Hepatocyte-based in vitro models and their application in pharmacotoxicology

The Use of Long-term Hepatocyte Cultures for Detecting Induction of Drug Metabolising Enzymes: The Current Status

In this report, metabolically competent in vitro systems have been reviewed, in the context of drug metabolising enzyme induction. Based on the experience of the scientists involved, a thorough survey of the literature on metabolically competent long-term culture models was performed. Following this, a prevalidation proposal for the use of the collagen gel sandwich hepatocyte culture system for drug metabolising enzyme induction was designed, focusing on the induction of the cytochrome P450 enzymes as the principal enzymes of interest. The ultimate goal of this prevalidation proposal is to provide industry and academia with a metabolically competent in vitro alternative for long-term studies. In an initial phase, the prevalidation study will be limited to the investigation of induction. However, proposals for other long-term applications of these systems should be forwarded to the European Centre for the Validation of Alternative Methods for consideration. The prevalidation proposal deals with several issues, including: a) species; b) practical prevalidation methodology; c) enzyme inducers; and d) advantages of working with independent expert laboratories. Since it is preferable to include other alternative tests for drug metabolising enzyme induction, when such tests arise, it is recommended that they meet the same level of development as for the collagen gel sandwich long-term hepatocyte system. Those tests which do so should begin the prevalidation and validation process.

Optimization of the isolation and cultivation of Cyprinus carpio primary hepatocytes

The aquatic environment is affected by numerous chemical contaminants. There is an increasing need to identify these chemicals and to evaluate their potential toxicity towards aquatic life. In this research we optimized techniques for primary cell culture of Cyprinus carpio hepatocytes as one adjunct model for ecotoxicological evaluation of the potential hazards of xenobiotics in the aquatic environment. In this study, Cyprinus carpio hepatocytes were isolated by mechanical separation, two-step collagenase perfusion, and pancreatin digestion. The hepatocytes or parenchymal cells could be separated from cell debris and from non-parenchymal cells by low-speed centrifugation (Percoll gradient centrifugation). The harvested hepatocytes were suspended in DMEM, M199 (cultured in 5% CO(2)), or L-15 (cultured without 5% CO(2)) medium then cultured at 17, 27, or 37 degrees C. Cell yield was counted by use of a hemocytometer, and the viability of the cells was assessed by use of the Trypan blue exclusion test. Results from these studies showed that the best method of isolation was pancreatin digestion (the cell yield was 2.7 x 10(8) per g (liver weight) and the viability was 98.4%) and the best medium was M199 (cultured in 5% CO(2)) or L-15 (cultured without 5% CO(2)). The optimum culture temperature was 27 degrees C. The primary hepatocytes culture of Cyprimus carpio grew well and satisfied requirements for most toxicological experiments in this condition.

Characterization of cytochrome P450 expression in murine embryonic stem cell-derived hepatic tissue system

An in vitro system for liver organogenesis from murine embryonic stem (ES) cells has been recently established. This system is expected to be applied to the development of a new drug metabolism assay system that uses ES cells as a substitute for animal experiments. The objective of this study was to elucidate the drug metabolism profiles of the murine ES cell-derived hepatic tissue system compared with those of primary cultures of murine adult and fetal hepatocytes. The expression of the genes of the cytochrome P450 (P450) family, such as Cyp2a5, Cyp2b10, Cyp2c29, Cyp2d9, Cyp3a11, and Cyp7a1, was observed in the murine ES cell-derived hepatic tissue system at 16 days and 18 days after plating (A16 and A18). To investigate the activities of these P450 family enzymes in the murine ES cell-derived hepatic tissue system at A16 and A18, testosterone metabolism in this system was analyzed. Testosterone was hydroxylated to 6beta-hydroxytestosterone (6beta-OHT), 16alpha-OHT, 2alpha-OHT, and 2beta-OHT in this system, and was not hydroxylated to 15alpha-OHT, 7alpha-OHT, and 16beta-OHT. This metabolism profile was similar to that of fetal hepatocytes and different from that of adult hepatocytes. Furthermore, pretreatment with phenobarbital resulted in a 2.5- and 2.6-fold increase in the production of 6beta-OHT and 16beta-OHT. Thus, evidence for drug metabolic activities in relation to P450s has been demonstrated in this system. These results in this system would be a stepping stone of the research on the development and differentiation to adult liver.

