- Email: [email protected]
Evaluation of a short-time, oxygen carrier-free perfusion model in rat liver: Mitochondrial energy metabolism and insulin effect
JOURNAL
OF SURGICAL RESEARCH d&69-76
(1986)
Evaluation of a Short-Time, Oxygen Carrier-Free Perfusion Model in Rat Liver: Mitochondrial
Energy Metabolism and Insulin Effect
YASUYUKI SHIMAHARA,M.D.,’ AND WOLF ISSELHARD, M.D. Institute of Experimental Medicine, University of Cologne, Robert-Koch-Strasse IO, D 5000 Cologne 41, Federal Republic of Germany Submitted for publication October 17, 1984 Livers of rats fasted for 18 hr were perfused for 60 min at 37,35, and 32°C in a simple noncirculating systemwith a carbogen(95%, 5% CO3 equilibmted, oxygen carrier-free,glucose-containing(200 mg .dl-‘) cristalloid per&ate (pH 7.4, 280 mosmol - kg-‘) without or with the addition of insulin (10 U * hr-‘), and analyzed for tissue ATP, ADP, AMP, acetoacetate,and &hydroxybutyrate as well as for parameters of the energy metabolism of isolated mitochondria. Perfusion without insulin at 37°C causedsignificant alterations: ATP and total adenine nucleotides (TAN) dropped to 62 and 74% of the control values (3.33 and 4.57 lmol . g-l), respectively. The energy charge potential ECP decreasedfrom 0.842 to 0.7 13. Also, the mitochondrial phosphorylation rate (PR), the respiratory control ratio (RC), and the state 3 respiration rate (ST 3) were significantly depressedto 62,54, and 72% of the control values. Perfusion with insulin maintained a normal metabolic pattern: ATP 1051, TAN 9796,ECP 0.877, PR 9496,RC 6996,and ST 3 105%.The metabolic deterioration during insulin-free perfusion was decreasedat 35/32”C: ATP 67/82%, PR 72/95%, RC 74187%.At 35”C, insulin ameliorated the metabolic pattern, but normal ranges were no longer reached. At 32°C insulin had no effect. 0 1986 AC&C& ~esr.IW.
INTRODUCTION
The perfusion of the isolated liver is a valuable method for the investigation of physiological, biochemical, and pharmacological aspects of integral functions of the liver, eliminating the influences of systemic reactions. Various techniques have been developed allowing for valuable insight into hepatic functions. However, approaches of liver preparation and perfusion techniques are sometimes very complicated, which may give rise to alterations in the regular hepatic functions. A simple perfusion approach capable of maintaining fully the viability of liver would be a definite achievement. The omission of an oxygen carrier in the perhtsate greatly simplifies the technical setup, and moreover hasthe advantageof eliminating the effectsdue to the metabolism of blood cells, serum elements, or unknown factors of chemical oxygen carriers such as fluorocarbon, which are also utilized for organ perfusion. ’ Y. Shimahara holds a researchscholarship granted by the Alexander von Humboldt-Foundation.
However, it has the disadvantage of an impending hypoxia, to which the liver is especially susceptible. Although no perfused organ preservation can claim to be physiological, there are certain accepted criteria of satisfactory perfusion. Besides simple criteria like appearance of the perfused organ, absence of swelling, portal pressure, and oxygen consumption, more liver-specific parameters like gluconeogenesis, ketogenesis,urea production, protein synthesis or bile production [5,8,10,14] generally serve to judge the efficacy of perfusion and the viability of the liver. But since function and maintenance of cellular integrity of the liver are linked to a sufficient aerobic energy supply [22,24,26], it is easyto investigate the hepatic energy status as a parameter of liver viability. The purpose of this study was to evaluate a short-time, simple perfusion model of the rat liver, in which the organ viability is well maintained in terms of energy metabolism. It will be demonstrated that in a very simple per&sion systememploying an electrolyte perfusate without an oxygen carrier and without colloi69
0022-4804/86 $1.50 ‘h&&t Q 1986by Academic I’m%,Inc. All rights of l-wlvdllciicm in any form r*med.
70
JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 1, JANUARY 1986
dal osmolarity, the perfused liver is capable of maintaining a high activity of mitochondrial function and quasi-normal tissue levels and ratios in the adenine nucleotide system, concomitant with a high oxido-reduction state of the mitochondria. Further, it will be evidenced, that insulin plays an important role in maintaining the energy status of the perfused liver by enhancing mitochondrial oxidative phosphorylation. MATERIALS
AND METHODS
Male albino Wistar rats (200-300 g) were allowed water ad libitum, but food was restricted for about 18 hr prior to the experiment. The rats were anesthetized by i.p. injection of 50 mg pentobarbital/kg body wt. The liver was perfused via the portal vein employing a modified Mortimore’s method [ 151 in an open system (Fig. 1). The perfiusatewas Krebs-Henseleit solution [7] containing 0.54 pMEDTA (ethylendiaminetetra acetate),glucose (200 mg . dl-I), and heparin (1.7 U * ml-‘), and equilibrated with carbogen(95% O2 + 5% Cq). The perfusion was started not
II
u
before the pOz exceded 500 mm Hg, and temperature and pH (7.4) were well controlled. The osmolarity of the solution averaged 280 mosmol kg-‘. The portal vein was cannulated and connected to the perfusion system without interrupting the portal flow. The initial flow rate was set to 6-7 ml mini. Immediately thereafter the infrahepatic vena cava was cannulated, and the suprahepatic vena cava was ligated just cranial the diaphragma. The washout of blood was finished within 5 min. Then the flow rate was increased to 36 ml. min-’ (about 5 ml - min-’ . g-’ hepatic tissue). The portal pressureaveraged25 cm H20. The perfusion lasted 60 min without or with an additional insulin infusion, and was performed at 37,35, and 32”C, respectively. Insulin (MCInsulin NOVO Rapitard) was diluted with the perfusate to a concentration of 1.6 U - ml-’ just before the onset of the perfusion, and infused at a rate of 10 Us hr-‘. Every 15 min, perfusate (1 ml) was sampled from both the inflow and outflow line of the system and analysed for gaseswith the help of an ABL2 Acid Base Laboratory (Radiometer, Copenhagen).
,... .. . ..Y . .._......_. . . . . . . .
l
l
Puma /
pH Meter
Thermostat
FIG.
1. Schema of the perfusion system.
SHIMAHARA
71
AND ISSELHARD: RAT LIVER PERFUSION
From the remnant of the liver, mitochonThe oxygenconsumption of the liver was calculated from the O2 differencesand the flow dria werepreparedaspreviouslyreported[ 171. Their oxygen consumption was measured porate. After 60 min of perfusion, the left anterior larographically [ 161 with a Clark-type elecliver lobe was grasped and pressedto about 2 trode (Yellow Spring, Model 53, USA). The mm thickness between stainless-steel tongs precooled in liquid nitrogen. The frozen tissue was weighed; about 2 g was homogenized in 6 ml ice-cold 6% HC104 using a tissue homogenizer “Ultraturrax” (Janke u. Kunkel Co., Staufen, FRG). The homogenate was centrifuged at 10,OOOgfor 15 min at 0-4”C. The precipitate was washed with ice-cold 3% HC104 and centrifuged at 15,OOOg for 15 min at 0-4”C. The supernatants were pooled and adjusted to pH 6 with cold 20% (W/V) KOH and recentrifuged at 10,OOOgfor 5 min at O4°C. In these extracts, the concentrations of adenine nucleotides (ATP, ADP, AMP), and ketone bodies-acetoacetate (AcAc) and fihydroxybutyrate (@-OHB)-were measured enzymatically [ 1,9, 13,291. The sum of ATP, ADP, and AMP is referred to as total adenine nucleotides (TAN). The energy charge potential (ECP) was calculated according to Atkinson [2], ECP = (ATP + 0.5 ADP)/TAN. The ketone body ratio, which indicates the hepatic mitochondrial oxido-reduction state (NAD+/ NADH) was expressedas the ratio of AcAc to ,&OHB.
