Myocardial glucose transporters and glycolytic metabolism during ischemia in hyperglycemic diabetic swine

Myocardial William

Glucose

C. Stanley,

Jennifer

Transporters and Glycolytic Hyperglycemic Diabetic L. Hail, Kevin R. Smith,

Gregory

Metabolism Swine

D. Car-tee, Timothy

During

A. Hacker,

Ischemia

and Judith

in

A. Wisneski

We assessed the effects of 4 weeks of streptozocin-induced diabetes on regional myocardial glycolytic metabolism during ischemia in anesthetized open-chest domestic swine. Diabetic animals were hyperglycemic (12.0 * 2.1 v 6.6 2 .5 mmol/L), and had lower fasting insulin levels (27 f 8 Y 79 2 18 pmol/L). Myocardial glycolytic metabolism was studied with coronary flow controlled by an extracorporeal perfusion circuit. Left anterior descending coronary artery (LAD) flow was decreased by 50% for 45 minutes and left circumflex (CFX) flow was constant. Myocardial glucose uptake and extraction were measured with o-[6-3H]-2-deoxyglucose (DG) and myocardial blood flow was measured with microspheres. The rate of glucose conversion to lactate and lactate uptake and output were assessed with a continuous infusion of [6-9Jglucose and [U-13C]lactate into the coronary perfusion circuit. Both diabetic and nondiabetic animals had sharp decreases in subendocardial blood flow during ischemia (from 1.21 f .I0 to 0.43 + .08 mL . g-l * min-’ in the nondiabetic group, and from 1.30 + .I5 to 0.55 2 .I1 in the diabetic group). Diabetes had no significant effect on myocardial glucose uptake or glucose conversion to lactate under either well-perfused or ischemic conditions. Forty-five minutes of ischemia resulted in significant glycogen depletion in the subendocardium in both nondiabetic and diabetic animals, with no differences between the two groups. Glycolytic metabolism is not impaired in hyperglycemic diabetic swine after 1 month of the disease when compared with that in normoglycemic nondiabetic animals. The myocardial content of the insulin-regulatable glucose transporter (GLUT 4) was measured in left ventricular biopsies. Diabetes resulted in significantly lower GLUT 4 levels in both the subepicardial and subendocardial layers (23% [P < .Ol] and 11% [P < .05], respectively). There were no subepicardial/subendocardial differences in GLUT 4 levels in either group. In conclusion, 1 month of streptozocin diabetes resulted in a modest decrease in myocardial GLUT 4 levels, but did not affect the rate of glycolysis during myocardial ischemia. Copyright o 1994 by W. B. Saunders Company

D

IABETIC PATIENTS have a greater risk for fatal myocardial infarction than nondiabetic patients. The increased death rate from heart disease in diabetic patients is due to a greater occurrence of coronary artery diseasele5 and dn impaired recovery of the myocardium from ischemic ecents.‘-h The Framingham study reported that the ageadjusted incidence of death from cardiovascular disease is twice as high in diabetic men and four to five times as high in diabetic women.’ Kuller et al7 recently reported that the percentage of diabetics among coronary heart disease deaths in men between 35 and 44 years of age increased from 6.5% in 1970 to 1972 to 23.0% in 1985 to 1986. Clearly the rates of mortality and morbidity associated with an acute myocardial infarction are significantly increased in the diabetic population. Measurements of myocardial glucose uptake in hyperglycemic human diabetic patients showed that the arterialvenous glucose difference is either normaV9 or impairedlo when compared with that in normoglycemic healthy people. Studies on the isolated perfused rat heart demonstrated that diabetic hearts have lower rates of glucose uptake than nondiabetic hearts under well-perfused conditions, with anoxia, and in response to insulin stimulation.11J2 Furthermore. diabetic dogs studied 4 to 6 weeks after injection of streptozocin have myocardial insulin resistance.13 Thus, diabstic myocardium appears to have an impaired ability to increase glucose uptake in response to either insulin or oxygen deficiency. The metabolic response to acute myocardial ischemia in diabetic patients is not well understood. In the nondiabetic heart, acute ischemia results in an accelerated rate of anaerobic glycolysis, as seen in a switch from net lactate uptake to myocardial lactate production and net glycogen breakdown.14J5 Myocardial glycolytic metabolism is not uniform across the left ventricular wall during reductions in Metabolism,

Vol43,

No

1 (January).1994:

~~61.69

coronary artery blood flow. Studies using radiolabeled microspheres have shown that acute myocardial ischemia results in a nonuniform blood flow distribution in the left ventricular wall.14~1sUnder conditions of normal coronary perfusion, subendocardial blood flow is similar to subepicardial blood flow. A significant decrease in coronary artery blood flow (40% to 50% of normal) results in little change in subepicardial blood flow, but will cause marked decreases (60% to 80%) in subendocardial blood flow. This results in a higher rate of glycogen breakdown in the subendocardium.ls We recently introduced a novel method for assessing subendocardial glucose uptake and arterialvenous glucose difference using deoxyglucose tracers and microspheres.ls We observed that a 60% reduction in left anterior descending coronary artery (LAD) blood flow in swine resulted in an 80% decrease in subendocardial blood flow with only a modest 40% decrease in subepicardial blood flow. Myocardial glucose uptake was not significantly

From the Biodynamics Laboratoy and the Section of Cardioloa, University of Wisconsin, Madtion, WI; the Department of Medicine, University of California, San Francisco; and the Veterans Affairs Medical Center, San Francisco. CA. Submitted September 28, 1992; accepted February I, 1993. Supported by the University of Wisconsin Graduate School, the American Hean Association-Wisconsin Ajiliute (W. C.S.), the Juvenile Diabetes Foundation International (W. C.S.), National Institutes of Health Award No. HL-47094 (W.C.S.), and the Medical Research Service of the Department of Veterans Aflairs, San Francisco (J.A. W.). Present address: W.C.S., Institute of Pharmacology (R2-IOI), Syntex Research, 3401 Hillview Ave, Palo Alto, CA 94303. Address reprint requests to William C. Stanley, PhD, Institute of Pharmacology (R2-IOI), Syntex Research, 3401 Hillview Ave. Palo Alto, CA 94303. Copyright 0 1994 by W.B. Saunders Cornpan?, 0026.049519414301~OOiO$O3.OOjO

