Metabolism of hyaluronic acid by liver endothelial cells: effect of ischemia-reperfusion in the isolated perfused rat liver

Liver slices in in vitro pharmacotoxicology with special reference to the use of human liver tissue

In the early years of research in in vitro pharmacotoxicology liver slices have been used. After a decline in the application of slices in favour of the use of isolated hepatocytes and the isolated perfused liver preparation, the development of the Krumdieck slicer in the 1980s led to a ;comeback' of the technique. This review will focus on the use of human liver, with special reference to the comparison of slices with isolated hepatocytes in in vitro pharmacotoxicology. In addition, an overview on the predictive value of these in vitro systems for drug disposition and toxicity in vivo will be given. Preservation techniques for liver slices and hepatocytes will also be discussed. These techniques ensure an efficient utilization of the scarce human material. For long-term storage of liver slices and hepatocytes, cryopreservation seems most promising. However, cryopreservation is still in its infancy, and reports mainly deal with drug metabolism studies after cryopreservation. Drug toxicity, metabolism and transport data determined in slices and isolated hepatocytes, from both human and animal liver showed good correlation with the corresponding parameters measured in vivo. Therefore, the results obtained in such studies may give rise to more in-depth research on the mechanisms of pharmactoxicology in the human liver.

Cyclosporin-A does not prevent cold ischemia/reperfusion injury of rat livers

Cyclosporin-A (CsA) has been reported to protect livers from warm ischemia/reperfusion (I/R) injury. To study if CsA has also a protective effect on cold I/R injury, two models were used: the isolated perfused rat liver (IPRL) and the orthotopic rat liver transplantation (ORLT). (1) IPRL: Livers were preserved for 24 h (5°C) in University of Wisconsin (UW) solution alone (group 1), with CsA (400 nM) dissolved in dimethylsulfoxide (50 μM) (group 2), and with dimethylsulfoxide (DMSO) alone (group 3). Livers were reperfused for 60 min (37°C) (n = 8/group). Cell necrosis was evaluated by trypan blue uptake and apoptosis by laddering and terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay, and by caspase-3 activation. Marked and similar sinusoidal endothelial cell necrosis was found in the three groups while apoptosis was found similarly deceased in groups 2 and 3 compared with group 1. (2) ORLT: Donors received either CsA (5 mg/kg) or corn oil 24 h before transplantation. Recipients were sacrificed after 240 min; cell necrosis and apoptosis were then evaluated. No difference was found between treated and control groups. The current data strongly suggest that CsA has no protective effect on hepatic cold I/R injury. Hepatocyte apoptosis can be reduced by antioxidants, as occurred with DMSO, but introduction of CsA does not provide additional protective effect.

Effect of cold ischemia-warm reperfusion on the cirrhotic rat liver

Cirrhosis is known to induce capillarization of the sinusoidal endothelial cells (SECs) and collagenization of the space of Disse, resulting in a reduced access of plasma and plasma-dissolved substances to hepatocytes due to their limited diffusion in the extravascular space. The aim of the present study was to use a well known effect of cold ischemia-warm reperfusion (CI-WR) on liver SECs, that is, their retraction and detachment, progressing to a denudation of the SEC lining. The disappearance of the capillarized SEC lining would improve the access of plasma and plasma-dissolved substances to the hepatocytes and consequently might improve the metabolic function of cirrhotic livers. This study was performed using the isolated perfused rat liver model subjected to 24-hour CI followed by a 60-minute WR in thioacetamide-induced cirrhosis. Liver microcirculation was evaluated using the multiple indicator dilution curve (MIDC) technique. Hepatocyte, SEC, and Kupffer cell functions were evaluated using specific elimination processes. As occurs in normal livers, CI-WR induced extensive SEC necrosis with a marked reduction of the hyaluronic acid elimination. By contrast, the hepatic microcirculation was not modified: vascular, extravascular, and the cellular spaces were similar before and following CI-WR. In addition, the hepatic metabolic functions, as evaluated by propranolol and taurocholate hepatic uptake, were neither improved nor decreased, as were Kupffer cell functions. The present data strongly suggest that capillarization of SECs plays a lesser role than collagenization of the space of Disse in the reduced exchange between sinusoids and hepatocytes in thioacetamide-induced cirrhotic rat livers, which appear to be quite resistant to CI-WR.

