Effects of method of preservation on functions of livers from fed and fasted rabbits

CRYOBIOLOGY

28, 227-236 (1991)

Effects of Method of Preservation on Functions of Livers from Fed and Fasted Rabbits KARIM

BOUDJEMA,

SUSANNE L. LINDELL, FOLKERT JAMES H. SOUTHARD

0. BELZER,

AND

Department of Surgery, Universi@ of Wisconsin, 600 Highland Avenue, Madison, Wisconsin 53792 Livers from fed, fasted (48 h) and glucose-fed rabbits were preserved for 24 and 48 h by either simple cold storage (CS) or continuous machine perfusion (MP) with the University of Wisconsin preservation solutions. After preservation liver functions were measured by isolated perfusion of the liver (at 37°C) for 2 h. Fasting caused an 85% reduction in the concentration of glycogen in the liver but no change in ATP or glutathione. Glucose feeding suppressed the loss of glycogen (3% loss). After 24 h preservation by CS livers from fed or fasted animals were similar including bile production (6.2 ? 0.5 and 5.6 2 0.4 ml/2 h, 100 g, respectively), hepatocellular injury (LDH release = %5 * 100 and 1049 + 2l34 Uihter), and concentrations of ATP (1.17 + 0.15 and 1.18 2 0.04 pmol/g, glutathione (1.94 2 0.51 and 2.35 + 0.26 umol/g, respectively), and K:Na ratio (6.7 f 1.0 and 7.7 + 0.5, respectively). After 48 h CS livers from fed animals were superior to livers from fasted animals including significantly more bile production (5.0 + 0.9 vs 2.0 2 0.3 ml/2 h, 100 g), less LDH release (1123 + 98 vs 3701 f 562 Uiliter), higher concentration of ATP (0.50 ? 0.16 vs 0.33 f 0.07 umol/g) and glutathione (0.93 ? 0.14 vs 0.30 2 0.13 pmol/g), and a larger K:Na ratio (7.4 vs 1.5). Livers from fed animals were also better preserved than livers from fasted animals when the method was machine perfusion. The decrease in liver functions in livers from fasted animals preserved for 48 h by CS or MP was prevented by feeding glucose. Glucose feeding increased bile formation after 48 h CS preservation from 2.0 f 0.3 (fasted) to 6.9 f 1.2 ml/2 h, 100 g; LDH release was reduced from 3701 f 562 (fasted) to 1450 f 154 Uihter; ATP was increased from 0.33 * 0.07 (fasted) to 1.63 + 0.18 pmol/g; glutathione was increased from 0.30 2 0.01 (fasted) to 2.17 + 0.30 umol/g; and K:Na ratio was increased from 1.5 f 0.9 to 5.3 f 1.0. This study shows that the nutritional status of the donor can affect the quality of liver preservation. The improvement in preservation by feeding rabbits only glucose suggests that glycogen is an important metabolite for successful liver preservation. Glycogen may be a source for ATP synthesis during the early period of reperlusion of preserved livers. 8 1991Academic press, hc.

Recently methods have been defined which allow preservation of the liver for 2 (8) to 3 days (19). Clinically, livers have been successfully preserved for up to 30 h (11, 29). However, in clinical liver transplantation the quality of preservation is variable and some livers have delayed function while others do not function at all, i.e., primary nonfunction (PNF), which requires urgent retransplantation. There does not appear to be a correlation between the degree of function of the transplanted liver

Received April 11, 1990; accepted August 14, 1990. ’ This work supported by a grant from the National Institutes of Health, DK 35143.

and time of preservation. In our center PNF has occurred in 7% of the patients with preservation times as short as 6 h and as long as 20 h. Many factors can affect the quality of a liver for transplantation including preservation time, preservation method, and condition of the donor or recipient. Even with the development of adequate methods to preserve the liver, the quality of preservation will depend upon the quality of the liver prior to procurement. Livers can be injured during the procurement, if exposed to a period of warm ischemia or if the donor is hypotensive. It is unclear how other donor factors (including the effects of brain death on the hormonal status of the patient (33) 227 OOll-2240/91 $3.00 Copyright All rights

0 1991 by Academic Ress, Inc. of reproduction in any form reserved.