61

62

STANLEY

increased by &hernia. However, ischemia caused a significant 13-fold increase in the arterial-venous glucose difference in the subendocardial layer; this far exceeded the threefold increase observed in the less ischemic subepicardial layer of the LAD perfusion bed. Glucose enters cells by facilitated diffusion via a glucose transporter (GLUT) protein. A family of GLUTS exists with at least five isoforms currently identified.l’j GLUT 4 is the primary isoform expressed in cardiac muscle.17 Camps et alI8 recently observed a decrease of 21% in the GLUT 4 protein level in the whole heart of 7-day streptozocindiabetic rats. To our knowledge, regional GLUT 4 levels have not been assessed in a large animal model. It is not known whether diabetic myocardium is capable of maintaining a high rate of glycolysis during ischemia. The purposes of this study were to assess the effects of 1 month of diabetes in swine on (1) the GLUT 4 protein content in the myocardium; (2) the rate of exogenous glucose uptake and extraction in the subepicardial, midmyocardial, and subendocardial layers of well-perfused and acutely ischemit myocardium; (3) the rate of glucose conversion to lactate and the rates of lactate uptake and output; and (4) glycogenolysis during ischemia. In this initial investigation, we elected to study the animals under their ambient metabolic conditions to simulate clinical conditions. Thus, hyperglycemic diabetic swine were compared with a group of healthy normoglycemic animals. MATERIALS

AND METHODS

Studies were performed on female domestic pigs. Two groups of animals were investigated, nondiabetic pigs aged 16.8 rt .4 weeks (n = 7) and diabetic pigs aged 17.6 f .4 weeks (n = 5). All animals were from the same vendor (Lonny Han, DeForest, WI). Animals in the diabetic group were received from the vendor at 11 to 12 weeks of age, and the nondiabetic group arrived 4 to 7 days before the terminal study. Diabetic Model We induced diabetes with streptozocin using methods modified from Wilson et al.19 First, animals were instrumented with an indwelling jugular venous catheter for infusion of streptozocin and monitoring of blood glucose levels. Pigs were premeditated with ketamine (10 mgikg subcutaneously [SC]), intubated, ventilated, and maintained with halothane (0.4% to 0.8%). Using sterile technique, a polyethylene cannula (6 French) was inserted into a jugular vein and tunneled under the skin to exit on the midline on the back of the neck. Porcine penicillin (500,000 IU) and streptomycin (500 mg) were administered intramuscularly immediately after surgery. The catheter was filled with saline containing heparin to prevent clotting, and was flushed daily. Three to 7 days after the surgery, animals received an intravenous (IV) injection of sterile, freshly prepared streptozocin (100 mg/kg) in citrate buffer (Zanosar, Upjohn, Kalamazoo, MI). This dose has been shown to induce basal hyperglycemia and remove the insulin response to a glucose load in swine.M Animals were fasted for 18 hours before the injection. Blood glucose level was measured immediately before the streptozocin injection, 4 to 7 hours postinjection, and 6 days per week in the morning with the animal in the fasting state. Blood glucose concentration was assayed using a hand-held clinical glucose analyzer (Ames Glucometer 3).

ET AL

Myocardial Metabolism Protocol Experiments were performed with the animals under general anesthesia, using an open-chest preparation. Overnight-fasted domestic swine were premeditated with ketamine (10 mg/kg SC) and anesthetized with halothane (4 to 8%). A tracheostomy was performed under deep general anesthesia, and the animals were ventilated to maintain arterial blood gases in the normal range (Paoz > 100 mm Hg, PacoZ 35 to 45 mm Hg, and pH 7.35 to 7.45); anesthesia was maintained with halothane (1.0 to 1.5%). The heart was exposed as previously described.15z21 Briefly, a transthoracotomy with bilateral rib resections was performed for wide exposure of the heart. Blood pressure was monitored in the aorta and left ventricle using a manometer-tipped pressure device (Millar Instruments). A venous sampling catheter was inserted in the left anterior intewentricular vein, which drains the perfusion territory of the LAD. The anterior inter-ventricular vein effluent was allowed to drain into the chest cavity. Blood from the chest cavity was continuously pumped into a reservoir, and then reinfused into the right femoral vein. The blood reinfusion rate was adjusted to give a left ventricular peak-systolic pressure of approximately 80 to 90 mm Hg. The animal was heparinized (20,000 U heparin IV bolus, followed by 10,000 U/h). Coronary blood flow was controlled through an extracorporeal circuit as previously described. 15,21Briefly, blood for the coronary perfusion circuit was withdrawn from the right femoral artery, passed through a mixing chamber, divided among three independently controlled perfusion pumps, filtered through blood filters (40 km), and infused into the left main, right, and LAD coronary arteries. Perfusion pressures were measured in each perfusion line and calibrated to accurately measure coronary artery pressure. The LAD perfusion pump flow was adjusted so that the interventricular vein coronary venous oxygen saturation was approximately 40%. Circumflex and right coronary artery flows were adjusted to approximately match mean aortic pressure. Ischemia was induced in the LAD perfusion bed by decreasing the flow on the LAD perfusion pump to 50% of the control value. The perfusion system allowed us to control the LAD blood flow independently of the left circumflex (CFX) and right coronary artery flows, so that ischemia in the LAD perfusion bed was induced without affecting blood flow to the CFX. Blood samples were drawn from the coronary artery perfusion line and the anterior interventricular vein for determination of the arterialvenous difference for oxygen, glucose, lactate, and free fatty acids at -10, -2,20,25, and 30 minutes of ischemia (Fig 1). Renstrom et al** showed in the same preparation with indocyanine green indicator that blood drained through this vein was primarily from

nm: Art.andveln ampbe:

?lztzii 0 t

_h)rim_

BlOPSY

303530

tt

45

t

Fig 1. Sampling protocol. At time 0 the blood flow on the LAD perfusion line was decreased by 50%. Arterial blood samples were drawn from the coronary perfusion circuit, and vein samples were taken from the anterior interventricular vein. [S-W]Glucose and [U-W]lactate were continuously infused into the mixing chamber of the coronary perfusion circuit from 40 minutes before the onset of ischemia through 30 minutes of ischemia.