Matrix metalloproteinase inhibition protects rat livers from prolonged cold ischemia-warm reperfusion injury

Matrix metalloproteinases (MMPs) have been implicated in the hepatic injury induced after cold ischemia-warm reperfusion (CI-WR), by altering the extracellular matrix (ECM), but their precise role remains unknown. The hepatic MMP expression was evaluated after 2 conditions of CI (4 degrees C for 24 and 42 hours: viable and nonviable livers) followed by different periods of WR, using isolated perfused rat livers. CI-WR induced moderate changes in hepatic MMP transcript levels not influenced by CI duration, whereas gelatinase activities accumulated in liver effluents. Therefore, the protective effect of a new phosphinic MMP inhibitor, RXP409, was tested after prolonged CI. RXP409 (10 microM) was added to the University of Wisconsin solution, and livers were preserved for 42 hours (4 degrees C), then reperfused for 1 hour in Krebs solution (37 degrees C), containing 20% erythrocytes. Liver viability parameters were recorded, and the extent of cell necrosis was evaluated on liver biopsies, using trypan blue nuclear uptake. Treatment with RXP409 significantly improved liver function (transaminase release and bile secretion) and liver injury. In particular, the MMP inhibitor significantly modified the extent of cell death from large clusters of necrotic hepatocytes as found in control livers (2%-60% of liver biopsies; mean, 26% +/- 9%) to isolated necrotic hepatocytes as found in treated livers (0.2%-12%; mean, 3% +/- 2%) (P < 0.05).

CONCLUSION

These data demonstrate that MMPs, by altering the ECM, play a major role in liver CI-WR injury leading to extensive hepatocyte necrosis and that their inhibition might prove to be a new strategy in improving preservation solutions.

Augmentation of Ischemia/Reperfusion Injury to Endothelial Cells by Cyclosporin A

Ischemia/reperfusion (I/R) carries significant injury to endothelial cells in transplanted organs and is an important factor in chronic rejection. Immunosuppressive drugs, notably cyclosporin A (CyA) and FK506, can potentially augment this injury. Here, our goal was to determine the combined effects of I/R and CyA or FK506 on endothelial cells. Transformed mouse endothelial cells (SVEC 4–10) were subjected to ischemia or I/R for 2–24 hours by incubating cells in 100 per cent N2 (ischemia) followed by 5 per cent CO2 and 95 per cent O2 (reperfusion) for 24 hours. In separate experiments, CyA or FK506 was added to cells subjected to ischemia or I/R. Nonviable cells were determined by Trypan blue exclusion assay. All experiments (done in triplicate) were analyzed by Student's t test. Increasing ischemia times resulted in a greater number of nonviable cells (2% nonviable cells at 0 hours and 57% at 24 hours of I/R). Addition of CyA significantly increased the number of nonviable cells when compared with the control (I/R only) group ( P = 0.014). Interestingly, FK506 did not increase the percentage of nonviable cells compared with the control group ( P = 0.2). Unlike FK506, CyA augments I/R injury to endothelial cells in vitro. These findings could be relevant in chronic rejection and transplantation.

Sinusoidal endothelial cell and hepatocyte death following cold ischemia-warm reperfusion of the rat liver

Cold ischemia-warm reperfusion (CI-WR) injury of the liver is characterized by marked alterations of sinusoidal endothelial cells (SECs), whereas hepatocytes appear to be relatively unscathed. However, the time course and mechanism of cell death remain controversial: early versus late phenomenon, necrosis versus apoptosis? We describe the occurrence and nature of cell death after different periods of CI with University of Wisconsin (UW) solution and after different periods of WR in the isolated perfused rat liver model. After 24- and 42-hour CI (viable and nonviable livers, respectively), similar patterns of liver cell death were seen: SEC necrosis appeared early after WR (10 minutes) and remained stable for up to 120 minutes. After 30 minutes of WR, apoptosis increased progressively with WR length. Based on morphological criteria, apoptotic cells were mainly hepatocytes within liver plates or extruded in the sinusoidal lumen. In addition, only after 42-hour CI were large clusters of necrotic hepatocytes found in areas of congested sinusoids. In these same livers, the hepatic microcirculation, evaluated by means of the multiple-indicator dilution technique, revealed extracellular matrix disappearance with no-flow areas. In conclusion, different time courses and mechanisms of cell death occur in rat livers after CI-WR, with early SEC necrosis followed by delayed hepatocyte apoptosis. These processes do not appear to be of major importance in the mechanism of graft failure because they are similar under both nonlethal and lethal conditions; this is not the case for the loss of the extracellular matrix found only under lethal conditions and associated with hepatocyte necrosis.

Prevention of ischemia-reperfusion-induced hepatic microcirculatory disruption by inhibiting stellate cell contraction using rock inhibitor

BACKGROUND

We demonstrated that hepatic stellate cells (HSCs) isolated from rat livers exposed to warm ischemia are significantly contractile when compared with HSCs from intact rat livers. This suggests that ischemia-reperfusion (IR)-induced impairment of sinusoidal microcirculation results, at least in part, from contraction of HSCs.