228

BOUDJEMA

affect the quality of organs harvested for transplantation. One factor which has not received much attention is the nutritional status of the donor. Many donors have prolonged hospital stays prior to hepatectomy with little attention given to the nutritional needs of the donor. The liver is an organ sensitive to changes in caloric intake (20,22) which can cause decreases in liver glycogen (24), adenine nucleotides (10, 12), and glutathione (20). Thus, nutritionally depleted livers may be more sensitive to preservationreperfusion injury than livers from healthy donors. This could be due to a depletion in the concentration of substrates available for energy metabolism, loss of high energy compounds (ATP), a depletion in the concentration of endogenous antioxidants (glutathione, vitamin E), or the loss of other metabolites. In this study we have investigated the effects of nutritional status (fasting) of the donor on the quality of liver preservation. Two methods of preservation were used, simple cold storage (CS) and machine perfusion (MP). Cold storage was selected because it is the more common method of clinical liver preservation. Machine perfusion was chosen because previous studies have shown that supporting hypothermic metabolism by continuous perfusion yields longer term and better quality liver preservation than cold storage (19, 27). In this study rabbits were fed or fasted (48 h) and livers preserved for 24 or 48 h. In addition, one group of rabbits were fed only glucose for 48 h. Following preservation liver functions were assessed by isolated perfusion. MATERIALS

AND

METHODS

Animals. New Zealand white rabbits (2 to 3 kg) were used in these experiments and were allowed free access to food (standard laboratory diet) or fasted (no food only water) for 48 h. Glucose-fed rabbits were given a 40% glucose solution in place of

ET

AL.

water but no solid food. Rabbits given only glucose consumed from 150 to 500 ml of the 40% glucose solution. Liver preservation. Livers were preserved by either cold storage (YC) after flush out with the UW cold storage solution (lactobionate solution (32) or by machine perfusion with the UW machine perfusion solution (gluconate solution (15)) as described previously (9, 26). In the studies reported here, however, the donor rabbits were not pretreated with chlorpromazine or methylprednisolone. Rabbit livers were perfused continuously through the portal vein at a perfusion pressure of 15 to 25 torr. The perfusate was equilibrated with 100% oxygen (p0 2 > 400 torr) by surface aeration of the perfusate. Isolated perfusion. The method of isolated perfusion of rabbit livers has been previously described (7, 25). Briefly, rabbit livers were perfused at a flow of 3 ml/mitt, g with Krebs-Hanseleit buffer containing 5 g% bovine serum albumin. The temperature was maintained at 37°C and the perfusate equilibrated with 100% oxygen, 5% carbon dioxide with a membrane oxygenator. Bile was collected through a bile duct cannula. Perfusate was collected after 2 h for analysis of lactate dehydrogenase (LDH) and aspartate aminotransaminase (AST) activity. At the end of perfusion liver samples were frozen in dry ice-acetone and stored at -70°C for latter analysis of ATP and glutathione. Methods of analysis. LDH and AST activities were determined by enzymatic methods using Sigma diagnostic kits (Sigma No. 500 and No. 505, Sigma Diagnostics, St. Louis, MO). Bile was collected into graduated test tubes for volume measurement. ATP was determined on acidextracted liver tissue by high performance liquid chromatography as described (23). Glutathione was determined on acidextracted liver tissue by the method of Griffith (4) as described previously (31). This method determines both oxidized and re-

RABBIT

LIVER

FUNCTION

duced glutathione and results given are total tissue glutathione. Glycogen was determined by the enzymatic hydrolysis of glycogen in freshly obtained liver tissue homogenates or homogenates prepared from preserved liver tissue and by determination of glucose as described (3). The potassium and sodium content of liver tissue was determined by flame photometry after digestion of the tissue in nitric acid (13). Results were analyzed using the Student t test with significance assumed for P < 0.05. The following groups of livers were studied by reperfusion after preservation: Group 1 = CS, 24 h, fed; Group 2 = CS, 24 h, fasted; Group 3 = CS, 48 h, fed; Group 4 = CS, 48 h, fasted; Group 5 = MP, 24 h, fed; Group 6 = MP, 24 h, fasted; Group 7 = MP, 48 h, fed; Group 8 = MP, 48 h, fasted. RESULTS

Effects of Cold Preservation

Alone

The effect of fasting for 48 h on the concentrations of glycogen, ATP, and~glutathione in rabbit livers is shown in Table 1. Glycogen was decreased by 85% in fasted rabbit livers. There were no significant changes in the concentrations of ATP or TABLE 1 Metabolite Concentrations in Livers from Fed and Fasted Rabbits

Fed (N = 3) Fasted (N = 3) Glucose (N = 3)

Glycogen b3k)

ATP (kmo%)

Glutathione (wnoW

52 f 8

1.64 2 0.08

2.21 f 0.18

8+3

1.66 f 0.08

2.19 f 0.02

32 + 9

N.D.

N.D.

Note. Rabbits were fed, fasted for 48 h, or fed only glucose for 48 h (glucose). The liver samples were obtained prior to interruption of blood flow and frozen in dry ice-acetone for latter determination of metabolites as described under Materials and Methods. Values are means (per g wet wt) and standard error of the means. N.D., not determined.

