- Email: [email protected]
Reperfusion Injury
33, 430–435 (1996) 0043
CRYOBIOLOGY ARTICLE NO.
Comparison of Isolated Hepatocytes and Tissue Slices for Study of Liver Hypothermic Preservation/Reperfusion Injury1 PAUL K. VREUGDENHIL, DIANE C. MARSH,
AND
JAMES H. SOUTHARD
Department of Surgery, University of Wisconsin, Madison, Wisconsin Simple models are needed that effectively test the variables that may be important in liver preservation. Two such models are isolated hepatocytes and tissue slices. In this study the effects of hypothermic preservation on the viability of hepatocytes (HC) and tissue slices (TS) from rat livers were measured by LDH leakage after cold storage and rewarming. We compared how glycine, calcium, and fasting, shown previously to affect preservation injury in hepatocytes, affected both HC and TS viability. Hepatocytes were cold-stored in University of Wisconsin organ preservation solution for up to 48 h and rewarmed in Krebs–Henseleit Bicarbonate (KHB) for 120 min. Tissue slices were studied in two ways. Either livers were cold-stored intact and then tissue slices (TS-A) prepared and rewarmed in KHB, or tissue slices were prepared from the fresh liver, cold-stored, and then rewarmed (TS-B). The latter method may be similar to cold storage of HC. Freshly prepared samples (HC, TS-A, or TS-B) showed õ15% LDH leakage during the rewarming phase. Cold storage for 24 h resulted in õ30% LDH leakage in all preparations. After 48 h cold storage there was a significant increase in LDH leakage (HC, 65.1 { 5.1%; TS-A, 52.9 { 0.8%, TS-B, 53.6 { 2.6%). Glycine (3 mM) or calcium (1.5 mM) included in the KHB significantly reduced LDH leakage from 48 h cold-stored HC to 20.7 { 1.8 and 26.3 { 2.4%, respectively. These agents caused a smaller decrease in LDH release from tissue slices (around 40%). Hepatocytes appear more susceptible to preservation/reperfusion damage than the more structurally intact tissue slices as suggested by the greater release of LDH. Another difference was that the agents which improved preservation quality of HC were not as effective in TS. Hepatocytes may be more vulnerable to preservation/reperfusion damage because of the harsh methods used in their preparation. The damage induced during preparation appears amenable to suppression by glycine or calcium. Tissue slices, which are intact pieces of liver tissue, may be more suitable for studies related to development of better methods for liver preservation. The intact cells in TS have not been exposed to harsh conditions and maintain a more natural cell–cell relationship. q 1996 Academic Press, Inc.
Optimal utilization of cadaveric organs for modifications affect tissue viability. The usetransplantation requires methods for success- fulness of this approach is dependent upon the ful organ preservation. Although current prac- reliability and simplicity of the model used to tice using simple cold storage (CS) with the test the various preservation solutions. A model University of Wisconsin organ preservation commonly used and considered as clinically solution (UW solution) is quite reliable for relevant is orthotopic transplantation in the dog most organs, better methods could lower the or pig. Success is judged by survival and organ incidence of preservation/reperfusion (P/R) function. Rat liver transplantation has also injury which occurs in 6 to 15% of trans- found widespread use (9, 23, 24). However, these models are complex, expensive, and labor planted livers (19). One goal in transplantation research is the intense, and do not allow rapid or efficient testdevelopment of improved preservation solu- ing of the myriad variables that could improve tions. A common approach to attain this goal organ preservation. To overcome these shortinvolves modification of the composition of comings, simplified models have been develcurrent preservation solutions to test how these oped including tissue slices, isolated or cultured cells, and isolated perfused organs. Isolated cell suspensions (hepatocytes, HC) and tissue slices Received September 19, 1995; accepted May 9, 1996. (TS) are simple models that allow performance 1 This work was supported by NIH Grant DK35143. of a large number of experiments in a relatively 430 0011-2240/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
CRYO 1957
/
a903$$$$61
07-18-96 22:53:33
cryoa
AP: CRYO
431
LIVER HYPOTHERMIC PRESERVATION/REPERFUSION INJURY
short period of time. Their utility in organ preservation studies has been demonstrated previously (1, 5, 7, 17) and discussed in various reviews (18, 21). We have used both systems (HC and TS) in our attempts to understand mechanisms of P/R injury to the liver and to develop improved methods of organ preservation (2, 12, 22). Recently, studies from a number of groups have suggested that P/R injury is more likely due to disruption of the structural organization of the tissue than injury to cell metabolism (10, 15, 16). Although cell metabolism was adversely affected by P/R, this injury appears more readily reversible than structural forms of injury. Some studies suggest that disruption of the cell cytoskeleton (4, 6) was a major component in irreversible ischemic/reperfusion injury. The disruption of the cytoskeleton was thought to involve deregulation of cell calcium and oxygen free radical metabolism. Studies to improve organ preservation therefore may require use of models in which the cytoskeleton and ultrastructure of the tissue are maintained. Thus, isolated or cultured cells (HC) may be less suitable for these studies than more structured tissue, such as tissue slices. Tissue slices may be hardier than HC. Slices are intact pieces of tissue which maintain relatively normal architecture including cell-to-cell contact and cell-to-cell communication. They have not been exposed to the disruptive conditions of isolation. Hepatocytes, on the other hand, may be more vulnerable to injury than slices due to the severity of the isolation conditions which include a brief exposure to calcium-deficient media and possible periods of hypoxia. Consequently, cytoprotective agents may be beneficial to the more vulnerable HC than the more intact tissue slices during periods of injury (i.e., ischemia) and reperfusion (i.e., rewarming). The greater vulnerability of HC versus TS to P/R injury was suggested from previous studies describing the cytoprotective effects of glycine in CS of the liver (3, 13, 14). We have shown that glycine, in the rewarming