respiratory control ratio (RC = state 3 respiration rate/state 4 respiration rate), state 3 respiration rate, ADP/O, and oxidative phosphorylative activity (PR = state 3 X ADP/O) were calculated from the polarographical trao ings according to Chance [3]. These measurements were performed at 22°C using glutamate as substrate. Mitochondrial protein was measured by the method of Lowry et al. [ 1I]. For the measurement of the dry weight of the liver samples, about 400 mg of the wet tissue was dried for 4-5 days in a heat box at 110°C. Normal fasted rat livers served as controls in comparison to the perfused livers. The statistical significances between the mean values were determined by Student’s t test. RESULTS
ATP and TAN (Table 1) were at the end of the insulin-free perfusion least altered at 32°C with averages of 82 and 80% of the control values, respectively. The alterations in ATP and TAN increased with a rise in perfusion temperature, averaging 67 and 74% at 35”C,
TABLE ADENINE NUCLEOTIDE
Groups
Number of rats
1
CONTENTS (amole/g WET TISSUE) AND ENERGY CHARGE POTENTIAL AND PERFUSED LIVERS OF RATS FASTED 18 hr (MEAN + SEM)
ATP
ADP
AMP
ATP + ADP + AMP
(ECP) IN
CONTROL
ECP
(8)
3.33 * 0.10
1.03 * 0.07
0.21 + 0.02
4.57 + 0.16
0.842 + 0.008
(5) (5)
2.08 + 0.14*** 3.51 * 0.02c
1.13 f 0.09 0.76 f 0.09**-
0.52 f 0.18* 0.17 + 0.03
3.37 * 0.178’ 4.44 f 0.26”
0.7 13 * 0.047** 0.877 * o.oll**~
35°C Insulin (-) Insulin (+)
(5) (5)
2.23 f 0.20*** 2.83 f 0.08***p
0.90 + 0.09 0.90 + 0.09
0.25 + 0.07 0.19 -+ 0.02
3.37 f 0.22*** 3.92 k 0.06.”
0.793 + 0.028* 0.838 2 0.013
32°C Insulin (-) Insulin (+)
(5) (5)
2.73 f 0.19** 2.83 ?I 0.27*
0.77 f 0.09 0.85 f 0.03
0.15 + 0.01’ 0.17 + 0.01
3.65 + 0.26”; 3.85 f 0.26*
0.854 Ik 0.011 0.843 + 0.013
Control 37°C Insulin (-) Insulin (+)
a P < 0.05, *P < 0.01, 'P < 0.001 as compared between Insulin (-) and Insulin (+) groups. * P < 0.05, l *P < 0.01, ***P -c 0.001 as compared to the control values.
72
JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 1, JANUARY 1986 TABLE 2
KETONEBODY CONTENTS@mole/g WET TISSUE)AND RATIO DRY WEIGHT(dw)/WET WEIGHT(ww) IN CONTROL AND PERFUSEDLIVERSOFRATSFASTED18 hr (MEAN + SEM)
Groups
Number of rats
ACAC
j3-OHB
dw. 100 ww
Total
(8)
0.2 16 + 0.037
0.564 + 0.068
0.780 + 0.099
29.6 + 0.2
(5) (5)
0.038 zkO.OlO** 0.043 f 0.010**
0.049 f 0.016*** 0.015 5 0.003***~
0.087 k 0.026*** 0.057 f 0.013***
27.1 + 0.8’** 28.1 + 0.7*
35°C Insulin (-) Insulin (+)
(5) (5)
0.049 + 0.010** 0.059 5 0.009**
0.025 + 0.013*** 0.0 13 + 0.002***
0.074 + 0.022*** 0.072 + O.OlO***
28.6 + 0.2** 28.0 f 0.4***
32°C Insulin (-) Insulin (+)
(5) (5)
0.073 + 0.014** 0.048 + 0.005**
0.02 1 * 0.003*+* 0.013 k 0.005***
0.094 3I 0.012*** 0.06 1 + 0.007***
28.5 + 0.6* 28.4 + 0.8
Control 37°C Insulin (-) Insulin (+)
OP< 0.05ascompared between Insulin(-) andInsulin(+) groups. * P < 0.05, **P < 0.01, ***P 0.001 as compared to the control values.
and 62 and 74% at 37”C, respectively. With insulin, higher temperaturesbecamefavorable: At 32”C, there were no differences between groups without and with insulin. At 35”C, ATP and TAN amounted to 85 and 86%, respectively. At 37”C, these values averaged 105 and 97% of the controls, respectively. The differences between groups without or with insulin were significant at 35 and 37°C. The advantageous effect of insulin at 35 and 37°C also became apparent in the normal values of the ECP. AcAc and @-OHB (Table 2) were significantly decreasedat the end of the perfusion as compared to the control values. The decreasein &OHB was more pronounced than that in AcAc. There was no relevant influence of temperature or the admixture of insulin. The ketone body ratio AcAc/&OHB (Fig. 2) was increased significantly in comparison to the control range. This increase enlarged with hypothermia and was more pronounced in insulin-perfused livers. The mitochondrial phosphorylation rate (Fig. 3) as an indicator of mitochondrial ATP synthesis decreased significantly during perfusion without insulin to 62 and 72% of the control value at 37 and 35”C, respectively. The infusion of insulin suppressedthis alteration
maintaining the phosphorylation rate at 94% of the control value. At 32”C, the phosphorylation rate remained unchanged even without insulin infusion, and the effect of insulin was not significant. The mitochondrial respiratory control ratio (RC), which reflectsthe integrity of the isolated mitochondria, the state 3 respiration rate (ST 3) indicating the O2 consumption in the presence of sufficient ADP and substrate, and the
Control
3vc
35oc
32oC
FIG. 2. Ketone body ratios in control and perbed livers
of rats fasted 18 hr without and with an insulin infusion during perfusion (mean k SEM); ***P i 0.001 for perfusion values vs control value and for insulin (-) vs insulin (+) groups.
SHIMAHARA
73
AND ISSELHARD: RAT LIVER PERFUSION
The oxygen consumption of the liver remained constant during the entire perfusion period and averaged (mean + SEM) without/ with the application ofinsulin 56.5 + M/55.0 + 5.0 111g-r wet tissue. min-’ at 37”C, 49.5 + 2.8/45.8 f 4.4 ~1. g-r min-’ at 35°C and 4 1.2 f 4.4/46.4 + 1.8 ~1. g-’ . min-’ at 32°C respectively. The differences between groups without and with insulin infusion were not significant. The ratio dry weight/wet weight of hepatic tissue (Table 2) was in perfused livers reduced as compared to the control value. Again, there were no significant differences between groups without and with insulin infusion. l
l
0) it rol
(5)
(5)
37%
(5)
(5)
35%
(5)
(5)
32OC
FIG. 3. Phosphorylation rate of mitcchondria from DISCUSSION control and perfused livers of rats fasted 18 hr without and with an insulin infusion during perfusion (mean A permanent, sufficient energy supply is the + SEM); **P < 0.01 perfused vs control livers. 'P< 0.05, prerequisite for the maintenance of cellular 'P-z0.01 insulin (-) vs insulin (+) groups.