PRESERVATION

EFFECTS

229

glutathione. Feeding only glucose for 48 h suppressed the loss of glycogen in the liver caused by fasting, (32 + 9 mg/g with glucose vs 8 f 3 mg/g with fasting). The effect of the method of preservation on the ATP concentration in livers preserved for up to 48 h is shown in Table 2. ATP decreased rapidly in livers preserved by CS. After 6 h CS ATP was reduced from about 1.6 bmol/g to 0.27 pmol/g (fed) and 0.07 pmol/g (fasted). ATP decreased further after 24 h CS and after 48 h CS was 3% or less of Time 0 values. The ATP concentrations were higher in livers after MP than those after CS. In fed livers the ATP concentration decreased to about 62% of control values after 6 and 24 h MP. After 48 h MP of livers from fed rabbits there was an increase in the ATP concentration to 130% (2.14 Fmol/g) of the 0 time concentration. This increase is most likely due to the availability of adenosine in the pet&sate. In fasted livers there was a decrease in ATP to 32% (6 h), 25% (24 h), and 11% (48 h) during MP even though adenosine was present in the perfusate. In the remaining studies to be described liver function and liver metabolite values were obtained in livers after 2 h isolated reperfusion. We did not repeat control (no preservation time) experiments but did determine liver functions and liver metabolites after 6 h preservation. Table 3 compares our previous control data (7) with results obtained after 6 h preservation of the rabbit liver. In this series bile production was greater after 6 h preservation and 2 h isolated reperfusion than that obtained in our previous controls, 2 h isolated perfusion of freshly procured livers. LDH release (and AST, results not shown) was slightly higher in 6-h-preserved livers than in control livers. Preservation by Simple Cold Storage (CS) and Normothermic Reperfusion The results in Table 4 show the effects of cold storage on liver function (b-de produc-

230

BOUDJEMA

ET AL.

TABLE 2 ATP Concentration in Rabbit Livers after Cold Storage or Machine Perfusion ATP (umol/g) Fed Time (h) 0 6 24 48

Fasted

cs 1.64 0.27 0.09 0.05

+ f + +

MP 0.08 0.08 0.03 0.01

1.64 1.01 1.02 2.14

f + 2 k

cs 0.08 0.12 0.45 0.10

1.66 0.07 0.03 0.02

f 2 2 f

MP 0.08 0.01 0.02 0.01

1.66 0.53 0.42 0.18

2 f ” +

0.08 0.21 0.17 0.04

Note. The ATP concentration was determined in livers after preservation by cold storage (CS) or machine perfusion (MP) as described under Materials and Methods. Three livers were analyzed for each group and the results presented are the means with standard error of the means.

tion), hepatocellular injury (LDH and AST release), and liver metabolic capabilities (ATP, glutathione, and K:Na) after 24 and 48 h preservation. After 24 h preservation of livers from both fed and fasted groups, there were no significant differences in bile production, LDH or AST release, glutathione concentration, or K:Na. Only the concentration of ATP was reduced in livers from fasted rabbits and was about 33% of the concentration in livers from fed rabbits after 24 h storage and 2 h perfusion. After 48 h preservation, however, there was a significant increase in liver injury in livers from fasted animals. Bile production was reduced by greater than 50% and there TABLE 3 Liver Functions on Reperfusion in Control and 6-h CS Rabbit Livers Functions Bile (ml/2 h, 100 f3) LDH (Units/liter) ATP (pmol/g) K:Na Glutathione OlmoW

Control 9.8 2 2.4 180 ” 40 N.D. N.D. 2.43 + 0.16

6hCS 14.7 325 1.61 9.2

f f 2 +

3.2 77 0.37 1.6

2.51 f 0.26

Note. Control values were obtained from previous studies in our laboratory (16). In this study rabbit livers were CS for 6 h in the UW solution and reperfused for 2 h at 37°C as described under Materials and Methods. Values are means and standard error of the means is obtained from at least three animals in each group.

was a significant increase in hepatocellular injury as indicated by the large release of LDH and AST in livers from fasted rabbits versus livers from fed rabbits. There was also a significant reduction in the concentrations of ATP and glutathione and lower K:Na in livers from fasted rabbits versus livers from fed rabbits. Preservation by Machine Perfusion and Normothermic Reperfusion

(MP)

