Enhanced Peripheral Glucose Utilization in Transgenic Mice ...

THE

JOURNALOF BIOLOGICAL CHEMISTRY

Vol. 269, No. 47, Issue of November 25, pp. 2995629961, 1994 Printed in U.S.A.

0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Enhanced Peripheral Glucose Utilizationin Transgenic Mice Expressing the Human GLUT4 Gene* (Received forpublication, June 27, 1994, and in revised form, August 22, 1994)

Judith L. TreadwayS, Diane M. Hargrove, Nancy A. Nardone, R. Kirk McPherson, Joseph F. Russo, Anthony J. Milicig, Hans A. Stukenbrokg, E. Michael Gibbs, Ralph W. Stevenson, and Jeffrey E. Pessinn From the Departments of Metabolic Diseasesand $Immunologyand Infectious Diseases, Central Research Division, Pfizer Znc., Groton, CT 06340 and the Wepartment of Physiology and Biophysics, University of Zowa, Zowa City, Zowa 52242

Human GLUT4 protein expression in muscle and adipose tissues of transgenic mice decreases plasma insulin and glucoselevels and improvesglucose tolerance comparedwith nontransgenic controls (Liu, M.-L., Gibbs, E. M., McCoid, S . C., Milici, A. J., Stukenbrok, H. A, McPherson, R. K., Treadway, J. L., and Pessin, J. E. (1993)Proc. NutZ. Acad. Sci. U. S. A. 90,11346-11350). We examined the basis of improvedglycemic control in hGLUT4 transgenic mice by determining glucose homeostasis and metabolic profilesin uiuo. Glucose turnover experiments indicated a 1.4-fold greater systemic glucose clearance in hGLUT4 mice relative to controls ( p < 0.05), whereas hepatic glucose production was similar despite 26% lower ( p c 0.05) glucose levels, Glucose infusion rateduring an euglycemic-hyperinsulinemic clamp was 2-fold greater ( p c 0.05) in hGLUT4mice versus controls, and skeletal muscle and heart glycogen content were increased =-fold ( p < 0.05). The increased peripheral glucose clearance in hGLUT4 micewas associated with increased (2532%) basal and insulin-st’1mulated glucose transport rate insoleus muscle ( p < 0.01), and increased muscleplasmamembrane-associated GLUT4 protein. Fed hGLUT4mice displayed 2 0 3 0 % lower plasma glucose and insulin levels ( p < 0.05) and 43% elevated glucagon levels( p c 0.001) compared with controls. Triglycerides,free fatty acids, and P-hydroxybutyrate were elevated 4343% ( p < 0.05) in hGLUT4 mice due to hypoinsulinemia-induced lipolysis. Free fatty acids and P-hydroxybutyratelevels in hGLUT4 mice increased further upon fasting, and skeletal muscle glycogen levelsdecreased markedly comparedwith controls. The data demonstrate that high level expression of hGLUT4 increases systemic glucose clearance and muscle glucoseutilization in vivo and also results in marked compensatory lipolysis and muscle glycogenolysis during a fast.

Deficiencies insulin-regulated in glucose transporter (GLUT4)l protein in skeletal muscle and adipose tissues have been implicated in the defective stimulation of glucose trans-

port by insulin in peripheral tissues of insulinopenic and Type 1diabetic rodents(1-9). The recentisolation and characterization of the human GLUT4 gene (10) has allowed the study of the molecular regulation of GLUT4 expression and pathophysiology in the human disease states of diabetes and insulin resistance. Toward this end, it has been observed that GLUT4 expression in transgenic mice under the control of the aP2 adipose tissue-specific promoter resulted in marked enhancement of insulin-stimulated glucose transportactivity (11). Similarly, a transgenic model has also been employed to determine the physiologic effects of expressing a human GLUT4 minigene under the control of the human GLUT4 promoter (12-14). Transgenic mice carrying 11.5 kilobases of human GLUT4 genomic DNA displayed up to10-fold increased exprestissues sion of GLUT4 in skeletalmuscle, cardiac, and adipose (12). The tissue-specific pattern of transgenic hGLUT4 expression and regulation under variousinsulin-deficient states was identical to the endogenous murine GLUT4 gene (12, 14). In addition, the hGLUT4-11.5 transgenic mice displayed a constitutively high level of basal glucose uptake in adipocytes in vitro, as well as hypoglycemia and insulinopenia relative to nontransgenic controls (13). Furthermore, blood glucose excursion during an oral glucose tolerance test was markedly reduced in the hGLUT4-11.5 transgenic mice, demonstrating the profound effect of the transgeneon glycemic control in vivo (13). To further examine the role of the GLUT4 protein in the regulation of glucose homeostasis invivo, we have performed a detailed physiologic characterization of the hGLUT4-11.5 transgenic mice. In the present study, we have examined the effects of high level hGLUT4 expression on: 1)glucose turnover and peripheral utilization in vivo; 2) glucose transport and metabolism by skeletal muscle; and 3) metabolic and hormonal consequences of the hGLUT4 transgenein fed andfasted states. The results of this study demonstrate that hGLUT411.5 transgenic mice have increasedsystemic glucose clearance and muscle glucose utilization in vivo. In addition, the transgenic mice display marked compensatory lipid mobilizationand muscle glycogenolysis during a fast t o maintainsubstrate availability in lightof the relativehypoglycemia observed.

