Feeding and fasting determine in isolated working rat hearts
postischemic
glucose utilization
CHRISTIAN A. SCHNEIDER, VAN T. B. NGUYEN, AND HEINRICH TAEGTMEYER Division of Cardiology, Department of Medicine, University of Texas Medical School at Houston and The Graduate School of Biomedical Sciences, University of Texas Health Science Center at Houston, Houston, Texas 77030
SCHNEIDER$HRISTIAN A., VAN T.B. NGUY~R,AND HEINRICH TAEGTMEYER. Feeding and fasting determine postischemic glucose utilization in isolated working rat hearts. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H542-H548, 1991.-To assess the effects of endogenous substrate on glucose utilization after 15 min of ischemia, we perfused isolated working rat hearts from fed and fasted (16 h) animals with glucose and the positron-emitting glucose analogue 2-[18F]fluoro-2-deoxy-~glucose (2-FDG). Hearts were perfused in a recirculating system with bicarbonate buffer containing glucose (10 mM) and 2FDG (0.5 &i/ml). Mechanical performance and 2-FDG uptake were measured on-line, and glucose and lactate metabolic rates were calculated. Fasting raised the glycogen content by 25% and the triglyceride content by 38%. Hearts in both groups recovered preischemic function. Rates of 2-FDG uptake during the preischemic period were the same in both groups. In contrast, during the postischemic period rates of 2-FDG uptake were significantly depressed in hearts of fed animals but were unchanged in hearts of fasted animals. Thus hearts of fasted animals took up more 2-FDG during the postischemic period than hearts of fed animals (P c 0.005). The lumped constant (range, 0.38-0.40) was the same in both groups before and after ischemia. Glucose utilization was suppressed during the postischemic period in hearts of fed animals, whereas at the same time lactate utilization was significantly increased. We conclude that 1) 2-FDG accurately traces glucose utilization independent of the nutritional state or ischemic insult; 2) reversibly ischemic, viable myocardium exhibits vastly different rates of glucose utilization depending on the nutritional state of the animal before ischemia; 3) lactate derived from glycolysis suppresses utilization of exogenously supplied glucose in the early reperfusion period without affecting postischemic performance. glycogen; lactate; ischemia; D-ghCOSe; lumped constant
reperfusion;
2- [ ‘“Flfluoro-2-deoxy-
FASTING the mammalian organism undergoes a series of metabolic adaptations to derive energy from stored fuels (4). Although these adaptations include changes in circulating free fatty acid and hormone levels (i.e., exogenous factors), it is also important to consider the effects of endogenous substrates on organ metabolism. This is especially important in the setting of myocardial ischemia when substrates are largely derived from endogenous fuel stores. To examine the role of endogenous metabolic fuels during myocardial ischemia, hearts from fed and fasted rats were perfused and glucose utilization before and after ischemia was studied. With fasting it is possible to raise myocardial glycogen and
triglyceride content (6, 8) without the exogenous supply of drugs or hormones. A preliminary report by Yamada et al. (33) demonstrated that hearts of fasted rats in vivo accumulated less of the glucose tracer 2-[lsF]fluoro-2-deoxy-~-glucose (2-FDG) than hearts of fed rats. In the present study we measured in vitro rates of myocardial 2-FDG uptake in conjunction with metabolic rates of glucose utilization derived by the Fick principle to further characterize glucose utilization of reversibly ischemic myocardium during reperfusion in hearts from fed and fasted rats. We were interested to learn whether the 2-FDG uptake in the isolated working rat heart is dependent on the nutritional state of the animal, whether tracer uptake is increased in reversibly ischemic heart muscle, as it was suggested earlier (1, 15), and whether this increase in tracer uptake represents a true increase in glucose metabolic rate or a change in the affinity of the glucose transporter and/or hexokinase to the tracer analogue (i.e., in the “lumped constant,” the definition of which is discussed below). Although we found that 2-FDG accurately traces myocardial glucose utilization before and after ischemia independent from nutritional states, we did not observe any increase in glucose utilization of the postischemic myocardium. Instead, real-time analysis of myocardial glucose metabolism with 2-FDG revealed depressed 2FDG uptake and glucose utilization by the fully functional, postischemic myocardium of fed animals and at the same time a predominant utilization of endogenously produced lactate. In contrast, hearts of fasted animals did not show any change in 2-FDG uptake or glucose utilization during the postischemic period when compared with the preischemic control period.
DURING
H542
0363-6135/91
$1.50 Copyright
MATERIALS
AND
METHODS
Animals Male Sprague-Dawley rats (300-350 g) were obtained from Sasco (Houston, TX) and were either fed ad libitum or fasted overnight (16-20 h) with free access to water before the experiment. Materials All chemicals were obtained from Fisher Scientific (Lexington, MA) or Sigma Chemical (St. Louis, MO).
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
LACTATE
AND
POSTISCHEMIC
Enzymes and cofactors were obtained from Boehringer Mannheim (Indianapolis, IN) unless indicated otherwise. The positron-emitting isotope 2-FDG (sp act 5,000 Ci/ mmol) was prepared by the method of Hamacher et al. (10) at the University of Texas Science Center at Houston Cyclotron Facility. The radiochemical purity was 99% as determined by high-performance liquid chromatography using an Aminex HPX-807C column (Bio-Rad, Richmond, CA). The mobile phase was water run at a rate of 0.8 ml/min and at a temperature of 85°C. Perfusion
Apparatus
The isolated working heart perfusion apparatus described by Taegtmeyer et al. (29) was used to measure the physiological performance of the heart (heart rate, aortic systolic and diastolic pressure, cardiac output), to assess glucose and lactate metabolic rates, and to record myocardial tracer uptake of 2-FDG. The perfusion chamber was modified to accommodate the placement of a pair of coincidence detectors on opposite sides of the heart (18). 2-FDG radioactivity in the perfusate was measured on-line from a sidearm of the aortic cannula and was returned into the apparatus after passing the detectors. Special care was taken to keep the tissue temperature of the heart at 37°C throughout the experiment by using a tungsten lamp and a Yellow Springs Instruments temperature probe (data not presented).