functions and integrity. In the liver, biosynthetic processes like gluconeogenesis, urearatio ADP/O (Table 3) were the more altered genesis, protein synthesis, bile formation as as compared to the control values the higher well as the many active membrane transport the perfusion temperature was. The changes processes require a sufficient ATP supply and were most pronounced in RC and least in sufficiently high ATP levels. An inadequate ADP/O. With insulin, the alterations were less energy supply will result in a progressive susmaintaining RC at 32°C and ST 3 at all tem- pension of functions, a progressive structural peratures within the control ranges. alteration, and eventually the irreversible cell TABLE 3 RESPIRATORY CONTROLRATIO (RC), STATE 3 RESPIRATION RATE (ST 3, n Atom O/mgprotein/min), ADP/O OF MITOCHONDRIA ISOLATEDFROMCONTROLAND PERFUSED LIVERY
ANDRATIO
OF RATS FASTED18 hr (MEAN f SEM)
Groups Control 37OC Insulin (-) Insulin (+)
Number of rats
RC
ST 3
ADP/O
(9)
4.86 + 0.25
30.2 f 2.1
2.15 It 0.08
(5) (5)
2.62 + 0.36*** 3.34 -t 0.20***
21.8 f 4.0* 31.8 -L 4.3
2.41 + 0.20 2.48 + 0.06*
35°C Insulin (-) Insulin (+)
(5) (5)
3.62 k 0.30** 3.95 + 0.18**
24.0 f 1.6’ 30.2 f 2.9
2.52 + 0.08 2.69 f 0.25
32°C Insulin (-) Insulin (+)
(5) (5)
4.22 -t OS%* 4.14 + 0.42
29.3 + 6.7 32.8 k 6.4
2.79 + 0.18 2.74 zk 0.11
* P < 0.05, **P c 0.01, ***P< 0.001 as compared to the control values. DitTerences between Insulin (-) and Insulin (+) groups were not significant.
74
JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 1, JANUARY 1986
damageand death. On the background of these interrelations, it seems to be meaningful to evaluate the hepatic energy metabolism, when functional capacity and viability of the liver are discussed. Since the liver has only little glycolytic capacity, the mitochondrial performance is of particular interest. The concept of the adenylate energy charge potential (ECP) proposed by Atkinson [2] regardsthe adenine nucleotide system as the energy currency of the cell, which reflects not only a balance between energy-generating and energy-consuming reactions, but also a regulatory parameter in the energy metabolism. The ECP is maintained at a high and constant level under normal aerobic conditions, it decreases,if energy production does not meet the energy requirements. In various pathological states,it is a more sensitive parameter than the tissue level of TAN [20, 211. In the livers perfused at 37 and 35°C without insulin, the ECP decreasedsignificantly. Since no extrahepatic load was added in these experiments, this decreasemore likely reflects an impaired energy production than enhanced energyconsuming reactions. The data on mitochondrial functions support this view. The activity of mitochondria in vivo is reflected in the mitochondrial redox (oxido-reduction) state (NAD+/NADH), which according to Williamson et al. [28] can be expressed by the ratio of ketone bodies in the liver tissue (mitochondrial NAD+/NADH = hepatic (AcAc/@-OHB) X l/K, where K is the equilibrium constant of /3-hydroxybutyrate dehydrogenase.) The ketone body ratio in the liver in situ of normally fed rats averages about 1. Due to fasting the rats, it was decreasedin the control series. All perfused livers exhibited a significantly higher value of the ketone body ratio than the control livers in situ, becausethe tissue level of /?-OHB was more drastically reduced than that of AcAc. Becauseof the absence of free fatty acids in the perfusate, the amount of ketone bodies of the perfused livers was significantly lower than that of the liver in situ. The high redox state together with the low number of ketone bodies reflects a situation in which hepatic energy metabolism is
supported only by glucose oxidation, and which possibly is exaggeratedby the absence of a normal ketone body formation. The ketone body ratio in the 37”C-Insulin (-) group was markedly lower than in the other groups. This may reflect an insufficient energy production under the experimental conditions resulting in a particularly pronounced decreasein the ECP. A decreasein the ketone body ratio is generally considered to result from (1) mitochondrial impairment, (2) oxygen deficiency, or (3) fatty acid oxidation. Since the oxygen consumption rate of this group did not differ from the other groups, and free fatty acids were not supplied via the perfusate, this low ketone body ratio is most likely indicative of a mitochondrial impairment. In support of this view, the isolated mitochondria of this group showed the most pronounced decreasein the phosphorylation rate, the respiratory control ratio, and the state 3 respiration rate. From these findings it can be concluded that in this model the perfused liver at 37“C without insulin failed to maintain its viability due to an impairment of mitochondria. On the contrary, the liver perfusion at 37°C with insulin infusion revealed to be satisfactory: The tissue levels of adenine nucleotides and the ECP were within the normal ranges at the end of the 60-min perfusion period, the mitochondrial phosphoryiation remained at 94% of the control value, the ketone body ratio was significantly increased, the mitochondrial respiratory control ratio (RC) and the ADP/ 0 ratio were less depressed, and the mitochondrial state 3 respiration rate (ST 3) corresponded to the control value. The maintenance of a normal energy status in the normothermic erythrocyte-free perfused liver seems to result from a stimulation of mitochondrial activities by insulin. Several investigations deal with the effects of insulin on the mitochondrial oxidative phosphorylative activity. Matsubara et al. reported that mitochondria from livers of diabetic rats exhibited a marked deterioration, which was completely reversedby insulin [ 121. In that situation, the mitochondrial derangement was attributed to an accumulation of free
SHIMAHARA
AND ISSELHARD: RAT LIVER PERFUSION
fatty acids, which act as an uncoupler of mitochondrial oxidativephosphorylation;insulin was thought to exert its effectsby removing the fatty acids with an utilization of glucose. In the present experiment, however, the mechanism of insulin action on mitochondrial activities appearsto be different, since no fatty acids were included in the perfusate and the perfusion period wastoo short for the synthesis of a large amount of fatty acids. Ozawa et al. demonstrated that insulin stimulated the oxidative phosphorylation of liver mitochondria and considered insulin to be an important portal factor regulating the viability of the liver [ 18, 191.He proposed that insulin has a direct action on any single rate-limiting step of either the electron transport or phosphorylation system. Consistent with this, Ida et al. found in the perfused guinea pig liver that insulin enhanced the mitochondrial oxidative phosphorylation, and also suggestedthat this effect is linked to a regulative mechanism such as conformational or configurational changes in an organized membrane system of respiratory assemblies [6]. On the other hand, it cannot be neglected to elucidate the mechanism of insulin action on mitochondrial activity in terms of regulatory enzymes or substrate currency in the metabolic pathway. It is known that insulin effectively activates the pyruvate dehydrogenase complex [27], which exists in mitochondria and catalizes the reaction between pyruvate and acetyl Co-A. This phenomenon in the initial step of the TCA cycle will induce an acceleration of the TCA cycle and may be related to an enhanced mitochondrial activity, which would be in keeping with the elevation of the oxide-reduction state of the mitochondria by insulin infusion of the present study. However, the mechanism of the insulin effect on mitochondria is not completely elucidated. Reports that the addition of insulin to isolated mitochondria had no effect on the oxidative phosphorylation [ 12, 191 are op posed by reports on positive effects [4]. It is not clear how insulin enters mitochondria. Sakamoto et al. suggestedthat mitochondria have no specific recepter for insulin, and that
