Metabolism of methoxymorpholino-doxorubicin in rat, dog and monkey liver microsomes: comparison with human microsomes.

The morpholino anthracycline, methoxymorpholino-doxorubicin (MMDx) is a novel anticancer agent. The metabolism of this highly lipophilic doxorubicin analogue is not fully elucidated. MMDx is metabolically activated in vivo, resulting in an 80-fold increase in potency over the parent drug. In this study, MMDx in vitro metabolism was compared in rat, dog, monkey and human liver microsomes. When microsomal fractions were incubated with MMDx, 6-8 metabolites were formed depending on the species and on the substrate concentrations. Among these eight metabolites, three comigrated with authentic standards, namely MMDx-ol, PNU156686 and PNU159682, and the five others are in the process of being characterized. Quantitatively, monkey and human metabolize MMDx with a higher rate than rat and dog. Qualitatively, MMDx metabolic profile in dog microsomes was different from the three other species. MMDx-ol was predominant in dog and only minor in other species. In conclusion, MMDx metabolism was species-different. Rat and monkey liver microsomes may be used as models to study MMDx metabolism in humans. Dog liver microsomes may be a good model for studying the formation of MMDx-ol.

Induction and inhibition of cicletanine metabolism in cultured hepatocytes and liver microsomes from rats.

Cicletanine, a racemic furopyridine derivative synthesized as racemate, is used as an antihypertensive agent. Its two enantiomers are involved in the pharmacological effects of the drug. Cicletanine is metabolized by conjugation enzyme systems (phase II) into sulfoconjugated or glucuroconjugated enantiomers. This study reports on the use of both the induction with 3-methylcholanthrene (3-MC) or phenobarbital (PB) and inhibition with selective compounds to determine and identify UGT isoenzymes involved in the metabolism of cicletanine enantiomers. PB and 3-MC both enhanced the cicletanine enantiomer glucuronidation. These two compounds being known as inducing agents of UGT2B1 and UGTIA6 isoforms, respectively, this suggests an implication of UGT2B1 and UGT1A6 isoforms in the metabolism of the two cicletanine enantiomers: ( + )-cicletanine and ( - )-cicletanine. The use of selective compounds for inhibition study evidenced, in addition to UGT2B1 and UGT1A6 isoforms, the involvement of other UGT isoforms such as UGT1A1, UGT2B7 and UGT2B15 in cicletanine metabolism.

Stereoselective biotransformation of cicletanine in cultured rat and human hepatocytes.

Cicletanine [(+/-)-C] is a racemic furopyridine derivative used as an antihypertensive agent. Pharmacokinetic and metabolic studies have shown that cicletanine is rapidly and almost fully metabolized into sulfo- and glucuro-conjugated metabolites. However, the stereoselective metabolism of cicletanine is not well-known in humans. In the present study, the stereoselective aspect of cicletanine metabolism was investigated in cultured hepatocytes from humans and rats. The two enantiomers of cicletanine were both strongly metabolized in rat hepatocytes. So, after 24 h of incubation, very low amounts of free cicletanine were found [1.2% for (+)-C and 2.7% for (-)-C], respectively. In addition (+/-)-C was mainly biotransformed into conjugated metabolites: (-)-C mainly transformed into (-)-C-glucuronide (78.3+/-6.4%) and (+)-C mainly into (+)-C-sulfate (87.4+/-3.3%). In human hepatocytes, inter-individual variations were more marked than in rat hepatocytes. In addition (+/-)-C biotransformation in human was lower than the one observed in rat. (-)-C enantiomer was more metabolized than (+)-C. After a 24-h incubation the percentages of free (+)-C and (-)-C were 32.3+/-16.4 and 8.2+/-10.3, respectively. On the contrary to rat hepatocytes, both cicletanine enantiomers in humans were mainly metabolized into glucuroconjugated metabolites. These results clearly demonstrated that cicletanine underwent stereospecific metabolism in both rat and human hepatocytes.

Effects of oxazepam and acetaminophen on cicletanine metabolism in rat hepatocytes and liver microsomes.

