Cell-based models to study hepatic drug metabolism and enzyme induction in humans.

Cell-based in vitro models are invaluable tools in elucidating the pharmacokinetic profile of a drug candidate during its drug discovery and development process. As biotransformation is one of the key determinants of a drug's disposition in the body, many in vitro models to study drug metabolism have been established, and others are still being developed and validated. This review is aimed at providing the reader with a concise overview of the characteristics and optimal application of established and emerging in vitro cell-based models to study human drug metabolism and induction of drug metabolising enzymes in the liver. The strengths and weaknesses of liver-derived models, such as primary hepatocytes, either freshly isolated or cryopreserved, and from adult or fetal donors, precision-cut liver slices, and cell lines, including immortalised cells, reporter cell lines, hepatocarcinoma-derived cell lines and recombinant cell lines, are discussed. Relevant cell culture configuration aspects as well as other models such as stem cell-derived hepatocyte-like cells and humanised animal models are also reviewed. The status of model development, their acceptance by health authorities and recommendations for the most appropriate use of the models are presented.

Determination of cytochrome P450 1A2 and cytochrome P450 3A4 induction in cryopreserved human hepatocytes.

Freshly prepared human hepatocytes are considered as the 'gold standard' for in vitro testing of drug candidates. However, several disadvantages are associated with the use of this model system. The availability of hepatocytes is often low and consequently the planning of the experiments rendered difficult. In addition, the quality of the available cells is in some cases poor. As an alternative, cryopreserved human hepatocytes were validated as a model to study cytochrome P450 1A2 (CYP1A2) and cytochrome P450 3A4 (CYP3A4) induction. In a single blinded experiment, hepatocytes from three separate lots were incubated with three concentrations of different compounds, and compared to non-treated cells and cells incubated with omeprazole or rifampicin. CYP1A2 and CYP3A4 induction was determined by measuring 7-ethoxyresorufin-O-deethylation activity and 6beta-hydroxytestosterone formation, respectively. CYP1A2 and CYP3A4 mRNA and protein expression were analyzed by TaqMan QRT-PCR and immunodetection. Cells responded well to the prototypical inducers with a mean 38.8- and 6.2-fold induction of CYP1A2 and CYP3A4 activity, respectively. Similar as with fresh human hepatocytes, high batch-to-batch variation of CYP1A2 and CYP3A4 induction was observed. Except for 1 and 10 microM rosiglitazone, the glitazones did not significantly affect CYP1A2. A similar result was observed for CYP3A4 activity although CYP3A4 mRNA and protein expression were dose-dependently upregulated. In conclusion, cryopreserved human hepatocytes may be a good alternative to fresh hepatocytes to study CYP1A and 3A induction.

Metabolism and excretion of RWJ-333369 [1,2-ethanediol, 1-(2-chlorophenyl)-, 2-carbamate, (S)-] in mice, rats, rabbits, and dogs.

The in vivo metabolism and excretion of RWJ-333369 [1,2-ethanediol, 1-(2-chlorophenyl)-, 2-carbamate, (S)-], a novel neuromodulator, were investigated in mice, rats, rabbits, and dogs after oral administration of (14)C-RWJ-333369. Plasma, urine, and feces samples were collected, assayed for radioactivity, and profiled for metabolites. In almost all species, the administered radioactive dose was predominantly excreted in urine (>85%) with less than 10% in feces. Excretion of radioactivity was rapid and nearly complete at 96 h after dosing in all species. Unchanged drug excreted in urine was minimal (<2.3% of the administered dose) in all species. The primary metabolic pathways were O-glucuronidation (rabbit > mouse > dog > rat) of RWJ-333369 and hydrolysis of the carbamate ester followed by oxidation to 2-chloromandelic acid. The latter metabolite was subsequently metabolized in parallel to 2-chlorophenylglycine and 2-chlorobenzoic acid (combined hydrolytic and oxidative pathways: rat > dog > mouse > rabbit). Other metabolic pathways present in all species included chiral inversion in combination with O-glucuronidation and sulfate conjugation (directly and/or following hydroxylation of RWJ-333369). Species-specific pathways, including N-acetylation of 2-chlorophenylglycine (mice, rats, and dogs) and arene oxidation followed by glutathione conjugation of RWJ-333369 (mice and rats), were more predominant in rodents than in other species. Consistent with human metabolism, multiple metabolic pathways and renal excretion were mainly involved in the elimination of RWJ-333369 and its metabolites in animal species. Unchanged drug was the major plasma circulating drug-related substance in the preclinical species and humans.

Expression and induction potential of cytochromes P450 in human cryopreserved hepatocytes.