The results in Table 5 show the effects of machine perfusion on liver function, hepatocellular injury, and liver metabolic capabilities after 24 and 48 h preservation. After 24 h preservation livers from fed rabbits exhibited better function and showed less injury than those from fasted rabbits. This included a significantly greater production of bile in livers from fed rabbits which was similar to the amount of bile produced in control or 6-h E&preserved livers. There was also less hepatocellular injury as indicated by the lower amount of LDH and AST released from the livers of fed rabbits versus those of fasted rabbits. The ATP concentrations were similar after 24 h MP and isolated perfusion. However, glutathione concentration was also significantly lower and the K:Na ratio was lower in livers from fasted rabbits versus fed rabbits. Livers from fed rabbits also showed significantly better functions after 48 h preser-

RABBIT

LIVER FUNCTION

PRESERVATION

231

EFFECTS

TABLE 4 Liver Functions and Metabolites in Fed and Fasted Rabbit Livers Preserved by CS and Reperfused Group

Animal

Time (W

Bile

LDH

Fed (N = 3)

24

6.2 f 0.5

%5 f 100

121 f 9

24 48

5.6 + 0.4 5.0 + 0.9

1049 + 284 1123 + 98

48

2.0 + 0.3*

3701 + 562*

1 2

Fasted

3 4

(N = 3) Fed (N = 3) Fasted (N = 3)

AST

ATP

Glutathione

K:Na

1.17 + 0.5

1.94 + 0.51

6.7 + 1.0

107 + 9 154 f 20

0.39 + 0.1* 0.082 + 0.16

1.82 2 0.28 0.93 5 0.14

5.8 + 1.1 7.4 f 0.6

749 + 171*

0.326 f 0.07*.**

0.30 f 0.01*

1.5 + 0.9*

Note. Livers from fed or fasted rabbits were CS for 24 or 48 h and functions tested by isolated perfusion for 2 h as described under Materials and Methods. Bile = mV2 h, 100 g; LDH = units/liter, 100 g; AST = Uiliter, 100 g. ATP and glutathione = p,mol/g. Values are means and standard error of means. Statistical analysis was by Student’s t test and differences were assumed if P < 0.05. *Statisticallv different when comoared to fed livers preserved for the same period of time. **Statistically different when compared to livers CS for the same peAod of time.

vation than livers from fasted rabbits. This included greater bile production, less LDH and AST release, a greater concentration of ATP and glutathione, and a larger K:Na.

better than those obtained in livers from fed rabbits, including bile production, ATP, and glutathione . Livers from glucose-fed rabbits were also significantly better preserved after 48-h preservation than livers from fed rabbits. This included greater bile production, a greater concentration of ATP and glutathione, and a larger K:Na.

Effects of Glucose Feeding on Liver Preservation by CS and Normothermic Perfusion Rabbits were fed only glucose for 48 h and the livers preserved for 48 h by either CS (Fig. 1) or MP (Fig. 2). The results in the figures were obtained after 2 h isolated perfusion and are compared to results obtained in livers from fed or fasted rabbits. Feeding rabbits glucose had a dramatic effect on the quality of preservation of the rabbit liver. Livers from glucose-fed rabbits preserved by CS (Fig. 1) produced significantly more bile than livers from fasted rabbits, had less LDH and AST release, produced significantly more ATP and glutathione, and had a greater K:Na. In fact, many liver functions in this group of livers were

Effects of Glucose Feeding on Liver Preservation by MP and Normothermic Perfusion Livers from glucose-fed rabbits preserved by MP were also superior to livers from fasted rabbits (Fig. 2). This included greater bile production, less LDH and AST release, greater synthesis of ATP and glutathione, and a greater K:Na. Bile production after 48-h MP of livers from glucosefed rabbits was, in fact, equivalent to the amount of bile produced by 6-h-preserved

TABLE Liver Group 5 6

Functions

and Metabolites

Animal

Time 00

Fed Fasted Fed Fasted

24 24 48 48

in Fed and Fasted

Bile 14.4 5.7 6.2 1.1

e f f f

5 Rabbit

LDH 0.9 3.9* 0.4 0.1*

300 3700 223 6538

f + f +

Livers

Preserved

AST 51 600’ 50 1216*

29 303 35 725

-t f + +

9 51’ 10 213*

1.18 1.37 1.37 0.37

by MP and Reperfusion

ATP

Glutathione

+ + f f

2.35 0.06 2.71 0.45

0.04 0.29 0.31 0.12*,**

-t + f f

0.26 0.5* 0.16 0.131

K:Na 7.7 2.9 8.3 1.5

+ f f f

0.5 0.8’ 0.8 0.9*

Note. Procedures are identical to those described in the legend to Table 3. ATP and glutathione = pmol/g. ** Statistically diierent when compared to livers CS for the same period of time. *Statistically different when compared livers from fed rabbits.

to

232

BOUDJEMA

25

37.0

r

r

al-

1 0

Fasted

-7

% ?