METHODS

Rho-associated coiled-coil forming protein serine/threonine kinase (ROCK) is one of the key regulators of HSCs motility. Therefore we investigated whether Y-27632, a p160ROCK-specific inhibitor, has beneficial effects on warm IR injury in an in vivo rat partial liver IR model and a rat orthotopic liver transplantation model.

RESULTS

After reperfusion following 90 min of warm ischemia, livers in untreated control rats had persistent congestion and impaired mitochondrial respiration, as demonstrated by increasing deoxy-hemoglobin and reduced cytochrome oxidase contents in the hepatic tissues using in vivo near-infrared spectroscopy. Serum levels of transaminase and endothelin (ET)-1 in these rats were markedly increased 1 hr after reperfusion. In contrast, when Y-27632 (3-30 mg/kg) was administered orally, hepatic tissue contents of deoxy-hemoglobin and cytochrome oxidase rapidly normalized. In such animals, the elevation of serum transaminase levels, but not that of ET-1 levels, was significantly suppressed. This is consistent with in vitro data demonstrating that Y-27632 causes HSCs to undergo relaxation even in the presence of ET-1. Moreover, in a rat orthotopic liver transplantation model, Y-27632 pretreatment dramatically improved the survival of recipients with liver grafts subjected to 45 min of warm ischemia.

CONCLUSIONS

Y-27632 attenuates IR-induced hepatic microcirculation disruption by inhibiting contraction of HSCs.

Fate of hepatocyte and sinusoidal lining cell function and kinetics after extended cold preservation and transplantation of the rat liver

We investigated the chronological profile of graft damage and recovery after liver cold ischemia-reperfusion (I/R) injury, with particular attention to the role of apoptosis on hepatocyte and sinusoidal endothelial cell (SEC) damage. Male Lewis rats underwent rearterialized orthotopic liver transplantation using grafts subjected to a short (University of Wisconsin [UW] solution for 1 hour [UW1h]) and prolonged period (UW16h) of cold preservation. Experiments were performed immediately after preservation and 4 hours, 24 hours, 3 days, and 7 days after reperfusion. At each time, graft function, incidence of apoptotic cells, expression of the epitope recognized by a monoclonal antibody specific to rat SECs (SE-1), and incidence of proliferating cells were estimated. In the UW16h group, the proportion of apoptotic SECs was markedly elevated at 4 hours. The incidence of hepatocyte apoptosis was very low, although massive hepatocyte necrosis was evident at 24 hours. The incidence of proliferating hepatocytes and SECs peaked at 3 days, then returned to normal by 7 days. SE-1 expression was reduced immediately after preservation, followed by a marked reduction at 4 and 24 hours after reperfusion, and expression returned to normal by 7 days. Although SEC apoptosis was induced in the early phase of cold I/R injury, hepatocyte damage developed without the occurrence of apoptosis. Regeneration of both hepatocytes and SECs after cold I/R injury peaked at 3 days and was complete by 7 days, whereas functional recovery of these cell populations was complete 3 days after reperfusion.

Effects of prostaglandin E(1) on the efficacy of xenogeneic extracorporeal pig liver perfusion in a canine model of acute liver failure

Xenogeneic extracorporeal liver perfusion (ECLP) has the potential to become an important tool in the management of patients with severe liver failure. We previously showed that xenogeneic pig liver perfusion may be prolonged for up to 9 hours by the administration of prostaglandin E(1) (PGE(1)). In this study, we used a canine model of acute liver failure to evaluate the effects of PGE(1) on the efficacy of ECLP as a liver-assist device. Liver failure was surgically induced in 12 beagle dogs, with a control group (group 1, n = 4) not connected to the ECLP circuit. Direct cross-circulation between the dogs and the ECLP circuit using a pig liver was performed without (group 2, n = 4) or with (group 3, n = 4) continuous administration of PGE(1) through the portal vein of the pig liver. The duration of cross-circulation in group 3 (9.4 +/- 1.2 hours) was significantly longer than in group 2 (4.3 +/- 1.0 hours). In addition, elevation of blood ammonia, total bile acid, and hyaluronic acid levels was less marked in group 3 compared with the other 2 groups. The ratio of branched-chain amino acids to aromatic amino acids was also improved in group 3. The mean survival time in group 3 (26.6 +/- 0.4 hours) was significantly longer than in group 1 (15.5 +/- 1.3 hours) or group 2 (17.1 +/- 2.9 hours). Continuous administration of PGE(1) to xenogeneic ECLP resulted in a significant improvement in both liver function and survival time of dogs with surgically induced liver failure.