GLYCOLYTIC

METABOLISM

63

IN THE DIABETIC HEART

the LAD perfusion bed during aerobic conditions (91% 2 1%) and with a 60% decrease in LAD blood flow (78% 2 2%). At time 0. the flow rate on the LAD perfusion pump was decreased by 50%; the total duration of ischemia was 45 minutes. After 45 minutes of &hernia, two sections of myocardium (-5 g each) were rapidly removed from the center of the LAD and CFX perfusion areas. The tissue samples were immediately sliced into three sections of approximately equal thickness (subendocardial, midmyocardial, and subepicardial). frozen in liquid nitrogen, and stored at -70°C until analysis. The rates of glucose uptake in the subepicardial, midmyocardial, and subendocardial layers were assessed using the deoxyglucose method, as previously described.‘” After 20 minutes of ischemia. we began an infusion of o-[6-3H]-2-deoxyglucose ([DG] American Radiochemical, St Louis, MO) for the measurement of regional myocardial glucose uptake. DG (1.3 PCiimin) was infused for 10 minutes into the mixing chamber of the coronary perfusion circuit upstream of the perfusion pumps, so that both LAD and CFX perfusion beds received the same specific radioactivity of tracer. Timed blood samples (1.0 mL) were withdrawn from the LAD perfusion line at 0.5, 1, 2,3,4, 6,8, and 10 minutes of DG infusion for determination of plasma glucose and DG concentration (Fig 1). The DG infusion was discontinued after 10 minutes, and additional coronary arterial blood samples were taken at 0.5, 1,2,3,4,5,7,10, and 15 minutes after the end of DG infusion. The metabolic fate of glucose and lactate extraction and release were assessed using the dual carbon-labeled isotope method of Wisneski et a1.14.23.Z4[6-14C]GI ucose (American Radiochemical) and [U-‘3C]lactate (Merck. Rahway, NJ) were infused into the mixing chamber of the coronary perfusion circuit beginning 40 minutes before the onset of ischemia at rates of 0.14 @/min and 17 ~molimin, respectively. Samples for measurement of arterial and venous glucose and lactate specific activities, [‘3C]lactate enrichment, and plasma free fatty acid and insulin levels were taken at - 10. -2.20.25, and 30 minutes of ischemia (Fig 1). Regional blood flow was measured using the radioactive microsphere method, with injections made at -6 and 26 minutes of &hernia. Radiolabeled spheres (15~pm diameter, New England Nuclear, Boston, MA) were injected over a 15-second period into the mixing chamber of the extracorporeal perfusion circuit. 141Ce-, ““Ru-, 4hS~-, and “‘Sn-labeled microspheres were used, and the order of injection was alternated between animals. Six and 12 p,Ci were injected during control and ischemia periods, respectively. A reference sample for the calculation of blood flow was withdrawn from the right coronary artery perfusion line at a rate of 1.5 ml’min for 3 minutes after the microsphere injection using a syringe withdrawal pump (Haward Apparatus). Tissue samples ( - 0.8 g each) for microsphere flow determinations were taken in triplicate from the region immediately surrounding the biopsies. In all ,:ases, the region circumventing the LAD biopsy had a decreased flow. and thus the biopsy used for analysis of deoxyglucose in the LAD hed was ischemic.

concentration was assayed in duplicate using a radioimmunoassay kit (Diagnostic Products, Los Angeles, CA).23 Plasma activities of DG were measured on 0.2-mL aliquots of plasma and plasma glucose using a hexokinase-glucose 6-phosphate-based assay (Sigma Chemical, St Louis, MO). [U-‘3C]Lactate was assayed using the method of Tserng et al.25 as previously modified by Wisneski et a1,21 using a gas chromatograph (model 5880, Hewlett Packard) interfaced to a mass spectrometer (model 597114, Hewlett Packard). 14C-Glucose and 14C-lactate specific radioactivities were assayed using ion-exchange chromatography and liquid scintillation counting, as previously described.24 Tissue DG 6-phosphate (DG 6-P) concentration was measured on approximately 300 mg tissue. Frozen tissue samples were crushed in a stainless steel tissue pulverizer cooled in liquid nitrogen. Samples were homogenized at 4°C in IO vol 1 mol/L perchloric acid in a glass tissue grinder. The sample was centrifuged, and the supernatant was neutralized with K2CO3 and applied to an anion-exchange column (Dowex 1, acetate form) to separate DG from DG 6-P.” DG was eluted with water, and DG 6-P with 1 mol/L HCI. Tissue glycogen was assayed on perchloric acid extracts using the amyloglucosidase method as described by Passoneau and Lauderdale.26 Myocardial GLUT 4 concentration was quantified in the subepicardial and subendocardial biopsies from the CFX perfusion bed. Aliquots of HES (20 mmol/L Hepes, 1 mmol/L EDTA, 250 mmol/L sucrose, pH 7.4) homogenate were diluted with Laemmli sample buffer. Thirty to 55 mg myocardial tissue was weighed, diluted 39-fold with ice-cold HES buffer, and homogenized on ice. Protein concentration was assessed spectrophotometrically with the Bicinchoninic acid (BCA) assay (Pierce Chemical, Rockford. IL). Triplicate aliquots of this mixture containing 30 pg protein were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a 10% resolving gel. Proteins were electrophoretically transferred to 0.45+m Immobilon-P membranes (Millipore, Bedford, MA). Membranes were blocked overnight at 4°C with TBST (50 mmol/L Tris, 200 mmol/L NaCl, 0.025% Tween 20. 0.02% Na azide, pH 7.5) containing 5% Carnation nonfat dry milk. Immobilon sheets were then incubated for 60 minutes at 37°C in TBST containing 1% milk and a 250-fold dilution of R820 anti-GLUT 4 polyclonal antisera (East-Acres Biologicals, Southbridge, MA). Membranes were subsequently rinsed twice at room temperature (RT) with TBST + 1% milk, followed by a ho-minute incubation at RT with TBST + 1% milk including 3 FCi ‘251-goat anti-rabbit IgG (ICN, Irvine, CA). The membranes were washed 3 X 15 minutes at RT in TBST + 1% milk, air-dried, and exposed to Kodak XAR-5 film (Eastman Kodak, Rochester, NY) at -70°C. The visualized bands were traced, excised, and counted in a gamma counter. Results were corrected for background determined in apparently unlabeled areas. Values are expressed relative to an internal control (myocardial tissue from Yucatan mini-pig), All myocardial homogenates and internal controls were run in triplicate.