75
the effectsof insulin on mitochondria might be attributable to some unknown factor derived secondarilyfrom the binding site of the plasma membrane [23]. The metabolic deterioration and the depression of mitochondrial activity in livers perfused without insulin were progressively moderated by hypothermia of 35 and 32°C respectively. The decreasein ATP and ECP was less pronounced or suspended, the parameters of mitochondrial activity-phosphorylation rate, RC, ST 3, and ADP/Owere improved at 35°C and quasi-normal at 32°C respectively, and the ketone body ratio was progressively increased. These data can be explained with a hypothermia-induced reduction in hepatic energy demand. In keeping with this interpretation, the oxygen consump tion of the liver dropped with a decrease in temperature. Insulin again exerted a positive effect on the mitochondrial activities and the hepatic metabolic status at 35“C, although the tissue levels of ATP and TAN were not as completely maintained as in the insulin (+) group at 37°C. At 32°C the effectsof insulin were neglectible. These results indicate, that the action of insulin is diminished with a progressivedecreasein temperature. Perfusionswithout oxygen carrier have been reported by several authors, but most work in this field apparently revealed such a perfusion as inadequate compared with the use of erythrocytes. Sugano et al., however, demonstrated a well designed oxygen carrier-free perfusion system,and proved that it was satisfactory and applicable to a variety of investigations, if the perfusion flow rate exceeds 3 ml - min-’ . g-r of liver [ 251.Their experiment was performed at 32°C without insulin infusion. The results ofthe present study at 32°C are consistent with those. As to the doses of insulin, further investigation might be necessaryfor this system, although Ida et al. reported that 10 U. hr-’ were optimal to induce mitochondrial enhancement in an erythrocyte-containing perfusion system [6]. From the results it may be concluded that it is possible to maintain the liver of rats viable at normothermia or quasi-normothermia for
76
JOURNAL OF SURGICAL RESEARCH:
at least 60 min in terms of the hepatic tissue levels of adenine nucleotides and mitochondrial activities employing an oxygen carrierfree electrolyte medium in a simple, noncirculating perfusion system, if insulin is adequately administered. REFERENCES 1. Adam, H. Adenosin-5diphosphate and adenosin-5monophosphate. In H. U. Bergmeyer (Ed.), Methods ofEnzymatic Analysis, p. 573. New York: Academic Press, 1965. 2. Atkinson, D. E. The energy charge of the adenylate pool as a regulatory parameter, interaction with feedback modifiers. Biochemistry 7: 4030, 1968. 3. Chance, B. Quantitative aspectson the control of oxygen utilization. In W. Cew and C. M. G’Connor, (Eds.), Ciba Foundation Symposium on Regulation of Cell Metabolism. p. 9 1. London: Churchill, 1959. 4. Hall, J. C., Sordahl, L. A., and Stelko, P. L. The effect of insulin on oxidative phosphorylation in normal and diabeticmitochondria. J. Eiol. Chem. 235: 1536,1959. 5. Hems, R., Ross, B. D., Berry, M. N., and Krebs, H. A. Gluconeogenesis in the perfused rat liver. B&hem. J. 101: 284, 1966. 6. Ida, T., Sato, M., Yamaoka, Y., Takeda, H., Kamiyama, Y., Kimura, K., Gzawa, K., and Honjo, I. Effect of insulin on mitochondrial oxidative phosphorylation and energy charge of the perfused guinea pig liver. J. Lab. Clin. Med. 87: 925, 1976. 7. Krebs, H. A., and Hen&it, K Untersuchungentiber die HarnstofIbildtmg in Tierkorper. Z. Physiol. Chem. 210: 33, 1932. 8. Krebs, H. A., Patricia, G., Wallace, R., Hems, R., and Freadland, R. A. Rates of ketone body formation in the perfused rat Liver. B&hem. J. 112: 595, 1969. 9. Lamprecht, W., and Trautschold, I. Determination with hexokinase and glucose-6-phosphate dehydrogenase.In H. U. Bergmeyer (Ed.), Methods ofEnzymatic Analysis, p. 543. New York: Academic Press, 1965. 10. Lee, D., and Walker, J. M. Maintenance of the functional state of isolated rat liver by hypothermic perfusion with an erythrocyte-free medium. Transplantation 23: 136, 1977. 11. Lowry, 0. H., Rosebrough, N. J., Frarr, A. L., and Randall, R. J. Protein measurement with Frolin phs no1reagent. J. Biol. Chem. 193: 265, 1951. 12. Matsubara, T., and Tochino, Y. Depression of respiratory activities in the liver mitochondria of diabetic rats and the restorative action of insulin. J. Biochem. 66: 397, 1969. 13. Mellanby, J., and Williamson, D. H. Acetoacetate. In H. U. Bergmeyer (Ed.), Methods of Enzymatic Analysis, p. 1840. New York: Academic Press, 1974. 14. Miller, L. L., Bly, C. G., Watson, M. L., and Bale,
JOL.40, NO. 1, JANUARY 1986
W. F. The dominant role of the liver in plasma protein synthesis. A direct study of the isolated perfused rat liver with the aids of lysine-C’4. J. Exp. Med. 94: 43 1, 1951. 15. Mortimore, G. E., Tietze, F., and Stetten, D. Metabolism of insulin-I’31. Studies in isolated perfbsed rat liver and hind-limb preparations. Diabetes 8: 307, 1959. 16. Ozawa, K., Takasan, H., and Kitamura, 0. Effect of ligation of portal vein on liver mitochondrial metabolism. J. Biochem. 70: 755, 1971. 17. Ozawa, K., Kitamura, O., and Mizukami, T. Human liver mitochondria. Clin. Chim. Acta 38: 385, 1972. 18. Gzawa, K., Yamada, T., and Honjo, I. Role of insulin as a portal factor in maintaining the viability of liver. Ann. Surg. 180: 716, 1974. 19. Gzawa, K., Yamaoka, Y., Nanbu, H., and Honjo, I. Insulin as the primary factor governing changes in mitochondrial metabolism leading to liver regeneration and atrophy. Amer. J. Surg. 127: 669, 1974. 20. Gzawa, K., Ida, T., Kamano, J., Garbus, J., and Cowley, R. A. Different responseof hepatic energy charge and adenine nucleotides concentrations to hemorrhagic shock. Res. Exp. Med. 169: 145, 1976. 21. Gzawa, K., Yamaoka, Y., Kimura, K., Kimiyama, Y., Sato, M., Ukikusa, M., and Tobe, T. Circulating hepatodepressantfactors decreasingthe energy charge levels of the remnant liver after hepatectomy. Eur. Surg. Rex 13: 444, 1981. 22. Gzawa, K. Energy metabolism. In R. A. Cowley and B. F. Trump (Eds.),Pathophysiologyof Shock,Anoxia, and Ischemia. p. 74. Baltimore: Waverly, 1982. 23. Sakamoto, Y., Kuzuya, T., and Kawanishi, K. Study of insulin action: Uptake of insulin into subcellular fractions and stimulation of pyruvate dehydrogenase in mouse adipocytes. Hormon to Rinsho 29: 1061, 1981. 24. Shimahara, Y., Ozawa, K., Ida, T., Ukikusa, M., and Tobe, T. Four stagesof mitochondrial deterioration in hemorrhagic shock. Res. Exp. Med. 179: 23,198 1. 25. Sugano, T., Suda, K., Shimada, M., and Gshino, N. Biochemical and ultrastructural evaluation of isolated rat liver systemsperfusedwith a hemoglubin-free me dium. J. Biochem. 83: 995, 1978. 26. Ukikusa, M., Ida, Y., Ozawa, K., and Tobe, T. The influence of hypoxia and hemorrhage upon adenylate energy charge and bile tlow. Surg. Gynecol. Obstet. 149: 346, 1979. 27. Weiss,L., Liiffler, G., and Wieland, 0. H. Regulation by insulin of adipose tissue pyruvate dehydrogenase. Hoppe-Sqler’s Z. Physiol. Chem. 355: 363, 1974. 28. Williamson, D. H., Lund, P. A., and Krebs, H. A. The redox state of free nicotinamide-adeninedinucleotide in the cytoplasm and mitochondria of rat liver. Biochem. J. 103: 514, 1967. 29. Williamson, D. H., and Mellanby, J. j%Hydroxybutyrate. In H. U. Bergmeyer (Ed.), Methods of Enzymatic Analysis, p. 1836. New York: Academic Press, 1974.