Role of hepatocytes and bile duct cells in preservation-reperfusion injury of liver grafts

In liver transplantation, it is currently hypothesized that nonparenchymal cell damage and/or activation is the major cause of preservation-related graft injury. Because parenchymal cells (hepatocytes) appear morphologically well preserved even after extended cold preservation, their injury after warm reperfusion is ascribed to the consequences of nonparenchymal cell damage and/or activation. However, accumulating evidence over the past decade indicated that the current hypothesis cannot fully explain preservation-related liver graft injury. We review data obtained in animal and human liver transplantation and isolated perfused animal livers, as well as isolated cell models to highlight growing evidence of the importance of hepatocyte disturbances in the pathogenesis of normal and fatty graft injury. Particular attention is given to preservation time-dependent decreases in high-energy adenine nucleotide levels in liver cells, a circumstance that (1) sensitizes hepatocytes to various stimuli and insults, (2) correlates well with graft function after liver transplantation, and (3) may also underlie the preservation time-dependent increase in endothelial cell damage. We also review damage to bile duct cells, which is increasingly being recognized as important in the long-lasting phase of reperfusion injury. The role of hydrophobic bile salts in that context is particularly assessed. Finally, a number of avenues aimed at preserving hepatocyte and bile duct cell integrity are discussed in the context of liver transplantation therapy as a complement to reducing nonparenchymal cell damage and/or activation.

Cold-preservation-induced sensitivity of rat hepatocyte function to rewarming injury and its prevention by short-term reperfusion

With increasing time of cold preservation, levels of high-energy nucleotides in the liver are reducing. The authors hypothesized that cold preservation sensitizes hepatocyte function to ischemic injury occurring during graft rewarming and that the injury can be prevented by short-term reperfusion. Rat livers were cold-preserved in University of Wisconsin solution for 0 to 18 hours and ischemically rewarmed for 0 to 45 minutes to simulate the implantation stage of transplantation. Hepatobiliary function was assessed using a blood-free perfusion model. In comparison with controls, neither 18-hour preservation nor 45-minute ischemic rewarming significantly influenced hepatocyte function. Compared with livers subjected to 45-minute ischemic rewarming, livers subjected to 9-hour preservation and 45-minute rewarming, and livers subjected to 18-hour preservation and 45-minute rewarming exhibited, respectively: 3.8 and 24 times reduced bile production, 4.3- and 116-fold decreased taurocholate excretion, and 3.1 and 42 times depressed bromosulfophthalein excretion. Thirty-minute oxygenated warm reperfusion after 9- and 18-hour preservation nearly completely blunted sensitization of hepatocyte function to rewarming ischemia. The authors found that short-term oxygenated reperfusion restored adenine nucleotides in liver tissue to the values found before organ preservation and that reperfusion with energy substrate containing solutions increased tissue adenosine triphosphate concentration to a higher level than that found before preservation. In conclusion, sensitization of hepatocyte function to rewarming ischemia increases disproportionally with storage time, suggesting that this phenomenon may play a role in graft dysfunctions with increasing liver preservation time. Short-term oxygenated reperfusion of the liver may protect hepatocyte functions against warm ischemic insult, even after extended preservation.

Serum hyaluronan as a predictor of hepatic regeneration after hepatectomy in humans

BACKGROUND

The capacity for hepatic regeneration after hepatectomy is important for allowing surgeons to determine the appropriate extent of resection. However, conventional preoperative liver function tests are unsatisfactory for estimating the post-operative regenerative capacity of the remnant liver. The aim of this study was to evaluate the relationship between preoperative serum hyaluronan and hepatic regeneration.

METHODS

Preoperative serum hyaluronan levels and the hepatic regeneration rate were estimated in 49 patients using computerized tomographic volumetry. The hepatic fibrotic rate was calculated with non-tumorous tissues stained with Azan-Mallory. Immunolocalization of factor VIII-related antigen (FVIIIAg) was examined as a marker for hepatic sinusoidal capillarization.

RESULTS

The serum hyaluronan level was significantly correlated with the hepatic regeneration rate (P < 0. 001). Patients with serum hyaluronan levels below 200 ng mL-1 exhibited a significant correlation between the hepatic regeneration rate and the hepatic fibrotic rate. However, patients with serum hyaluronan levels above 200 ng mL-1 did not demonstrate a distinct correlation. The hepatic regeneration rate of patients with FVIIIAg in the liver and serum hyaluronan levels above 200 ng mL-1 were very low compared with those without FVIIIAg (P < 0.001). Multiple regression analysis revealed that serum hyaluronan was a significant predictor of post-operative hepatic regeneration among several clinical variables (r = 0.857, R2 = 0.735).