AID
CRYO 1957
/
a903$$$$62
media, dramatically improved the viability (LDH leakage) of isolated hepatocytes. However, when we tested glycine in reperfusion of cold-stored rat livers, as well as in orthotopic dog liver transplantation, it showed little effect. This suggested to us that the beneficial effects of glycine in P/R injury may be dependent upon the model used. Thus we have carried out this study to investigate the relationship between HC and TS and the suppression of P/R injury by glycine. Calcium (Ca) has also been shown to be cytoprotective to CS hepatocytes (25), and the effects of Ca on HC and TS were also studied. MATERIALS AND METHODS
Animals. Livers from chloral hydrate (7.2%, ip)-anesthetized Sprague–Dawley rats (Harlan–Sprague–Dawley, Madison, WI) weighing 300 to 325 g were used for preparation of HC or TS. The rats were either fed a conventional laboratory diet or fasted (water ad libitum) for 24 h prior to use. Hepatocytes. Hepatocytes were isolated by a modification of the collagenase digestion method of Seglen (20) as previously described by Marsh et al. (12). After isolation, the hepatocytes were either suspended in a modified Krebs–Henseleit buffer (mKHB), containing NaCl (118 mM) KCl (4.7 mM) MgCl2 (1.2 mM) glucose (11.1 mM), NaHCO3 (10 mM), and HEPES (20 mM) at 377C and pH 7.4, or placed in cold (47C) UW solution. The UW solution had been bubbled with nitrogen (100%) for 10 min at 47C prior to use for suspension of HC. In some experiments the mKHB also contained CaCl2 (1.5 mM) or glycine (3 mM). Hepatocytes suspended in UW solution were cold-stored at 47C for 24 or 48 h without agitation to simulate the anoxic cold storage of a whole liver. After CS the HC were agitated, sedimented by centrifugation (600g, 3 min), and resuspended in mKHB with or without CaCl2 or glycine. Reperfusion was carried out on HC in mKHB by incubation for 120 min at 377C under an atmosphere of room air. Protein (Pc) was measured by the Biuret reaction us-
07-18-96 22:53:33
cryoa
AP: CRYO
432
VREUGDENHIL, MARSH, AND SOUTHARD
ing bovine serum albumin as a standard. All HC suspensions were adjusted to 5 mg/ml Pc for both CS and reperfusion. Cell viability was determined by LDH leakage into the medium (12). Total cellular LDH was measured in the supernate from ultrasonically disrupted cells. Tissue slice preparation. Livers were flushed out immediately with 25 ml cold (47C) UW solution through the portal vein and used for tissue slice preparation either immediately or after CS of the liver for 24 and 48 h. Tissue slices prepared immediately after flushout of the liver were either suspended in mKHB (controls, for LDH leakage) or suspended in UW solution (equilibrated with nitrogen) and stored at 47C for 24 or 48 h. This group is identified as TS-B. When the whole liver was CS intact, tissue slices were prepared after 24 or 48 h. This group is identified as TS-A. Tissue slices weighing on average 0.1 g each were made using a Stadie Rigg’s tissue slicer (22). For reperfusion studies, slices (five per flask containing 5 ml mKHB) were given an initial 15-min incubation at 377C under an atmosphere of 95% O2 and 5% CO2 and shaken at 100 rpm (LabLine Orbit Shaker Bath). The initial incubation was to remove any LDH from broken cells as well as other cellular debris resulting from sample preparation. The slices were then placed in fresh mKHB for 120 min and LDH was measured in the incubation medium (free LDH) and in the tissue slices (bound LDH) which were homogenized in fresh mKHB with a Potter–Elvehjem-type Teflon pestle tissue grinder (22). All reagents were obtained from Sigma Chemical Co. (St. Louis, MO). Collagenase (type D) was obtained from Boehringer Mannheim Biochemicals (Indianapolis, IN). Data are presented as means { SEM from four to eight livers. Groups were compared using the Students t test (InStat; GraphPad Software, Inc., San Diego, CA) with P õ 0.05 indicating significance. RESULTS
In this study P/R injury was assessed by the amount of LDH that leaked out of HC or TS
AID
CRYO 1957
/
a903$$$$62
FIG. 1. Effect of cold storage on LDH release from HC, TS-A, and TS-B. Methods of preparation, cold storage, and rewarming of hepatocytes (HC), tissue slices prepared from cold-stored livers (TS-A), and tissue slices coldstored for 24 or 48 h (TS-B) are described under Materials and Methods. After preservation for the indicated time the samples were rewarmed for 120 min in mKHB with or without glycine (3 mM) or CaCl2 (1.5 mM). Values are means with vertical bars showing standard error of the mean. *P õ 0.05 vs 24-hr control; †P õ 0.05 vs 48hr control.
during 2 h of normothermic (377C) incubation. LDH leakage was significantly greater after 48 h than 24 h cold storage for each of the preparations analyzed (Fig. 1). After 48 h cold storage the total LDH release had increased from approximately 15% (0 h preservation, for TS or HC) to 65.1 { 5.1% (HC), 51.8 { 0.9% (TS-A), and 53.6 { 2.6% (TS-B). The amount released from HC was greater than that from TS, although only significant for TSA (P õ 0.05). Therefore, HC in suspension appear slightly more sensitive to P/R injury than TS. Another difference between preparations was the extent of cytoprotection by glycine or calcium (Fig. 1). Greater cytoprotection was seen in HC than in the other preparations. In 48-h CS HC, the cytoprotective agents, when added to the resuspension medium, lowered LDH release by 40 to 45%. Calcium significantly decreased LDH release from 65.1 { 5.1 to 26.3 { 2.4% (P õ 0.05) and glycine to 20.7 { 1.8% (P õ 0.05), after 48 h cold storage. In the presence of these agents the amount of LDH released was similar to that in freshly isolated HC (14.4 { 1.4%, n Å 6).
07-18-96 22:53:33
cryoa
AP: CRYO
433
LIVER HYPOTHERMIC PRESERVATION/REPERFUSION INJURY
FIG. 2. Effect of nutritional status of the rat on viability of HC or tissue slices (TS-A) after 24 or 48 hr cold storage. Methods are described under Materials and Methods and in the legend to Fig. 1. Rats were either fed a standard laboratory diet or fasted overnight prior to liver harvest. *P õ 0.05 vs fed; †P õ 0.05 vs fasted.