Cicletanine, a racemic furopyridine derivative synthesized as racemate, is used as an antihypertensive agent. Its two enantiomers are involved in the pharmacological effects of the drug. Cicletanine is metabolized by conjugation enzyme systems (phase II) into sulfoconjugated or glucuroconjugated enantiomers. As oxazepam and acetaminophen are widely prescribed, especially to elderly patients, these two drugs may be co-administered with cicletanine. The metabolic profile and the kinetics of biotransformation were studied by using rat hepatocytes and liver microsomes. Cicletanine was extensively metabolized by rat hepatocytes. More than 80% of the drug was biotransformed after a 3 h incubation. The formation of glucuroconjugated metabolites was characterized by the following kinetic parameters, i.e. Vmax = 2.05 +/- 0.21 nmol/min/mg protein and Km = 287 +/- 6.7 microM for (-)-cicletanine, and Vmax = 1.44 +/- 0.12 nmol/min/mg protein and K(m) = 171 +/- 4.1 microM for (+)-cicletanine. Oxazepam inhibited the glucuronidation of cicletanine in both rat hepatocytes and liver microsomes with a competitive-type inhibition, i.e. K(i) = 129 +/- 7.5 and 152 +/- 19.7 microM for (-)-cicletanine and (+)-cicletanine, respectively. The co-incubation of acetaminophen with cicletanine showed that only sulfoconjugation was inhibited in rat hepatocytes. Glucuronidation was not modified by acetaminophen. As natriuric activity is due to sulfoconjugated (+)-cicletanine, acetaminophen could potentially modulate in vivo the pharmacological effect of cicletanine. The data of the in vitro study reported here suggested an interaction between cicletanine and oxazepam or cicletanine and acetaminophen. However, the clinical impact of such a drug interaction needs further evaluation.

Myelosuppressive activity of two herbicides, atrazine and dinoterb, on human haematopoietic progenitor cells: An in vitro assay to evaluate the effects of intermediate or long-term exposure.

The effects of chronic exposure to phytosanitary products are difficult to determine because of their use in combination with other products and their variety of formulations containing additives or contaminants. In order to evaluate, at the cellular level, the risk of myelosuppressive effects caused by two widely used herbicides, atrazine and dinoterb, we performed in vitro assays on human granulo-monocytic progenitor-cells (CFU-GM) and also granulomonocytic expansion in liquid media. Both of these techniques were carried out in the presence of each molecule. Seven stable environmental metabolites of atrazine were studied using the above techniques in addition to supernatants of rat hepatocytes preincubated with atrazine and dinoterb for 24 hr. Parent atrazine and dinoterb showed similar moderate-direct toxicity on CFU-GM. In cells grown in liquid media for a period longer than 14 days, dinoterb toxicity appeared delayed but increased when compared with atrazine. 2-chloro-diamino-atrazine was found to be as toxic as atrazine on CFU-GM. Supernatants of rat hepatocyte preincubated for 24 hr with dinoterb exhibited a 150-fold increase in toxicity compared with the parent molecule, while toxicity remained unchanged for atrazine. This phenomenon was directly correlated to toxicity on rat hepatocytes. The present study will be useful in defining tissue-specific toxicities of phytosanitary products, including environmental or biotransformed metabolites.

Glucuronidation of SN-38, the active metabolite of irinotecan, by human hepatic microsomes.

We have investigated the glucuronidation in vitro of SN-38, the active metabolite of irinotecan, a semi-synthetic anticancer drug derived from 20(S)camptothecin. Preparations of human hepatic microsomes (final concentration : 1 mg prot./ml), were incubated for 1 hr in 0.1 M Tris buffer, pH 7.4, containing 10 mM MgCl2, in the presence of UDP-glucuronic acid (4 mM), saccharolactone (4 mM), and a detergent. Microsomes from five livers were studied individually or as a pooled preparation. SN-38, either in its lactone or its carboxylate form, was added at a range of concentrations. The SN-38 beta-glucuronide formed was measured by HPLC with fluorometric detection. The glucuronidation reaction appeared linear over 1 hr in these conditions and Brij 35 at 0.5 mg/mg prot. was the best activator. The apparent parameters of the reaction were independent of the molecular form of the substrate. The half-saturation constant was 17-20 microM and Vmax was 60-75 pmol/min./mg prot. The interindividual variation of SN-38 glucuronidation was relatively low (ratio of 1.8 between extreme values). In addition, the effect of twelve drugs currently associated with irinotecan in clinics was evaluated in this system (drug concentration: 100 microM; SN-38 concentration: 5 microM). These produced little if any interference with SN-38 glucuronidation. Therefore, major interferences of this transformation by comedications are unlikely to occur in vivo.