CONCLUSION

It has been suggested that hepatic regeneration is closely related to both hepatic fibrosis and hepatic sinusoidal capillarization. The serum hyaluronan level is regarded as a useful predictor for hepatic regeneration after hepatectomy.

Effects of acute physical exercise on hepatocyte volume and function in rat

The goal of the present experiment was to measure the volume of the different compartments in liver of exercised rats and to get some insights into the appropriate working of the hepatic function following exercise. Hence, livers from male rats were isolated and perfused after treadmill exercise or rest. This procedure was performed on rats that were overnight semifasted (50% food restriction) or well fed. To evaluate the hepatocyte cell volume, the multiple-indicator dilution curve technique was used after 40 min of perfusion. Radioactive tracers for red blood cells, sucrose, and water were used to measure liver vascular space, liver interstitial space, and water cellular space, respectively. The hepatocyte function was assessed by taurocholate and propanolol clearance. Oxygen consumption, intrahepatic resistance, bile secretion, and lactate dehydrogenase release estimated liver viability. Liver viability and hepatocyte function were not changed following exercise either in the fed or in the semifasted animals. As expected, liver glycogen levels were significantly (P < 0.01) reduced in the food-restricted rats. Consequently, liver glycogen levels following exercise were decreased significantly (P < 0.01) only in the fed rats. Despite this, exercise decreased the hepatocyte water space in both food-restricted and fed groups ( approximately 15%; P < 0.01) without altering the sinusoidal and interstitial space. The present data show that acute exercise decreased the hepatocyte volume and that this volume change is not entirely linked to a decrease in hepatic glycogen level.

Role of endothelin in obstructive jaundice

Mediators responsible for renal changes in obstructive jaundice are not specified. This study is designed to study the role of endothelin-1 (ET-1) in obstructive jaundice in rats. Animals were randomly placed into five experimental groups. Group 1 (N = 3) was the sham-operated group. Group 2 (N = 8) after common bile duct (CBD) ligation, received bosentan, which is a nonselective endothelin receptor blocker, 50 mg/kg/day for seven days. Group 3 (N = 7) received 1 microg/kg/day captopril. Group 4 (N = 7) was given both drugs orally for seven days. Group 5 (N = 6) after CBD ligation, received Arabic gum as the vehicle. Blood was drawn from the infrahepatic vena cava for the determination of ET-1, bilirubin, creatinine, protein oxidation products, hyaluronic acid, and beta-N-acetyl-hexosaminase. Liver tissue samples were obtained to determine glutathione levels. ET-1, protein oxidation products, hyaluronic acid, bilirubin, and creatinine levels increased significantly in the control group when compared with sham. Bosentan effectively prevented ET-1 elevation but could not reverse creatinine or bilirubin elevation. Captopril with or without bosentan was cytoprotective but did not reverse increased creatinine levels. It is concluded that increased ET-1 in obstructive jaundice may be one of the contributing factors of renal damage.

Effects of exogenous superoxide anion and nitric oxide on the scavenging function and electron microscopic appearance of the sinusoidal endothelium in the isolated, perfused rat liver

BACKGROUND/AIMS

Functional and morphological alterations of the hepatic sinusoidal endothelial cell occur in several models of experimental liver injury and in clinical settings. The causes of these alterations are multiple. The aim of this study was to test the hypothesis that the early functional impairment and morphological alterations of the sinusoidal endothelial cell and hepatic sinusoid associated with liver injury are mediated by free radical species, such as superoxide anion and nitric oxide.

METHODS

Isolated rat livers were perfused by recirculation with hemoglobin-free, Krebs-Henseleit bicarbonate buffer and presented with a source of superoxide anion (xanthine oxidase+hypoxanthine) or nitric oxide (S-nitroso-N-acetyl penicillamine). Hyaluronan uptake (an index of sinusoidal endothelial cell scavenging function), thiobarbituric acid-reactive substances content of the tissue (a marker of lipid peroxidation), reduced and oxidized glutathione (a marker of the thiol system oxidation/reduction state), lactate dehydrogenase and alanine aminotransferase activities (markers of cytolysis), as well as scanning and transmission electron microscopic appearance of the sinusoid were evaluated.