In both TS preparations the extent of suppression of LDH leakage by glycine or calcium (6 to 14% reduction) was not as dramatic as in HC (40 to 45%) after 48 h cold storage (Fig. 1). LDH leakage was reduced by glycine from about 52 to 37.9 { 1.4% (TS-A) and 42.4 { 1.4% (TS-B) and by calcium to 45.6 { 1.5% (TS-A) and 38.9 { 2.2% (TS-B). The % LDH released in the presence of glycine or calcium after 48 h cold storage was significantly greater than in 0-h preserved TS (15.1 { 1.1, n Å 6). An even more dramatic difference in the extent of P/R injury between HC and TS was seen when livers from fasted rats were used (Fig. 2). Hepatocytes from 24-h-fasted rats showed significantly greater P/R injury than HC from fed rats after 24 h (LDH leakage 82.5 { 2.8 versus 27.7 { 5.9%, P õ 0.05) and 48 h CS (86.6 { 2.7 versus 65.1 { 5.1%, P õ 0.05). However, TS from fasted rats showed a similar response to P/R injury as TS from fed rats (Fig. 2). Glycine suppressed P/ R injury in HC from fasted rats but had little effect on P/R injury in TS from fasted rats. DISCUSSION
This study originated from our observations that glycine suppressed reperfusion injury in cold-stored HC but was less effective in mod-
AID
CRYO 1957
/
a903$$$$62
els using the whole organ or in orthotopic liver transplantation. For instance, we showed that HC cold stored in the UW solution for up to 6 days released a large amount of LDH on reperfusion. However, reperfusion with glycine (3 mM) in the rewarming medium significantly suppressed P/R injury (13). This suggested that glycine could improve liver preservation/transplantation. However, when we tested glycine in the isolated perfused liver or liver transplantation models the results were disappointing and glycine showed little benefit (3). Glycine, at various concentrations, did not improve significantly the quality or extend the safe duration of liver preservation. We concluded that the beneficial effects of glycine may be specific to the model (isolated cells) and may not be cytoprotective in more structurally intact systems. This conclusion was strengthened by the demonstration of cytoprotection by glycine in isolated renal proximal tubules exposed to hypoxia or toxic chemicals (26, 27). However, there was a lack of effect in more clinically relevant intact tissue models (8, 11) where glycine was only marginally protective in the cold-stored kidney or the kidney exposed to experimental warm ischemia/reperfusion. Weinberg et al. (28) suggested that the concentration of glycine normally present in whole kidney tissue is high and therefore additional protection by glycine in the reperfusion medium would not be readily seen. Glycine was only cytoprotective in cells that had lost glycine through the processes of tissue preparation. During the isolation of cells (renal tubules or HC) there could be a loss of glycine. Consequently, the addition of glycine to the suspending medium would provide cytoprotection. In this study we compared the cytoprotective effects of glycine, calcium, and fasting in HC and liver TS. We consider TS to be more representative of the whole liver. They have relatively normal cell–cell contact, contain an intact cytoskeleton anchored to the plasma membrane and other cells, and have not been exposed to extensive washing procedures in
07-18-96 22:53:33
cryoa
AP: CRYO
434
VREUGDENHIL, MARSH, AND SOUTHARD
the preparation. Thus, how P/R affects injury to TS may be more closely related to events occurring in the intact tissue used for transplantation. Hepatocytes, on the other hand, have undergone extensive washing and exposure to collagenase and low calcium medium, and do not maintain a normal cell–cell relationship. The hepatocyte suspension, therefore, may be more sensitive to P/R injury because of the stresses of isolation. Culturing HC prior to use may stimulate repair of the injury caused by the isolation and give a more hardy preparation; however, the lack of naturally occurring interactions between cells as well as other changes induced by culturing (i.e., exposure to enriched culture medium) were considered to preclude our interest in using these as accurate models of P/R injury in the liver. The results show some significant differences in sensitivity of hepatocytes to P/R injury compared to tissue slices. These differences were true for slices prepared from coldstored livers (TS-A) or slices prepared from fresh livers and then cold-stored (TS-B). There was a greater release of LDH from coldstored HC than from TS during rewarming, suggesting greater sensitivity of membranes of isolated cells to P/R injury than intact tissue. This difference was especially noticeable when HC from fasted rats were compared with TS from fasted rats. Fasting caused a large increase in the sensitization of HC to preservation injury after 24 h cold storage. However, fasting showed no adverse effect on how TS responded to P/R. Finally, glycine and calcium showed a significant and extensive cytoprotection of HC and suppressed the P/R-induced release of LDH after 48 h cold storage. The degree of cytoprotection in TS was noticeably less extensive. We conclude, therefore, that the methods used for the preparation of HC sensitize the membranes to P/R injury to a greater degree than occurs in intact tissue preparations. Changes that occur during their preparation make it possible to demonstrate cytoprotection by some agents in HC that are less effective
AID
CRYO 1957
/
a903$$$$62