GLUCOSE
In mals FDG, mine FDG
a separate series of experiments hearts of fed aniwere perfused for 80 min with glucose (10 mM), 2and increasing concentrations of lactate to deterthe effects of lactate on the rates of myocardial 2uptake.
Real Time Measurement
Heart Preparation
The working heart preparation has been described in detail earlier (29). Briefly, rats were anesthetized with pentobarbital sodium (10 mg/lOO g body wt, ip). After injection of heparin (200 IU) into the inferior vena cava, the heart was rapidly removed and mounted on the aortic cannula. A brief period of retrograde perfusion (~4 min) was necessary to wash out any blood and to perform the left atria1 cannulation. Hearts were then perfused at 37°C with recirculating Krebs-Henseleit bicarbonate buffer (200 ml) containing glucose (10 mM) as substrate. The perfusion medium was gassed with 95% Og-5% C02. Data acquisition of 2-FDG activity was started immediately after the addition of 2-FDG (radioactivity, 100-150 &i/ 200 ml perfusate; injection volume, 100-300 ~1) to the oxygenator port of the perfusion system. All experiments were carried out with an afterload of 100 cmH20 and a preload of 15 cmHZO. Aortic flow and coronary flow were measured every 10 min. Cardiac output was calculated as the sum of aortic and coronary flow. Heart rate and systolic and diastolic pressure were continuously measured using a Hewlett-Packard transducer and recording system. Perfusion
Protocol
The perfusion protocol consisted of three parts. Hearts were perfused under normoxic conditions for 30 min; after this initial perfusion period hearts were subjected to global, normothermic no-flow ischemia for 15 min and then reperfused for another 30 min. Ischemia was induced by clamping both the aortic and the atria1 lines, and reperfusion was accomplished by opening the same.
of Radioactivity
External detection of 2-FDG radioactivity in tissue and perfusate was accomplished as described earlier (18). Briefly, a pair of gamma photon detectors (bismuth germanate) connected in coincidence were placed on opposite sides of the heart to measure myocardial2-FDG uptake. A separate delayed timing window measured accidental counts directly. After the electronic signals were amplified and processed, the data were sent to a VAX 11-780 computer (Digital Equipment, Maynard, MA) for storage and data analysis. 2-FDG radioactivity in the recirculating perfusate (input function) was recorded continuously by a device connected to the aortic cannula. The processed electronic signals of the input function were also sent to the VAX computer for data storage and analysis. The conversion of tissue counts (counts per second) to microcuries per milliliter was performed with a bar phantom similar in size and shape to a rat heart and containing a known amount of [IsF] radioactivity. Analysis
Working
H543
UTILIZATION
of Tracer
Uptake
The curves were analyzed using the graphical analysis of Patlak and Blasberg (20). The slope of the Patlak plot represents the fractional rate of transfer and phosphorylation of 2-FDG from the extracellular space to an intracellular, irreversible compartment. The slope is expressed as milliliters of perfusate per milliliter of tissue per minute. The glucose metabolic rate (pmol h-l. g dry wt-‘) using the slope of the Patlak plot (K) was calculated as follows: (Km g,. 60)/g dry wt, where g, is the actual glucose concentration in the perfusate. The lumped constant (LC, 28) was calculated as the ratio between the glucose metabolic rate derived from the Patlak graphical analysis and the glucose metabolic rate derived by measuring the rate of glucose removal from the recirculating perfusate (Fick principle) 2-FDG-derived metabolic rate (Patlak) LC glucose metabolic rate (Fick) Biochemical
Methods
Perfusate glucose and lactate determination. Perfusate samples (1 ml) were withdrawn at 5- to IO-min intervals during the perfusion to measure the concentration of glucose and lactate. Lactate concentration was determined in the coronary effluent collected at 1-min intervals for the 1st 3 min immediately after the start of reperfusion. Subsequently, perfusate sampling resumed for the duration of the 30-min reperfusion period as described earlier. Glucose and lactate were measured with a glucose-lactate analyzer (2300 STAT, Yellow Springs Instrument, Yellow Springs, OH). Rates of glucose utilization and lactate production or utilization were
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
H544
LACTATE
AND
POSTISCHEMIC
determined by disappearance or appearance of glucose or lactate in the perfusion medium as described earlier (29) Tissue extraction and analytical procedure. In separate experiments myocardial glycogen and triglyceride contents were measured in vivo. Rats were anesthetized with pentobarbital(l0 mg/lOO g body wt) and the hearts were quickly removed and freeze-clamped as described earlier (30). In some experiments myocardial glycogen content was also measured in vitro, at the end of ischemia or at the end of reperfusion. Glycogen was determined by the modified method of Walaas and Walaas (32) using amyloglucosidase. Tissue triglycerides were extracted with chloroform-ethanol (2:1, vol/vol) (9) and determined by the method of Eggstein and Kuhlmann (7). A small portion of tissue was used to obtain the wet weight-dry weight ratio. The tissue was weighed and then dried in an oven (60°C) for 72 h to constant weight (dry wt). Data for glycogen are presented as micromoles glucose per gram dry weight, data for triglycerides as micromoles per gram dry weight. Statistical
Analysis
All data are presented as means t SE. To compare the pre- and postischemic periods of hearts from fed and fasted rats a two-way analysis of variance test was employed and, if significant, followed by Newman-Keuls post hoc analysis. Differences were considered statistically significant when P < 0.05. RESULTS
Triacylglycerol Content, Glycogen Content, and Performance of Working Hearts The in vivo tissue triacylglycerol content from hearts of fasted rats was significantly higher than in hearts of fed rats (151.2 t 9.5 vs. 108.9 t 14, P < 0.05). The in vivo glycogen content and the glycogen content at the end of ischemia in hearts from fasted rats were significantly greater than in hearts from fed rats (Table 1). Steady-state cardiac performance during the pre- and postischemic period was the same in hearts from fed and fasted animals (Table 2). Myocardial 2FDG
Uptake
The curves depicted in Figs. l-3 show representative time-activity curves for 2-FDG uptake by isolated working rat hearts. The curves were generated from the sampled total coincidence counts in the tissue (Figs. l1. Glycogen content in vivo and at end of ischemia
Significance vs. fed Values
are means
n
In Vivo
At End of Ischemia
6 5
89t5 125t5
39-+6 66t4
P < 0.002
P c 0.01
t SE in pmol/g
dry wt.
