R. Glöckner

New Hepatocyte In Vitro Systems for Drug Metabolism: Metabolic Capacity and Recommendations for Application in Basic Research and Drug Development, Standard Operation Procedures

Primary hepatocytes represent a well-accepted in vitro cell culture system for studies of drug metabolism, enzyme induction, transplantation, viral hepatitis, and hepatocyte regeneration. Recently, a multicentric research program has been initiated to optimize and standardize new in vitro systems with hepatocytes. In this article, we discuss five of these in vitro systems: hepatocytes in suspension, perifusion culture systems, liver slices, co-culture systems of hepatocytes with intestinal bacteria, and 96-well plate bioreactors. From a technical point of view, freshly isolated or cryopreserved hepatocytes in suspension represent a readily available and easy-to-handle in vitro system that can be used to characterize the metabolism of test substances. Hepatocytes in suspension correctly predict interspecies differences in drug metabolism, which is demonstrated with pantoprazole and propafenone. A limitation of the hepatocyte suspensions is the length of the incubation period, which should not exceed 4 hr. This incubation period is sufficiently long to determine the metabolic stability and to allow identification of the main metabolites of a test substance, but may be too short to allow generation of some minor, particularly phase II metabolites, that contribute less than 3% to total metabolism. To achieve longer incubation periods, hepatocyte culture systems or bioreactors are used. In this research program, two bioreactor systems have been optimized: the perifusion culture system and 96-well plate bioreactors. The perifusion culture system consists of collagen-coated slides allowing the continuous superfusion of a hepatocyte monolayer with culture medium as well as establishment of a constant atmosphere of 13% oxygen, 82% nitrogen, and 5% CO2. This system is stable for at least 2 weeks and guarantees a remarkable sensitivity to enzyme induction, even if weak inducers are tested. A particular advantage of this system is that the same bioreactor can be perfused with different concentrations of a test substance in a sequential manner. The 96-well plate bioreactor runs 96 modules in parallel for pharmacokinetic testing under aerobic culture conditions. This system combines the advantages of a three-dimensional culture system in collagen gel, controlled oxygen supply, and constant culture medium conditions, with the possibility of high throughput and automatization. A newly developed co-culture system of hepatocytes with intestinal bacteria offers the possibility to study the metabolic interaction between liver and intestinal microflora. It consists of two chambers separated by a permeable polycarbonate membrane, where hepatocytes are cultured under aerobic and intestinal bacteria in anaerobic conditions. Test substances are added to the aerobic side to allow their initial metabolism by the hepatocytes, followed by the metabolism by intestinal bacteria at the anaerobic side. Precision-cut slices represent an alternative to isolated hepatocytes and have been used for the investigation of hepatic metabolism, hepatotoxicity, and enzyme induction. A specific advantage of liver slices is the possibility to study toxic effects on hepatocytes that are mediated or modified by nonparenchymal cells (e.g., by cytokine release from Kupffer cells) because the physiological liver microarchitecture is maintained in cultured slices. For all these in vitro systems, a prevalidation has been performed using standard assays for phase I and II enzymes. Representative results with test substances and recommendations for application of these in vitro systems, as well as standard operation procedures are given.

Application of cryopreserved precision-cut liver slices in pharmacotoxicology — principles, literature data and own investigations with special reference to CYP1A1-mRNA induction

Principle steps necessary for cryopreservation of precision-cut liver slices as currently applied by different groups are summarized including own results concerning mode of freezing. Now we use rapid freezing by immersion in liquid nitrogen after exposure to 10% DMSO as the cryoprotectant for rat liver slices. The results indicate well-maintained cytochrome P450 (CYP)-dependent deethylation rates in slice homogenate after short-term incubation. ECOD rate in intact thawed slices was even higher than in fresh ones after 2 h incubation. In contrast to fresh slices all parameters except protein content decreased to marginal levels during long-term incubation of thawed slices for 24 h. The first preliminary experiments on albumin secretion by thawed rat liver slices, measured between the 2nd and the 4th hour of incubation, showed partial maintenance of this liver specific differentiated function. Trials to induce CYP1A1 in thawed rat liver slices in vitro by beta-naphthoflavone (BNF) resulted in increased expression of CYP1A1-mRNA within 6 h as shown by RT-PCR and quantified by competitive RT-PCR. The decline of deethylation rates, determined in slice homogenates, and of viability within 24 h incubation was not prevented by exposure to BNF or DMSO. The results derived from one sample of cryopreserved human liver slices indicate a quite acceptable maintenance of function up to 6 h, if the same protocol as developed for rat liver slices was used.