when tested in intact models of P/R, such as TS or the whole liver. Caution should be used in judging how preservation and cytoprotective agents affect viability of HC exposed to cold preservation/rewarming, realizing that what is effective in these preparations may not be effective in more intact models of liver preservation. Suspensions of HC or cultured cells are a valuable model for studying liver metabolism; however, the disruption of the normal tissue architecture that is inherent in their preparation may limit the usefulness of these preparations, especially if disruption of tissue architecture is a key event in injury to the liver. The irreversible injury caused by cold preservation of the liver may be more related to changes in the structural organization (cytoskeleton, microtubules, membranes, vascular organization) of the tissue than to injury to the metabolic machinery. Thus, models that maintain not only the metabolic competency but also the structural integrity of the liver may be better for developing improved methods of organ preservation. Tissue slices may, therefore, be more suitable for some studies in organ preservation than isolated hepatocyte suspensions. This study is not a direct comparison of how preservation affects HC versus TS, but of how preservation maneuvers affect cell viability in two different models. The results suggest that preservation maneuvers that are beneficial in the HC preparation (calcium, glycine) are not of significant benefit when the TS model is used. This difference may be important in future experiments designed to study methods to improve preservation of the liver. REFERENCES 1. D’Alessandro, A. M., Southard, J. H., Kalayoglu, M., and Belzer, F. O. Comparison of cold storage and perfusion of dog livers on function of tissue slices. Cryobiology 23, 161–167 (1986). 2. D’Alessandro, A. M., Southard, J. H., Kalayoglu, M., and Belzer, F. O. Effect of drug treatment on liver slice function following 72 hour hypothermic perfusion. Cryobiology 23, 415–521 (1986). 3. den Butter, G., Lindell, S. L., Gandolf, D., Schilling, M. K., Southard, J. H., and Belzer, F. O. The effect
07-18-96 22:53:33
cryoa
AP: CRYO
435
LIVER HYPOTHERMIC PRESERVATION/REPERFUSION INJURY
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
of glutathione and glycine in kidney preservation. Transplant. Proc. 25, 1633–1634 (1993). Ganote, C., and Armstrong, S. Ischaemia and the myocyte cytoskeleton: Review and speculation. Cardiovasc. Res. 27, 1387–1403 (1993). Guyomard, C., Chesne, C., Meunier, B., Fautrel, A., Clerc, C., Morel, F., Rissel, M., Campion, J., and Guillouzo, A. Primary culture of adult rat hepatocytes after 48-hour preservation of the liver with cold UW solution. Hepatology 12, 1329–1336 (1990). Hall, S. M., Evans, J., and Haworth, S. G. Influence of cold preservation on the cytoskeleton of cultured pulmonary arterial endothelial cells. Am. J. Respir. Cell Mol. Biol. 9, 106–114 (1993). Hawkins, H. E., Clark, P., Lippert, E. C., and Karow, A. M. Functional preservation of the mammalian kidney. J. Surg. Res. 38, 281–288 (1985). Heyman, S. N., and Epstein, F. H. Glycine and cytopreservation: Disappointments and promises. J. Lab. Clin. Med. 121, 199–200 (1993). Kamada, N., and Calne, R. Y. A surgical experience with five hundred thirty liver transplants in the rat. Surgery 93, 64–69 (1983). Lemasters, J. J., Bunzendahl, H., and Thurman, R. G. Reperfusion injury to donor livers stored for transplantation. Liver Transplant. Surg. 1(2), 124– 138 (1995). Mangino, M. J., Murphy, M. K., Grabau, G. G., and Anderson, C. B. Protective effects of glycine during hypothermic renal ischemia-reperfusion injury. Am. J. Physiol. 261, F841–F848 (1991). Marsh, D. C., Lindell, S. L., Fox, L. E., Belzer, F. O., and Southard, J. H. Hypothermic preservation of hepatocytes. I. Role of cell swelling. Cryobiology 26, 524–534 (1989). Marsh, D. C., Hjelmhaug, J. A., Vreugdenhil, P. K., Belzer, F. O., and Southard, J. H. Glycine prevention of cold ischemic injury in isolated hepatocytes. Cryobiology 28, 105–109 (1991). Marsh, D. C., Vreugdenhil, P. K., Mack, V. E., Belzer, F. O., and Southard, J. H. Glycine protects hepatocytes from injury caused by anoxia, cold ischemia and mitochondrial inhibitors, but not injury caused by calcium ionophores or oxidative stress. Hepatology 17, 91–98 (1992). Marzi, I., Zhong, Z., Lemasters, J. J., and Thurman, R. G. Evidence that graft survival is not related to parenchymal cell viability in rat liver transplantation. The importance of nonparenchymal cells. Transplantation 48, 463–468 (1989). McKeown, C. M. B., Edwards, V., Phillips, M. J., Harvey, P. R. C., Petrunka, C. N., and Strasberg,
AID
CRYO 1957
/
a903$$$$62
17.
18.
19.
20. 21. 22.
23.
24.
25.
26.
27.
28.
S. M. Sinusoidal lining cell damage: the critical injury in cold preservation of liver allografts in the rat. Transplantation 46(2), 178–191 (1988). Michell, I. D., Chipman, J. K., and McMaster, P. The isolated hepatocyte preservation model: A comparison of hypertonic citrate and lactobionate solutions. Transplant. Proc. 21, 1312–1313 (1989). Page, R. A., Stowell, K. M., Hardman, M. J., and Kitson, K. E. The assessment of viability in isolated rat hepatocytes. Anal. Biochem. 200, 171– 175 (1992). Ploeg, R. J., D’Alessandro, A. M., Knechtle, S. J., Stegall, M. D., Pirsch, J. D., Hoffmann, R. M., Sasaki, T., Sollinger, H. W., Belzer, F. O., and Kalayoglu, M. Risk factors for primary dysfunction after liver transplantation: A multivariate analysis. Transplantation 55, 123–129 (1993). Seglen, P. O. Preparation of isolated rat liver cells. Methods Cell Biol. 13, 30–83 (1976). Southard, J. H. Viability assays in organ preservation. Cryobiology 26, 232–238 (1989). Southard, J. H., Kuniyoshi, M., Lutz, M. F., Ametani, M., and Belzer, F. O. Comparison of the effects of 3- and 5-day hypothermic perfusion of dog kidneys on metabolism of tissue slices. Cryobiology 21, 285–295 (1984). Sumimoto, R., Jamieson, N. V., Wake, K., and Kamada, N. 24-hour rat liver preservation using UW solution and some simplified variants. Transplantation 48, 1–5 (1989). Sumimoto, R., Lindell, S., Gambiez, L., Southard, J. H., and Belzer, F. O. The effect of pretransplant flushout of the liver with a solution containing albumin on survival and organ function. Transplant. Proc. 25, 3001–3003 (1993). Umeshita, K., Monden, M., Fujimori, T., Sakai, T., Gotoh, M., Okamura, J., and Mori, T. Extracellular calcium protects cultured rat hepatocytes from injury caused by hypothermic preservation. Cryobiology 25, 102–109 (1988). Weinberg, J. M., Davis, J. A., Abarzua, M., and Rajan, T. Cytoprotective effects of glycine and glutathione against hypoxic injury to renal tubules. J. Clin. Invest. 80, 1446–1454 (1987). Weinberg, J. M., Davis, J. A., Abarzua, M., Kiani, T., and Kunkel, R. Protection by glycine of proximal tubules from injury due to inhibitors of mitochondrial ATP production. Am. J. Physiol. 258, C1127– C1140 (1990). Weinberg, J. M., Nissim, I., Roeser, N. F., Davis, J. A., and Schultz, S. Relationships between intracellular amino acid level and protection against injury to isolated proximal tubules. Am. J. Physiol. 260, F410–F419 (1991).