OF SURGICAL RESEARCH d&69-76
(1986)
Evaluation of a Short-Time, Oxygen Carrier-Free Perfusion Model in Rat Liver: Mitochondrial
Energy Metabolism and Insulin Effect
YASUYUKI SHIMAHARA,M.D.,’ AND WOLF ISSELHARD, M.D. Institute of Experimental Medicine, University of Cologne, Robert-Koch-Strasse IO, D 5000 Cologne 41, Federal Republic of Germany Submitted for publication October 17, 1984 Livers of rats fasted for 18 hr were perfused for 60 min at 37,35, and 32°C in a simple noncirculating systemwith a carbogen(95%, 5% CO3 equilibmted, oxygen carrier-free,glucose-containing(200 mg .dl-‘) cristalloid per&ate (pH 7.4, 280 mosmol - kg-‘) without or with the addition of insulin (10 U * hr-‘), and analyzed for tissue ATP, ADP, AMP, acetoacetate,and &hydroxybutyrate as well as for parameters of the energy metabolism of isolated mitochondria. Perfusion without insulin at 37°C causedsignificant alterations: ATP and total adenine nucleotides (TAN) dropped to 62 and 74% of the control values (3.33 and 4.57 lmol . g-l), respectively. The energy charge potential ECP decreasedfrom 0.842 to 0.7 13. Also, the mitochondrial phosphorylation rate (PR), the respiratory control ratio (RC), and the state 3 respiration rate (ST 3) were significantly depressedto 62,54, and 72% of the control values. Perfusion with insulin maintained a normal metabolic pattern: ATP 1051, TAN 9796,ECP 0.877, PR 9496,RC 6996,and ST 3 105%.The metabolic deterioration during insulin-free perfusion was decreasedat 35/32”C: ATP 67/82%, PR 72/95%, RC 74187%.At 35”C, insulin ameliorated the metabolic pattern, but normal ranges were no longer reached. At 32°C insulin had no effect. 0 1986 AC&C& ~esr.IW.
INTRODUCTION
The perfusion of the isolated liver is a valuable method for the investigation of physiological, biochemical, and pharmacological aspects of integral functions of the liver, eliminating the influences of systemic reactions. Various techniques have been developed allowing for valuable insight into hepatic functions. However, approaches of liver preparation and perfusion techniques are sometimes very complicated, which may give rise to alterations in the regular hepatic functions. A simple perfusion approach capable of maintaining fully the viability of liver would be a definite achievement. The omission of an oxygen carrier in the perhtsate greatly simplifies the technical setup, and moreover hasthe advantageof eliminating the effectsdue to the metabolism of blood cells, serum elements, or unknown factors of chemical oxygen carriers such as fluorocarbon, which are also utilized for organ perfusion. ’ Y. Shimahara holds a researchscholarship granted by the Alexander von Humboldt-Foundation.
However, it has the disadvantage of an impending hypoxia, to which the liver is especially susceptible. Although no perfused organ preservation can claim to be physiological, there are certain accepted criteria of satisfactory perfusion. Besides simple criteria like appearance of the perfused organ, absence of swelling, portal pressure, and oxygen consumption, more liver-specific parameters like gluconeogenesis, ketogenesis,urea production, protein synthesis or bile production [5,8,10,14] generally serve to judge the efficacy of perfusion and the viability of the liver. But since function and maintenance of cellular integrity of the liver are linked to a sufficient aerobic energy supply [22,24,26], it is easyto investigate the hepatic energy status as a parameter of liver viability. The purpose of this study was to evaluate a short-time, simple perfusion model of the rat liver, in which the organ viability is well maintained in terms of energy metabolism. It will be demonstrated that in a very simple per&sion systememploying an electrolyte perfusate without an oxygen carrier and without colloi69
0022-4804/86 $1.50 ‘h&&t Q 1986by Academic I’m%,Inc. All rights of l-wlvdllciicm in any form r*med.
70
JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 1, JANUARY 1986
dal osmolarity, the perfused liver is capable of maintaining a high activity of mitochondrial function and quasi-normal tissue levels and ratios in the adenine nucleotide system, concomitant with a high oxido-reduction state of the mitochondria. Further, it will be evidenced, that insulin plays an important role in maintaining the energy status of the perfused liver by enhancing mitochondrial oxidative phosphorylation. MATERIALS
AND METHODS
Male albino Wistar rats (200-300 g) were allowed water ad libitum, but food was restricted for about 18 hr prior to the experiment. The rats were anesthetized by i.p. injection of 50 mg pentobarbital/kg body wt. The liver was perfused via the portal vein employing a modified Mortimore’s method [ 151 in an open system (Fig. 1). The perfiusatewas Krebs-Henseleit solution [7] containing 0.54 pMEDTA (ethylendiaminetetra acetate),glucose (200 mg . dl-I), and heparin (1.7 U * ml-‘), and equilibrated with carbogen(95% O2 + 5% Cq). The perfusion was started not
II
u
before the pOz exceded 500 mm Hg, and temperature and pH (7.4) were well controlled. The osmolarity of the solution averaged 280 mosmol kg-‘. The portal vein was cannulated and connected to the perfusion system without interrupting the portal flow. The initial flow rate was set to 6-7 ml mini. Immediately thereafter the infrahepatic vena cava was cannulated, and the suprahepatic vena cava was ligated just cranial the diaphragma. The washout of blood was finished within 5 min. Then the flow rate was increased to 36 ml. min-’ (about 5 ml - min-’ . g-’ hepatic tissue). The portal pressureaveraged25 cm H20. The perfusion lasted 60 min without or with an additional insulin infusion, and was performed at 37,35, and 32”C, respectively. Insulin (MCInsulin NOVO Rapitard) was diluted with the perfusate to a concentration of 1.6 U - ml-’ just before the onset of the perfusion, and infused at a rate of 10 Us hr-‘. Every 15 min, perfusate (1 ml) was sampled from both the inflow and outflow line of the system and analysed for gaseswith the help of an ABL2 Acid Base Laboratory (Radiometer, Copenhagen).
,... .. . ..Y . .._......_. . . . . . . .
l
l
Puma /
pH Meter
Thermostat
FIG.
1. Schema of the perfusion system.
SHIMAHARA
71
AND ISSELHARD: RAT LIVER PERFUSION
From the remnant of the liver, mitochonThe oxygenconsumption of the liver was calculated from the O2 differencesand the flow dria werepreparedaspreviouslyreported[ 171. Their oxygen consumption was measured porate. After 60 min of perfusion, the left anterior larographically [ 161 with a Clark-type elecliver lobe was grasped and pressedto about 2 trode (Yellow Spring, Model 53, USA). The mm thickness between stainless-steel tongs precooled in liquid nitrogen. The frozen tissue was weighed; about 2 g was homogenized in 6 ml ice-cold 6% HC104 using a tissue homogenizer “Ultraturrax” (Janke u. Kunkel Co., Staufen, FRG). The homogenate was centrifuged at 10,OOOgfor 15 min at 0-4”C. The precipitate was washed with ice-cold 3% HC104 and centrifuged at 15,OOOg for 15 min at 0-4”C. The supernatants were pooled and adjusted to pH 6 with cold 20% (W/V) KOH and recentrifuged at 10,OOOgfor 5 min at O4°C. In these extracts, the concentrations of adenine nucleotides (ATP, ADP, AMP), and ketone bodies-acetoacetate (AcAc) and fihydroxybutyrate (@-OHB)-were measured enzymatically [ 1,9, 13,291. The sum of ATP, ADP, and AMP is referred to as total adenine nucleotides (TAN). The energy charge potential (ECP) was calculated according to Atkinson [2], ECP = (ATP + 0.5 ADP)/TAN. The ketone body ratio, which indicates the hepatic mitochondrial oxido-reduction state (NAD+/ NADH) was expressedas the ratio of AcAc to ,&OHB.