EXPERIMENTALPROCEDURES Bansgenic Mice-Transgenicmice carrying 11.5kilobases of the * This work wassupported by National Institutes of Health Research human GLUT4 glucosetransporter genomic DNA(hGLUTP11.5) were Grants DK42452, DK44612, andDK25925 (to J. E. P.).The costs of produced as described in Ref. 12. The male and femaletransgenic mice publication of this article were defrayed inpart by the payment of page and age-matched nontransgeniclitter mate controls used in the study charges. This article must therefore be hereby marked “advertisement” were obtained from two hemizygotic transgenic lines(hGLUT4-11.5B in accordance with 18 U.S.C. Section 1734 solely toindicate this fact. 12 and 19 weeks of $ To whom correspondence should be addressed: Dept. of Metabolic and hGLUT4-11.5C) and were analyzed between 9). ThehGLUT4-11.5 transgenic mice &isDiseases, Central Research Div., Pfizer Inc.,Eastern Point Rd., Groton, age(bodyweight18-29 played an approximate 10-fold increased expression of GLUT4 protein CT 06340. The abbreviations used: GLUT4, glucosetransporter 4 (muscle/fat- relative to endogenous murine levels in relevant tissues, e.g. adipose, specific insulin-regulated facilitative glucose transporter isoform);FFA, skeletal muscle, and heart (12). In Vivo Bacer Kinetics and Euglycemic Clamps-Studies were confree fatty acids; p-OHB, P-hydroxybutyrate; R,, rate of glucose appearance; R,,rate of glucose disappearance. ducted in fasted (16 h) male hGLUT4-11.5B mice during anesthesia