Monooxygenation, cytochrome P450-mRNA expression and other functions in precision-cut rat liver slices

Precision-cut rat liver slices (KRUMDIECK slicer, slice thickness 200-250 microm) were incubated in rollers containing modified Williams medium E at 37 degrees C for 2, 24 and 48 hrs. Protein, DNA, potassium and glutathione concentrations did not decrease during 48 hrs. Lactate dehydrogenase (LDH) leakage into the medium was relatively marked during the first 2 hrs of incubation, from the 2nd to the 48th hr LDH leakage was very low. The same is true of the release of thiobarbituric acid-reactive substances. Albumin synthesis and transport into the medium decreased to about 70% after 48 hrs. Cytochrome P450 (CYP)-dependent 7-ethoxycoumarin O-deethylation rate was relatively stable up to 48 hrs, whereas testosterone hydroxylation decreased significantly without alterations of the proportions of the 7 quantified hydroxylated metabolites. After exposure of the slices to beta-naphthoflavone for 6 hrs CYP1A1-mRNA expression, measured by competitive RT-PCR, was increased by a factor of at least 1000. Precision-cut liver slices are a useful tool for the study of various hepatic functions, drug metabolism and its induction in vitro.

Induction of cytochrome P450 2B1-mRNA and pentoxyresorufin O-depentylation after exposure of precision-cut rat liver slices to phenobarbital

Precision-cut rat liver slices were prepared from male Wistar rats with a Krumdieck slicer and cultured in Williams medium E for up to 24 h. In untreated control slices, CYP2B1-mRNA concentration, which was quantified by competitive RT-PCR, did not decrease during this time. After exposure of the slices to 100 microM phenobarbital, CYP2B1-mRNA increased by about 10- or 60-fold after 6 or 24 h, respectively. The extent of this in vitro induction was similar to that after in vivo administration of 60 mg/kg phenobarbital. Pentoxyresorufin O-depentylation (PROD) was also inducible in vitro after 24 h, but to a lesser extent than the corresponding CYP-mRNA. Precision-cut liver slices proved to be a simple and reliable in vitro system for the sensitive detection of an induction by phenobarbital.

Monooxygenation, conjugation and other functions in cryopreserved rat liver slices until 24 h after thawing

For the extensive use of precision-cut liver slices (particularly of human origin) for toxicological investigations successful cryopreservation is necessary. But so far, survival of thawed slices was limited to few hours. This was now overcome by modification of previous procedures. The concentration of DMSO as a cryoprotectant was enhanced to 30%, and washing steps after rapid thawing were omitted. The slices were frozen in liquid nitrogen, thawed at 38 degrees C and incubated immediately in Williams medium E. Protein and potassium contents were stable until 24 h. Glutathione content, amounting to nearly 50% of fresh slices, increased during incubation. High initial lactate dehydrogenase leakage dropped after medium change to less than half during 2-24 h. Testosterone hydroxylation and 7-ethoxycoumarin O-deethylation rates were similar to fresh slices, the latter reaction was inducible by beta-naphthoflavone within 24 h. Methylumbelliferone glucuronidation and p-nitrophenol glucuronidation and sulfation were well measurable and either maintained or decreased by about 50% until 24 h.Altogether, the results are encouraging for further experiments to standardise cryopreservation conditions and to investigate the suitability of this cryopreservation protocol with human liver slices.

Use of Fresh and Cryopreserved Human Liver Slices in Toxicology with Special Reference to In vitro Induction of Cytochrome P450

Human liver slices were prepared with a Krumdieck slicer from macroscopically healthy surgical waste after partial hepatectomy. They were incubated without or with the addition of the inducer beta-naphthoflavone (BNF) (25 mum) either immediately after preparation (fresh slices) for up to 24 hours or after cryopreservation in liquid nitrogen (thawed slices) for up to 6 hours. Potassium concentration was well maintained in fresh and thawed slices over 24 and 6 hours, respectively, but at lower levels than in rat liver slices. Albumin secretion showed relatively large interindividual differences. Both parameters were lower in thawed slices than in fresh ones, but indicated a certain number of viable cells. In untreated fresh slices CYP1A1-mRNA was not detectable; however, it increased distinctly within 6 hours of exposure to BNF. The amounts of induced CYP1A1-mRNA differed by a factor of more than 100 among six human livers and were lower than in fresh rat liver slices. Even in thawed human slices, CYP1A1-mRNA expression could be induced in vitro by BNF, although at a very low level and preferentially in those specimens with comparably high inducibility already before freezing.