07-18-96 22:53:33
cryoa
AP: CRYO
CRYOBIOLOGY ARTICLE NO.
Comparison of Isolated Hepatocytes and Tissue Slices for Study of Liver Hypothermic Preservation/Reperfusion Injury1 PAUL K. VREUGDENHIL, DIANE C. MARSH,
AND
JAMES H. SOUTHARD
Department of Surgery, University of Wisconsin, Madison, Wisconsin Simple models are needed that effectively test the variables that may be important in liver preservation. Two such models are isolated hepatocytes and tissue slices. In this study the effects of hypothermic preservation on the viability of hepatocytes (HC) and tissue slices (TS) from rat livers were measured by LDH leakage after cold storage and rewarming. We compared how glycine, calcium, and fasting, shown previously to affect preservation injury in hepatocytes, affected both HC and TS viability. Hepatocytes were cold-stored in University of Wisconsin organ preservation solution for up to 48 h and rewarmed in Krebs–Henseleit Bicarbonate (KHB) for 120 min. Tissue slices were studied in two ways. Either livers were cold-stored intact and then tissue slices (TS-A) prepared and rewarmed in KHB, or tissue slices were prepared from the fresh liver, cold-stored, and then rewarmed (TS-B). The latter method may be similar to cold storage of HC. Freshly prepared samples (HC, TS-A, or TS-B) showed õ15% LDH leakage during the rewarming phase. Cold storage for 24 h resulted in õ30% LDH leakage in all preparations. After 48 h cold storage there was a significant increase in LDH leakage (HC, 65.1 { 5.1%; TS-A, 52.9 { 0.8%, TS-B, 53.6 { 2.6%). Glycine (3 mM) or calcium (1.5 mM) included in the KHB significantly reduced LDH leakage from 48 h cold-stored HC to 20.7 { 1.8 and 26.3 { 2.4%, respectively. These agents caused a smaller decrease in LDH release from tissue slices (around 40%). Hepatocytes appear more susceptible to preservation/reperfusion damage than the more structurally intact tissue slices as suggested by the greater release of LDH. Another difference was that the agents which improved preservation quality of HC were not as effective in TS. Hepatocytes may be more vulnerable to preservation/reperfusion damage because of the harsh methods used in their preparation. The damage induced during preparation appears amenable to suppression by glycine or calcium. Tissue slices, which are intact pieces of liver tissue, may be more suitable for studies related to development of better methods for liver preservation. The intact cells in TS have not been exposed to harsh conditions and maintain a more natural cell–cell relationship. q 1996 Academic Press, Inc.
Optimal utilization of cadaveric organs for modifications affect tissue viability. The usetransplantation requires methods for success- fulness of this approach is dependent upon the ful organ preservation. Although current prac- reliability and simplicity of the model used to tice using simple cold storage (CS) with the test the various preservation solutions. A model University of Wisconsin organ preservation commonly used and considered as clinically solution (UW solution) is quite reliable for relevant is orthotopic transplantation in the dog most organs, better methods could lower the or pig. Success is judged by survival and organ incidence of preservation/reperfusion (P/R) function. Rat liver transplantation has also injury which occurs in 6 to 15% of trans- found widespread use (9, 23, 24). However, these models are complex, expensive, and labor planted livers (19). One goal in transplantation research is the intense, and do not allow rapid or efficient testdevelopment of improved preservation solu- ing of the myriad variables that could improve tions. A common approach to attain this goal organ preservation. To overcome these shortinvolves modification of the composition of comings, simplified models have been develcurrent preservation solutions to test how these oped including tissue slices, isolated or cultured cells, and isolated perfused organs. Isolated cell suspensions (hepatocytes, HC) and tissue slices Received September 19, 1995; accepted May 9, 1996. (TS) are simple models that allow performance 1 This work was supported by NIH Grant DK35143. of a large number of experiments in a relatively 430 0011-2240/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
CRYO 1957
/
a903$$$$61
07-18-96 22:53:33
cryoa
AP: CRYO
431
LIVER HYPOTHERMIC PRESERVATION/REPERFUSION INJURY
short period of time. Their utility in organ preservation studies has been demonstrated previously (1, 5, 7, 17) and discussed in various reviews (18, 21). We have used both systems (HC and TS) in our attempts to understand mechanisms of P/R injury to the liver and to develop improved methods of organ preservation (2, 12, 22). Recently, studies from a number of groups have suggested that P/R injury is more likely due to disruption of the structural organization of the tissue than injury to cell metabolism (10, 15, 16). Although cell metabolism was adversely affected by P/R, this injury appears more readily reversible than structural forms of injury. Some studies suggest that disruption of the cell cytoskeleton (4, 6) was a major component in irreversible ischemic/reperfusion injury. The disruption of the cytoskeleton was thought to involve deregulation of cell calcium and oxygen free radical metabolism. Studies to improve organ preservation therefore may require use of models in which the cytoskeleton and ultrastructure of the tissue are maintained. Thus, isolated or cultured cells (HC) may be less suitable for these studies than more structured tissue, such as tissue slices. Tissue slices may be hardier than HC. Slices are intact pieces of tissue which maintain relatively normal architecture including cell-to-cell contact and cell-to-cell communication. They have not been exposed to the disruptive conditions of isolation. Hepatocytes, on the other hand, may be more vulnerable to injury than slices due to the severity of the isolation conditions which include a brief exposure to calcium-deficient media and possible periods of hypoxia. Consequently, cytoprotective agents may be beneficial to the more vulnerable HC than the more intact tissue slices during periods of injury (i.e., ischemia) and reperfusion (i.e., rewarming). The greater vulnerability of HC versus TS to P/R injury was suggested from previous studies describing the cytoprotective effects of glycine in CS of the liver (3, 13, 14). We have shown that glycine, in the rewarming