Both cytochromes P450 2E1 and 3A are involved in the O-hydroxylation of p-nitrophenol, a catalytic activity known to be specific for P450 2E1.

4-Nitrophenol 2-hydroxylation activity was previously shown to be mainly catalyzed by P450 2E1 in animal species and humans. As this chemical compound is widely used as an in vitro probe for P450 2E1, this study was carried out to test its catalytic specificity. First, experiments were carried out on liver microsomes and hepatocyte cultures of rat treated with different inducers. Liver microsomes from pyrazole- and dexamethasone-treated rats hydroxylated p-nitrophenol with a metabolic rate increased by 2.5- and 2.7-fold vs control. Dexamethasone treatment increased the hepatic content of P450 3A but not that of P450 2E1. Two specific inhibitors of P450 3A catalytic activities, namely, ketoconazole and troleandomycin (TAO), inhibited up to 50% of 4-nitrophenol hydroxylation in dexamethasone-treated rats but not in controls. Hepatocyte cultures from dexamethasone-treated rats transformed p-nitrophenol into 4-nitrocatechol 7.8 times more than controls. This catalytic activity was inhibited by TAO. Similarly, hepatocyte cultures from pyrazole-treated rats hydroxylated p-nitrophenol with a metabolic ratio increased by about 8-fold vs control. This reaction was inhibited by diethyl dithiocarbamate and dimethyl sulfoxide, both inhibitors of P450 2E1. Second, the capability of human P450s other than P450 2E1 to catalyze the formation of 4-nitrocatechol was examined in a panel of 13 human liver microsomes. Diethyl dithiocarbamate and ketoconazole reduced 4-nitrophenol hydroxylase activity by 77% (+/- 11) and 13% (+/- 16), respectively. Furthermore, the residual activity following diethyl dithiocarbamate inhibition was significantly correlated with seven P450 3A4 catalytic activities. Finally, the use of human cell lines genetically engineered for expression of human P450s demonstrated that P450 2E1 and 3A4 hydroxylated 4-nitrophenol with turnovers of 19.5 and 1.65 min-1, respectively. In conclusion, P450 3A may make a significant contribution to 4-nitrophenol hydroxylase activity in man and rat.

Metabolism and toxicity of coumarin on cultured human, rat, mouse and rabbit hepatocytes.

We compared the cytotoxic effect of coumarin and its derivatives, 7-hydroxycoumarin (7-OHC), 4-hydroxycoumarin (4-OHC), o-hydroxyphenyl acetic acid (OHPAA) and o-coumaric acid (CA), on cultured hepatocytes from human, rat, mouse and rabbit liver. At 10(-5) and 5 x 10(-5) M, coumarin and its derivatives did not give rise to any signs of toxicity on cultured hepatocytes of the four species. At 10(-4) M, coumarin, but not its derivatives, induced release of lactate dehydrogenase (LDH) into the medium, especially in rat hepatocyte cultures. Intracellular LDH activities were correspondingly reduced. The cytotoxic effect of coumarin in cultured rat hepatocytes was evidenced on morphological examination and from the results of the 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium (MTT) reduction test. At higher concentrations (5 x 10(-4) M), 7-OHC and CA were also found to be cytotoxic in cultured rat hepatocytes. The cytotoxic effect of coumarin (5 x 10(-4) M) was decreased in the presence of SKF 525-A, a cytochrome P450 inhibitor. Interspecies comparisons showed that rat hepatocytes were the most sensitive to the toxicity of coumarin and its derivatives, whereas human hepatocytes were the most resistant. Our results suggest that the cytotoxicity of coumarin is metabolism and species-dependent. Thus, the rat may not be a suitable model for evaluating the pharmacological hazards of coumarin in humans.

Both cytochromes P450 2E1 and 1A1 are involved in the metabolism of chlorzoxazone.