UTILIZATION
3, top tracings labeled tissue), single counts of the ZFDG activity in the recirculating perfusate (Figs. l-3, middle tracings labeled perfusate), and random accidental counts in the system (Figs. l-3 bottom). All counts were decay corrected to the time of 2-FDG injection. Rates of 2-FDG uptake during the preischemic period were the same in both groups (fed 0.148 vs. fasted 0.177, NS; Figs. 1 and 2, and Table 3). During the postischemic period rates of 2FDG uptake were significantly depressed in hearts of fed rats (Fig. 1). Quantitative analysis by the Patlak graphical method showed a suppression of tracer uptake by 57% during the postischemic period (Table 3; from preischemic 0.148 to postischemic 0.064, P < 0.05). In contrast, in hearts from fasted animals, rates of 2-FDG uptake during the postischemic period (Fig. 2) did not change significantly (Table 3; from preischemic 0.177 to postischemic 0.174, NS). The postischemic rates of 2-FDG uptake by hearts from fed animals were significantly lower when compared with postischemic rates of 2-FDG uptake by hearts from fasted animals (0.064 vs. 0.174, P < 0.05). Thus the average rate of 2-FDG uptake during the postischemic period by hearts from fed animals was less than one-half the preischemic uptake rate (-57%), whereas in the fasted group the rate of 2-FDG uptake was essentially unchanged (-2%, P < 0.005 when compared with fed; Table 3). Glucose and Lactate Metabolic Rates In the preischemic period hearts from both groups used exogenous glucose as substrate and there was a net production of lactate (Table 3). During the postischemic period glucose utilization in hearts from fed animals was markedly reduced (P c 0.35, Table 3). During the first 3 min immediately after reperfusion, lactate release from the postischemic myocardium into the coronary effluent was significantly greater in hearts from fed rats than in hearts from fasted rats (1.8 t 0.26 vs. 0.43 t 0.17 mM, P < O.Ol), leading to significant higher lactate concentration in the perfusate (0.91 t 0.19 vs. 0.23 t 0.03 mM, P < 0.01). At the end of the 30-min reperfusion period, lactate levels in the recirculating perfusate of both groups were the same. The greater perfusate concentration of lactate brought about a net utilization of lactate during the postischemic period in hearts from fed animals that was significantly greater than in hearts from fasted animals (P < 0.01, Table 3). In contrast, in hearts from fasted animals there was no change in glucose utilization. Moreover, glucose utilization was significantly greater in hearts from fasted animals than in hearts from fed animals (Table 3). Lumped Constant
TABLE
Fed Fasted
GLUCOSE
Significance vs. In Vivo P < 0.01
P < 0.01
Neither ischemia and reperfusion nor the nutritional state of the animals caused a significant change in the LC (Table 3). Glucose utilization by the isolated working rat hearts in either group was accurately traced by 2FDG as analyzed by the Patlak graphical analysis. Effects of Lactate on Myocardial 2-FDG Uptake To test the hypothesis that the postischemic reduction in 2-FDG uptake by hearts of fed animals was the result
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
LACTATE
TABLE
AND
POSTISCHEMIC
GLUCOSE
H545
UTILIZATION
2. Cardiac performance of perfused hearts from fed and fasted rats n
AF
CF ml - min-l
HR, beats/min
co
Systolic
Diastolic
- g dry wt-l
mmHg
Fed Preischemic Postischemic
10 10
133t21 147tl7
89t18 103t19
Preischemic Postischemic
5 5
158tl7 154tl8
76tl3 86tl9
222t32 249t47
232t8 234t7
95t2 97t3
72t3 74t3
234*49 240t47
235t6 260t8
108-1-2 llOt4
88t4 88k4
Fasted
Values are means t SE. Data reported were obtained 30 min after start of perfusion (postischemic). AF, aortic flow; CF, coronary flow; CO, cardiac output; HR, heart rate.
(preischemic)
and 30 min
after
start
of reperfusion
Tissue
Perfusate /
I 30
I 45
I 60
I 75
0
Time (min) FIG.
tracing 15 min during tissue
1. Representative of isolated working of no-flow ischemia. postischemic period. 2-FDG radioactivity;
2-[“Flfluoro-2-deoxy-D-glucose (2-FDG) rat heart from fed animal before and after Note significant decreased 2-FDG uptake cps, counts per second; tissue, myocardial perfusate, perfusate 2-FDG radioactivity.
of the availability of lactate as competing substrate during the postischemic period, hearts from fed animals (n = 4) were perfused with glucose (10 mM) and increasing amounts of sodium-lactate were added to the perfusate. Figure 3 shows a representative 2-FDG tracing of the effects of increasing lactate concentrations on 2-FDG uptake by a heart from a fed rat. Increasing concentrations of lactate caused a progressive decrease in 2-FDG uptake. We note that at 1 mM lactate the Patlak slope was similar to the Patlak slope of hearts from fed rats in the postischemic period (0.07 vs. 0.064). In addition, there is a good correlation between the percent decrease of the Patlak slope and increasing lactate concentrations in the perfusate (Fig. 4). Cardiac performance was stable throughout these experiments and was not significantly different from the cardiac performance of the other experiments (cardiac output ranging from 220 to 250 ml min-l . g dry wt-‘; data not presented). DISCUSSION
The two salient findings of the present study are suppressed glucose uptake and increased lactate utiliza-
I
15
30
I
45
I
60
1
75
Time (min) 2. Representative 2-FDG tracing of isolated working rat heart from fasted animal before and after 15 min of no-flow ischemia. Note essentially unchanged 2-FDG uptake before and after ischemia. FIG.