respiratory control ratio (RC = state 3 respiration rate/state 4 respiration rate), state 3 respiration rate, ADP/O, and oxidative phosphorylative activity (PR = state 3 X ADP/O) were calculated from the polarographical trao ings according to Chance [3]. These measurements were performed at 22°C using glutamate as substrate. Mitochondrial protein was measured by the method of Lowry et al. [ 1I]. For the measurement of the dry weight of the liver samples, about 400 mg of the wet tissue was dried for 4-5 days in a heat box at 110°C. Normal fasted rat livers served as controls in comparison to the perfused livers. The statistical significances between the mean values were determined by Student’s t test. RESULTS
ATP and TAN (Table 1) were at the end of the insulin-free perfusion least altered at 32°C with averages of 82 and 80% of the control values, respectively. The alterations in ATP and TAN increased with a rise in perfusion temperature, averaging 67 and 74% at 35”C,
TABLE ADENINE NUCLEOTIDE
Groups
Number of rats
1
CONTENTS (amole/g WET TISSUE) AND ENERGY CHARGE POTENTIAL AND PERFUSED LIVERS OF RATS FASTED 18 hr (MEAN + SEM)
ATP
ADP
AMP
ATP + ADP + AMP
(ECP) IN
CONTROL
ECP
(8)
3.33 * 0.10
1.03 * 0.07
0.21 + 0.02
4.57 + 0.16
0.842 + 0.008
(5) (5)
2.08 + 0.14*** 3.51 * 0.02c
1.13 f 0.09 0.76 f 0.09**-
0.52 f 0.18* 0.17 + 0.03
3.37 * 0.178’ 4.44 f 0.26”
0.7 13 * 0.047** 0.877 * o.oll**~
35°C Insulin (-) Insulin (+)
(5) (5)
2.23 f 0.20*** 2.83 f 0.08***p
0.90 + 0.09 0.90 + 0.09
0.25 + 0.07 0.19 -+ 0.02
3.37 f 0.22*** 3.92 k 0.06.”
0.793 + 0.028* 0.838 2 0.013
32°C Insulin (-) Insulin (+)
(5) (5)
2.73 f 0.19** 2.83 ?I 0.27*
0.77 f 0.09 0.85 f 0.03
0.15 + 0.01’ 0.17 + 0.01
3.65 + 0.26”; 3.85 f 0.26*
0.854 Ik 0.011 0.843 + 0.013
Control 37°C Insulin (-) Insulin (+)
a P < 0.05, *P < 0.01, 'P < 0.001 as compared between Insulin (-) and Insulin (+) groups. * P < 0.05, l *P < 0.01, ***P -c 0.001 as compared to the control values.
72
JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 1, JANUARY 1986 TABLE 2
KETONEBODY CONTENTS@mole/g WET TISSUE)AND RATIO DRY WEIGHT(dw)/WET WEIGHT(ww) IN CONTROL AND PERFUSEDLIVERSOFRATSFASTED18 hr (MEAN + SEM)
Groups
Number of rats
ACAC
j3-OHB
dw. 100 ww
Total
(8)
0.2 16 + 0.037
0.564 + 0.068
0.780 + 0.099
29.6 + 0.2
(5) (5)
0.038 zkO.OlO** 0.043 f 0.010**
0.049 f 0.016*** 0.015 5 0.003***~
0.087 k 0.026*** 0.057 f 0.013***
27.1 + 0.8’** 28.1 + 0.7*
35°C Insulin (-) Insulin (+)
(5) (5)
0.049 + 0.010** 0.059 5 0.009**
0.025 + 0.013*** 0.0 13 + 0.002***
0.074 + 0.022*** 0.072 + O.OlO***
28.6 + 0.2** 28.0 f 0.4***
32°C Insulin (-) Insulin (+)
(5) (5)
0.073 + 0.014** 0.048 + 0.005**
0.02 1 * 0.003*+* 0.013 k 0.005***
0.094 3I 0.012*** 0.06 1 + 0.007***
28.5 + 0.6* 28.4 + 0.8
Control 37°C Insulin (-) Insulin (+)
OP< 0.05ascompared between Insulin(-) andInsulin(+) groups. * P < 0.05, **P < 0.01, ***P 0.001 as compared to the control values.
and 62 and 74% at 37”C, respectively. With insulin, higher temperaturesbecamefavorable: At 32”C, there were no differences between groups without and with insulin. At 35”C, ATP and TAN amounted to 85 and 86%, respectively. At 37”C, these values averaged 105 and 97% of the controls, respectively. The differences between groups without or with insulin were significant at 35 and 37°C. The advantageous effect of insulin at 35 and 37°C also became apparent in the normal values of the ECP. AcAc and @-OHB (Table 2) were significantly decreasedat the end of the perfusion as compared to the control values. The decreasein &OHB was more pronounced than that in AcAc. There was no relevant influence of temperature or the admixture of insulin. The ketone body ratio AcAc/&OHB (Fig. 2) was increased significantly in comparison to the control range. This increase enlarged with hypothermia and was more pronounced in insulin-perfused livers. The mitochondrial phosphorylation rate (Fig. 3) as an indicator of mitochondrial ATP synthesis decreased significantly during perfusion without insulin to 62 and 72% of the control value at 37 and 35”C, respectively. The infusion of insulin suppressedthis alteration
maintaining the phosphorylation rate at 94% of the control value. At 32”C, the phosphorylation rate remained unchanged even without insulin infusion, and the effect of insulin was not significant. The mitochondrial respiratory control ratio (RC), which reflectsthe integrity of the isolated mitochondria, the state 3 respiration rate (ST 3) indicating the O2 consumption in the presence of sufficient ADP and substrate, and the
Control
3vc
35oc
32oC
FIG. 2. Ketone body ratios in control and perbed livers
of rats fasted 18 hr without and with an insulin infusion during perfusion (mean k SEM); ***P i 0.001 for perfusion values vs control value and for insulin (-) vs insulin (+) groups.
SHIMAHARA
73
AND ISSELHARD: RAT LIVER PERFUSION
The oxygen consumption of the liver remained constant during the entire perfusion period and averaged (mean + SEM) without/ with the application ofinsulin 56.5 + M/55.0 + 5.0 111g-r wet tissue. min-’ at 37”C, 49.5 + 2.8/45.8 f 4.4 ~1. g-r min-’ at 35°C and 4 1.2 f 4.4/46.4 + 1.8 ~1. g-’ . min-’ at 32°C respectively. The differences between groups without and with insulin infusion were not significant. The ratio dry weight/wet weight of hepatic tissue (Table 2) was in perfused livers reduced as compared to the control value. Again, there were no significant differences between groups without and with insulin infusion. l
l
0) it rol
(5)
(5)
37%
(5)
(5)
35%
(5)
(5)
32OC
FIG. 3. Phosphorylation rate of mitcchondria from DISCUSSION control and perfused livers of rats fasted 18 hr without and with an insulin infusion during perfusion (mean A permanent, sufficient energy supply is the + SEM); **P < 0.01 perfused vs control livers. 'P< 0.05, prerequisite for the maintenance of cellular 'P-z0.01 insulin (-) vs insulin (+) groups.