Chlorzoxazone, a centrally acting muscle relaxant, was previously shown to be hydroxylated on carbon 6 specifically by cytochrome P450 2E1. Accordingly, this drug has been proposed as a potential noninvasive in-vivo probe for screening the hepatic P450 2E1 activity. This study was carried out to test the specificity of such a substrate when first experiments conducted by using human hepatocyte cultures showed that the chlorzoxazone 6-hydroxylation activity increased after 3-methylcholanthrene treatment of cells. Indeed, the ability of both rat and human hepatocytes to metabolize chlorzoxazone significantly increased after treatments by 3-methylcholanthrene alone or plus ethanol, suggesting the involvement of P450 1A enzymes in this oxidative reaction. Identical results were obtained by in-vivo treatment of rats with four inducers of P450 1A enzymes, namely, beta-naphthoflavone, isosafrole, Arochlor 1254, and 3-methylcholanthrene. Furthermore, the chlorzoxazone 6-hydroxylation activity was inhibited by both alpha-naphthoflavone and dimethyl sulfoxide, both known to inhibit P450 1A and P450 2E1 activities, respectively. Finally, the use of yeasts genetically engineered for expression of human P450 1A1, 1A2, 2C9, and 3A4 demonstrated that P450 1A1 was significantly involved in this catalytic activity. In conclusion, these results taken together suggest that chlorzoxazone should be used with precaution as in-vivo tool for evaluating P450 2E1. However, the relative Km of P450 1A1 and 2E1 for chlorzoxazone and, on the other hand, the relative levels of these two enzymes in the human liver suggest that P450 2E1 would generally be the major form metabolizing chlorzoxazone in-vivo.

Modulation of multidrug resistance gene expression in rat hepatocytes maintained under various culture conditions.

P-glycoprotein (P-gp), the multidrug resistance gene product, is overexpressed in normal adult rat hepatocytes under standard culture conditions. We have studied the modulation of P-gp expression in this in vitro model in the presence of both epidermal growth factor and pyruvate, which favor hepatocyte growth, as well as in the presence of either dimethyl sulfoxide (DMSO) or nicotinamide, which favors maintenance of differentiated functions. P-gp overexpression, estimated by northern blotting and doxorubicin-mediated drug efflux analyses, was similarly observed during culture in both standard and proliferating conditions, while it was delayed, but not inhibited, in the presence of DMSO or nicotinamide. These results suggest that the functional P-gp overexpression occurring in rat hepatocytes when exposed to an unfamiliar environment is at least partly not related to cell proliferation or the degree of cell differentiation in vitro.

Overexpression of the multidrug resistance gene product in adult rat hepatocytes during primary culture.

Expression of P-glycoprotein (P-gp), the product of multidrug resistance gene(s), was investigated in primary cultures of normal adult rat hepatocytes. Levels of P-gp mRNAs determined by Northern blotting and of P-gp measured by immunoblotting increased in parallel with time in culture. As in normal liver, P-gp was found to be localized on the membrane of bile canaliculus-like structures. This increased expression of P-gp was associated with decreased intracellular retention of doxorubicin, which could be restored by compounds such as verapamil and cyclosporin; doxorubicin (and also vincristine) was more cytotoxic to early than to late cultures. As in preneoplastic and neoplastic liver, overexpression of P-gp in cultured hepatocytes was associated with differential changes in drug-metabolizing enzymes, including increased glutathione S-transferase 7-7. Functional P-gp over-expression was observed in the absence of xenobiotic exposure or cell division; it could be linked to cellular stress occurring during cell isolation and plating. Increased expression of P-gp was blocked by actinomycin D, indicating its dependence on increased transcription, while cycloheximide led to a superinduction suggesting negative regulation by a protein factor.

Metabolism of doxorubicin, daunorubicin and epirubicin in human and rat hepatoma cells.

The metabolism of three anthracyclines, namely daunorubicin, doxorubicin and epirubicin, was studied in two human (HepG2 and HBG2) and one rat (FAO) hepatoma cell lines. Both species and substrate differences in anthracycline metabolism were retained by these hepatoma cell lines with regard to the aldoketoreductase pathway. However, glucuronidation of epirubicin normally found in vivo and in human hepatocyte cultures was not recovered in human hepatoma cells, further indicating that these cells express only some of the liver functions. Nevertheless, these cell lines could be a suitable model to investigate the still poorly understood aldoketoreductase activity.

Intralobular distribution and quantitation of cytochrome P-450 enzymes in human liver as a function of age.