tion in postischemic myocardium from fed animals. The results seem at first glance inconsistent with two widely held concepts. The first concept is that reversibly ischemic myocardium exhibits increased rates of glucose uptake and, possibly, also increased rates of glucose utilization (2, 5). The second concept is that glycogen loading increases substrate readily available for accelerated anaerobic energy production through preferential utilization of glycogen as source for lactate production from glycolysis (18). Both concepts require further discussion in the light of our present findings. We show that after a brief period of global ischemia and in the presence of a single exogenously supplied substrate (glucose), 2-FDG uptake and rates of glucose utilization are suppressed in hearts of fed animals. The present data expand recent reports on the delayed recovery of glucose metabolism (3) after longer periods of ischemia and on the unaffected glucose metabolism (25) after acute ischemia and reperfusion in vivo. It is of note that all previous studies (2, 5) have been carried out in the presence of hormones (e.g., insulin, glucagon, and catecholamines) and competing substrates of largely unknown concentrations. In the isolated working rat heart
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
H546
LACTATE
AND
POSTISCHEMIC
GLUCOSE
$ 1500-l
a 0 2 a
IOOOPerfusate
5000 +--t--7--0
1
IO
20
30
.
,-;.
1
40 50 TIME (min)
1
I
I
60
70
80
FIG. 3. Representative 2FDG tracing of isolated working rat heart of fed animal perfused with glucose (10 mM) and increasing lactate concentrations. Note progressive decrease in myocardial 2FDG uptake with increasing lactate concentrations in perfusate. At high lactate concentrations (5-10 mM) 2-FDG uptake and glucose utilization was almost completely suppressed (data not shown). Numbers below curve, fractional rate of 2-FDG uptake (Patlak slope).
preparation these variables can be controlled to the extent of their addition to the perfusate. We provide convincing evidence that lactate, produced by the heart itself, can serve as the preferred fuel for respiration during reoxygenation of the previously ischemic myocardium. This observation suggeststhat fully functional, postischemic myocardium oxidizes the fuel most readily available. Furthermore, flux through pyruvate dehydrogenase, the gatekeeper for acetyl-CoA production and, hence, pyruvate oxidation in the citric acid cycle, could not have been significantly inhibited after 15 min of global no-flow ischemia in our model. These results are in agreement with recent experiments in reperfused swine heart in situ by Renstrom et al. (24), who demonstrated nearly normal postischemic pyruvate oxidation in the presence of an inhibitor of long-chain fatty acid transfer across the inner mitochondrial membrane. The decrease in pyruvate dehydrogenase activity associated with reperfusion after 10 min of low-flow ischemia (1 ml/min) in rat hearts reported by Kobayashi and Neely (12) is probably the result of procedural differences. Furthermore, Olsen et al. (19) have pointed out that with physiological perturbations the assay of iso-
UTILIZATION
lated enzyme systems in vitro may not reflect the actual regulatory effects on the enzyme in the intact organ. Although our data in postischemic hearts from both fed and fasted animals strongly suggest pyruvate dehydrogenase activation, postischemic rates of glucose utilization in hearts from fed animals equally strongly suggest an inhibition of glycolysis by lactate restricting flux through the glycolytic pathway, most likely at the level of glyceraldehyde-3-phosphate dehydrogenase (27) However, in contrast to the findings by these earlier workers, we did not observe any functional impairment with lactate. Thus endogenously produced lactate may become the preferred fuel for energy production without adverse effects on cardiac performance. A surprising result of our study was that glycogen-rich hearts from fasted animals released and utilized less lactate than hearts from fed animals. A possible explanation for this finding could be the differential breakdown of glycogen in hearts from fed and fasted animals. It seems reasonable to speculate that under conditions of increased lactate production more glycogen is broken down than under conditions of less lactate production. Although the glycogen content at the end of ischemia was still significantly higher in hearts from fasted rats, the amounts of glycogen broken down during ischemia were similar (Table 1). Thus differences in glycogen breakdown during ischemia and, consequently, supply of endogenous substrate, cannot explain our findings. Although our data are similar to data by Kilgour and Riggs (11) and Tani and Neely (31) showing that in glycogenrich hearts from diabetic animals the lactate content was significantly less at the end of ischemia and at the end of reperfusion, we have no explanation why hearts from fasted rats released less lactate during reperfusion than hearts from fed rats. Isotopic labeling studies will be required to identify the exact source of the lactate (glycogen vs. exogenously supplied glucose) released and oxidized by postischemic hearts from fed animals. Although differences in the triacylglycerol content in hearts from fed and fasted rats could contribute to the observed differences in the glucose utilization, data by van Bilsen et al. (1) have shown that neither the total triacylglycerol content during ischemia nor the oxidation of fatty acids during reperfusion is changed. It is therefore unlikely that changes in the triacylglycerol content lead to the observed change in glucose and lactate utilization during reperfusion.
TABLE 3. Patlak slopesof 2-FIX&tracing and metabolic rates for glucose and lactate of hearts from fed and fasted animals before and after 15 min of ischemia Fed (n = 10)
Patlak slope Patlak slope ( % Preischemic) Glucose metabolic Lactate metabolic Lumped constant
rate rate
Fasted
(n = 5)
Preischemic
Postischemic
Preischemic
Postischemic
0.148t0.03
0.064*0.05” -57k9
0.177t0.07
0. 174+0.08t’ -2t3’
-978k283 278t135 0.39kO.04
-305+3@ -976:219” 0.39kO.05
-784t162 112&85 0.4tO.l
-637k57t’ -160*186’ 0.38*0.08
Values are means & SE. 2-FDG, 2-[‘“Flfluoro-2-deoxy-D-glucose; wtt’ . h? a P < 0.05 vs. preischemic fed; ” P < 0.05 vs. postischemic 0.03 vs. preischemic fed; f‘ P < 0.01 vs. postischemic fed.