Calculations Anulytical Methods Arterial and venous pH, PCOZ, and POT were determined on a blood gas analyzer (Radiometer ABL2), and hemoglobin concentratlon and saturation on a saturation meter (Radiometer OMI). Blood samples for glucose and lactate analysis were immediately deproteinized in ice-cold 1 mol/L perchloric acid (1:2 vol/vol), weighed, centrifuged, and stored at -70” until analyzed in triplicate for glucose and lactate using previously described enzymatic spectrophotometric methods. 23.Z4Plasma free fatty acid levels were measured in duplicate using a commercial enzymatic spectrophotometric kit (Wake Chemicals, Richmond, VA). Plasma insulin

The rate of myocardial glucose uptake (Rs) was calculated tissue concentration of DG 6-P (dpmig) divided by the area the arterial plasma DG specific radioactivity (DG to glucose versus time curve during and following the period of infusion.

R,(ymol g-’

min-‘)

= tissue DG 6-P concentration The integral

as the under ratio) tracer

IS2’:DG specific

of the DG specific activity curve (dpm

min

activity.

. Arnold’)

64

STANLEY ET AL

was taken from the onset of DG infusion (time 0) to the time of the tissue biopsy (25 minutes after the onset of DG infusion) as previously described.15 Regional myocardial blood flow was calculated from the microsphere data using the reference sample withdrawal method of Buckberg et al.27 Regional myocardial arterial-venous glucose

concentration difference (mmol/L was calculated from the DGmeasured Rg and the microsphere blood flow data, ie, arterialvenous glucose = R,/myocardial blood flow. The percent of the arterial glucose extracted by a given region was calculated as (arterial-venous glucose difference x lOO)/arterial glucose concentration). The net extraction of glucose, lactate, and free fatty acids for the LAD perfusion bed was taken as the arterial - venous concentration difference. The rate of net uptake by the LAD bed (mL . g-l min-‘) was calculated as the product of the net arterial - venous concentration difference and transmural LAD perfusion bed blood flow as measured using radiolabeled microspheres. The rate of exogenous glucose converted to lactate by the myocardium was calculated as previously describedz4 from the appearance of secondarily labeled 14C-lactate in the interventricular venous blood. The tracer-measured rate of lactate uptake was assessed from the myocardial uptake of isotopically labeled 13Clactate tracer.24 Tracer-measured

rate of lactate uptake = blood flow

x a lactate cOnc x (a 13C-lactate cone - Y 13C-lactate cone) a 13C-lactate cone where cone is concentration, and a is arterial and v is interventricular venous blood. Tracer-measured lactate output was taken as the difference between tracer-measured lactate uptake and net lactate uptake.24 Statistics Values are presented as the mean 2 standard error. A paired t test was used to compare the average of the two control period samples with the average of the 20-, 25, and 30-minute samples during the ischemic period. Diabetic and nondiabetic values were compared using unpaired t tests. Within-group comparisons in regional myocardial blood flow, glycogen concentration, glucose

uptake, arterial-venous glucose difference, and percent glucose extraction for the LAD and CFX beds were made using a one-way

ANOVA and the Bonferroni post hoc test for multiple comparisons.

RESULTS

Diabetic Model No major complications were encountered with the induction of diabetes or with the maintenance of these animals. The fasting venous glucose concentration increased from 4.8 t .2 mmol/L immediately before the streptozocin injection to 12.5 ? 2.9, 13.3 ? 1.5, 15.9 ? 2.1, 14.7 ? 1.6, and 14.2 ? 1.8 mmol/L at 24 hours and 1, 2, 3, and 4 weeks after the injection, respectively; glucose levels were significantly higher (P < .OS) than preinjection values during weeks 1 through 4. Overnight fasting plasma insulin values during the terminal surgery were significantly lower than in control animals (Table 2). Body weight was not significantly affected by diabetes (45.6 * 3.4 and 50.3 2 3.3 kg for diabetic and nondiabetic groups, respectively). Left Ventricular Function

The decrease in LAD blood flow resulted in significant reductions in the rate of myocardial oxygen consumption in the LAD perfusion bed in nondiabetic and diabetic animals, respectively (Table 1). No significant differences were observed between diabetic and nondiabetic groups in myocardial oxygen consumption, heart rate, left ventricular end diastolic pressures, or peak left ventricular dp/dt under either aerobic or ischemic conditions (Table 1). The peak systolic pressure we observed is lower than that observed in awake swine (unpublished observation), but is typical for this preparation.15J1J2 Left ventricular peak systolic pressure was higher in the diabetic group during the aerobic period; however, this difference was not significant during the ischemic period. Ischemia did not affect left ventricular end diastolic or peak systolic pressure in either group, but peak left ventricular DP/DT decreased significantly from aerobic to ischemic conditions in both diabetic and nondiabetic groups (P < .04; Table 1). Heart rate increased

Table 1. Hemodynamic Data (mean + SE) P (aerobic v Aerobic Period

bed myocardial O2 consumption (mL Nondiabetic

LAD

lschemic Period

ischemic period)

g-’ . min-l)

Diabetic Heart rate (beatslmin) Nondiabetic Diabetic Left ventricular end diastolic pressure (mm Hg) Nondiabetic Diabetic Left ventricular peak systolic pressure (mm Hg) Nondiabetic Diabetic Peak left ventricular dpldt (mm Hg/s) Nondiabetic Diabetic *Significantly different from nondiabetic group, P < .05.

