Analysis of multivariables during porcine liver digestion to improve hepatocyte yield and viability for use in bioartificial liver support systems.

In order to achieve optimal BALSS function, preparation of porcine hepatocytes with high yield, viability, and P450 activity is known to be important. To date hepatocyte yields have varied from 0.58 x 10(10) to 3.45 x 10(10) and viabilities from 75% to 95% within and between laboratories, even when using the same digestion methods and procedures, indicating that hepatocyte isolation during porcine liver digestion is not fully optimized. The aim of this work was to identify the critical parameters affecting cell recovery during porcine liver harvesting by investigating 21 variables involved in the process, including pig body and liver weight, different digestion times of perfusates, pH, a range of concentrations of sodium and chloride in EDTA, and collagenase perfusates. Univariate and multivariate analysis of a retrospective study (n = 23) revealed that low perfusate pH during the process of digestion had a positive effect on hepatocyte yield (p < 0.05), while high (relative) concentrations of sodium and chloride in the perfusates had significant negative effects on hepatocyte viability (both p < 0.05). Sodium and chloride had narrow optimal ranges for achieving a >90% viability. These findings were then tested in a prospective study (n = 10) and further verified. High hepatocyte viabilities (91.8+/-1.6% p = 0.036) and yields (2.56+/-0.48 x 10(10)) were achieved consistently, and P450IA1 activity was increased after sodium and chloride concentrations and pH in the perfusates were controlled. The physiological mechanism by which sodium and chloride affects hepatocyte viability during porcine liver digestion is discussed.

Factors affecting hepatocyte viability and CYPIA1 activity during encapsulation.

Hepatocytes encapsulated in alginate-poly-1-lysine-alginate (APA) are used in transplantation studies and in bioartificial liver support systems. Loss of cell viability in the process of APA encapsulation is usually 20-30% while the effect on cytochrome CYP450 activity is rarely reported. This work investigates the negative influences on hepatocyte viability and CYPIA1 activity during APA encapsulation, and reports methods to alleviate these influences by incorporating certain reagents into the encapsulation solution. The results show that loss of hepatocyte viability and CYPIA1 activity was caused almost entirely by extracellular calcium toxicity rather than by mechanical damage (p < 0.05). Use of 10 mM instead of 100 mM calcium chloride (CaCl2) in the encapsulation process improved CYPIA1 activity (p < 0.05), but did not improve hepatocyte viability (p > 0.05) or result in satisfactory microcapsules. Hepatocyte viability was 25% higher (p < 0.05) in CaCl2 than in calcium lactate (CaLa) when the cells were gelled by contact with these calcium solutions at room temperature (RT). Hepatocyte viability showed little improvement by processing at 4 degrees C than at RT in CaCl2 (p > 0.05) but was 23% higher at 4 degrees C than at RT in CaLa (p < 0.05). Calcium used in the process of encapsulation caused cell necrosis rather than apoptosis. Addition of Dulbecco's modified Eagle's medium (containing 10% foetal bovine serum) or 20 mM fructose to the calcium solution did not improve cell survival. However, nifedipine at a final concentration of 25 mM modestly improved hepatocyte survival in solution containing 100 mM CaCl2 (p = 0.003). Glutathione and taurine in certain concentrations showed protective effects against loss of CYPIA1 activity (p < 0.05 and <0.01 respectively). In conclusion, to optimise the use of calcium during the process of encapsulation, CaCl2 is preferred to CaLa and inclusion of nifedipine, glutathione or taurine in 100 mM CaCl2 solution is recommended.

Cadaveric liver procurement using aortic perfusion only.