Patlak slope, ml plasma-ml tissue-‘. min-‘; metabolic rates, fed; ’ P < 0.005 vs. postischemic fed; ’ P < 0.05 vs. preischemic
pmo1.g dry fed; ’ P
postischemic
glucose utilization
CHRISTIAN A. SCHNEIDER, VAN T. B. NGUYEN, AND HEINRICH TAEGTMEYER Division of Cardiology, Department of Medicine, University of Texas Medical School at Houston and The Graduate School of Biomedical Sciences, University of Texas Health Science Center at Houston, Houston, Texas 77030
SCHNEIDER$HRISTIAN A., VAN T.B. NGUY~R,AND HEINRICH TAEGTMEYER. Feeding and fasting determine postischemic glucose utilization in isolated working rat hearts. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H542-H548, 1991.-To assess the effects of endogenous substrate on glucose utilization after 15 min of ischemia, we perfused isolated working rat hearts from fed and fasted (16 h) animals with glucose and the positron-emitting glucose analogue 2-[18F]fluoro-2-deoxy-~glucose (2-FDG). Hearts were perfused in a recirculating system with bicarbonate buffer containing glucose (10 mM) and 2FDG (0.5 &i/ml). Mechanical performance and 2-FDG uptake were measured on-line, and glucose and lactate metabolic rates were calculated. Fasting raised the glycogen content by 25% and the triglyceride content by 38%. Hearts in both groups recovered preischemic function. Rates of 2-FDG uptake during the preischemic period were the same in both groups. In contrast, during the postischemic period rates of 2-FDG uptake were significantly depressed in hearts of fed animals but were unchanged in hearts of fasted animals. Thus hearts of fasted animals took up more 2-FDG during the postischemic period than hearts of fed animals (P c 0.005). The lumped constant (range, 0.38-0.40) was the same in both groups before and after ischemia. Glucose utilization was suppressed during the postischemic period in hearts of fed animals, whereas at the same time lactate utilization was significantly increased. We conclude that 1) 2-FDG accurately traces glucose utilization independent of the nutritional state or ischemic insult; 2) reversibly ischemic, viable myocardium exhibits vastly different rates of glucose utilization depending on the nutritional state of the animal before ischemia; 3) lactate derived from glycolysis suppresses utilization of exogenously supplied glucose in the early reperfusion period without affecting postischemic performance. glycogen; lactate; ischemia; D-ghCOSe; lumped constant
reperfusion;
2- [ ‘“Flfluoro-2-deoxy-
FASTING the mammalian organism undergoes a series of metabolic adaptations to derive energy from stored fuels (4). Although these adaptations include changes in circulating free fatty acid and hormone levels (i.e., exogenous factors), it is also important to consider the effects of endogenous substrates on organ metabolism. This is especially important in the setting of myocardial ischemia when substrates are largely derived from endogenous fuel stores. To examine the role of endogenous metabolic fuels during myocardial ischemia, hearts from fed and fasted rats were perfused and glucose utilization before and after ischemia was studied. With fasting it is possible to raise myocardial glycogen and
triglyceride content (6, 8) without the exogenous supply of drugs or hormones. A preliminary report by Yamada et al. (33) demonstrated that hearts of fasted rats in vivo accumulated less of the glucose tracer 2-[lsF]fluoro-2-deoxy-~-glucose (2-FDG) than hearts of fed rats. In the present study we measured in vitro rates of myocardial 2-FDG uptake in conjunction with metabolic rates of glucose utilization derived by the Fick principle to further characterize glucose utilization of reversibly ischemic myocardium during reperfusion in hearts from fed and fasted rats. We were interested to learn whether the 2-FDG uptake in the isolated working rat heart is dependent on the nutritional state of the animal, whether tracer uptake is increased in reversibly ischemic heart muscle, as it was suggested earlier (1, 15), and whether this increase in tracer uptake represents a true increase in glucose metabolic rate or a change in the affinity of the glucose transporter and/or hexokinase to the tracer analogue (i.e., in the “lumped constant,” the definition of which is discussed below). Although we found that 2-FDG accurately traces myocardial glucose utilization before and after ischemia independent from nutritional states, we did not observe any increase in glucose utilization of the postischemic myocardium. Instead, real-time analysis of myocardial glucose metabolism with 2-FDG revealed depressed 2FDG uptake and glucose utilization by the fully functional, postischemic myocardium of fed animals and at the same time a predominant utilization of endogenously produced lactate. In contrast, hearts of fasted animals did not show any change in 2-FDG uptake or glucose utilization during the postischemic period when compared with the preischemic control period.
DURING
H542
0363-6135/91
$1.50 Copyright
MATERIALS
AND
METHODS
Animals Male Sprague-Dawley rats (300-350 g) were obtained from Sasco (Houston, TX) and were either fed ad libitum or fasted overnight (16-20 h) with free access to water before the experiment. Materials All chemicals were obtained from Fisher Scientific (Lexington, MA) or Sigma Chemical (St. Louis, MO).
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
LACTATE
AND
POSTISCHEMIC
Enzymes and cofactors were obtained from Boehringer Mannheim (Indianapolis, IN) unless indicated otherwise. The positron-emitting isotope 2-FDG (sp act 5,000 Ci/ mmol) was prepared by the method of Hamacher et al. (10) at the University of Texas Science Center at Houston Cyclotron Facility. The radiochemical purity was 99% as determined by high-performance liquid chromatography using an Aminex HPX-807C column (Bio-Rad, Richmond, CA). The mobile phase was water run at a rate of 0.8 ml/min and at a temperature of 85°C. Perfusion
Apparatus
The isolated working heart perfusion apparatus described by Taegtmeyer et al. (29) was used to measure the physiological performance of the heart (heart rate, aortic systolic and diastolic pressure, cardiac output), to assess glucose and lactate metabolic rates, and to record myocardial tracer uptake of 2-FDG. The perfusion chamber was modified to accommodate the placement of a pair of coincidence detectors on opposite sides of the heart (18). 2-FDG radioactivity in the perfusate was measured on-line from a sidearm of the aortic cannula and was returned into the apparatus after passing the detectors. Special care was taken to keep the tissue temperature of the heart at 37°C throughout the experiment by using a tungsten lamp and a Yellow Springs Instruments temperature probe (data not presented).