0.105 * ,010

,054 + 0.022

,001

.I24 2 ,017

0.061 2 0.007

,003

134.5 + 12.2

145.4 + 13.5

.02

154.7 t 18.2

149.0 + 9.4

NS

5.6 + 0.7

9.9 + 3.5

NS

3.1 * 1.0

7.0 + 3.7

NS

80.3 + 3.6

78.8 2 3.8

NS

93.7 * 2.91

88.5 + 3.5

NS

2,004 + 224

1,659 + 239

.04

2,438 ? 283

1,791 + 149

.04

GLYCOLYTIC METABOLISM

65

IN THE DIABETIC HEART

significantly from aerobic to ischemic conditions nondiabetic group (P < .02: Table 1).

in the

Regional Myocardial Blood Flow No significant differences were observed between nondiabetlc and diabetic groups in regional myocardial blood flow (Fig 2). As expected, ischemia in the LAD perfusion bed resulted in no significant change in blood flow in the CFX percusion bed, but caused significant decreases in blood flow in the LAD bed. The most pronounced decrement occurred in the subendocardial region (Fig 2). Both nondiabetic and diabetic animals had significantly lower blood flow in the ischemic LAD bed in all three myocardial layers whr n compared with the aerobic CFX bed.

GLlcose and Lactate Exchange in the LAD Pe@sion Bed Arterial glucose concentrations were approximately twice as high in the diabetic animals than in the nondiabetic group (Table 2). The diabetic group did not differ significandy from the control group in glucose exchange in the LAD perfusion bed. Ischemia resulted in a significant

NONDIABETIC 0

AEROBIC

PERIOD

q ISCHEMIC

PERIOD

2 T

EPI

MID

END0

LAD

I EPI

MID

END0

CFX

3

DlABElTC

n AEROBIC 1 q ISCHEMIC

PERIOD PERIOD

_ T

increase in the arterial-interventricular venous glucose difference in both groups (P < .02; Table 2). The transmural rate of glucose uptake in the LAD bed increased in both groups, but was not statistically significant (Table 2). Glucose conversion to lactate was calculated from the appearance of secondarily labeled 14C-lactate in the venous effluent from the LAD perfusion bed. No differences were observed between nondiabetic and diabetic groups in either aerobic or ischemic conditions. However, ischemia resulted in a significant increase in the rate of glucose conversion to lactate in both groups (Table 2). The arterial-venous lactate difference and the rate of net lactate uptake during the aerobic period were both greater in nondiabetic animals (Table 2). No differences between the two groups were observed during ischemia. Ischemia resulted in an increased rate of anaerobic glycolysis in the myocardium, as seen in the switch from net lactate uptake during the control period to net lactate production during ischemia (Table 2). Lactate uptake was also assessed using isotopically labeled Y-lactate tracer. There were no significant differences between groups in tracer-measured lactate uptake. Tracer-measured lactate uptake decreased from control to ischemic conditions (P < .Ol; Table 2); however, this decrease was not significant in the nondiabetic group. Tracer-measured lactate output was taken as the difference between tracer-measured lactate uptake and net lactate uptake. There were no differences in tracer-measured lactate output between diabetic and nondiabetic groups under either condition. Ischemia resulted in significant increases in tracer-measured lactate release in both groups (Table 2). We observed high lactate levels in both groups (Table 2), which could artificially cause high rates of lactate uptake and decrease glucose uptake. 24We recently observed similar lactate values in nondiabetic animals using the same preparation.15 Guth et all4 reported values that were 60% to 70% lower using a less extensive surgical preparation. The increased lactate levels that we observed may be the result of surgical trauma and/or greater peripheral hypoperfusion.

T Regional Glucose Metabolism Assessed With DG

Ischemia resulted in a significant increase in myocardial glucose uptake, arterial-venous glucose difference, and percent glucose extraction in the subendocardial layer of the LAD bed relative to the subendocardial layer of the nonischemic CFX bed (Fig 3). No significant differences were observed between diabetic and nondiabetic groups in any of the three parameters. ”

EPI

MID

LAD

END0

EPI

MID

END0

CFX

Fig 2. Myocardial blood flow during the aerobic and ischemic periods for the subepicardial (EPI), midmyocardial (MID), and subendocardial (ENDO) layers of the LAD and CFX coronary artery perfusion beds. The LAD perfusion bed was subjected to ischemia, and the CFX bed had normal blood flow. No significant differences were observed between diabetic and nondiabetic groups. *P c .005, V < ,001, VJ < ,005, V c .OOOl, significantly different from the aerobic period value in the same layer.

Myocardial Glvcogen Concentration

Regional glycogen concentrations were not different between diabetic and nondiabetic groups (Fig 4). Glycogen levels in the ischemic LAD bed were significantly lower than in the corresponding region in the CFX bed. Furthermore, there was a gradient in the concentration of glycogen in the LAD bed, with the lowest levels being found in the more ischemic subendocardial layer (Fig 4). Thus, the net

STANLEY ET AL

66

Table 2. Glucose and Lactate Metabolism in the LAD Perfusion Bed (mean 2 SE) P (aerobicv ischemicperiod)

Aerobic Period

lschemicPeriod

Nondiabetic

79 + 19

78 + 28

NS

Diabetic

27 + a*

37 * 11

NS

5.65 + A7

NS

11.67 t 2.13$

NS

Arterial insulin (pmol/L)

Arterial glucose (mmol/L) Nondiabetic

6.65 + 50

Diabetic

12.0 ? 2.1t

Arterial-venous glucose difference (mmol/L) Nondiabetic

0.07 + .02

0.52 2 0.17

.02

Diabetic

0.07 ? 0.06

0.43 2 0.13

.02

Nondiabetic

0.07 2 .03

0.26 ir .12

NS

Diabetic

0.13 + .09

0.29 ‘- .09

NS

Nondiabetic

0.02 + .Ol

0.22 2 .07

.02

Diabetic

0.04 + .02

0.18 + .06

.04

Nondiabetic

6.19 + .68

6.03 2 .59

NS

Diabetic

3.96 +

.ai

4.11 ? .J6

NS

Glucose uptake (bmol . 9-l.