Characterization of cytochrome P450 expression in murine embryonic stem cell-derived hepatic tissue system

An in vitro system for liver organogenesis from murine embryonic stem (ES) cells has been recently established. This system is expected to be applied to the development of a new drug metabolism assay system that uses ES cells as a substitute for animal experiments. The objective of this study was to elucidate the drug metabolism profiles of the murine ES cell-derived hepatic tissue system compared with those of primary cultures of murine adult and fetal hepatocytes. The expression of the genes of the cytochrome P450 (P450) family, such as Cyp2a5, Cyp2b10, Cyp2c29, Cyp2d9, Cyp3a11, and Cyp7a1, was observed in the murine ES cell-derived hepatic tissue system at 16 days and 18 days after plating (A16 and A18). To investigate the activities of these P450 family enzymes in the murine ES cell-derived hepatic tissue system at A16 and A18, testosterone metabolism in this system was analyzed. Testosterone was hydroxylated to 6beta-hydroxytestosterone (6beta-OHT), 16alpha-OHT, 2alpha-OHT, and 2beta-OHT in this system, and was not hydroxylated to 15alpha-OHT, 7alpha-OHT, and 16beta-OHT. This metabolism profile was similar to that of fetal hepatocytes and different from that of adult hepatocytes. Furthermore, pretreatment with phenobarbital resulted in a 2.5- and 2.6-fold increase in the production of 6beta-OHT and 16beta-OHT. Thus, evidence for drug metabolic activities in relation to P450s has been demonstrated in this system. These results in this system would be a stepping stone of the research on the development and differentiation to adult liver.

Differential effects of histone deacetylase inhibitors in tumor and normal cells-what is the toxicological relevance?

Histone deacetylase (HDAC) inhibitors target key steps of tumor development: They inhibit proliferation, induce differentiation and/or apoptosis, and exhibit potent antimetastatic and antiangiogenic properties in transformed cells in vitro and in vivo. Preliminary studies in animal models have revealed a relatively high tumor selectivity of HDAC inhibitors, strenghtening their promising potential in cancer chemotherapy. Until now, preclinical in vitro research has almost exclusively been performed in cancer cell lines and oncogene-transformed cells. However, as cell proliferation and apoptosis are essential for normal tissue and organ homeostasis, it is important to investigate how HDAC inhibitors influence the regulation of and interplay between proliferation, differentiation, and apoptosis in primary cells as well. This review highlights the discrepancies in molecular events triggered by trichostatin A, the reference compound of hydroxamic acid-containing HDAC inhibitors, in hepatoma cells and primary hepatocytes (which are key targets for drug-induced toxicity). The implications of these differential outcomes in both cell types are discussed with respect to both toxicology and drug development. In view of the future use of HDAC inhibitors as cytostatic drugs, it is highly recommended to include both tumor cells and their healthy counterparts in preclinical developmental studies. Screening the toxicological properties of compounds early in their development process, using a battery of different cell types, will enable researchers to discard those compounds bearing undesirable adverse activity before entering into expensive clinical trials. This will not only reduce the risk for harmful exposure of patients but also save time and money.

Secretion of functional plasma haemostasis proteins in long-term primary cultures of human hepatocytes

This study was designed to investigate the ability of long-term primary cultures of adult human hepatocytes to secrete the main haemostasis proteins. Factors II, V, VII, VIII, PIVKA-II (protein induced by vitamin K 1 absence or antagonist II), fibrinogen and antithrombin were quantified in culture medium by immunological methods and by measuring the coagulant activity of factors II, V and VII. All the haemostasis protein antigens except the factor VIII antigen (FVIII:Ag) were found in the culture medium throughout the culture period. The clotting activity of each factor correlated well with antigen level. In addition, fibrinogen and fibrin were detected in the fibrillar material following incubation of the culture medium with thromboplastin. Moreover, adding vitamin K 1 to the culture medium resulted in a significant increase of factors II and VII and a reciprocal decrease of the PIVKA-II, and adding von Willebrand factor resulted in a drastic increase of the level of FVIII:Ag. We conclude that, in our culture system, normal adult human hepatocytes retain their capacity to secrete haemostasis proteins for at least 30 days.