RESULTS

At the high concentrations used, both free radical generating systems suppressed hyaluronan uptake, increased malondialdehyde content of the tissue, enhanced the release of both liver enzymes, decreased the total glutathione content of the liver, and altered the ratio of reduced/oxidized glutathione. Both free radical species induced dose-dependent morphological alterations of the sinusoid, consisting of the appearance of large gaps replacing the sieve-plated fenestration.

CONCLUSIONS

The free radical species-induced functional impairment and morphological alterations of the liver sinusoid, presented in this study, closely resemble the early in vivo changes associated with liver injury under a variety of conditions, such as preservation and reperfusion, or administration of hepatotoxicants such as D-galactosamine, Gram-negative bacterial lipopolysaccharides, acetaminophen, alcohol and others. Therefore, we suggest that early liver sinusoid injury, observed under these conditions, can be attributed to the action of free radicals, such as superoxide anion and nitric oxide.

Decreased survival in rat liver transplantation with extended cold preservation: role of portal vein clamping time

Primary liver graft dysfunction is currently related to cold ischemia-reperfusion injury, although a wide survival range has been reported using 24-hour preservation in cold University of Wisconsin (UW) solution. We hypothesized that the portal vein clamping time (PVCT) played a more important role than cold preservation injury in the postoperative outcome. Rat liver transplantation was performed using different clamping times after 24-hour cold ischemia in the UW solution. Survival rates, plasma tumor necrosis factor (TNF), and nitrate/nitrite levels were examined. Subsequently, the effect of clamping time was evaluated on hepatocyte and sinusoidal endothelial cell (SEC) function using isolated perfused livers. Survival rate was directly related to clamping time length. Marked increases in TNF and nitrate/nitrite levels were found after surgery, particularly after long clamping times. In perfusion studies, the SEC function was already markedly altered after preservation alone and was not further modified by transplantation. By contrast, the hepatocyte function was moderately altered after transplantation, irrespective of clamping times, even when rats operated with long clamping times were in terminal conditions. In rats, 24-hour preservation in cold UW solution is not a severely compromising condition leading to primary liver nonfunction. Long PVCTs are associated with an endotoxemia-like syndrome more related to a warm intestinal ischemia than to cold ischemia injury of the liver.

Early hepatocyte, endothelial, and bile duct cell injury after pediatric liver transplantation from cadaveric or living-related donors

BACKGROUND

When compared with cadaveric grafts (Cad), the potential advantages of pediatric orthotopic liver transplantation (OLT) from living-related (LR) donors may include better graft quality, shorter ischemic time, appropriate preparation of the recipient, and better immunologic compatibility.

METHODS

The aim of this study was to analyze early hepatocyte, endothelial, and bile duct cell injury following pediatric OLT using LR (n=15) or uncomplicated Cad reduced-size (n=10) grafts. Median (range) total ischemic times were 190 min (105-261) versus 760 min (418-948) in LR and Cad groups, respectively (P<0.001).

RESULTS

The post-OLT cytolytic profile, assessed daily during the first 7 days using both plasma glutamate-pyruvate transaminase and serum alpha-glutathione S-transferase, showed significantly higher levels of both parameters for the 10 uncomplicated Cad cases when compared with the 15 LR grafts (P<0.001). The evaluation of hepatic endothelial cell function during the first week after OLT, using serum hyaluronic acid levels, suggested lower endothelial injury in the LR grafts, when compared with the Cad grafts (P=0.059). Bile duct cell injury, as assessed using plasma gamma-glutamyl transferase levels, was similar in both groups, with a progressive increase at the end of the first week after OLT, which was correlated with a similar incidence of early acute rejection in both groups (80% in the LR group vs. 62% in the Cad group, NS).

CONCLUSION

(1) The hepatocellular and endothelial cell damage was reduced after OLT with LR grafts, which may be related to shorter ischemic time when compared with Cad grafts; (2) the putative immunologic advantage for LR grafts was not confirmed in terms of incidence of acute rejection.

Assessment of liver allograft function by hyaluronic acid and endothelin levels

BACKGROUND

Sinusoidal endothelial cells are considered the primary site of preservation-reperfusion injury occurring during cold storage and reperfusion. Hyaluronic acid (HA) and endothelin-1 (ET) are markers of endothelial cell integrity. The aim of this study was to evaluate the possible correlation between these markers and early graft function following liver transplantation.

MATERIALS AND METHODS

Blood samples were collected from 85 adult orthotopic liver transplant recipients at the following time points: before surgery, just before reperfusion, first and last 20 cc of caval effluent, 30, 60, and 120 min after reperfusion, and on postoperative day (POD) 1. Levels of both HA and ET were measured by sandwich enzyme-binding assay. HA uptake was also defined for every sample as the ratio of the actual measurement to the value obtained just before reperfusion.