min-‘1

Rate of glucose conversion to lactate (Pm01 . g-l

min-‘)

Arterial lactate (mmol/L)

Arterial-venous lactate difference (mmol/L) Nondiabetic

0.53 + .ll

-0.79

t .25

,003

Diabetic

0.18 k .OJ*

-0.61

2 .23

.Ol

Nondiabetic

0.78 k .I 7

-0.44

-t .16

,004

Diabetic

0.30 + .11*

-0.40

+ .14

.Ol .Ol

Net lactate uptake (Pmol

Tracer-measured

g-l

min-I)

lactate uptake (pmol

g-’

min-1)

Nondiabetic

0.96 + .19

0.33 + .05

Diabetic

0.55 + .25

0.29 + .04

Nondiabetic

0.21 * .05

0.77 A .20

.02

Diabetic

0.26 ? .14

0.70 k .13

.03

Tracer-measured

lactate output (pmol . g-l

NS

min-‘)

*Significantly differentfrom

nondiabetic group,P

< .05.

tsignificantly differentfrom

nondiabetic group,P

< .Ol.

*Significantly different from nondiabetic group, P < .005.

glycogenolytic rate was proportional ischemia in both groups.

to the severity of

Free Fatty Acid Metabolism

Diabetic animals had significantly higher arterial plasma free fatty acid levels during the aerobic period and a higher arterial-venous difference and net rate of free fatty acid uptake during the ischemic period compared with the nondiabetic group (Table 3). Ischemia resulted in a decrease in the rate of net free fatty acid uptake; however, the difference was only significant in the nondiabetic group. Myocardial GLUT 4 Levels

Total GLUT 4 concentration was expressed as a percent of an internal control run on each gel. GLUT 4 concentrations in biopsies from the nonischemic CFX bed were decreased in both the subepicardial and subendocardial layers of diabetic animals (Table 4). This represents decreases of 23% and 11% in total GLUT 4 concentration jn the subepicardium and subendocardium, respectively, in the diabetic group (Table 4). No regional difference was detected in GLUT 4 content between the subepicardial and subendocardial regions of the normal or diabetic groups.

GLUT 4 content was not measured in the midmyocardium or in the LAD perfusion bed. DISCUSSION

This is the first study to investigate the effects of diabetes on myocardial glycolytic activity during ischemia in a large animal model. The animals were studied under ambient metabolic conditions to simulate the conditions of clinical myocardial ischemia, thus hyperglycemic diabetic swine were compared with a group of healthy normoglycemic animals. Diabetes did not adversely affect ventricular hemodynamic function, and there were no differences in glycolytic metabolism during ischemia between groups. Thus, 1 month of streptozocin-induced diabetes did not result in any obvious impairment in myocardial function or metabolism under well-perfused or ischemic conditions when compared with that in normoglycemic nondiabetic animals. One should use caution when interpreting our results in terms of the effects of diabetes on the capacity for myocardial glycolytic metabolism during ischemia. Our results demonstrate that myocardial glycolytic rates are the same in hyperglycemic diabetic animals as in normoglycemic controls. We do not know the effects of hyperglycemia on

GLYCOLYTIC METABOLISM

67

IN THE DIABETIC HEART

NONDIABETIC

EPI

MID

END0

EPI

MID

END0 U

CFX

LAD

I

MID END0

EPI

EPI

MID END0

CEX

LAD

Fig 4. Myocardial glycogen concentration plotted for the subepicardial (EPI), midmyocardial (MID), and subendocardial (ENDO) layers of the LAD and CFX coronary artery perfusion beds. The LAD perfusion bed was subjected to ischemia, and the CFX bed had normal blood flow. No significant differences were observed between diabetic and nondiabetic groups. *P c .05, ‘P < ,002, significantly different from the same layer in the CFX bed. “Significantly different from the EPI layer of the LAD bed at P < .05.

EPI

MID

END0

EPI

MID

END0

CFX

LAD t

‘%-

EPI

MID

LAD

END0

EPI

MID

END0

CFX

Rate of glucose uptake, arterial-venous glucose difference, and percent glucose extraction plotted for the subepicardial (EPI), midmyocardial (MID), and subendocardial (ENDO) layers of the LAD and CFX coronary artery perfusion beds. The LAD perfusion bed was subjected to ischemia, and the CFX bed had normal blood flow. No significant differences were observed between diabetic and nondiabetic groups. “P c .05, nP < .Ol, significantly different from the same layer in the CFX bed. Significantly different from the EPI layer of the LAD bed at P c .05.

parison of glycolytic rates under identical metabolic conditions. We observed that 1 month of streptozocin-induced diabetes results in significant decreases in total GLUT 4 content in the subepicardium and the subendocardium of the left ventricle. Decreases in GLUT 4 content in the diabetic model have been reported for various insulin-sensitive tissues including heart, lx skeletal muscle.‘x~“9 and adipose tissue.3”J’ Our tissue biopsies were heterogenous, containing a high content of capillary endothelium in addition to cardiomyocytes. We did not attempt to identify the site of the decrease in GLUT 4 levels, and thus it is not known whether there is a uniform decrease in GLUT 4 levels in all GLUT 4-containing cell types in the biopsy. The mechanisms responsible for the decreased number of glucose transporters are not well understood. The diabetic model of this study exhibited modest hyperglycemia (80% increased) and a 65%~ decrease in plasma insulin Table 3. Free Fatty Acid Metabolism in the LAD Perfusion Bed (mean + SE) P (aerobic Y Aerobic

lschemlc

ischemic

Period

Penod

period)

Arterial free fatty acids (mmol/L)

glycolytic metabolism in nondiabetic animals. It seems likely that the decrease in GLUT 4 levels and the antiglycolytic effects of increased plasma free fatty acid levels in the diabetic animals were compensated for by the hyperglycemia. The net effect was that diabetes did not effect the rate of glycolytic metabolism during ischemia. Clearly, future studies should include an additional group of control animals with hyperglycemia, elevated free fatty acid levels, and decreased insulin levels. This would eliminate the confounding variable of hyperglycemia and allow for com-

Nondiabetic

0.43 ? .08

0.43 2 .09

NS

Diabetic

0.70 + .13*

0.79 I!z .21

NS

Arterial-venous free fatty acids (mmol/L) Nondiabetic

0.09 ? 0.03

0.10 * 0.03

NS

Diabetic

0.18 ? 0.06

0.24 + 0.07*

NS

Nondiabetic

0.09 * .03

0.04 i .02

Diabetic

0.21 2 .08

0.11 -t .03*

Net free fatty acid uptake (Km01 g-l

min-‘)

*Significantly different from nondiabetic group, P < .05.