29956

Physiologic Profile

of

hGLUT4 Dansgenic

with 1 mlkg intraperitoneal Innovar Vet (Pitman Moore, Mundelein, IL) and 15 mglkg diazepam (Hofkann-La Roche, Nutley,NJ). APE 10 catheter was implanted in theright jugularvein forinfusions. The mice wereplacedon a heating pad for the duration of the infusion experiment. In the tracer kinetic studies. 3-PHlglucose (DuF’ont NENf was injected (0.7 pCi) and continuously infused (1pCih) via the jugular vein catheter. Retro-orbital blood samples (25 pl) were obtained at 60, 90, and 120 min for the determination of specific activity. Blood glucose concentration was determined using a One-Touch I1 glucose meter (Lifescan Inc. Milpitas, CA). Blood [3Hlglucose was measured by scintillation counting using samples deproteinized with ZnSO, and Ba(OH),,followedby evaporation to dryness, and reconstitution in water. Glucose appearance (E,)and disposal (R,) were calculated by the nonsteady state equations of Steele (15). Glucose clearance was calculated by dividing R, by the prevailing glucose concentration. In the euglycemic clamp studies, the mice were primed (13 milliunits) and then continuously infused with insulin (U-100 R porcine/ bovine insulin, Eli Lilly Co. Indianapolis, IN) at a rate of 0.2 milliunitsl min (100 pVh) for 2 h. Blood samples were obtained every 5 min from the tailfor determination of glucose concentration; for this, the tail was cut and then clamped with a small hemostat to prevent bleeding between samples. Blood glucose concentration was measured using a One Touch I1 glucose meter. When blood glucose levels dropped below 100 mgldl, an intravenous infusion of a 25% glucosesolution was initiated. The glucose infusion rate was then adjusted to maintain the blood glucose concentration at 100 mgldl. At the end of the clamp, a 400-pl blood sample was obtained from the retro-orbital sinus for plasma insulin determination. 2-Deozy-o-g~ucoseUptakein Soleus Muscle-Soleus musclesfrom female hGLUT4-11.5B transgenic mice and age-matched nontransgenic controls were isolated and incubated for 15 min at 37 “C in KrebsRinger bicarbonate (KRB: 119 mM NaCl, 4.7 mM KCI, 1.2 mM KH2P0,, pH 7.4, 1.2 m~ MgSO,, 24.9 m~ NaHCO,, and 1.25 mM CaC1,) buffer supplemented with 1.5%bovine serum albumin. Soleus muscles were then preincubated for 15 min at 37 “C in KRB, 1.5%bovine serum albumin plus 2 mx pyruvate, 1 mM sorbitol, and 200 I” insulin as indicated. Muscle glucose uptake was initiated by transfer to media containing 2-deoxy-~-[’~C]glucose (0.2 pCi/ml,1mM), l-[3Hlsorbitol(0.5 pCi/ml, 1mM), and insulin as indicated. The reaction was terminated 30 min later by removal of the muscles, rapid blotting in KRB buffer, and freezing in liquid nitrogen. Muscles were solubilized in 1 N KOH, followedby dual-label liquid scintillation counting. Specific deo oxy-^[‘4Clglucose uptake (pmol/mglmin) wascalculated following correction for the extramyofibrillar trapping of l-[3Hlsorbitol. ltssue GZycogen Determination-Tissue samples were obtained from female hGLUT4-11.5B transgenic and nontransgenic mice in the fed state or following a 16-h fast. Quadriceps, diaphragm, heart, and liver samples were isolated and frozen in liquid nitrogen and thensolubilized in 30% KOH. Glycogenwas precipitated with ethanol and then hydrolyzed to glucose with 5 N HC1. The sample was neutralized and subjected to spectrophotometric determination of glucose via the anthrone method (16).The results are expressed as pmol of glucose equivalent of glycogedg of tissue weight. Immunofluorescence-Wild type or hGLUTP11.5B transgenic mice were fasted overnight or fed ad libitum and then given an intraperitoneal injection of glucose (1gkg) and insulin (8unitskg) asdescribed in Ref. 17. The mice were anesthetized 30 min later and perfused through the left ventricle with 25 ml of 4% formaldehyde plus 0.2% glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.2, at 23 “C. Following fixation in situ, quadriceps muscles were removed and fixed by immersectioning (18)with 4.8 pglml sion for1h and thenprocessed for frozen of rabbit anti-human GLUT4 IgG (19) for 1 h followed by incubation with 5 pglml of donkey anti-rabbit IgG conjugated to Texasred ( Jackson ImmunoResearch Laboratory, Inc.). Exposure-matched fluorescent micrographs were taken on a Nikon FXA microscope. Plasma Analyses”Runk blood fromfedor fasted (16 h) female hGLUT4-11.5B and hGLUT4-11.5C transgenic mice and wild type controls was collectedinto tubes containing sodium fluoride and potassium oxalate and the resultant serum was mixed with aprotinin (0.5 trysin inhibitor unitdml) for analysis of plasma metabolites and hormones or extracted with 6%perchloric acid forlactate analysis. Glucose, triglyceride, and total cholesterol levels were measured using the VP Super System chemistry analyzer (Abbott Laboratories). Lactate and P-OHB concentrations were determined spectrophotometricallyusing kits from Sigma. FFA concentration was determined utilizing a kit from Amano International Enzyme Co., Inc. Insulin and glucagon levels weredetermined by radioimmunoassay using kits fromBinax and Amersham

29957

Mice

TABLEI Whole body glucose turnover in hGLUT4-11.5B transgenic fGLUT4-B) and nontransgenic wild type mice Fasted hGLUT4-11.5B and wild type male mice were subjected to a continuous intravenous infusion of 3-[3Hlglucosefor120min under anesthesia in order to determine plasma glucoseconcentration (Glc),R,, and R, in vivo. Glucose clearance rate was calculated as the rate of glucose disposal relative to ambient glucose concentration (R,IGlc). Values presented are the mean k S.E. of n = 13-14 observationdgroup. Glucose Homeostasis in hGLUT4-11.5BMale Mice

Re

Rd

mglkglmin mglkglmin mg

Wild type 8.7 k 0.7 8.9 GLUT4-B 9.6 ? 0.9

Glc %

0.7 9.3 k 0.9 82 5

111k 8 k 8”

Glc Clearance mllminlkg

7.0 5 0.5 10.0 -C 1.2’