RESULTS

Patients with severe liver disease displayed significantly higher pretransplant HA and ET levels compared with those of less advanced illness. Poor early graft function (PEGF) was defined as peak enzyme levels on the first three PODs higher than 2500 U/L and prothrombin time longer than 16 s on POD 2. Patients with PEGF (n = 9) showed significantly lower HA uptake than patients with good graft function (n = 76) at 60 and 120 min after reperfusion. There was also a significant difference in the average uptake values measured in the last 20 cc of caval effluent between the two groups. We could not find any difference in ET levels between these two groups.

CONCLUSIONS

Perioperative HA uptake measurement may be an additional marker to evaluate early graft function.

The protective effects of prostaglandin E1 on sinusoidal endothelial cells in xenogeneic pig liver perfusion

The effects of prostaglandin E1 (PGE1) on hepatic sinusoidal endothelial cells (SEC) in the xenogeneic immunoreaction were investigated. Porcine livers were perfused with fresh human blood via the portal vein (PV) and the hepatic artery (HA) either with the administration of PGE1 (Group PG) or without PGE1 (Group C). The creatine kinase-BB component (CK-BB) in the perfusate was measured to assess SEC damage. SEC activation and complement activation were evaluated immunohistochemically by the expression of von Willebrand factor (vWF) and by the deposition of membrane attack complex (MAC), respectively. Xenoperfusion in Group C was discontinued between 4 and 6 hr due to the rapid elevation of HA pressures and the massive loss of perfusate. In Group PG, both PV and HA pressures were kept stable for up to 9 hr. In Group C, severe interlobular bleeding and diffuse extrasinusoidal hemorrhage were observed at 4 hr histologically, while in Group PG, the hepatic architecture was maintained without hemorrhage at 6 hr. MAC was markedly deposited on SEC and parenchymal cells at 3 hr in both groups. The amount of vWF, however, was expressed on SEC in large amounts at 1 hr in Group C, while small amounts were expressed at 1 hr in Group PG. In Group PG, CK-BB release was significantly lower than in Group C (P < 0.01). These results suggest that PGE1 suppressed SEC activation and protected the impairment of hepatic SEC during xenoperfusion without suppressing complement activation, resulting in the prolongation of xenogeneic liver perfusion.

Effect of calcium antagonists on rat liver during extended cold preservation-reperfusion

BACKGROUND

Nisoldipine, a calcium antagonist, has been reported to improve the quality of grafted rat livers. We thus assessed the protective effect of two calcium antagonists, nisoldipine and nickel, during extended cold ischemia-reperfusion.

METHODS

Rat livers were isolated and perfused before or after 24 hr of cold ischemia in University of Wisconsin solution (4 degrees C) with or without nisoldipine or nickel. Sinusoidal endothelial cell and hepatocyte functions were measured by hyaluronic acid and taurocholate elimination, respectively.

RESULTS

Similar alterations in hepatocyte and sinusoidal cell functions were found in all groups after cold ischemia with or without calcium antagonists. In a second set of experiments, liver transplantation was performed in two groups of rats with livers stored under identical conditions with or without nisoldipine. Seven of 12 animals (62.5%) in both groups survived for over 10 days after 24-hr preservation in University of Wisconsin solution. Survival rates were similar in both groups.

CONCLUSIONS

Calcium antagonists do not appear to have a direct protective effect on sinusoidal endothelial cell and hepatocyte functions, nor on the overall liver preservation after extended cold preservation-reperfusion.

No attenuation of ischemic and reperfusion injury in Kupffer cell-depleted, cold-preserved rat livers

