman, S. W. (1990) J. Biol. Chem Bell, d. I., Ka ano, T., Buse, J. B., Burant, C.F., Takeda, J., Lin, D., Fukumoto, 8, and Seino, S. (1990) Diabetes Care 13, Birnbaum, M. J. (1992) Int. Reu. Cytol. 137A, Czech, M. P., Clancy, B. M., Pessino, A., Woon, C.W., and Harrison, S. (1992) Trends Biochem. Sci. 17, Calderhead, D. M., Kita awa, K, Tanner, L. I., Holman, G. D., and Lienhard, G. E. (1990) j., Biol. Chem. 265, Carruthers, A. (1990) Physwl Rev. 70, Whitesell, R. R., and Gliemann, J. (1979) J. BWL Chem. 254, Taylor, L. P., and Holman, G. D. (1981) Biochim. Biophys. Acta 642, 325??E UYU 14. Keller, K., Strube, M., and Mueckler, M. (1989) J. Biol. Chem. 264, Gould, G. W., and Lienhard, G. E. (1989) Biochemistry 28, Gould, G. W., Thomas, H. M., Jess, T. J., andbell, G. I. (1991) Biochemistry 30, Birnbaum, M. J., Haspel, H. C., and Rosen, 0. M. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, Melton, D. A,, Krieg, P. A,, Rebagliati, M. R., Maniatis, T., Zinn, K., and Green, M. R. (1984) Nucleic Aclds Res. 12, Colman, A. (1984) in Expression of Exogenous DNA in Xenopus Oocytes in Transcription and Translation: a Practical Approach (Hames, B. D., and Higgins, S. J., eds) pp. 4970, IRL Press, Oxford 20. Karnieli, E., Zarnowski, M. J., Hissin, P. J., Simpson, I. A., Salans, L. B., and Cushman, S. W. (1981) J. Biol. Chem. 256, Nishimura, H., Saltis, J., Habberfleld, A. D., Garty, N. B., Greenber S., Cushman, S. W., Londos, C., and Simpson, I. A. (1991) Proc. fii: Acad. Sci. U. S. A. 88, Wall, D. A., and Patel, S. (1989) J. Membr. Biol. 107, Haspel, H. C., Birnbaum, M. J., Wilk, E. W., and Rosen, 0. M. (1985) J. Bwl. Chem. 260, Hofstee, B. H. J. (1959) Nature 184, Calderhead, D. M., Kitagawa, K., Lienhard, G. E., and Gould, G. W. (1990) Biochem. J. 269, Katagiri, H., Asano, T., Shibasaki, Y., Lin, J. L., Tsukuda, K., Ishihara, H., Akanuma, Y., Takaku, F., and Oka, Y. (1991) J. Biol. Chem. 266, "" Palfreyman, R. W., Clark, A., E., Denton, R. M., and Holman, G. D. (1992) Biochem. J. 284, Kozka, I. J., Clark, A. E., and Holman, G. D. (1991) J. Biol. Chem. 25, Clancy, B. M., Harrison, S. A., Buxton, J. M., and Czech, M. P. (1991) J. Biol. Chem. 266, Garcia de Herreros, A., and Birnbaum, M. J. (1989) J. Biol. Chem. 264, "." ~ "" 31. Kaestner, K. H., Flores, R. J. R., McLenithan, J. C., Janicot, M., and Lane, M. D. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, Harrison, S. A., Clancy., B. M., Pessino, A., and Czech, M. P. (1992) J. Biol. Chem. 267,37833; Martz, A,, Mookerjee, B. K., and Jung, C. Y. (1986) J. Biol. Chem. 261, 1 " xm1 "?.fins 34. Toyoda, N., Flanagan, J. E., and Kono, T. (1987) J. Biol. Chem. 262, W&h, R. R., Regen, D. M., and Abumrad, N. A. (1989) Biochemistry 28, Vannucci, S. J., Nishimura, H.? Satoh, S., Cushman, S. W., Holman, G. D., and Sim son, I. A. (1992) Bmchem. J. 288, Harrison, ff. A,, Buxton, J. M., and Czech, M. P. (1991) Proc. Natl. Acad. Sci. U. S. A. 88,