15-

f g

lo-

a 5-

Osue

LDH

AST

ATP

KlNa

GSH

1. Liv rer functions and metabolite concentration in 48-h CS livers from fed, fasted, and glucose-fed rabbits. Rabbit livers were cold stored for 48 h and functions and metabolites determined by isolated perfusion as described under Materials and Methods. Three livers were analyzed per group. Relative value refers to the following units: Bile = ml/2 h, g; LDH = units x lO*/liter; AST = units x IO/liter; ATP = umol x 10-‘/g; K:Na = ratio as indicated; and GSH = glutathione as pmol x 10-‘/g. The number 37.0 on the graft indicates the mean value for LDH release which exceeded the units on the graft. Bars are means with standard error of the means. FIG.

or control rabbit livers suggesting minimal preservation injury in these livers. Comparison of MP versus CS Livers from fed rabbits were better preserved by MP than CS as shown by a comparison of the results in Tables 4 and 5. This included greater bile production (14.4 vs 6.2 ml/2 h, 100 g at 24 h and 6.2 vs 5.0 ml/2 h, 100 g at 48 h for MP and CS, respectively). There was also significantly less enzyme release from livers MP than those CS (LDH

AST

ATP

K/Na

GSli

FIG. 2. Liver functions and metabolite concentrations in 48-h MP livers from fed, fasted, and ghrcosefed rabbits. Rabbit livers were machine perfused for 48 h and functions and metabolites determined by isolated perfusion as described in the legend to Fig. 1. Presentation of results and units of measurements are identical to that described in the legend to Fig. 1.

ET AL.

= 300 vs 965 U/liter, 100 g at 24 h and 223 vs 1123 U/liter at 48 h for-MP and CS, respectively). After 24-h preservation there was little differences in ATP, glutathione, or K:Na in livers MP or CS. However, there was a significant increase in these values in livers MP for 48 h vs those CS 48 h, including ATP (1.37 vs 0.50 pmol/g) and glutathione (2.71 vs 0.93 pmol/g). The K:Na values were, however, similar (8.3 vs 7.4). Livers from fasted rabbits were, in general, less well preserved by MP than by CS and this is opposite to the effects obtained with livers from fed rabbits. After 24-h preservation bile production was similar for livers CS or MP. However, after 48 h, bile production was slightly greater in livers CS (2.0 + 0.3 ml/2 h, 100 g) than those MP (1.1 -+ 0.1 ml/2 h, 100 g) and the difference was significant. There was a large difference in the degree of hepatocellular injury in livers from fasted rabbits CS or MP. Livers MP released more LDH (3700 vs 1049 U/liter at 24 h and 6538 vs 3701 U/liter at 48 h for CS versus MP, respectively). Although there was greater ATP synthesis in livers from fasted rabbits preserved for 24 h which were MP versus those CS, the glutathione concentration was greater in livers CS (1.82 vs 0.86 Fmol/g) as was the K:Na (5.8 vs 2.9). This suggests that livers from fasted rabbits CS were better able to carry out a biosynthetic reaction (glutathione synthesis) and ion pump activity (K:Na) after rewarming. These differences were not apparent after 48-h preservation and similar concentrations of ATP, glutathione, and K:Na were obtained in livers CS or MP. DISCUSSION

In clinical liver transplantation, liver function following reimplantation can vary from immediate function to no function (PNF). Some livers have delayed function which can contribute to a decrease in long term graft survival (5). The exact causes of

RABBIT

LIVER

FUNCTION

PNF or delayed function are not known but do not appear related to preservation time per se and are often thought to relate to donor or recipient conditions. One donor condition recently shown to be a contraindication for liver transplantation is a fatty liver (28). Another donor condition that could affect the quality of liver preservation and postoperative liver function is the nutritional status of the donor. Liver donors often remain in the hospital for days before liver procurement. Little attention has been given, in the past, to the nutritional support of these donors which could result in donors being essentially starved prior to liver procurement. A period of starvation or fasting causes changes in the liver, particularly a depletion of glycogen which can decrease by 90% within 48 h (24). In the rat there is also a decrease in the concentration of ATP (10) and glutathione (22). In this study we have investigated how fasting affected the quality of liver preservation. In previous studies we have shown that the isolated perfused liver is a good model for determining preservation injury, bile production and enzyme release into the perfusate are reasonably good indicators of the degree of preservation injury (7, 9, 26). In this study livers from fed or fasted rabbits showed a decrease in liver functions after 24 h CS, including a decrease in bile production and an increase in release of LDH and AST compared to freshly harvested and perfused livers. The decreases in liver functions were similar in livers from both fed and fasted rabbits. Therefore, fasting does not induce liver injury after only 24 h preservation in this model. However, after 48 h CS livers from fasted donors showed greater injury than livers from fed rabbits. Bile production was reduced by about 50% and LDH release was increased by about threefold indicating hepatocellular injury. Additionally, livers from fed rabbits regenerated greater amounts of ATP and glutathione and maintained a high K:Na after 2 h perfusion than livers from fasted