We have used immunohistochemical, immunoblotting and messenger RNA blotting approaches to study the distribution and quantitation of three cytochrome P-450 enzymes, namely P-450 IA2, P-450 IIC and P-450 IIIA and, for comparison, epoxide hydrolase and NADPH-cytochrome P-450 reductase in human liver. Age-related changes in both the amounts and the intralobular distributions of these enzymes were demonstrated, and the enzymes differ in these regards: In fetal liver, P-450 IA2 and P-450 IIC were very low, when present at all, whereas P-450 IIIA, epoxide hydrolase and the reductase were already abundant and found in all the hepatocytes. During the postnatal period, P-450 IIC dramatically increased and was observed in all hepatocytes, the centrilobular ones being more intensely stained. P-450 IIIA was restricted to centrilobular and midzonal hepatocytes in normal adult liver. P-450 IA2 showed this same intralobular distribution; however, its presence was detected only several weeks or months after birth as judged both by immunohistochemical and immunoblotting techniques. Epoxide hydrolase and NADPH-cytochrome P-450 reductase were easily visualized in all hepatocytes regardless of the age of the donor; in child and adult livers, centrilobular hepatocytes were more intensely stained. Immunoreactive protein contents and corresponding messenger RNA levels correlated well with immunohistochemical observations. No major modification was seen in fibrotic liver, whereas both positive and negative cells were observed in cirrhotic liver nodules for all enzymes studied.

Evidence for the involvement of several cytochromes P-450 in the first steps of caffeine metabolism by human liver microsomes.

Caffeine biotransformation and four monooxygenase activities involving cytochrome P-450IA2, namely ethoxy- and methoxyresorufin O-dealkylases, phenacetin O-deethylase, and acetanilide 4-hydroxylation were studied in 25 human liver microsomes. All these activities were highly significantly intercorrelated (r greater than 0.72, p less than 0.001) and correlated with the level of immunoreactive P-450IA2 content (r greater than 0.65; p less than 0.001). P-450IA content was measured by immunoblotting with anti-rat P-450 beta-naphthoflavone-B, an antibody that detects only a single band corresponding to P-450IA2. The formation rate of two caffeine metabolites, namely paraxathine and theobromine, was correlated with the four monooxygenase activities measured and P-450IA2-specific content (r greater than 0.75). However, inhibition studies of caffeine metabolism by phenacetin, a specific substrate of P-450IA2, clearly indicated that only the N-3 demethylation of caffeine was supported by this enzyme. These in vitro data demonstrate that P-450IA2 is predominantly responsible for the major metabolic pathway of caffeine and that the formation of other demethylated metabolites is mediated, at least partly, by other P-450 enzymes.

In vitro assessment of stereoselective hepatic metabolism of disopyramide in humans: comparison with in vivo data.

Metabolism of disopyramide (DP) enantiomers has been investigated in primary cultures of adult human hepatocytes. Results were compared with in vivo data obtained from a previous pharmacokinetic study (Le Corre et al. Drug Metab. Dispos. 16:858-864 1988). Metabolism of DP enantiomers as a function of incubation time showed constant velocity over time. The intracellular/extracellular distribution of both DP and mono-N-desisopropyldisopyramide did not appear to be stereoselective. Metabolism of DP enantiomers as a function of substrate concentration followed a first order kinetics. The average fractions of (-)-(R)-DP and (+)-(S)-DP metabolized in vitro (4.7 +/- 2.7 and 7.1 +/- 4.2%, respectively, n = 4) were about 5-fold lower than the fractions metabolized in vivo (26.0 +/- 6.0 and 40.2 +/- 8.8%, respectively, n = 6). The stereoselective index [(+)-(S)/(-)-(R)] of the N-dealkylation pathway obtained in vitro (1.51 +/- 0.11, n = 4) was very close to the one obtained in vivo (1.55 +/- 0.10, n = 6). These results highlight the interest of hepatocyte cultures in the evaluation of drug metabolism and especially in the assessment of stereoselectivity.

Expression of cytochrome P-450 enzymes in cultured human hepatocytes.

Hepatocytes from adult and newborn humans were put into primary culture and exposed to phenobarbital, 3-methylcholanthrene, or rifampicin, three well-known inducers of cytochrome P-450 in animals. The expression of four cytochrome P-450 enzymes (or groups of enzymes, namely P-450 IIIA, P-450 IIC8/9/10, P-450 IIE1, and P-450 IA2) was investigated. These enzymes were found to remain expressed during the period of culture studied. Treatment with the inducers for three days resulted in different responses, depending upon the inducer and the enzyme. Phenobarbital and rifampicin increased P-450 IIC8/9/10 mRNA transcripts and the corresponding protein, while 3-methylcholanthrene was ineffective. Both P-450 IIIA mRNA and protein were strongly induced by rifampicin. All of the hepatocytes were found to synthesize P-450 IIIA in response to rifampicin, as shown by immunoperoxidase staining. P-450 IIIA expression was not affected by phenobarbital and was decreased by 3-methylcholanthrene. P-450s IA2 and IIE1 decreased to 25-50% of the initial level during these cultures. P-450 IA2 and ethoxyresorufin O-deethylase activity (which is a monooxygenase activity related to P-450 IA family) were increased only by 3-methylcholanthrene and did not respond to the other inducers. P-450 IIE1 was not induced by any of these compounds. P-450 IIC8/9/10 and P-450 IIIA mRNA levels were also measured in human hepatocytes from one newborn. P-450 IIC8/9/10 was barely expressed in freshly isolated cells but increased dramatically with time in culture. P-450 IIIA transcripts were abundant in both freshly isolated and cultured cells derived from a newborn. These results clearly demonstrate that human hepatocytes continue to express cytochrome P-450 enzymes and respond to inducers in culture. This model system provides a useful approach for investigating the effects of drugs on maturation and expression of drug-metabolizing enzymes in human liver.