.02 NS t.08)

68

STANLEY ET AL

Table 4. Total GLUT 4 Content Expressed as Percent of Control (mean f SE) % Decrease Region

Nondiabetic

Diabetic

With Diabetes

Subepicardium

99.0 + 5.6

75.8 + 6.8X

23%

Subendocardium

94.7 it: 4.1

83.3 k a.4t

11%

‘Significantly less than nondiabetic group, P < .Ol. tsignificantly less than nondiabetic group, P < .05.

values compared with the normal healthy animals. Previous studies suggest that the decrease in skeletal muscle glucose transporters is primarily due to hyperglycemia in diabetic ratsz9 and in muscle cell cultures exposed to hyperglycemic medium.32x33These data suggest that chronic hyperglycemia, not hypoinsulinemia, is primarily responsible for the decrease in glucose transporter number in diabetic animals. The rate of glucose uptake is largely determined by the number and activity of GLUT 4 in the sarcolemmal membrane. It has been shown that either anoxia or insulin causes a translocation of myocardial glucose transporters from an intracellular site to the sarcolemmal membrane.34 Theoretically, this would increase the myocytes’ ability to take up glucose and use it anaerobically when the oxygen supply is limited, as in myocardial ischemia. In this study, we used crude myocardial homogenates to assess total GLUT 4 content, which includes both sarcolemmal and microsomal transporters and does not necessarily reflect sarcolemmal glucose transport activities. We did not measure GLUT 4 content separately in the sarcolemmal and microsomal membranes. Thus, we cannot eliminate the possibility that the myocardium from the diabetic group had a greater fraction of its total GLUT 4 protein residing in the sarcolemmal membranes compared with that from the nondiabetic animals. Previous reports have shown impaired glucose transport in response to physiologic stimuli in rats with chemically induced diabetes. Isolated buffer-perfused rat hearts from animals with chemically induced diabetes have impaired 3-O-methyl glucose transport in response to anoxia or insulin.12 We did not observe this phenomenon in vivo under our experimental conditions. However, the present investigation differs from the standard rodent model of chemically induced diabetes in that (1) we did not have extremely low plasma insulin levels, and (2) arterial substrate and insulin levels were not matched during the

measurements of glucose metabolism. We observed a 50% decrease in arterial insulin levels as compared with a 90% decrease in untreated streptozocin-diabetic rats.35 This greater insulin concentration may be sufficient to maintain normal or near-normal glucose metabolism in the heart. Myocardial glycogen levels were not increased by 1 month of diabetes in swine; this runs counter to the observations made on rats. Chen et aP6 observed a 200% increase in myocardial glycogen levels after 1 month of streptozocin-induced diabetes. This observation suggests a species difference. The ambient metabolic conditions in our animals were conducive to glycogen storage in the ratelevated plasma glucose and free fatty acid levels, and the presence of insulin, albeit at lower levels. However, the degree of hyperglycemia and the increase in free fatty acid levels in our diabetic pigs was not as severe as in the diabetic rat model. This might result in lower rates of glycogen storage and lower myocardial glycogen levels. We selected to maintain our diabetic animals for 1 month in this initial study. This may not be enough time to elicit pathologic changes described with longer-term diabetes.37.38In addition, it may take longer to develop diabetic cardiomyopathies in large animals than in rodents, due to the lower metabolic rate and hemodynamic stress. In conclusion, our results suggest that glycolytic metabolism in well-perfused myocardium is not impaired in hyperglycemic diabetic swine when compared with normoglycemic nondiabetic animals. This occurred despite a significant ( N 20%) decrease in myocardial glucose transporter concentration. Diabetes had no significant effect on myocardial glucose uptake, glucose conversion to lactate, lactate production, or glycogen depletion under ischemic conditions. Thus, glycolytic metabolism is not impaired in hyperglycemic diabetic swine after 1 month of the disease when compared with normoglycemic nondiabetic animals. ACKNOWLEDGMENT The authors would like to thank Drs A.J. Liedtke and B. Renstrom of the Cardiology Section and Drs T. Kern and R. Engerman of the Department of Ophthalmology of the University of Wisconsin for their encouragement and advice. We are grateful to A.M. Eggleston, J. Henderson, C.R. Kidd, D. Paulson, S.G. Voss, and L.F. Whitesell of the University of Wisconsin and G. Cassefer and M. Mayr of the University of California for their expert technical assistance. The streptozocin used in this study was a gift from the Upjohn Company, Kalamazoo, MI.

REFERENCES 1. Kannel WB, McGee DL: Diabetes and cardiovascular disease. The Framingham study. JAMA 241:2035-2038,1979 2. Czyzk A, Krolewski AS, Szablowska S: Clinical course of myocardial infarction among diabetic patients. Diabetes Care 3:526,1980 3. Partamian JO, Bradley RF: Acute myocardial infarction in 258 cases of diabetes: Immediate mortality and five year survival. N Engl J Med 273:455-461.1965 4. Rytter L, Troelsen S, Beck-Nielsen H: Prevalence and mortality of acute myocardial infarction in patients with diabetes. Diabetes Care 8:230-234, 1985 5. Soler NG, Pentecost BL, Bennett MA, et al: Coronary care for myocardial infarction in diabetics. Lancet 1:475-477, 1974 6. Stone PH. Muller JE, Hartwell T, et al: The effect of diabetes

mellitus on prognosis and serial left ventricular function after acute myocardial infarction: Contribution of both coronary disease and diastolic left ventricular dysfunction to the adverse prognosis. J Am Coll Cardiol 14:49-57, 1989 7. Kuller LH, Traven ND, Rutan GH, et al: Marked decline of coronary heart disease mortality in 35-44 year old white men in Allegheny County, Pennsylvania. Circulation 80:261-266,1989 8. Goodale WT, Olson RE, Hackel DB: The effects of fasting and diabetes mellitus on myocardial metabolism in man. Am J Med 27:212-220, 1959 9. Ungar I, Gilbert M, Siegel A, et al: Studies on myocardial metabolism. IV. Myocardial metabolism in diabetes. Am J Med 18:385-396. 1955 10. Avogaro A, Nosadini R, Doria A, et al: Myocardial metabo-