PRESERVATION

EFFECTS

233

rabbits. Thus, fasting sensitizes livers to preservation injury after 48 h preservation by CS. Machine perfusion is a better method to preserve livers than cold storage (19, 27) and in this study livers from fed rabbits which were MP retained a better functional profile than livers CS after both 24 and 48 h preservation. This included greater bile production, less enzyme release, greater resynthesis of ATP and glutathione, and a higher K:Na. Livers from fasted rabbits were less well preserved than livers from fed rabbits after both 24 and 48 h when the method of preservation was MP. Bile production was less, there was a greater release of enzymes into the perfusate, ATP and glutathione concentrations were lower, and the K:Na was lower in fasted versus fed rabbit livers. Thus, fasting increases the sensitivity of the liver to preservation injury after 24 h preservation when MP is the method of preservation. We determined the concentrations of three metabolites in livers from fed and fasted rabbits, glycogen, ATP, and glutathione. Only glycogen content decreased and after 48 h fasting was reduced by 85% from control values. When rabbits were fed only glucose for 48 h liver glycogen was decreased by only about 35%. Livers from glucose-fed rabbits tolerated preservation by both MP and CS for 24 to 48 h and liver functions were similar or better than those obtained with livers from fed rabbits. Thus, it appears that the liver content of glycogen may be an important factor in successful liver preservation. In recent studies it was shown that livers and hepatocytes from fasted rats were more sensitive to anoxic or ischemic injury than livers or hepatocytes from fed animals (1,2, 14, 18). However, when fructose, a source of anaerobic glycolytic production of ATP, was administered to the livers or hepatocytes from fasted animals, ischemic and anoxic injury were reduced (1, 30). Thus, it was concluded that the generation of ATP

234

BOUDJEMA

during the injury period was important in maintaining cell viability. Fasting was considered, therefore, to reduce the amount of glycogen available for anaerobic ATP synthesis and led to increased liver injury due to ischemia or anoxia. However, the role of glycogen in increasing the tolerance of the liver to warm and cold ischemic injury may not be due to the production of ATP during the period of injury but the production of ATP during the initial period of reperfusion. In CS the concentration of ATP at the end of preservation in both fed and fasted rabbit livers was low (1 to 3%) after 24 and 48 h preservation. However, livers from fed rabbits retained greater liver functions and less hepatocellular injury during perfusion. In livers from fed rabbits reperfusion stimulated ATP production and ATP increased from 0.09 to 1.17 Fmol/g after 24 h preservation and perfusion, respectively, and from 0.05 to 0.5 Fmol/g after 48 h preservation and perfusion, respectively. In livers from fasted animals ATP after perfusion was only 0.39 and 0.33 p,mol/g after 24 and 48 h preservation, respectively. This suggests that glycogen may be an important source of substrates for glycolysis and aerobic ATP synthesis during reperfusion of preserved livers. That fact that glycogen can be a source of ATP in the liver is also suggested from the effects of MP on the ATP content of livers from fed and fasted rabbits. During machine perfusion the concentration of ATP in livers from fasted rabbits continuously decreased and after 48 h was 0.18 kmol/g. However, in livers from fed rabbits the concentration of ATP after 48 h was 2.14 pmol/g. In this study livers from both fed and fasted rabbits were perfused with an identical perfusate containing glucose (10 rrN) which suggests that glucose is not an ideal substrate for ATP production in the hypothermically perfused rabbit liver. This has also been shown in hepatocytes and livers at normothermia (17). Also, Polombo et

ET AL.