AID
CRYO 1957
/
a903$$$$62
media, dramatically improved the viability (LDH leakage) of isolated hepatocytes. However, when we tested glycine in reperfusion of cold-stored rat livers, as well as in orthotopic dog liver transplantation, it showed little effect. This suggested to us that the beneficial effects of glycine in P/R injury may be dependent upon the model used. Thus we have carried out this study to investigate the relationship between HC and TS and the suppression of P/R injury by glycine. Calcium (Ca) has also been shown to be cytoprotective to CS hepatocytes (25), and the effects of Ca on HC and TS were also studied. MATERIALS AND METHODS
Animals. Livers from chloral hydrate (7.2%, ip)-anesthetized Sprague–Dawley rats (Harlan–Sprague–Dawley, Madison, WI) weighing 300 to 325 g were used for preparation of HC or TS. The rats were either fed a conventional laboratory diet or fasted (water ad libitum) for 24 h prior to use. Hepatocytes. Hepatocytes were isolated by a modification of the collagenase digestion method of Seglen (20) as previously described by Marsh et al. (12). After isolation, the hepatocytes were either suspended in a modified Krebs–Henseleit buffer (mKHB), containing NaCl (118 mM) KCl (4.7 mM) MgCl2 (1.2 mM) glucose (11.1 mM), NaHCO3 (10 mM), and HEPES (20 mM) at 377C and pH 7.4, or placed in cold (47C) UW solution. The UW solution had been bubbled with nitrogen (100%) for 10 min at 47C prior to use for suspension of HC. In some experiments the mKHB also contained CaCl2 (1.5 mM) or glycine (3 mM). Hepatocytes suspended in UW solution were cold-stored at 47C for 24 or 48 h without agitation to simulate the anoxic cold storage of a whole liver. After CS the HC were agitated, sedimented by centrifugation (600g, 3 min), and resuspended in mKHB with or without CaCl2 or glycine. Reperfusion was carried out on HC in mKHB by incubation for 120 min at 377C under an atmosphere of room air. Protein (Pc) was measured by the Biuret reaction us-
07-18-96 22:53:33
cryoa
AP: CRYO
432
VREUGDENHIL, MARSH, AND SOUTHARD
ing bovine serum albumin as a standard. All HC suspensions were adjusted to 5 mg/ml Pc for both CS and reperfusion. Cell viability was determined by LDH leakage into the medium (12). Total cellular LDH was measured in the supernate from ultrasonically disrupted cells. Tissue slice preparation. Livers were flushed out immediately with 25 ml cold (47C) UW solution through the portal vein and used for tissue slice preparation either immediately or after CS of the liver for 24 and 48 h. Tissue slices prepared immediately after flushout of the liver were either suspended in mKHB (controls, for LDH leakage) or suspended in UW solution (equilibrated with nitrogen) and stored at 47C for 24 or 48 h. This group is identified as TS-B. When the whole liver was CS intact, tissue slices were prepared after 24 or 48 h. This group is identified as TS-A. Tissue slices weighing on average 0.1 g each were made using a Stadie Rigg’s tissue slicer (22). For reperfusion studies, slices (five per flask containing 5 ml mKHB) were given an initial 15-min incubation at 377C under an atmosphere of 95% O2 and 5% CO2 and shaken at 100 rpm (LabLine Orbit Shaker Bath). The initial incubation was to remove any LDH from broken cells as well as other cellular debris resulting from sample preparation. The slices were then placed in fresh mKHB for 120 min and LDH was measured in the incubation medium (free LDH) and in the tissue slices (bound LDH) which were homogenized in fresh mKHB with a Potter–Elvehjem-type Teflon pestle tissue grinder (22). All reagents were obtained from Sigma Chemical Co. (St. Louis, MO). Collagenase (type D) was obtained from Boehringer Mannheim Biochemicals (Indianapolis, IN). Data are presented as means { SEM from four to eight livers. Groups were compared using the Students t test (InStat; GraphPad Software, Inc., San Diego, CA) with P õ 0.05 indicating significance. RESULTS
In this study P/R injury was assessed by the amount of LDH that leaked out of HC or TS
AID
CRYO 1957
/
a903$$$$62
FIG. 1. Effect of cold storage on LDH release from HC, TS-A, and TS-B. Methods of preparation, cold storage, and rewarming of hepatocytes (HC), tissue slices prepared from cold-stored livers (TS-A), and tissue slices coldstored for 24 or 48 h (TS-B) are described under Materials and Methods. After preservation for the indicated time the samples were rewarmed for 120 min in mKHB with or without glycine (3 mM) or CaCl2 (1.5 mM). Values are means with vertical bars showing standard error of the mean. *P õ 0.05 vs 24-hr control; †P õ 0.05 vs 48hr control.
during 2 h of normothermic (377C) incubation. LDH leakage was significantly greater after 48 h than 24 h cold storage for each of the preparations analyzed (Fig. 1). After 48 h cold storage the total LDH release had increased from approximately 15% (0 h preservation, for TS or HC) to 65.1 { 5.1% (HC), 51.8 { 0.9% (TS-A), and 53.6 { 2.6% (TS-B). The amount released from HC was greater than that from TS, although only significant for TSA (P õ 0.05). Therefore, HC in suspension appear slightly more sensitive to P/R injury than TS. Another difference between preparations was the extent of cytoprotection by glycine or calcium (Fig. 1). Greater cytoprotection was seen in HC than in the other preparations. In 48-h CS HC, the cytoprotective agents, when added to the resuspension medium, lowered LDH release by 40 to 45%. Calcium significantly decreased LDH release from 65.1 { 5.1 to 26.3 { 2.4% (P õ 0.05) and glycine to 20.7 { 1.8% (P õ 0.05), after 48 h cold storage. In the presence of these agents the amount of LDH released was similar to that in freshly isolated HC (14.4 { 1.4%, n Å 6).
07-18-96 22:53:33
cryoa
AP: CRYO
433
LIVER HYPOTHERMIC PRESERVATION/REPERFUSION INJURY
FIG. 2. Effect of nutritional status of the rat on viability of HC or tissue slices (TS-A) after 24 or 48 hr cold storage. Methods are described under Materials and Methods and in the legend to Fig. 1. Rats were either fed a standard laboratory diet or fasted overnight prior to liver harvest. *P õ 0.05 vs fed; †P õ 0.05 vs fasted.