functions and integrity. In the liver, biosynthetic processes like gluconeogenesis, urearatio ADP/O (Table 3) were the more altered genesis, protein synthesis, bile formation as as compared to the control values the higher well as the many active membrane transport the perfusion temperature was. The changes processes require a sufficient ATP supply and were most pronounced in RC and least in sufficiently high ATP levels. An inadequate ADP/O. With insulin, the alterations were less energy supply will result in a progressive susmaintaining RC at 32°C and ST 3 at all tem- pension of functions, a progressive structural peratures within the control ranges. alteration, and eventually the irreversible cell TABLE 3 RESPIRATORY CONTROLRATIO (RC), STATE 3 RESPIRATION RATE (ST 3, n Atom O/mgprotein/min), ADP/O OF MITOCHONDRIA ISOLATEDFROMCONTROLAND PERFUSED LIVERY
ANDRATIO
OF RATS FASTED18 hr (MEAN f SEM)
Groups Control 37OC Insulin (-) Insulin (+)
Number of rats
RC
ST 3
ADP/O
(9)
4.86 + 0.25
30.2 f 2.1
2.15 It 0.08
(5) (5)
2.62 + 0.36*** 3.34 -t 0.20***
21.8 f 4.0* 31.8 -L 4.3
2.41 + 0.20 2.48 + 0.06*
35°C Insulin (-) Insulin (+)
(5) (5)
3.62 k 0.30** 3.95 + 0.18**
24.0 f 1.6’ 30.2 f 2.9
2.52 + 0.08 2.69 f 0.25
32°C Insulin (-) Insulin (+)
(5) (5)
4.22 -t OS%* 4.14 + 0.42
29.3 + 6.7 32.8 k 6.4
2.79 + 0.18 2.74 zk 0.11
* P < 0.05, **P c 0.01, ***P< 0.001 as compared to the control values. DitTerences between Insulin (-) and Insulin (+) groups were not significant.
74
JOURNAL OF SURGICAL RESEARCH: VOL. 40, NO. 1, JANUARY 1986
damageand death. On the background of these interrelations, it seems to be meaningful to evaluate the hepatic energy metabolism, when functional capacity and viability of the liver are discussed. Since the liver has only little glycolytic capacity, the mitochondrial performance is of particular interest. The concept of the adenylate energy charge potential (ECP) proposed by Atkinson [2] regardsthe adenine nucleotide system as the energy currency of the cell, which reflects not only a balance between energy-generating and energy-consuming reactions, but also a regulatory parameter in the energy metabolism. The ECP is maintained at a high and constant level under normal aerobic conditions, it decreases,if energy production does not meet the energy requirements. In various pathological states,it is a more sensitive parameter than the tissue level of TAN [20, 211. In the livers perfused at 37 and 35°C without insulin, the ECP decreasedsignificantly. Since no extrahepatic load was added in these experiments, this decreasemore likely reflects an impaired energy production than enhanced energyconsuming reactions. The data on mitochondrial functions support this view. The activity of mitochondria in vivo is reflected in the mitochondrial redox (oxido-reduction) state (NAD+/NADH), which according to Williamson et al. [28] can be expressed by the ratio of ketone bodies in the liver tissue (mitochondrial NAD+/NADH = hepatic (AcAc/@-OHB) X l/K, where K is the equilibrium constant of /3-hydroxybutyrate dehydrogenase.) The ketone body ratio in the liver in situ of normally fed rats averages about 1. Due to fasting the rats, it was decreasedin the control series. All perfused livers exhibited a significantly higher value of the ketone body ratio than the control livers in situ, becausethe tissue level of /?-OHB was more drastically reduced than that of AcAc. Becauseof the absence of free fatty acids in the perfusate, the amount of ketone bodies of the perfused livers was significantly lower than that of the liver in situ. The high redox state together with the low number of ketone bodies reflects a situation in which hepatic energy metabolism is
supported only by glucose oxidation, and which possibly is exaggeratedby the absence of a normal ketone body formation. The ketone body ratio in the 37”C-Insulin (-) group was markedly lower than in the other groups. This may reflect an insufficient energy production under the experimental conditions resulting in a particularly pronounced decreasein the ECP. A decreasein the ketone body ratio is generally considered to result from (1) mitochondrial impairment, (2) oxygen deficiency, or (3) fatty acid oxidation. Since the oxygen consumption rate of this group did not differ from the other groups, and free fatty acids were not supplied via the perfusate, this low ketone body ratio is most likely indicative of a mitochondrial impairment. In support of this view, the isolated mitochondria of this group showed the most pronounced decreasein the phosphorylation rate, the respiratory control ratio, and the state 3 respiration rate. From these findings it can be concluded that in this model the perfused liver at 37“C without insulin failed to maintain its viability due to an impairment of mitochondria. On the contrary, the liver perfusion at 37°C with insulin infusion revealed to be satisfactory: The tissue levels of adenine nucleotides and the ECP were within the normal ranges at the end of the 60-min perfusion period, the mitochondrial phosphoryiation remained at 94% of the control value, the ketone body ratio was significantly increased, the mitochondrial respiratory control ratio (RC) and the ADP/ 0 ratio were less depressed, and the mitochondrial state 3 respiration rate (ST 3) corresponded to the control value. The maintenance of a normal energy status in the normothermic erythrocyte-free perfused liver seems to result from a stimulation of mitochondrial activities by insulin. Several investigations deal with the effects of insulin on the mitochondrial oxidative phosphorylative activity. Matsubara et al. reported that mitochondria from livers of diabetic rats exhibited a marked deterioration, which was completely reversedby insulin [ 121. In that situation, the mitochondrial derangement was attributed to an accumulation of free
SHIMAHARA
AND ISSELHARD: RAT LIVER PERFUSION
fatty acids, which act as an uncoupler of mitochondrial oxidativephosphorylation;insulin was thought to exert its effectsby removing the fatty acids with an utilization of glucose. In the present experiment, however, the mechanism of insulin action on mitochondrial activities appearsto be different, since no fatty acids were included in the perfusate and the perfusion period wastoo short for the synthesis of a large amount of fatty acids. Ozawa et al. demonstrated that insulin stimulated the oxidative phosphorylation of liver mitochondria and considered insulin to be an important portal factor regulating the viability of the liver [ 18, 191.He proposed that insulin has a direct action on any single rate-limiting step of either the electron transport or phosphorylation system. Consistent with this, Ida et al. found in the perfused guinea pig liver that insulin enhanced the mitochondrial oxidative phosphorylation, and also suggestedthat this effect is linked to a regulative mechanism such as conformational or configurational changes in an organized membrane system of respiratory assemblies [6]. On the other hand, it cannot be neglected to elucidate the mechanism of insulin action on mitochondrial activity in terms of regulatory enzymes or substrate currency in the metabolic pathway. It is known that insulin effectively activates the pyruvate dehydrogenase complex [27], which exists in mitochondria and catalizes the reaction between pyruvate and acetyl Co-A. This phenomenon in the initial step of the TCA cycle will induce an acceleration of the TCA cycle and may be related to an enhanced mitochondrial activity, which would be in keeping with the elevation of the oxide-reduction state of the mitochondria by insulin infusion of the present study. However, the mechanism of the insulin effect on mitochondria is not completely elucidated. Reports that the addition of insulin to isolated mitochondria had no effect on the oxidative phosphorylation [ 12, 191 are op posed by reports on positive effects [4]. It is not clear how insulin enters mitochondria. Sakamoto et al. suggestedthat mitochondria have no specific recepter for insulin, and that