al. (16) have shown that glucose is not ef-

fectively metabolized in the cold-stored rat liver. The results suggest, therefore, that fasting depletes liver glycogen which is necessary for the rapid regeneration of ATP during the initial period of perfusion of coldpreserved livers. It is unclear why the liver would require glycogen as a source of substrates for ATP synthesis and not utilize other sources such as fatty acids which normally are the primary substrates for oxidative metabolism. One explanation is that preservation injury is partly due to a disruption of the capabilities of the liver to effectively use fatty acids for ATP synthesis. This could occur by a depletion of the cofactors required for fatty acid oxidation. Finally, we noted that livers from fasted rabbits were better preserved by CS than by MP. This is contrary to the effects seen in livers from fed rabbits. In these livers MP was superior to CS as a method of preservation. Why MP would increase preservation injury in livers from fasted rabbits but not fed rabbits is not clear. However, the difference may be due to the lack of glycogen in livers from fasted rabbits. As discussed above livers from fed animals produced ATP during MP but livers from fasted rabbit livers lost ATP. During machine perfusion metabolism is supported by the constant delivery of oxygen and substrates to the liver and removal of end products of metabolism. Thus, during machine perfusion a source of ATP may be important for the maintenance of active metabolism and this source appears to be glycogen. The lack of ATP during machine perfusion may decrease the activity of the membrane ion pumps leading to cell swelling or cause activation of degradative enzymes which injure the cell. This study shows that fasting the donor affects the quality of liver preservation in the rabbit. How fasting would affect the quality of liver preservation in other species is not clear. The ultimate test of the

RABBIT

LIVER FUNCTION

effects of variables, such as fasting, on liver preservation is liver function and animal survival in the orthotopic transplant model. Recently we have shown that livers from fasted pigs are more sensitive to preservation injury than livers from fed pigs, as tested in the orthotopic transplant model. Pigs receiving livers from fed donors and preserved for 12 h had 83% survival (96). Pigs receiving livers from fasted donors preserved for 12 h had only 50% survival (3/6). Thus, it appears that in both the isolated perfusion model of liver preservation and in the orthotopic transplant model, fasting increases the sensitivity of the liver to preservation injury.

PRESERVATION

9.

10.

11.

12.

REFERENCES 1. Anundi, I., King, J., Owen, D. A., Schneider, N., Lemasters, J. J., and Thurman, R. J. Fructose prevents hypoxic cell death in liver. Am. J. Physiol. 253, G39O-G396 (1987). 2. Becker, G. L., Hensel, P., Holland, A. D., Miletich, D. J., and Albrecht, R. F. Energy deficits in hepatocytes isolated from phenobarbitaltreated or fasted rats and briefly exposed to halothane and hypoxia in vitro. Anesthesiology 65, 379-384 (1986). 3. Carr, R. S., and Neff, J. M. Quantitative semiautomated enzymatic assay for tissue glycogen. Comp. Biochem. Physiol. 77, 447-449 (1984). 4. Griffith, 0. W. Determination of glutathione and glutathione disultide using glutathione reductase and 2-vinylpyridine. Anal. Biochem. 106, 207-212 (1980). 5. Howard, T. K., Klintmalm, B. G., Cofer, J. B., Husberg, B. S., Gordstein, R. M., and Gouwa, T. A. The influence of preservation injury or rejection in the hepatic transplant recipient. Transplantation 49, 103-107 (1989). 6. Jaeger, R. J., Conolly, R. B., and Murphy, S. D. Effect of 18 h fasting and glutathione depletion on I-dichloroethylene-induced hepatotoxicity and lethality in rats. Exp. Mol. Pathol. 20, 187198 (1974). 7. Jamieson, N. V., Lindell, S., Sundberg, R., Southard, J. H., and Belzer, F. 0. An analysis of the components in the UW solution using the isolated perfused rabbit liver. Transplantation 46, 512-516 (1988). 8. Jamieson, N. V., Sundberg, R., Lindell, S., Southard, J. H., and Belzer, F. 0. Preservation of the canine liver for 24-48 hours using simple

13.

14.

15.

16.

17.

18.

19.

20.