In both TS preparations the extent of suppression of LDH leakage by glycine or calcium (6 to 14% reduction) was not as dramatic as in HC (40 to 45%) after 48 h cold storage (Fig. 1). LDH leakage was reduced by glycine from about 52 to 37.9 { 1.4% (TS-A) and 42.4 { 1.4% (TS-B) and by calcium to 45.6 { 1.5% (TS-A) and 38.9 { 2.2% (TS-B). The % LDH released in the presence of glycine or calcium after 48 h cold storage was significantly greater than in 0-h preserved TS (15.1 { 1.1, n Å 6). An even more dramatic difference in the extent of P/R injury between HC and TS was seen when livers from fasted rats were used (Fig. 2). Hepatocytes from 24-h-fasted rats showed significantly greater P/R injury than HC from fed rats after 24 h (LDH leakage 82.5 { 2.8 versus 27.7 { 5.9%, P õ 0.05) and 48 h CS (86.6 { 2.7 versus 65.1 { 5.1%, P õ 0.05). However, TS from fasted rats showed a similar response to P/R injury as TS from fed rats (Fig. 2). Glycine suppressed P/ R injury in HC from fasted rats but had little effect on P/R injury in TS from fasted rats. DISCUSSION
This study originated from our observations that glycine suppressed reperfusion injury in cold-stored HC but was less effective in mod-
AID
CRYO 1957
/
a903$$$$62
els using the whole organ or in orthotopic liver transplantation. For instance, we showed that HC cold stored in the UW solution for up to 6 days released a large amount of LDH on reperfusion. However, reperfusion with glycine (3 mM) in the rewarming medium significantly suppressed P/R injury (13). This suggested that glycine could improve liver preservation/transplantation. However, when we tested glycine in the isolated perfused liver or liver transplantation models the results were disappointing and glycine showed little benefit (3). Glycine, at various concentrations, did not improve significantly the quality or extend the safe duration of liver preservation. We concluded that the beneficial effects of glycine may be specific to the model (isolated cells) and may not be cytoprotective in more structurally intact systems. This conclusion was strengthened by the demonstration of cytoprotection by glycine in isolated renal proximal tubules exposed to hypoxia or toxic chemicals (26, 27). However, there was a lack of effect in more clinically relevant intact tissue models (8, 11) where glycine was only marginally protective in the cold-stored kidney or the kidney exposed to experimental warm ischemia/reperfusion. Weinberg et al. (28) suggested that the concentration of glycine normally present in whole kidney tissue is high and therefore additional protection by glycine in the reperfusion medium would not be readily seen. Glycine was only cytoprotective in cells that had lost glycine through the processes of tissue preparation. During the isolation of cells (renal tubules or HC) there could be a loss of glycine. Consequently, the addition of glycine to the suspending medium would provide cytoprotection. In this study we compared the cytoprotective effects of glycine, calcium, and fasting in HC and liver TS. We consider TS to be more representative of the whole liver. They have relatively normal cell–cell contact, contain an intact cytoskeleton anchored to the plasma membrane and other cells, and have not been exposed to extensive washing procedures in
07-18-96 22:53:33
cryoa
AP: CRYO
434
VREUGDENHIL, MARSH, AND SOUTHARD
the preparation. Thus, how P/R affects injury to TS may be more closely related to events occurring in the intact tissue used for transplantation. Hepatocytes, on the other hand, have undergone extensive washing and exposure to collagenase and low calcium medium, and do not maintain a normal cell–cell relationship. The hepatocyte suspension, therefore, may be more sensitive to P/R injury because of the stresses of isolation. Culturing HC prior to use may stimulate repair of the injury caused by the isolation and give a more hardy preparation; however, the lack of naturally occurring interactions between cells as well as other changes induced by culturing (i.e., exposure to enriched culture medium) were considered to preclude our interest in using these as accurate models of P/R injury in the liver. The results show some significant differences in sensitivity of hepatocytes to P/R injury compared to tissue slices. These differences were true for slices prepared from coldstored livers (TS-A) or slices prepared from fresh livers and then cold-stored (TS-B). There was a greater release of LDH from coldstored HC than from TS during rewarming, suggesting greater sensitivity of membranes of isolated cells to P/R injury than intact tissue. This difference was especially noticeable when HC from fasted rats were compared with TS from fasted rats. Fasting caused a large increase in the sensitization of HC to preservation injury after 24 h cold storage. However, fasting showed no adverse effect on how TS responded to P/R. Finally, glycine and calcium showed a significant and extensive cytoprotection of HC and suppressed the P/R-induced release of LDH after 48 h cold storage. The degree of cytoprotection in TS was noticeably less extensive. We conclude, therefore, that the methods used for the preparation of HC sensitize the membranes to P/R injury to a greater degree than occurs in intact tissue preparations. Changes that occur during their preparation make it possible to demonstrate cytoprotection by some agents in HC that are less effective
AID
CRYO 1957
/
a903$$$$62