GLYCOLYTIC

METABOLISM

69

IN THE DIABETIC HEART

lism in insulin-deficient diabetic humans without coronary artery disease. Am J Physiol258:E606-E618, 1990 11. Randle PJ. Newsholme E. Garland PB: Regulation of glucose uptake by muscle. Biochem J 93:632-665, 1964 12. Morgan HE, Cadenas E. Regan DM, et al: Regulation of glucose uptake in muscle. II. Rate-limiting steps and effects of insulin and anoxia in heart muscle from diabetic rats. J Biol Chem 236:262-268, 1961 13. Barrett EJ, Schwartz RG, Young LH, et al: Effect of chronic diabetes on myocardial fuel metabolism and insulin sensitivity. Diabetes 37:943-948, 1988 14. Guth BD, Wisneski JA, Neese RA, et al: Myocardial lactate release during ischemia in swine. Relation to regional blood flow. Circulation 81:1948-1958, 1990 15. Stanley WC, Hall JL, Stone CK, et al: Acute myocardial &hernia causes a transmural gradient in glucose extraction but not glucose uptake. Am J Physiol 262:H91-H96, 1992 16. Bell Gl, Kayano T, Buse JB, et al: Molecular biology of mammalian glucose transporters. Diabetes Care 13:198-208,199O 17. James DE, Strube M, Mueckler M: Molecular cloning and characterization of an insulin-regulatable glucose transporter. Nature 338:83-X7, lY89 18 Camps M. Castello A, Munoz P, et al: Effect of diabetes and fasting on GLUT-4 (muscle/fat) glucose-transporter expression in insulin-sensitive tissues. Biochem J 282:765-772, 1992 19 Wilson JD, Dhall DP, Sineonovic CJ. et al: Induction and management of diabetes mellitus in the pig. Aust J Exp Biol Med Sci 61:489-500, 1986 70 Marshall M: Induction of chronic diabetes by streptozotocin in the miniature pig. Res Exp Med 175:187-196. 1979 21 Liedtke AJ. Nellis SH, Mjos OD: Effects of reducing fatty acid metabolism on mechanical function in regionally ischemic hearts. Am J Physiol247:H387-H394. 1984 22 Renstrom B, Nellis SH. Liedtke AJ: Metabolic oxidation of pyruvate and lactate during early myocardial reperfusion. Circ Res 66:282-288. 1YYO 73. Wisneski JA, Stanley WC. Neese RA. et al: Effects of acute hyperglycemia on myocardial glycolytic activity in humans. J Clin Invest 85:1648-1656. 1990 24. Wisneski JA, Gertz EW. Neese RA, et al: Metabolic fate of extracted glucose in normal human myocardium. J Clin Invest 76:18lY-1827. 1985

25. Tserng

KY, Gilfillan

CA,

Kalhan

SC: Determination

of

carbon-13 labeled lactate in blood by gas chromatography/mass spectrometry. Anal Chem 56:517-523,1984 26. Passoneau JV, methods of glycogen 60:405-412, 1974

Lauderdale VR: A comparison measurements in tissues. Anal

of three Biochem

27. Buckberg GD, Fixler DE, Archie JP, et al: Experimental subendocardial ischemia in dogs with normal coronary arteries. Circ Res 30:67-81, 1972 28. Kahn BB. Rossetti L, Lodish HF. et al: Decreased in vivo glucose uptake but normal expression of GLUT 1 and GLUT 4 in skeletal muscle of diabetic rats. J Clin Invest 87:2197-2206, 1991 29. Ramlal T, Rastogi S, Vranic M, et al: Decrease in glucose transporter number in skeletal muscle of mildly diabetic (streptozotocin-treated) rats. Endocrinology 125:890-897, 198Y 30. Garvey WT, Huecksteadt TP, Birnhaum MJ: Pretranslational suppression of an insulin-responsive glucose transporter in rats with diabetes mellitus. Science 24.5:60-63, 1989 31. Kahn BB, Charron MJ, Lodish HF: Differential regulation of two glucose transporters in adipose cells from diabetic and insulin-treated diabetic rats. J Clin Invest 84:404-411. 1989 32. Walker P, Ramlal T. Donovan JA. et al: Molecular regulation by insulin and glucose of the glucose transport system in the Lh rat muscle cell line. J Biol Chem 264:6587-6591. 1989 33. Sarabia V, Lam L, Burdett E. et al: Glucose transport in human skeletal muscle cells in culture. J Clin Invest YO:l386-13’35. 1992 34. Wheeler TJ: Translocation of glucose transporters in response to anoxia in heart. J Biol Chem 263:19447-19454, 1988 35. Chen V, Ianuzzo CD: Metabolic muscle of chronically streptozotocin-diabetic Biophys 217:131-138, 1982

alterations in skeletal rats. Arch Biochem

36. Chen V. Ianuzzo CD, Fong BC, et al: The etfects of acute and chronic diabetes on myocardial metabolism in rats. Diabetes 33:107X-108X,1984 37. Marshall M, Oberhofer H, Staubesand J: Early micro- and macro-angiopathy in the streptozotocin diabetic minipig. Res Exp Med 177~145-158, 1980 38. Regan TJ, Ettinger PO, Khan MI, et al: Altered myocardial function and metabolism in chronic diabetes mellitus without ischemia in dogs. Circ Res 35:222-237. 1974