75
the effectsof insulin on mitochondria might be attributable to some unknown factor derived secondarilyfrom the binding site of the plasma membrane [23]. The metabolic deterioration and the depression of mitochondrial activity in livers perfused without insulin were progressively moderated by hypothermia of 35 and 32°C respectively. The decreasein ATP and ECP was less pronounced or suspended, the parameters of mitochondrial activity-phosphorylation rate, RC, ST 3, and ADP/Owere improved at 35°C and quasi-normal at 32°C respectively, and the ketone body ratio was progressively increased. These data can be explained with a hypothermia-induced reduction in hepatic energy demand. In keeping with this interpretation, the oxygen consump tion of the liver dropped with a decrease in temperature. Insulin again exerted a positive effect on the mitochondrial activities and the hepatic metabolic status at 35“C, although the tissue levels of ATP and TAN were not as completely maintained as in the insulin (+) group at 37°C. At 32°C the effectsof insulin were neglectible. These results indicate, that the action of insulin is diminished with a progressivedecreasein temperature. Perfusionswithout oxygen carrier have been reported by several authors, but most work in this field apparently revealed such a perfusion as inadequate compared with the use of erythrocytes. Sugano et al., however, demonstrated a well designed oxygen carrier-free perfusion system,and proved that it was satisfactory and applicable to a variety of investigations, if the perfusion flow rate exceeds 3 ml - min-’ . g-r of liver [ 251.Their experiment was performed at 32°C without insulin infusion. The results ofthe present study at 32°C are consistent with those. As to the doses of insulin, further investigation might be necessaryfor this system, although Ida et al. reported that 10 U. hr-’ were optimal to induce mitochondrial enhancement in an erythrocyte-containing perfusion system [6]. From the results it may be concluded that it is possible to maintain the liver of rats viable at normothermia or quasi-normothermia for
76
JOURNAL OF SURGICAL RESEARCH:
at least 60 min in terms of the hepatic tissue levels of adenine nucleotides and mitochondrial activities employing an oxygen carrierfree electrolyte medium in a simple, noncirculating perfusion system, if insulin is adequately administered. REFERENCES 1. Adam, H. Adenosin-5diphosphate and adenosin-5monophosphate. In H. U. Bergmeyer (Ed.), Methods ofEnzymatic Analysis, p. 573. New York: Academic Press, 1965. 2. Atkinson, D. E. The energy charge of the adenylate pool as a regulatory parameter, interaction with feedback modifiers. Biochemistry 7: 4030, 1968. 3. Chance, B. Quantitative aspectson the control of oxygen utilization. In W. Cew and C. M. G’Connor, (Eds.), Ciba Foundation Symposium on Regulation of Cell Metabolism. p. 9 1. London: Churchill, 1959. 4. Hall, J. C., Sordahl, L. A., and Stelko, P. L. The effect of insulin on oxidative phosphorylation in normal and diabeticmitochondria. J. Eiol. Chem. 235: 1536,1959. 5. Hems, R., Ross, B. D., Berry, M. N., and Krebs, H. A. Gluconeogenesis in the perfused rat liver. B&hem. J. 101: 284, 1966. 6. Ida, T., Sato, M., Yamaoka, Y., Takeda, H., Kamiyama, Y., Kimura, K., Gzawa, K., and Honjo, I. Effect of insulin on mitochondrial oxidative phosphorylation and energy charge of the perfused guinea pig liver. J. Lab. Clin. Med. 87: 925, 1976. 7. Krebs, H. A., and Hen&it, K Untersuchungentiber die HarnstofIbildtmg in Tierkorper. Z. Physiol. Chem. 210: 33, 1932. 8. Krebs, H. A., Patricia, G., Wallace, R., Hems, R., and Freadland, R. A. Rates of ketone body formation in the perfused rat Liver. B&hem. J. 112: 595, 1969. 9. Lamprecht, W., and Trautschold, I. Determination with hexokinase and glucose-6-phosphate dehydrogenase.In H. U. Bergmeyer (Ed.), Methods ofEnzymatic Analysis, p. 543. New York: Academic Press, 1965. 10. Lee, D., and Walker, J. M. Maintenance of the functional state of isolated rat liver by hypothermic perfusion with an erythrocyte-free medium. Transplantation 23: 136, 1977. 11. Lowry, 0. H., Rosebrough, N. J., Frarr, A. L., and Randall, R. J. Protein measurement with Frolin phs no1reagent. J. Biol. Chem. 193: 265, 1951. 12. Matsubara, T., and Tochino, Y. Depression of respiratory activities in the liver mitochondria of diabetic rats and the restorative action of insulin. J. Biochem. 66: 397, 1969. 13. Mellanby, J., and Williamson, D. H. Acetoacetate. In H. U. Bergmeyer (Ed.), Methods of Enzymatic Analysis, p. 1840. New York: Academic Press, 1974. 14. Miller, L. L., Bly, C. G., Watson, M. L., and Bale,
JOL.40, NO. 1, JANUARY 1986
W. F. The dominant role of the liver in plasma protein synthesis. A direct study of the isolated perfused rat liver with the aids of lysine-C’4. J. Exp. Med. 94: 43 1, 1951. 15. Mortimore, G. E., Tietze, F., and Stetten, D. Metabolism of insulin-I’31. Studies in isolated perfbsed rat liver and hind-limb preparations. Diabetes 8: 307, 1959. 16. Ozawa, K., Takasan, H., and Kitamura, 0. Effect of ligation of portal vein on liver mitochondrial metabolism. J. Biochem. 70: 755, 1971. 17. Ozawa, K., Kitamura, O., and Mizukami, T. Human liver mitochondria. Clin. Chim. Acta 38: 385, 1972. 18. Gzawa, K., Yamada, T., and Honjo, I. Role of insulin as a portal factor in maintaining the viability of liver. Ann. Surg. 180: 716, 1974. 19. Gzawa, K., Yamaoka, Y., Nanbu, H., and Honjo, I. Insulin as the primary factor governing changes in mitochondrial metabolism leading to liver regeneration and atrophy. Amer. J. Surg. 127: 669, 1974. 20. Gzawa, K., Ida, T., Kamano, J., Garbus, J., and Cowley, R. A. Different responseof hepatic energy charge and adenine nucleotides concentrations to hemorrhagic shock. Res. Exp. Med. 169: 145, 1976. 21. Gzawa, K., Yamaoka, Y., Kimura, K., Kimiyama, Y., Sato, M., Ukikusa, M., and Tobe, T. Circulating hepatodepressantfactors decreasingthe energy charge levels of the remnant liver after hepatectomy. Eur. Surg. Rex 13: 444, 1981. 22. Gzawa, K. Energy metabolism. In R. A. Cowley and B. F. Trump (Eds.),Pathophysiologyof Shock,Anoxia, and Ischemia. p. 74. Baltimore: Waverly, 1982. 23. Sakamoto, Y., Kuzuya, T., and Kawanishi, K. Study of insulin action: Uptake of insulin into subcellular fractions and stimulation of pyruvate dehydrogenase in mouse adipocytes. Hormon to Rinsho 29: 1061, 1981. 24. Shimahara, Y., Ozawa, K., Ida, T., Ukikusa, M., and Tobe, T. Four stagesof mitochondrial deterioration in hemorrhagic shock. Res. Exp. Med. 179: 23,198 1. 25. Sugano, T., Suda, K., Shimada, M., and Gshino, N. Biochemical and ultrastructural evaluation of isolated rat liver systemsperfusedwith a hemoglubin-free me dium. J. Biochem. 83: 995, 1978. 26. Ukikusa, M., Ida, Y., Ozawa, K., and Tobe, T. The influence of hypoxia and hemorrhage upon adenylate energy charge and bile tlow. Surg. Gynecol. Obstet. 149: 346, 1979. 27. Weiss,L., Liiffler, G., and Wieland, 0. H. Regulation by insulin of adipose tissue pyruvate dehydrogenase. Hoppe-Sqler’s Z. Physiol. Chem. 355: 363, 1974. 28. Williamson, D. H., Lund, P. A., and Krebs, H. A. The redox state of free nicotinamide-adeninedinucleotide in the cytoplasm and mitochondria of rat liver. Biochem. J. 103: 514, 1967. 29. Williamson, D. H., and Mellanby, J. j%Hydroxybutyrate. In H. U. Bergmeyer (Ed.), Methods of Enzymatic Analysis, p. 1836. New York: Academic Press, 1974.