Monooxygenation, cytochrome P4501A1 and P4501A1-mRNA in rat liver slices exposed to beta-naphthoflavone and dexamethasone in vitro

Precision-cut liver slices (0.5 mm) were incubated at 30 degrees C in a modified Williams Medium E for up to 48 hrs. During the incubation, K+ and GSH/GSSG concentrations did not decrease. Cytochrome P450-dependent dealkylation rates of 7-ethoxycoumarin (ECOD), 7-allyloxycoumarin (ACOD) and 7-ethoxyresorufin (EROD) decreased to 1/3, 1/2 or did not change at all, respectively, after a 48 hrs incubation period. Exposure of the slices to 25 microM beta-naphthoflavone (beta NF) resulted in about 3 times higher monooxygenation rates. An exposure to a combination beta NF and dexamethasone (10(-6)M) caused a marked induction (6 times higher rates) after 48 hrs. Simultaneously an increase in P4501A1 content was observed. P4501A1-mRNA expression (measured by RT-PCR) was distinctly increased following beta NF exposure for 6 or 24 hrs. DMSO (0.2%) and dexamethasone alone modified monooxygenation rates, but did not have significant effects on P4501A1 content or, in the case of DMSO, P4501A1 gene expression (for dexamethasone not determined). Liver slices are a useful and simple tool for the detection of a beta NF-like induction within a few hours after preparation of the slices.

Ethoxycoumarin O-deethylation (ECOD) activity in rat liver slices exposed to beta-naphthoflavone (BNF) in vitro.

Summary Precision-cut rat liver slices (0.5 mm) were incubated at 30°C in Williams Medium E up to 24 hrs. Our incubation conditions seem to be suitable for maintaining slice function, indicated by constant contents of tissue protein, potassium and glutathione. Thiobarbituric acid reagible substances (TBARS) released into the incubation medium did not significantly increase. Addition of DMSO (0.2 % v/v) or BNF (50 μM) to the incubation medium had no influence on most parameters described above except for increased TBARS release. If ECOD activity was determined in intact liver slices without addition of any cofactor, but substrate only, the main amount of metabolite was found in the medium, and the amount of metabolite retained within the tissue could be neglected. In slices incubated for 24 hrs, no significant changes of ECOD activity occurred for control and DMSO groups, compared with slices incubated for 2 hrs, but in the BNF group activities were more than 3.5 times as high. If ECOD activity was determined in slice homogenate, i. e. with addition of cofactors, decreased activities were measured in all groups after 24 hrs of incubation. This decrease was highest in the control group, lowest in the BNF group. We conclude that intact liver slices can be used as a simple tool to investigate in vitro enzyme induction of BNF type.

In vitro induction of cytochrome P4503A1-mRNA and testosterone hydroxylation in precision-cut liver slices from male and female rats

Cytochrome P450 (CYP) 3A is constitutively highly expressed in the liver. Thus detection of induction might be more difficult than shown for scarcely expressed CYP families. In this paper the suitability of rat liver slices to prove CYP3A inducibility was demonstrated. CYP3A dependent basal testosterone hydroxylation (TH) at positions 15beta, 6beta and 2beta was lower in liver slices from female than male rats, but was more markedly induced by 10(-6) M dexamethasone (DEX) within 24 h (mean induction factors 12.5, 18.3 and 140, respectively, for female slices and 3.7, 2.3 and 3.5, respectively, for male slices). Basal expression of CYP3A1-mRNA was stable in vitro until 24 h and did not differ between male and female rats. In liver slices from male rats this mRNA was induced about 14 fold by both DEX and pregnenolone 16alpha-carbonitrile (PCN) within 24 h. In one sample of a female rat a similar range of CYP3A1-mRNA induction was reached by DEX. Altogether, CYP3A induction can be detected more sensitively in liver slices from female than male rats, if TH rates are used as indicators. With liver slices from male rats CYP3A1-mRNA reacts more sensitively to inducers than TH.

Cryopreserved precision-cut rat liver slices: morphology and cytochrome P450 isoforms expression after prolonged incubation

Precision-cut liver slices are an accepted in vitro system for toxicological investigations. However, cryopreservation of slices would make a more efficient utilisation, particularly of human liver tissue possible. In the present study sections of cryopreserved male rat liver slices were examined immunohistochemically for cytochrome P450 (CYP) isoforms expression after prolonged incubation and after exposure to typical inducers. Morphologically, with just thawed slices no major alterations were seen, but remarkable cell damage was observed even after 2 h of incubation mainly in the middle of the slices and in the periportal and intermediate regions of the lobules. After 24 h of incubation, viable cells were only observed at the edges of the slices or around bigger vessels. In the viable cells of the cryopreserved liver slices after 2 h of incubation CYP expression pattern was similar to that in normal liver specimens: a low CYP1A1, but a strong CYP2B1 and 3A2 expression predominantly in the central and intermediate lobular zones. After 24 h, the immunostaining for CYP2B1 and 3A2 in the viable cells was reduced, but that for CYP1A1 was increased. Incubation with beta-naphthoflavone further elevated CYP1A1 and 2B1 expression. Phenobarbital caused an enhanced CYP2B1 and 3A2 and dexamethasone and pregnenolone 16 alpha-carbonitrile an increased CYP3A2 immunostaining. These results show that also in cryopreserved liver slices and after a prolonged incubation, a distinct expression pattern and an in vitro induction of phase I enzymes can be demonstrated immunohistochemically.