Early oxidative damage in primary cultured trout hepatocytes: a time course study

The aim of this study was to evaluate the influence of the two-step hepatocyte isolation procedure on primary cultured trout (Oncorhynchus mykiss) hepatocytes over time. We characterised the possible changes of a variety of some cellular parameters within the first 24-48 h after seeding. We followed the time dependent changes of these parameters during subsequent culture times in order to see if the cells maintained a differentiated status. Scanning electron microscopy revealed bleb formation and 20% cell damage in freshly isolated hepatocytes. During subsequent culture times the bleb dimension appear to be reduced. Heat shock proteins 70 and 50 (HSP70, HSP50) were induced by hepatocyte isolation. During the first 4 h of culture, the hepatocytes showed a variation in mitochondrial activity, an increase in free radical species (ROS), and a decrease in both glutathione (GSH) content and catalase (CAT) activity; the generation of free radicals led to an increase in the formation of 8-hydroxydeoxyguanosine (8-OHdG) in the DNA. The cells showed detectable ethoxyresorufin-O-deethylase activity after 4 h of culture, which had rapidly increased by the 24th hour. After 24 h, mitochondrial and CAT activity, free radical production, and the content of GSH and 8-OHdG returned to their original levels. P450 activity was retained for at least 48 h after seeding. Our data show that trout hepatocytes suffer significant cell injury as a result of the isolation procedure, but primary cultured cells metabolically recover from this stress after a few hours: they are capable of repairing their damaged surfaces, recovering their antioxidant defences and retaining their ability to repair DNA. Our results also confirm that trout hepatocytes in a primary culture maintain their in vivo-like metabolic activities for 3-8 days.

Effects of medium composition on the morphology and function of rat hepatocytes cultured as spheroids and monolayers

Primary hepatocytes cultured as monolayers or as spheroids were studied to compare the effects of four different culture media (Williams' E, Chee's, Sigma Hepatocyte, and HepatoZYME medium). Rat hepatocytes were cultured as conventional monolayers for 3 d or as spheroids for 2 wk. For spheroid formation a method was emplOyed that combined the use of a nonadherent substratum with rotation of cultures. Hepatocyte integrity and morphology were assessed by light and electron microscopy and by reduced glutathione content. Hepatocyte function was measured by albumin secretion and 7-ethoxycoumarin metabolism. Chee's medium was found to be optimal for maintenance of hepatocyte viability and function in monolayers, but it failed to support spheroid formation. For spheroid formation and for the maintenance of spheroid morphology and function, Sigma HM was found to be optimal. These results demonstrate that the medium requirements of hepatocytes differ markedly depending on the culture model employed. Spheroid culture allowed better preservation of morphology and function of hepatocytes compared with conventional monolayer culture. Hepatocytes in spheroids formed bile canaliculi. and expressed an actin distribution resembling that found in hepatocytes in vivo. Albumin secretion was maintained at the same level as that found during the first d in primary culture, and 7-ethoxycoumarin metabolism was maintained over 2 wk in culture at approximately 30% of the levels found in freshly isolated hepatocytes. The improved morphology and function of hepatocyte cultures as spheroids may provide a more appropriate in vitro model for certain applications where the maintenance of liver-specific functions in long-term culture is crucial.

Cytochrome P450 activity in control and induced long-term cultures of rat hepatocyte spheroids