GLUCOSE
In mals FDG, mine FDG
a separate series of experiments hearts of fed aniwere perfused for 80 min with glucose (10 mM), 2and increasing concentrations of lactate to deterthe effects of lactate on the rates of myocardial 2uptake.
Real Time Measurement
Heart Preparation
The working heart preparation has been described in detail earlier (29). Briefly, rats were anesthetized with pentobarbital sodium (10 mg/lOO g body wt, ip). After injection of heparin (200 IU) into the inferior vena cava, the heart was rapidly removed and mounted on the aortic cannula. A brief period of retrograde perfusion (~4 min) was necessary to wash out any blood and to perform the left atria1 cannulation. Hearts were then perfused at 37°C with recirculating Krebs-Henseleit bicarbonate buffer (200 ml) containing glucose (10 mM) as substrate. The perfusion medium was gassed with 95% Og-5% C02. Data acquisition of 2-FDG activity was started immediately after the addition of 2-FDG (radioactivity, 100-150 &i/ 200 ml perfusate; injection volume, 100-300 ~1) to the oxygenator port of the perfusion system. All experiments were carried out with an afterload of 100 cmH20 and a preload of 15 cmHZO. Aortic flow and coronary flow were measured every 10 min. Cardiac output was calculated as the sum of aortic and coronary flow. Heart rate and systolic and diastolic pressure were continuously measured using a Hewlett-Packard transducer and recording system. Perfusion
Protocol
The perfusion protocol consisted of three parts. Hearts were perfused under normoxic conditions for 30 min; after this initial perfusion period hearts were subjected to global, normothermic no-flow ischemia for 15 min and then reperfused for another 30 min. Ischemia was induced by clamping both the aortic and the atria1 lines, and reperfusion was accomplished by opening the same.
of Radioactivity
External detection of 2-FDG radioactivity in tissue and perfusate was accomplished as described earlier (18). Briefly, a pair of gamma photon detectors (bismuth germanate) connected in coincidence were placed on opposite sides of the heart to measure myocardial2-FDG uptake. A separate delayed timing window measured accidental counts directly. After the electronic signals were amplified and processed, the data were sent to a VAX 11-780 computer (Digital Equipment, Maynard, MA) for storage and data analysis. 2-FDG radioactivity in the recirculating perfusate (input function) was recorded continuously by a device connected to the aortic cannula. The processed electronic signals of the input function were also sent to the VAX computer for data storage and analysis. The conversion of tissue counts (counts per second) to microcuries per milliliter was performed with a bar phantom similar in size and shape to a rat heart and containing a known amount of [IsF] radioactivity. Analysis
Working
H543
UTILIZATION
of Tracer
Uptake
The curves were analyzed using the graphical analysis of Patlak and Blasberg (20). The slope of the Patlak plot represents the fractional rate of transfer and phosphorylation of 2-FDG from the extracellular space to an intracellular, irreversible compartment. The slope is expressed as milliliters of perfusate per milliliter of tissue per minute. The glucose metabolic rate (pmol h-l. g dry wt-‘) using the slope of the Patlak plot (K) was calculated as follows: (Km g,. 60)/g dry wt, where g, is the actual glucose concentration in the perfusate. The lumped constant (LC, 28) was calculated as the ratio between the glucose metabolic rate derived from the Patlak graphical analysis and the glucose metabolic rate derived by measuring the rate of glucose removal from the recirculating perfusate (Fick principle) 2-FDG-derived metabolic rate (Patlak) LC glucose metabolic rate (Fick) Biochemical
Methods
Perfusate glucose and lactate determination. Perfusate samples (1 ml) were withdrawn at 5- to IO-min intervals during the perfusion to measure the concentration of glucose and lactate. Lactate concentration was determined in the coronary effluent collected at 1-min intervals for the 1st 3 min immediately after the start of reperfusion. Subsequently, perfusate sampling resumed for the duration of the 30-min reperfusion period as described earlier. Glucose and lactate were measured with a glucose-lactate analyzer (2300 STAT, Yellow Springs Instrument, Yellow Springs, OH). Rates of glucose utilization and lactate production or utilization were
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
H544
LACTATE
AND
POSTISCHEMIC
determined by disappearance or appearance of glucose or lactate in the perfusion medium as described earlier (29) Tissue extraction and analytical procedure. In separate experiments myocardial glycogen and triglyceride contents were measured in vivo. Rats were anesthetized with pentobarbital(l0 mg/lOO g body wt) and the hearts were quickly removed and freeze-clamped as described earlier (30). In some experiments myocardial glycogen content was also measured in vitro, at the end of ischemia or at the end of reperfusion. Glycogen was determined by the modified method of Walaas and Walaas (32) using amyloglucosidase. Tissue triglycerides were extracted with chloroform-ethanol (2:1, vol/vol) (9) and determined by the method of Eggstein and Kuhlmann (7). A small portion of tissue was used to obtain the wet weight-dry weight ratio. The tissue was weighed and then dried in an oven (60°C) for 72 h to constant weight (dry wt). Data for glycogen are presented as micromoles glucose per gram dry weight, data for triglycerides as micromoles per gram dry weight. Statistical
Analysis
All data are presented as means t SE. To compare the pre- and postischemic periods of hearts from fed and fasted rats a two-way analysis of variance test was employed and, if significant, followed by Newman-Keuls post hoc analysis. Differences were considered statistically significant when P < 0.05. RESULTS
Triacylglycerol Content, Glycogen Content, and Performance of Working Hearts The in vivo tissue triacylglycerol content from hearts of fasted rats was significantly higher than in hearts of fed rats (151.2 t 9.5 vs. 108.9 t 14, P < 0.05). The in vivo glycogen content and the glycogen content at the end of ischemia in hearts from fasted rats were significantly greater than in hearts from fed rats (Table 1). Steady-state cardiac performance during the pre- and postischemic period was the same in hearts from fed and fasted animals (Table 2). Myocardial 2FDG
Uptake
The curves depicted in Figs. l-3 show representative time-activity curves for 2-FDG uptake by isolated working rat hearts. The curves were generated from the sampled total coincidence counts in the tissue (Figs. l1. Glycogen content in vivo and at end of ischemia
Significance vs. fed Values
are means
n
In Vivo
At End of Ischemia
6 5
89t5 125t5
39-+6 66t4
P < 0.002
P c 0.01
t SE in pmol/g
dry wt.