when tested in intact models of P/R, such as TS or the whole liver. Caution should be used in judging how preservation and cytoprotective agents affect viability of HC exposed to cold preservation/rewarming, realizing that what is effective in these preparations may not be effective in more intact models of liver preservation. Suspensions of HC or cultured cells are a valuable model for studying liver metabolism; however, the disruption of the normal tissue architecture that is inherent in their preparation may limit the usefulness of these preparations, especially if disruption of tissue architecture is a key event in injury to the liver. The irreversible injury caused by cold preservation of the liver may be more related to changes in the structural organization (cytoskeleton, microtubules, membranes, vascular organization) of the tissue than to injury to the metabolic machinery. Thus, models that maintain not only the metabolic competency but also the structural integrity of the liver may be better for developing improved methods of organ preservation. Tissue slices may, therefore, be more suitable for some studies in organ preservation than isolated hepatocyte suspensions. This study is not a direct comparison of how preservation affects HC versus TS, but of how preservation maneuvers affect cell viability in two different models. The results suggest that preservation maneuvers that are beneficial in the HC preparation (calcium, glycine) are not of significant benefit when the TS model is used. This difference may be important in future experiments designed to study methods to improve preservation of the liver. REFERENCES 1. D’Alessandro, A. M., Southard, J. H., Kalayoglu, M., and Belzer, F. O. Comparison of cold storage and perfusion of dog livers on function of tissue slices. Cryobiology 23, 161–167 (1986). 2. D’Alessandro, A. M., Southard, J. H., Kalayoglu, M., and Belzer, F. O. Effect of drug treatment on liver slice function following 72 hour hypothermic perfusion. Cryobiology 23, 415–521 (1986). 3. den Butter, G., Lindell, S. L., Gandolf, D., Schilling, M. K., Southard, J. H., and Belzer, F. O. The effect
07-18-96 22:53:33
cryoa
AP: CRYO
435
LIVER HYPOTHERMIC PRESERVATION/REPERFUSION INJURY
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
of glutathione and glycine in kidney preservation. Transplant. Proc. 25, 1633–1634 (1993). Ganote, C., and Armstrong, S. Ischaemia and the myocyte cytoskeleton: Review and speculation. Cardiovasc. Res. 27, 1387–1403 (1993). Guyomard, C., Chesne, C., Meunier, B., Fautrel, A., Clerc, C., Morel, F., Rissel, M., Campion, J., and Guillouzo, A. Primary culture of adult rat hepatocytes after 48-hour preservation of the liver with cold UW solution. Hepatology 12, 1329–1336 (1990). Hall, S. M., Evans, J., and Haworth, S. G. Influence of cold preservation on the cytoskeleton of cultured pulmonary arterial endothelial cells. Am. J. Respir. Cell Mol. Biol. 9, 106–114 (1993). Hawkins, H. E., Clark, P., Lippert, E. C., and Karow, A. M. Functional preservation of the mammalian kidney. J. Surg. Res. 38, 281–288 (1985). Heyman, S. N., and Epstein, F. H. Glycine and cytopreservation: Disappointments and promises. J. Lab. Clin. Med. 121, 199–200 (1993). Kamada, N., and Calne, R. Y. A surgical experience with five hundred thirty liver transplants in the rat. Surgery 93, 64–69 (1983). Lemasters, J. J., Bunzendahl, H., and Thurman, R. G. Reperfusion injury to donor livers stored for transplantation. Liver Transplant. Surg. 1(2), 124– 138 (1995). Mangino, M. J., Murphy, M. K., Grabau, G. G., and Anderson, C. B. Protective effects of glycine during hypothermic renal ischemia-reperfusion injury. Am. J. Physiol. 261, F841–F848 (1991). Marsh, D. C., Lindell, S. L., Fox, L. E., Belzer, F. O., and Southard, J. H. Hypothermic preservation of hepatocytes. I. Role of cell swelling. Cryobiology 26, 524–534 (1989). Marsh, D. C., Hjelmhaug, J. A., Vreugdenhil, P. K., Belzer, F. O., and Southard, J. H. Glycine prevention of cold ischemic injury in isolated hepatocytes. Cryobiology 28, 105–109 (1991). Marsh, D. C., Vreugdenhil, P. K., Mack, V. E., Belzer, F. O., and Southard, J. H. Glycine protects hepatocytes from injury caused by anoxia, cold ischemia and mitochondrial inhibitors, but not injury caused by calcium ionophores or oxidative stress. Hepatology 17, 91–98 (1992). Marzi, I., Zhong, Z., Lemasters, J. J., and Thurman, R. G. Evidence that graft survival is not related to parenchymal cell viability in rat liver transplantation. The importance of nonparenchymal cells. Transplantation 48, 463–468 (1989). McKeown, C. M. B., Edwards, V., Phillips, M. J., Harvey, P. R. C., Petrunka, C. N., and Strasberg,
AID
CRYO 1957
/
a903$$$$62
17.
18.
19.
20. 21. 22.
23.
24.
25.
26.
27.
28.
S. M. Sinusoidal lining cell damage: the critical injury in cold preservation of liver allografts in the rat. Transplantation 46(2), 178–191 (1988). Michell, I. D., Chipman, J. K., and McMaster, P. The isolated hepatocyte preservation model: A comparison of hypertonic citrate and lactobionate solutions. Transplant. Proc. 21, 1312–1313 (1989). Page, R. A., Stowell, K. M., Hardman, M. J., and Kitson, K. E. The assessment of viability in isolated rat hepatocytes. Anal. Biochem. 200, 171– 175 (1992). Ploeg, R. J., D’Alessandro, A. M., Knechtle, S. J., Stegall, M. D., Pirsch, J. D., Hoffmann, R. M., Sasaki, T., Sollinger, H. W., Belzer, F. O., and Kalayoglu, M. Risk factors for primary dysfunction after liver transplantation: A multivariate analysis. Transplantation 55, 123–129 (1993). Seglen, P. O. Preparation of isolated rat liver cells. Methods Cell Biol. 13, 30–83 (1976). Southard, J. H. Viability assays in organ preservation. Cryobiology 26, 232–238 (1989). Southard, J. H., Kuniyoshi, M., Lutz, M. F., Ametani, M., and Belzer, F. O. Comparison of the effects of 3- and 5-day hypothermic perfusion of dog kidneys on metabolism of tissue slices. Cryobiology 21, 285–295 (1984). Sumimoto, R., Jamieson, N. V., Wake, K., and Kamada, N. 24-hour rat liver preservation using UW solution and some simplified variants. Transplantation 48, 1–5 (1989). Sumimoto, R., Lindell, S., Gambiez, L., Southard, J. H., and Belzer, F. O. The effect of pretransplant flushout of the liver with a solution containing albumin on survival and organ function. Transplant. Proc. 25, 3001–3003 (1993). Umeshita, K., Monden, M., Fujimori, T., Sakai, T., Gotoh, M., Okamura, J., and Mori, T. Extracellular calcium protects cultured rat hepatocytes from injury caused by hypothermic preservation. Cryobiology 25, 102–109 (1988). Weinberg, J. M., Davis, J. A., Abarzua, M., and Rajan, T. Cytoprotective effects of glycine and glutathione against hypoxic injury to renal tubules. J. Clin. Invest. 80, 1446–1454 (1987). Weinberg, J. M., Davis, J. A., Abarzua, M., Kiani, T., and Kunkel, R. Protection by glycine of proximal tubules from injury due to inhibitors of mitochondrial ATP production. Am. J. Physiol. 258, C1127– C1140 (1990). Weinberg, J. M., Nissim, I., Roeser, N. F., Davis, J. A., and Schultz, S. Relationships between intracellular amino acid level and protection against injury to isolated proximal tubules. Am. J. Physiol. 260, F410–F419 (1991).
07-18-96 22:53:33
cryoa
AP: CRYO