EFFECTS

235

cold storage with UW solution. Transplantation 46, 517-525 (1988). Jamieson, N. V., Sundberg, R., Lindell, S., Southard, J. H., and Belzer, F. 0. A comparison of cold storage solutions for hepatic preservation using the isolated perfused rabbit liver. Cryobiology 25, 300-310 (1988). Jennische, E. Effects of ischemia on the hepatic cell membrane potential in the rat: Differences between fed and fasted animals. Acta Physiol. Stand. 118, 69-73 (1983). Kalayoglu, M., Sollinger, H. W., Stratta, R. J., D’Alessandro, A. M., Hoffman, R. M., Pirsch, J. D., and Belzer, F. 0. Extended preservation of the liver for clinical transplantation. Lancer 2, 617-619 (1988). Kashiwagura, T., Wilson, D. F., and Erecinski, M. Oxygen dependence of cellular metabolism: The effect of 0, tension on gluconeogenesis and urea synthesis in isolated rat hepatocytes. .I. Cell Physiol. 120, 13-18 (1984). Little, J. R. Determination of water and electrolytes in tissue slices. Anal. Biochem. 7, 87-95 (1964). Marotto, M. S., Thurman, R. G., and Lemasters, J. J. Early midzonal cell death during low-flow hypoxia in the isolated, perfused rat liver: Protection by allopurinol. Hepatology 8, 585-590 (1988). McAnulty, J. F., Ploeg, R. J., Southard, J. H., and Belzer, F. 0. Successful five-day perfusion preservation of the canine kidney. Trunsplanration 47, 37-41 (1989). Palombo, J. D., Hirschberg, Y., Pomposelli, J. J., Blackbum, G. L., Zeisel, S. H., and Bistrian, B. R. Decreased loss of liver adenosine triphosphate during hypothermic preservation in rats pretreated with glucose: Implications for organ donor management. Gastroenterology 95, 1043-1049 (1988). Palombo, J. D., Pomposelli, J. J., Hirschberg, Y., Blackbum, G. L., and Bistrian, B. R. Glycolytic support of adenine nucleotides in rat liver flushed-preserved with UW or Collins’ II: Importance of donor nutritional status. Transplnnration 48, 901-905 (1989). Pessayre, D., Dolder, A., and Antigon, J. Y. Effect of fasting on metabolite-mediated hepatoxicity in the rat. Gastroenterology 77, 264-271 (1979). Pienaar, B. H., Lindell, S. L., van Gulik, T. M., Southard, J. H., and Belzer, F. 0. 72-hour preservation of the canine liver by machine perfusion. Transplantation, in press. Schumer, W., Kuttner, R. E., Sugai, T., Yamashita, K., and Apantaku, L. M. Hepatic glyco-

236

21.

22.

23.

24.

25.

26.

27.

BOUDJEMA

lytic intermediates in fed and fasted rats after severe hemorrhage. J. Trauma 26, 1009-1012. Seglen, P. 0. Autoregulation of glycolysis, respiration, gluconeogenesis, and glycogen synthesis in isolated parenchymal rat liver cells under aerobic and anaerobic conditions. Biochem. Biophys. Acta 338, 317-336 (1974). Shi, E. C. P., Fisher, R., McEvoy, M., Vantol, R., Rose, M., Ham, J. M. Factors influencing hepatic glutathione concentrations: A study in surgical patients. C/in. Sci. 62, 279-283 (1982). Southard, J. H., Lutz, M. F., Amentani, M. S., and Belzer, F. 0. Stimulation of ATP synthesis in hypothermically perfused dog kidneys by adenosine and phosphate. Cryobiology 21, 13-19 (1984). Start, C., and Newsholme, E. A. The effects of starvation and alloxan-diabetes on the content of citrate and other metabolic intermediates in the rat liver. Biochem. J. 107, 411-415 (1%8). Sundberg, R., Ar’Rajab, A., Ahren, B., and Bengmark, S. Improvements of liver preservation quality with UW solution by chlorpromazine pretreatment of the donor in an experimental model. Transplantation 48, 742-744 (1989). Sundberg, R., Lindell, S., Jamieson, N. V., Southard, J. H., and Belzer, F. 0. Effects of chlorpromazine and methylprednisone on perfusion preservation of rabbit livers. Cryobiology 25, 417-429 (1988). Tamaki, T., Kamada, N., Wight, D. G. W., and

ET AL.

28.

29.

30.

31.

32.

33.

Pegg, D. E. Successful 48-hour preservation of the rat liver by continuous hypothermic perfusion with haemaccel-isotonic citrate solution. Transplantation 43, 468-471 (1987). Todo, S., Demetia, A. J., Makowkg, L., Teperman, L., Pudestia, L., Shaver, T., Tzakis, A., and Starzite, T. E. Primary nonfunction of hepatic allografts with preexisting fatty inliltrations. Transplantation 47, 903-905 (1989). Todo, S., Nery, J., Yanaga, K., Podesta, L., Gordon, R. O., and Starzl, T. E. Extended preservation of human liver grafts with UW solution. J. Am. Med. Assoc. 261, 711-714 (1989). Vind, C., and Grunnet, N. The reversibility of cytosolic dehydrogenase in hepatocytes from starved and fed rats: Effect of fructose. Biothem. J. 222, 437-446 (1984). Vreugdenhil, P. K., Evans, W., Belzer, F. O., and Southard, J. H. Glutathione depletion in cold stored organs. Transplant Proc. (1990). Wahlberg, J. A., Southard, J. H., and Belzer, F. 0. Development of a cold storage solution for pancreas preservation. Cryobiology 23,477483 (1986). Wicomb, W. N., Cooper, D. K. C., Lanza, R. P., Novitzky, D., and Isaacs, S. The effects of brain death and 24 hours storage by hypothermic perfusion of donor heart function in the pig. J. Thorac. Cardiovasc. Surg. 91, 896-909 (1986).