Long-term events such as enzyme induction or chronic toxicity require long-term liver culture models that maintain activity of xenobiotic metabolising enzymes. The levels of these enzyme activities and their responsiveness to chemical induction was studied in rat hepatocyte spheroids, a potential long-term hepatocyte culture model. In comparison with other long-term liver culture models, the basal metabolic activity of spheroids has not been well studied. Additionally, no existing data on the induction of CYP3A activity in spheroids could be found. The basal xenobiotic metabolising activity of rat hepatocyte spheroids was monitored over 14 days in culture, using testosterone as a probe substrate. When spheroids from days 2-14 in culture were compared to 24-h control spheroids, there was a differential maintenance of basal CYP activity. CYP2A and CYP3A activities were maintained over the culture period, while there were time-related decreases in CYP2C11 and CYP2C/CYP2B1/2 activities. The responsiveness of rat hepatocyte spheroids to chemical induction was studied following treatment with phenobarbitone (PB) or dexamethasone (DEX). PB treatment induced CYP2A, CYP2C, CYP2B1/2 and CYP3A activities. DEX treatment resulted in an induction of CYP3A and CYP2C11 activities. The results demonstrate that rat hepatocyte spheroids retained some of the liver-specific functions essential in a long-term hepatocyte culture model, thus making spheroids comparable to other long-term culture models available.

Effects of medium composition on the morphology and function of rat hepatocytes cultured as spheroids and monolayers

Primary hepatocytes cultured as monolayers or as spheroids were studied to compare the effects of four different culture media (Williams' E, Chee's, Sigma Hepatocyte, and HepatoZYME medium). Rat hepatocytes were cultured as conventional monolayers for 3 d or as spheroids for 2 wk. For spheroid formation a method was emplOyed that combined the use of a nonadherent substratum with rotation of cultures. Hepatocyte integrity and morphology were assessed by light and electron microscopy and by reduced glutathione content. Hepatocyte function was measured by albumin secretion and 7-ethoxycoumarin metabolism. Chee's medium was found to be optimal for maintenance of hepatocyte viability and function in monolayers, but it failed to support spheroid formation. For spheroid formation and for the maintenance of spheroid morphology and function, Sigma HM was found to be optimal. These results demonstrate that the medium requirements of hepatocytes differ markedly depending on the culture model employed. Spheroid culture allowed better preservation of morphology and function of hepatocytes compared with conventional monolayer culture. Hepatocytes in spheroids formed bile canaliculi. and expressed an actin distribution resembling that found in hepatocytes in vivo. Albumin secretion was maintained at the same level as that found during the first d in primary culture, and 7-ethoxycoumarin metabolism was maintained over 2 wk in culture at approximately 30% of the levels found in freshly isolated hepatocytes. The improved morphology and function of hepatocyte cultures as spheroids may provide a more appropriate in vitro model for certain applications where the maintenance of liver-specific functions in long-term culture is crucial.

Metabolic activity in primary cultures of fish hepatocytes

In aquatic toxicology, isolated liver cells from fish can be used as a tool to generate initial information on the hepatic metabolism of xenobiotics, and on the mechanisms of xenobiotic activation or deactivation. This isolation of teleost liver cells is achieved by enzymic dissociation, and monolayer cultures of fish hepatocytes in serum-free medium maintain good viability for 3-8 days. During in vitro culture, fish liver cells express stable levels of phase I and phase II enzymes, such as cytochrome P4501A or glutathione S-transferase, and the cells show an induction of biotransformation enzymes after exposure to xenobiotics. The xenobiotic metabolite pattern produced by fish hepatocytes in vitro is generally similar to that observed in vivo. Limitations to more-intensive application of cultured fish hepatocytes as a screen in aquatic hazard assessment are partly due to the rather limited scope of existing studies, i.e. the focus on one particular species (rainbow trout), and on one particular biotransformation enzyme (cytochrome P4501A), as well as a lack of comparative in vitro/in vivo studies.

The use of liver spheroids as an in vitro model for studying induction of the stress response as a marker of chemical toxicity