Fresh human hepatocytes are still considered as the "gold standard" to screen in vitro for cytochrome P450 (P450) induction. However, sparse availability of good quality human liver tissue for research purposes and the demand for standardized cell populations, together with the need for proper storage of the cells not immediately required, have resulted in the development of cryopreservation techniques that provide adequate viability and plateability of hepatocytes after thawing. This study aimed at validating cryopreserved human hepatocytes as a model to investigate P450 induction. Cryopreserved cells from four different donors were plated and cultured for 48 h, followed by incubation in the presence of typical P450 inducers. During the experiments, quality of the cultured cells was monitored both physiologically and morphologically. Concomitantly, the activity of CYP1A2, 2B6, 2C9, 2E1, and 3A4 was measured together with their mRNA and protein expression. Determination of CYP1A2, 2B6, 2C9, 2E1, and 3A4 activity in control versus prototypical inducer-treated hepatocytes revealed a maximal significant mean 11.6-, 2.8-, 1.9-, 1.5-, and 9.0-fold induction over their basal expression, respectively. Protein expression analysis of these P450s confirmed these results. Moreover, a mean 44.9-, 3.5-, 3.2-, and 13.8-fold induction of CYP1A2, 2B6, 2C9, and 3A4 mRNA was observed. Our data demonstrate that cryopreserved human hepatocytes are a valuable tool to study the induction of CYP1A2, 2B6, 2C9, 2E1, and 3A4.

Pharmacokinetics of galantamine, a cholinesterase inhibitor, in several animal species.

In this publication, single and repeated dose experiments in rats, mice, rabbits and dogs are reported to assess the pharmacokinetics of galantamine (CAS-1953-04-4), a tertiary alkaloid with reversible cholinesterase inhibiting and nicotinic receptor modulatory properties developed for the treatment of Alzheimer's disease in humans. Rats received single i.v. and single and repeated oral administrations of various doses, up to 160 mg/kg/day. In mice, only repeated oral administration of galantamine was investigated, up to 40 mg/kg/day. Galantamine single and repeated oral doses up to 32 mg/kg/day were administered to female pregnant rabbits. Beagle dogs received single i.v. and single and repeated oral administrations of doses up to 8 mg/kg/day. Generally, oral absorption was rapid, with maximal plasma levels reached within 2 h in all species. Absolute oral bioavailability of a gavage dose was high in rat (77%) and dog (78%). In mice and rats, the bioavailability of galantamine administered via the food was lower than of galantamine administered by gavage. Elimination half-life of galantamine was relatively large in rat and dog and smaller in mouse and rabbit. In general, galantamine displayed dose-proportional to somewhat more than dose-proportional kinetics. In rats, plasma levels were lower in females than in males, whereas in mice, females showed higher levels than males. No gender differences were observed in dogs. No relevant differences in exposure to galantamine were found in rats and dogs upon oral administration of galantamine obtained as a natural extract or from chemical synthesis. The exposure to the active metabolite norgalantamine in plasma of the different animal species was low, except in the dog where the steady-state norgalantamine exposure was approximately 75% of galantamine exposure. Galantamine plasma levels after single and repeated administration of 10 mg/kg/day in all species investigated except female rat and rabbit were much higher than mean therapeutic plasma levels of galantamine obtained in humans. The pharmacokinetic profile of galantamine after repeated oral administration in rats was most similar to the profile obtained after repeated administration of 12 mg b.i.d. in man.

Pharmacokinetics and tissue distribution of galantamine and galantamine-related radioactivity after single intravenous and oral administration in the rat.

The plasma kinetics and tissue distribution of galantamine hydrobromide [4aS-(4a alpha,6beta,8aR*)]-4a,5,9,10,11,12-hexahydro-3-methoxy-11-methyl-6H-benzofuro-[3a,3,2-ef] [2benzazepin-6-ol hydrobromide, CAS-1953-04-4], a reversible acetylcholinesterase inhibitor, were studied in male and female non-pregnant and pregnant SPF Wistar rats and in male Fisher x Copenhagen pigmented rats. Most studies were performed using 3H-labelled galantamine hydrobromide, measuring unchanged drug (UD) and non-volatile radioactivity (NVR) in plasma and tissues by high-performance liquid chromatography (HPLC), liquid scintillation counting and quantitative whole-body autoradiography (QWBA). Plasma levels after single intravenous administration of UD (1.25-2.5 mg/kg) declined bi- or triphasically, with an elimination half-life of 3.5 h in male, and 5.1 h in female rats. The plasma clearance (Cl) averaged 1.9 l/kg/h (male rats) and 0.9 l/kg/h (female rats), and the volume of distribution (VdSS) was about 5 l/kg for both male and female rats. Following oral administration (2.5-10 mg/kg), galantamine was rapidly absorbed in both sexes, with an absolute oral bioavailability of 77%. Distribution studies after oral administration of 3H-galantamine showed an almost immediate equilibrium between plasma and tissues, with highest tissue levels of NVR and UD in liver, kidney, salivary glands, adrenal glands and, for the female rat, spleen, and lowest in white fat. To most tissues and especially to brain, the distribution of UD was more pronounced than that of its metabolites. Tissue concentrations of UD and NVR declined at a similar rate as plasma, showing no undue retention. QWBA in the pigmented rat showed the same distribution and elimination pattern of NVR. Only in hair follicles and choroid some retention of NVR was seen, but the calculated half-life was less than one day. In the female pregnant SPF Wistar rat, maternal tissue distribution of NVR was similar to that of the non-pregnant rat. NVR tissue levels in the foetus were similar to those found in maternal blood during the whole experiment, indicating a rapid equilibrium without accumulation.