BACKGROUND: To test the effectiveness of a simpler surgical technique for cadaveric liver procurement for liver transplantation, a prospective randomized study was carried out between August 1994 and December 1995, to compare aortic perfusion only (APO) for flush-preservation of the liver with the conventional combined aortic and portal perfusion (APP) technique. METHODS: Forty multiple organ donors were enrolled with 20 in each arm of the trial. Donor parameters (age, bodyweight, liver function tests), surgeons performing the operations, the involvement of other procurement teams and the total ischaemic times were similar in the two groups. The liver recipients had a wide range of native liver pathology but were of similar age, sex and bodyweight in the two groups. RESULTS: The mean procurement operation times for the APO and APP groups were 126.7+/-38.6 and 137.8+/-55.9 min, respectively (P=ns). The perfusion took longer to complete in the APO group (10.2+/-1.7 vs 7.2+/-1.4 min (APP), P < 0.001). The liver temperature fell to its lowest level (12.5+/-3.4 degrees C (APO) vs 11+/-3 degrees C (APP), P=ns) in a similar time (11.9+/-3.8 min (APO) vs 9.3+/-3.4 mins (APP), P=ns). There was no graft primary non-function or graft arterial injury in either group. There was no significant difference between the APO and APP initial graft outcomes. The 3-month patient survival rate was identical in the two groups (95%); 81% of renal grafts from the APO donors functioned well from the time of transplantation as did 76% of those from APP donors. CONCLUSIONS: It is concluded that the APO procurement technique produces equivalent results to those achieved with the APP method. The simplicity of the APO technique makes it the preferred technique.

Positive biochemical effects of a bioartificial liver support system (BALSS) in a porcine fulminant hepatic failure (FHF) model.

This study describes biochemical changes in the plasma and blood of pigs with devascularised livers treated in a bioartificial liver support system (BALSS). Porcine hepatic cells were incubated with collagen-coated dextran microspheres (CDM) for 3 hours and the medium tested to determine cellular metabolic activity. Incubation continued for a further 18 hours during which the hepatic cells attach to the CDM. The CDM-attached cells were inoculated into a hollow fibre bioreactor which was part of an extracorporeal support system. Hepatic cell content of the bioreactor was 6 x 10(9) cells. The system was tested in a controlled trial in pigs prepared in a surgical model of fulminant hepatic failure (FHF). When plasma from FHF pigs was circulated through the device containing hepatic cells, there was significantly less increase in the accumulation of ammonia and most amino acids, together with a decrease in plasma lactate and of one amino acid, compared to control experiments when hepatic cells were excluded. We conclude that primary porcine hepatocytes can contribute beneficial metabolic function in a BALSS.

Preclinical trial of a bioartificial liver support system in a porcine fulminant hepatic failure model.

BACKGROUND: This study describes the pre-clinical trials of an extracorporeal bioartificial liver support system (BALSS). It includes the biochemical changes which occur in the plasma and blood of pigs with devascularized livers when the plasma is treated in a BALSS, and the testing of the system for presence or absence of infective agents, pyrogens and for toxicity. METHODS: Hepatic cells were prepared from littermate juvenile white landrace pigs with a double-step collagenase digest technique. The cell preparations were incubated with collagen-coated dextran microspheres (CDM) for 3 h and the medium was tested to determine cellular metabolic activity. Incubation continued for a further 20 h during which the hepatic cells attach to the CDM. The CDM-attached cells were inoculated into a hollow fibre bioreactor which was part of an extracorporeal liver support system. RESULTS: Hepatic cell content of the bioreactor was 6 x 10(9) +/- 3 x 10(8) cells, equivalent to those present in half a pig's liver. The system was tested in a controlled trial with the plasma of pigs with fulminant hepatic failure (FHF) due to devascularized livers. When plasma from FHF pigs was circulated through the device there was significantly less of an increase in the accumulation of ammonia, lactate and most amino acids when hepatic cells were included in the circuit compared with those in control experiments when they were excluded. Similar changes occurred in procine blood. There were few infections diagnosed and an absence of pyrogens, endotoxins and toxicity in the bioreactor contents or in the terminating reservoir or animal blood samples. CONCLUSIONS: We believe that the results, demonstrating function of the porcine hepatic cells in the circuit, together with low risks, justify a clinical trial of use of the BALSS in Australia.