UTILIZATION
3, top tracings labeled tissue), single counts of the ZFDG activity in the recirculating perfusate (Figs. l-3, middle tracings labeled perfusate), and random accidental counts in the system (Figs. l-3 bottom). All counts were decay corrected to the time of 2-FDG injection. Rates of 2-FDG uptake during the preischemic period were the same in both groups (fed 0.148 vs. fasted 0.177, NS; Figs. 1 and 2, and Table 3). During the postischemic period rates of 2FDG uptake were significantly depressed in hearts of fed rats (Fig. 1). Quantitative analysis by the Patlak graphical method showed a suppression of tracer uptake by 57% during the postischemic period (Table 3; from preischemic 0.148 to postischemic 0.064, P < 0.05). In contrast, in hearts from fasted animals, rates of 2-FDG uptake during the postischemic period (Fig. 2) did not change significantly (Table 3; from preischemic 0.177 to postischemic 0.174, NS). The postischemic rates of 2-FDG uptake by hearts from fed animals were significantly lower when compared with postischemic rates of 2-FDG uptake by hearts from fasted animals (0.064 vs. 0.174, P < 0.05). Thus the average rate of 2-FDG uptake during the postischemic period by hearts from fed animals was less than one-half the preischemic uptake rate (-57%), whereas in the fasted group the rate of 2-FDG uptake was essentially unchanged (-2%, P < 0.005 when compared with fed; Table 3). Glucose and Lactate Metabolic Rates In the preischemic period hearts from both groups used exogenous glucose as substrate and there was a net production of lactate (Table 3). During the postischemic period glucose utilization in hearts from fed animals was markedly reduced (P c 0.35, Table 3). During the first 3 min immediately after reperfusion, lactate release from the postischemic myocardium into the coronary effluent was significantly greater in hearts from fed rats than in hearts from fasted rats (1.8 t 0.26 vs. 0.43 t 0.17 mM, P < O.Ol), leading to significant higher lactate concentration in the perfusate (0.91 t 0.19 vs. 0.23 t 0.03 mM, P < 0.01). At the end of the 30-min reperfusion period, lactate levels in the recirculating perfusate of both groups were the same. The greater perfusate concentration of lactate brought about a net utilization of lactate during the postischemic period in hearts from fed animals that was significantly greater than in hearts from fasted animals (P < 0.01, Table 3). In contrast, in hearts from fasted animals there was no change in glucose utilization. Moreover, glucose utilization was significantly greater in hearts from fasted animals than in hearts from fed animals (Table 3). Lumped Constant
TABLE
Fed Fasted
GLUCOSE
Significance vs. In Vivo P < 0.01
P < 0.01
Neither ischemia and reperfusion nor the nutritional state of the animals caused a significant change in the LC (Table 3). Glucose utilization by the isolated working rat hearts in either group was accurately traced by 2FDG as analyzed by the Patlak graphical analysis. Effects of Lactate on Myocardial 2-FDG Uptake To test the hypothesis that the postischemic reduction in 2-FDG uptake by hearts of fed animals was the result
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
LACTATE
TABLE
AND
POSTISCHEMIC
GLUCOSE
H545
UTILIZATION
2. Cardiac performance of perfused hearts from fed and fasted rats n
AF
CF ml - min-l
HR, beats/min
co
Systolic
Diastolic
- g dry wt-l
mmHg
Fed Preischemic Postischemic
10 10
133t21 147tl7
89t18 103t19
Preischemic Postischemic
5 5
158tl7 154tl8
76tl3 86tl9
222t32 249t47
232t8 234t7
95t2 97t3
72t3 74t3
234*49 240t47
235t6 260t8
108-1-2 llOt4
88t4 88k4
Fasted
Values are means t SE. Data reported were obtained 30 min after start of perfusion (postischemic). AF, aortic flow; CF, coronary flow; CO, cardiac output; HR, heart rate.
(preischemic)
and 30 min
after
start
of reperfusion
Tissue
Perfusate /
I 30
I 45
I 60
I 75
0
Time (min) FIG.
tracing 15 min during tissue
1. Representative of isolated working of no-flow ischemia. postischemic period. 2-FDG radioactivity;
2-[“Flfluoro-2-deoxy-D-glucose (2-FDG) rat heart from fed animal before and after Note significant decreased 2-FDG uptake cps, counts per second; tissue, myocardial perfusate, perfusate 2-FDG radioactivity.
of the availability of lactate as competing substrate during the postischemic period, hearts from fed animals (n = 4) were perfused with glucose (10 mM) and increasing amounts of sodium-lactate were added to the perfusate. Figure 3 shows a representative 2-FDG tracing of the effects of increasing lactate concentrations on 2-FDG uptake by a heart from a fed rat. Increasing concentrations of lactate caused a progressive decrease in 2-FDG uptake. We note that at 1 mM lactate the Patlak slope was similar to the Patlak slope of hearts from fed rats in the postischemic period (0.07 vs. 0.064). In addition, there is a good correlation between the percent decrease of the Patlak slope and increasing lactate concentrations in the perfusate (Fig. 4). Cardiac performance was stable throughout these experiments and was not significantly different from the cardiac performance of the other experiments (cardiac output ranging from 220 to 250 ml min-l . g dry wt-‘; data not presented). DISCUSSION
The two salient findings of the present study are suppressed glucose uptake and increased lactate utiliza-
I
15
30
I
45
I
60
1
75
Time (min) 2. Representative 2-FDG tracing of isolated working rat heart from fasted animal before and after 15 min of no-flow ischemia. Note essentially unchanged 2-FDG uptake before and after ischemia. FIG.