Stress protein induction has been advocated as a sensitive indicator of compound-induced toxicity. In monolayer cultures of primary hepatocytes, however, the two stress proteins, Hsp25 and Hsp72/3 are up-regulated, probably due to the effect of the isolation procedure and adaptation of the cells to the culture conditions. The aim of the current studies was to determine whether liver spheroids would provide an improved experimental model for the study of heat shock protein induction in vitro. Primary rat hepatocytes were cultured as liver spheroids and the expression of Hsp25 and Hsp72/3 measured along with the levels of ATP, GSH and albumin secretion. Hsp72/3 was initially increased in spheroid culture but returned to in vivo levels after 3 days of culture. Hsp25 was maintained at in vivo levels until day 6 of culture, after which levels increased slightly. The effects of the two hepatotoxins, hydrazine and cadmium chloride (CdCl(2)), were therefore measured on day 6 of spheroid culture. CdCl(2) had no effect on Hsp25 but increased Hsp72/3 at concentrations that affected other biochemical parameters. Hydrazine caused a rapid reduction in ATP levels and albumin secretion, but did not affect Hsp72/3. Hsp25 was slightly induced by hydrazine at later sampling times at concentrations, however, that affected other biochemical parameters. It can be concluded that liver spheroids provide a model for studying stress protein expression. However, the increase in stress proteins appears to be a relatively insensitive parameter compared to other more conventionally used toxicity endpoints and the response appears to vary with individual toxins under study.

Long-term primary cultures of adult human hepatocytes

In this work we have investigated a system of long-term primary cultures of adult human hepatocytes which, in contrast to those previously described, has the advantage of requiring neither the use of additive cells as in co-cultures, nor of matrix component preparations like Matrigel or collagen sandwich. This system has been used previously for long-term cultures of hepatocytes from young baboon, and some modifications have been introduced here to take into account the specificity of adult human hepatocytes. In this system, hepatocytes are plated at confluence on collagen-coated dishes and cultured in a serum-free medium consisting of Williams'E supplemented with hormones and growth factors. Proteins secreted specifically by the liver, including albumin, alpha-1 antitrypsin, plasminogen, fibrinogen, lipoproteins ApoA1 and ApoB100, have been quantified in the extracellular medium as a function of time, either by immunoblot or ELISA. In addition, the expression and inducibility of CYP proteins of the CYP1, CYP2 and CYP3 families in response to their prototypical inducers including 2,3,7,8-tetrachlorodibenzo(p)dioxin and rifampicin, have been evaluated by immunoblot analysis of microsomes or cell lysates. Moreover, the oxidative metabolism of cyclosporin A, a monoxygenase activity depending on CYP3A4, has been monitored directly on the cultured cells by HPLC analysis of extracellular medium. Our results show that, under these culture conditions, adult human hepatocytes retain these phenotypical characteristics for at least 35 days. This system meets the requirements for use as a model for screening CYP protein inducers.

Activity and inducibility of drug-metabolizing enzymes in immortalized hepatocyte-like cells (mhPKT) derived from a L-PK/Tag1 transgenic mouse

This report describes the establishment and characterization of the mhPKT cell line derived from the liver of a transgenic mouse harboring the simian virus (SV40) large T and small t antigens placed under the control of the 5' regulatory sequence of the rat L-type pyruvate kinase (L-PK) gene. mhPKT cells had a prolonged life span, expressed the SV40-encoded nuclear large T antigen when grown in glucose-enriched medium, and induced tumors when injected subcutaneously into athymic (nu-nu) mice. Growth on petri dishes or filters yielded multiple layers of cuboid cells, with numerous spaces between adjacent cells that were closed by junctional complexes. These bile canaliculi-like structures exhibited numerous microvilli in which villin, an actin-binding brush-border protein, colocalized with actin. These bile canaliculi-like structures appeared to be functional as they accumulated fluorescein. mhPKT cells conserved the expression of the liver-specific transcription factors HNF1, HNF3, HNF4, and DBP together with substantial levels of L-PK and albumin but not alpha-fetoprotein mRNA transcripts. mhPKT cells mainly metabolized testosterone into androstenedione and 6beta-hydroxytestosterone, as in vivo. 3-Methylcholanthrene and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) markedly increased ethoxyresorufin-O-deethylase activity and the related cytochrome P450 (CYP) 1A1/2 protein, whereas alpha-naphtoflavone antagonized the TCDD-elicited induction. Phenobarbital slightly increased the CYP2B-mediated activities of pentoxyresorufin-O-depentylase, 2beta- and 16beta-testosterone hydroxylase. mhPKT cells also had substantial sulfotransferase, UDP-glucuronyltransferase, and glutathione S-transferase activities. This model may serve as a tool for long-term in vitro studies of xenobiotic metabolism, potent CYP inducers, and hepatocyte damage due to drugs and other factors.