Changes in expression of mRNA coding for glutathione S-transferase subunits 1-2 and 7 in cultured rat hepatocytes.

mRNA hybridizing to probes for glutathione S-transferase (GST) subunits 1 and 2 (probe pGSTr 155) and subunit 7 (probe pGSTr 7) has been measured by Northern blot analysis in adult rat hepatocytes both in conventional monoculture and in co-culture with epithelial cells. In addition, several media conditions were used, namely with and without fetal calf serum (FCS) and with and without nicotinamide or dimethyl sulfoxide (DMSO). In monoculture, mRNA coding for subunits 1 and 2 was extensively reduced in the presence of FCS. In the absence of FCS, after an initial decrease, an increase of subunits 1 and 2 mRNA was noticed on day 6. When nicotinamide or DMSO was added to the medium, the GST subunits 1 and 2 mRNA level increased during the culture period. In co-culture, an initial reduction in levels of mRNA encoding subunits 1 and 2 was less marked and the values measured increased with co-culture time. Nicotinamide tended to reduce these mRNA levels, whereas DMSO increased them. In contrast, in conventional culture, mRNA encoding subunit 7 was expressed de novo and this induction was prevented by DMSO but not by nicotinamide. Similar results were obtained with co-culture.

Methylcholanthrene but not phenobarbital enhances caffeine and theophylline metabolism in cultured adult human hepatocytes.

Biotransformation of caffeine and theophylline and the effect of two well-known inducers of P-450 isozymes, namely phenobarbital (PB) and methylcholanthrene (3-MC) were studied in cultured hepatocytes from six human adult donors. Hepatocytes co-cultured with rat liver epithelial cells maintained a higher metabolic capacity than pure cultures. PB treatment of cultured hepatocytes for 3 days slightly increased the rate of caffeine metabolism 1.4 +/- 0.5-fold (N = 6) vs controls, and theophylline metabolism 1.2 +/- 0.4-fold (N = 6), whereas 3-MC treatment increased metabolism markedly 5.8 +/- 2.3- and 3.3 +/- 1.1-fold (N = 6) vs controls for caffeine and theophylline, respectively. Paraxanthine and theophylline formations from caffeine were the most induced by 3-MC. Their increase was significantly correlated (rs = 0.89, P less than 0.007) but not with TB formation, suggesting that at least two isozymes of the P-450IA family are involved in the first demethylations of caffeine. In addition, the N-1 demethylation of theophylline (mean increase of 554% vs controls) was not correlated with the N-1 demethylation of caffeine (mean to increase 247% vs controls) for the same donor after 3-MC treatment, suggesting that these two demethylations are mediated by a different P-450.

Long-term culture of functional hepatocytes.

Recent studies have clearly demonstrated that the hepatocyte requires a complex and well defined environment to survive and maintain differentiated functions in vitro. Soluble factors as well as cell-matrix and cell-cell interactions have been found to affect markedly hepatocyte functions. Thus co-culturing hepatocytes with another rat liver cell type results in a prolonged expression of liver functions including phase I and phase II drug-metabolizing enzymes. Addition of corticosteroids to the co-culture medium is a prerequisite, and accumulation of insoluble matrix components is observed within a few days primarily between the two cell types. Hepatocyte cultures have been widely used for pharmacology and toxicology studies during recent years, but most studies deal with short-term investigations. Although specific functions are not completely stabilized the use of long-term hepatocyte cultures represents a promising tool to investigate enzyme induction and inhibition, and drug chronic toxicity.