Activated Kupffer cells (KC) have been implicated in the damage sustained by preserved liver grafts during ischemia and reperfusion. The aim of this study was to compare ischemia/reperfusion injury in preserved, KC-depleted rat livers and preserved control livers, with special regard to sinusoidal endothelial cell (SEC) injury. Wistar rats were injected with liposome-encapsulated dichloromethylene diphosphonate, 48 hr before hepatectomy, to eliminate KC, or were withheld this pretreatment (controls). Livers were flushed with cold University of Wisconsin solution and after 0, 8, 16, or 24 hr of storage at 4 degrees C, were reperfused in a recirculation system with 200 ml of oxygenated Krebs-Henseleit solution at 37 degrees C for 90 min. Damage to SEC was measured by the uptake of hyaluronic acid (HA) from the perfusate and release of purine nucleoside phosphorylase (PNP). Perfusate samples were, furthermore, analyzed for aspartate aminotransferase (AST) and tumor necrosis factor-alpha. Carbon particles were infused in the perfusate to determine the phagocytotic capacity of KC. Biopsies were taken for histological examination and sections were stained with ED2 monoclonal antibodies to confirm the absence of KC. After 90 min of reperfusion, immediately after cold flush (t0), the uptake of HA was 72.2+/-2.3% and 69.3+/-1.3% in KC-depleted livers and in control livers, respectively (n.s.). After 8 hr of storage, HA uptake was 21.6+/-4.5% and 34.6+/-8.0%, respectively (n.s.). After 16 and 24 hr of storage and reperfusion, no uptake of HA was found in either KC-depleted or control livers, indicating abolished SEC function. PNP activities in the perfusate were higher in control livers (after 8 and 24 hr of storage), presumably due to release from damaged KC. No difference was found in AST and no tumor necrosis factor-alpha was measured in the perfusates of normal and KC-depleted livers. Electron microscopic studies showed that after 8 and 24 hr of storage and reperfusion, KC were activated and were able to phagocytose colloidal carbon. Our conclusion was that the elimination of Kupffer cells did not result in reduction of ischemic and reperfusion damage in livers preserved up to 24 hr, as assessed in vitro by SEC uptake of HA, PNP release, and AST release.

Elevated plasma levels of hyaluronic acid indicate endothelial cell dysfunction in the initial stages of alcoholic liver disease in the rat

BACKGROUND/AIMS

We used the intragastric feeding rat model for alcoholic liver disease to evaluate the relationship between morphologic and functional indicators of endothelial cell dysfunction.

METHODS

Twelve groups of rats (4-5 rats/group) were fed the following diets: saturated fat and dextrose (SD), saturated fat and ethanol (SE), corn oil and dextrose (CD), corn oil and ethanol (CE). Four of the 12 groups were sacrificed at 2 weeks, four groups at 4 weeks and remaining four groups at 8 weeks. The following were evaluated at sacrifice: pathologic changes in the liver, endothelial cell proliferation using a monoclonal antibody to proliferating cell nuclear antigen, factor VIII-related antigen staining of endothelial cells in liver, plasma endotoxin, hyaluronan and prostaglandin F2 alpha.

RESULTS

Only CE rats at 4 and 8 weeks showed pathologic changes. The plasma levels of HA were significantly higher in the CE groups compared to the other groups at all time intervals studied. In the CE rats, a significant correlation was obtained between plasma endotoxin and hyaluronan (r = 0.84, p < 0.01). Endotoxin levels also correlated significantly with the number of G1/S arrested hepatic sinusoidal endothelial cell (r = 0.61, p < 0.05). A role for prostaglandin F2 alpha, in causing endothelial dysfunction, was suggested by a significant correlation between plasma hyaluronan and prostaglandin F2 alpha levels (r = 0.95, p < 0.01). Positive factor VIII related antigen staining of hepatic endothelial cells was seen in rats with high plasma hyaluronan levels.

CONCLUSION

We propose that endotoxin, mediating part of its effect through prostaglandin F2 alpha, plays a role in hepatic sinusoidal endothelial cell G1/S arrest. This morphologic change, associated with increased plasma hyaluronan levels, precedes capillarization in this model of alcoholic liver injury.

Role of Kupffer cells in cold ischemia/reperfusion injury of rat liver

BACKGROUND & AIMS

Kupffer cell activation is hypothesized to play an etiopathogenic role in storage-related graft failure after liver transplantation. The aim of this study was to verify whether the elimination of Kupffer cells modifies the magnitude of cold ischemia/reperfusion injury of the liver.

METHODS

Rat Kupffer cells were eliminated by an intravenous injection of liposome-encapsulated dichloromethylene diphosphonate. Livers from control and treated rats were isolated and perfused before and after 24-hour cold ischemia in the University of Wisconsin solution (4 degrees C). Hepatocyte and sinusoidal endothelial cell functions were evaluated by taurocholate and hyaluronic acid elimination, respectively. Liver transplantation was also performed using control and treated donor livers stored under identical conditions.

RESULTS

Compared with baseline values, similar alterations were found in both groups after cold ischemia for hepatocyte function (intrahepatic resistance, bile secretion, lactate dehydrogenase release, oxygen consumption, and taurocholate intrinsic clearance) and for sinusoidal endothelial cell function (hyaluronic acid intrinsic clearance). The 10-day survival rate of animals undergoing transplantation was not different between the groups (6 of 15 vs. 4 of 15, control vs. treated donor livers, respectively).

CONCLUSIONS

The presence or absence of Kupffer cells does not modify the effect of 24-hour cold ischemia/reperfusion on the rat liver.