tion in postischemic myocardium from fed animals. The results seem at first glance inconsistent with two widely held concepts. The first concept is that reversibly ischemic myocardium exhibits increased rates of glucose uptake and, possibly, also increased rates of glucose utilization (2, 5). The second concept is that glycogen loading increases substrate readily available for accelerated anaerobic energy production through preferential utilization of glycogen as source for lactate production from glycolysis (18). Both concepts require further discussion in the light of our present findings. We show that after a brief period of global ischemia and in the presence of a single exogenously supplied substrate (glucose), 2-FDG uptake and rates of glucose utilization are suppressed in hearts of fed animals. The present data expand recent reports on the delayed recovery of glucose metabolism (3) after longer periods of ischemia and on the unaffected glucose metabolism (25) after acute ischemia and reperfusion in vivo. It is of note that all previous studies (2, 5) have been carried out in the presence of hormones (e.g., insulin, glucagon, and catecholamines) and competing substrates of largely unknown concentrations. In the isolated working rat heart
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
H546
LACTATE
AND
POSTISCHEMIC
GLUCOSE
$ 1500-l
a 0 2 a
IOOOPerfusate
5000 +--t--7--0
1
IO
20
30
.
,-;.
1
40 50 TIME (min)
1
I
I
60
70
80
FIG. 3. Representative 2FDG tracing of isolated working rat heart of fed animal perfused with glucose (10 mM) and increasing lactate concentrations. Note progressive decrease in myocardial 2FDG uptake with increasing lactate concentrations in perfusate. At high lactate concentrations (5-10 mM) 2-FDG uptake and glucose utilization was almost completely suppressed (data not shown). Numbers below curve, fractional rate of 2-FDG uptake (Patlak slope).
preparation these variables can be controlled to the extent of their addition to the perfusate. We provide convincing evidence that lactate, produced by the heart itself, can serve as the preferred fuel for respiration during reoxygenation of the previously ischemic myocardium. This observation suggeststhat fully functional, postischemic myocardium oxidizes the fuel most readily available. Furthermore, flux through pyruvate dehydrogenase, the gatekeeper for acetyl-CoA production and, hence, pyruvate oxidation in the citric acid cycle, could not have been significantly inhibited after 15 min of global no-flow ischemia in our model. These results are in agreement with recent experiments in reperfused swine heart in situ by Renstrom et al. (24), who demonstrated nearly normal postischemic pyruvate oxidation in the presence of an inhibitor of long-chain fatty acid transfer across the inner mitochondrial membrane. The decrease in pyruvate dehydrogenase activity associated with reperfusion after 10 min of low-flow ischemia (1 ml/min) in rat hearts reported by Kobayashi and Neely (12) is probably the result of procedural differences. Furthermore, Olsen et al. (19) have pointed out that with physiological perturbations the assay of iso-
UTILIZATION
lated enzyme systems in vitro may not reflect the actual regulatory effects on the enzyme in the intact organ. Although our data in postischemic hearts from both fed and fasted animals strongly suggest pyruvate dehydrogenase activation, postischemic rates of glucose utilization in hearts from fed animals equally strongly suggest an inhibition of glycolysis by lactate restricting flux through the glycolytic pathway, most likely at the level of glyceraldehyde-3-phosphate dehydrogenase (27) However, in contrast to the findings by these earlier workers, we did not observe any functional impairment with lactate. Thus endogenously produced lactate may become the preferred fuel for energy production without adverse effects on cardiac performance. A surprising result of our study was that glycogen-rich hearts from fasted animals released and utilized less lactate than hearts from fed animals. A possible explanation for this finding could be the differential breakdown of glycogen in hearts from fed and fasted animals. It seems reasonable to speculate that under conditions of increased lactate production more glycogen is broken down than under conditions of less lactate production. Although the glycogen content at the end of ischemia was still significantly higher in hearts from fasted rats, the amounts of glycogen broken down during ischemia were similar (Table 1). Thus differences in glycogen breakdown during ischemia and, consequently, supply of endogenous substrate, cannot explain our findings. Although our data are similar to data by Kilgour and Riggs (11) and Tani and Neely (31) showing that in glycogenrich hearts from diabetic animals the lactate content was significantly less at the end of ischemia and at the end of reperfusion, we have no explanation why hearts from fasted rats released less lactate during reperfusion than hearts from fed rats. Isotopic labeling studies will be required to identify the exact source of the lactate (glycogen vs. exogenously supplied glucose) released and oxidized by postischemic hearts from fed animals. Although differences in the triacylglycerol content in hearts from fed and fasted rats could contribute to the observed differences in the glucose utilization, data by van Bilsen et al. (1) have shown that neither the total triacylglycerol content during ischemia nor the oxidation of fatty acids during reperfusion is changed. It is therefore unlikely that changes in the triacylglycerol content lead to the observed change in glucose and lactate utilization during reperfusion.
TABLE 3. Patlak slopesof 2-FIX&tracing and metabolic rates for glucose and lactate of hearts from fed and fasted animals before and after 15 min of ischemia Fed (n = 10)
Patlak slope Patlak slope ( % Preischemic) Glucose metabolic Lactate metabolic Lumped constant
rate rate
Fasted
(n = 5)
Preischemic
Postischemic
Preischemic
Postischemic
0.148t0.03
0.064*0.05” -57k9
0.177t0.07
0. 174+0.08t’ -2t3’
-978k283 278t135 0.39kO.04
-305+3@ -976:219” 0.39kO.05
-784t162 112&85 0.4tO.l
-637k57t’ -160*186’ 0.38*0.08
Values are means & SE. 2-FDG, 2-[‘“Flfluoro-2-deoxy-D-glucose; wtt’ . h? a P < 0.05 vs. preischemic fed; ” P < 0.05 vs. postischemic 0.03 vs. preischemic fed; f‘ P < 0.01 vs. postischemic fed.
Patlak slope, ml plasma-ml tissue-‘. min-‘; metabolic rates, fed; ’ P < 0.005 vs. postischemic fed; ’ P < 0.05 vs. preischemic
pmo1.g dry fed; ’ P