Indicates significant difference ( p < 0.05) in comparison to agematched wild type litter mates. Corp., respectively. Statistical analyses-Statistical comparisonsbetween and within groupsweremadewith Student’s t test or analysis of variance with Student-Newman-Kedsanalysis. Data are presented as the mean k S.E. RESULTS

Dacer Kinetic Studies-To determine the effect of hGLUT4 expression on systemic glucose homeostasis in vivo, glucose turnover was assessed by a tracer kinetic technique in fasted male hGLUT4-11.5B mice and nontransgenic controls (Table I). As observed previously (131, fasting glucose concentration was 26% lower ( p < 0.05) in hGLUT4-11.5B transgenic mice relative t o controls. Analysis of [3Hlglucose tracer kinetics in vivo indicated that the rateof hepatic glucose production (R,) and the rate of systemic glucose disposal (R,) did not significantly differ between the hGLUTP11.5B and wild type mice (Table I). However, the glucose clearance rate was 1.4-fold greater ( p < 0.05) in thehGLUT4-11.5B transgenic mice compared with the nontransgenic controls. Thus, hGLUT4-11.5 transgenic mice displayed enhanced systemic glucose clearance presumably due to increased peripheral glucose utilization. Since fasting hepatic glucose production was unaltered in the hGLUT4 transgenic mice, the hypoglycemic state observed relative to thewild type control mice is directly accounted for by an increased peripheral clearance of glucose. It is not apparent why R, was unchanged despite mild hypoglycemia in the hGLUT4-11.5 mice, but one possibility is a reduction in the availability of lactate asa gluconeogenic substrate in the transgenic mice in the fasteduersus the fed state (seeTable 111). Euglycemic-HyperinsulinemicClamp-To determine thebasis of the increased glucose clearance rate observed in the hGLUT4 transgenic mice, a euglycemic-hyperinsulinemic clamp procedure was performed to directly assess peripheral glucose utilization in uiuo. Thestandard euglycemic-hyperinsulinemic clamp procedure, originally designed for rats (e.g. >200 g), was modified for smaller rodents (e.g.-25 g) as described under “Experimental Procedures” and recently in Ref. 20. The nontransgenic and hGLUT4-11.5B transgenic mice subjected to theeuglycemic-hyperinsulinemicclamp procedure achieved steady statelevels of glucose infusion during the first 60 min of the study (Fig. 1,A and B).Mice were then maintained at steady statefor an additional60 min t o assess differences between the two groups. The glucose infusion rate required to maintain blood glucose at -100 mg/dl during the last 60 min of the clamp was 2-fold higher ( p < 0.05) in the hGLUT4-11.5B transgenic mice relative to the nontransgenic controls (Fig. 1 B ) . Plasma insulin concentrations at the end of the glucose/insulininfusion procedure were 292 29 and 215 2 21 microunitdm1 in the control and transgenic mice, respectively. These datadirectly demonstrate an increased glucose clearance rate in the transgenic mice despite slightly lower insulin exposure compared with nontransgenic animals.

29958

Physiologic Profile of hGLUT4 Pansgenic Mice TABLEI1 Comparison of muscle and liver glycogen content between hGLUT4-11.5B transgenic (GLUT4-B)and nontransgenic wild type mice in fed and fasted (16 h) states Quadriceps, diaphragm, heart, and liver samples fromhGLUT4-B and wild type female mice were isolated and frozen in liquid nitrogen, and glycogen content was determined as indicated under "Experimental Procedures." Values are the mean f S.E. for n = 3-13 observations per group. Fed

pmol lg 0

2

m

0

Fasted

GLUT4-B

type Wild

Wild type

GLUT4-B

pmollg

Diaphragm 7.1 -c 1.3 27.5 f 5.8" 2.6 f 0.6 0.8 +0.2*,' 7.2 2 0.7 22.0 f 2.7" 2.4 0.2d 4.0 f 0.3a,d Quadriceps Heart 6.2 f 0.8 31.8 f2.7' 6.8 f 1.4 23.0 f4.8' Liver 406.7 f 24.2395.5 f 24.035.7 -c 7.8d20.2 f 2.gd

3 60

30

Time

90

(min)

120

a Indicates significant difference ( p < 0.01) comparedwithagematched wild type litter mates. *Indicates significant difference ( p < 0.05) compared with agematched wild type litter mates. ' Indicates significant difference ( p < 0.05) between fed and fasted groups. Indicates significant difference ( p