In silico platform for xenobiotics ADME-T pharmacological properties modeling and prediction. Part I: Beyond the reduction of animal model use.

There is an urgent need for efficient in silico ADME-T prediction tools for the selection of potent therapeutic drugs as well as the elimination of toxic compounds. This is particularly important in view of the high costs and ethical issues inherent to the use of animal models for drugs filtering. To achieve this mission, not only does the accuracy of in silico tools need to be improved, but also new experts in the field with skills in theoretical chemistry, clinical and fundamental biology have to be trained. Similarly, clinical biologists committed to the obligation of means and legally responsible for the results they generate could establish a legal framework that defines legal responsibilities when performing in silico predictions.

Establishment of functional human cytochrome P450 monooxygenase systems in Escherichia coli.

Cytochromes P450 (CYP) have been expressed in a variety of systems such as mammalian cells, yeast, and bacteria. The bacterial system is technically the least demanding and provides large amounts of catalytically active P450s for metabolic and structural studies relating to preclinical drug development. This chapter provides a detailed technical description of the processes that allow the coexpression of various CYP isoforms together with CYP reductase in Escherichia coli and gives some examples of the results that can be achieved for the expression of human P450s.

Directional trans-epithelial transport of organic anions in porcine LLC-PK1 cells that co-express human OATP1B1 (OATP-C) and MRP2.

The transcellular transport of many compounds, which cannot readily cross the lipid bilayer, is mediated by drug uptake and efflux transporters. Human OATP1B1 and MRP2 have been implicated in the hepato-biliary transport of many endogenous and exogenous compounds. Here, we have established epithelial porcine kidney LLC-PK1 derived cell lines, that express both transporters in a polarized fashion, as a model to predict hepato-biliary transport. Immunological identification of OATP1B1 in the recombinant cell lines was greatly facilitated by its C-terminal tagging with a peptide sequence derived from hemagglutinin (HA) avoiding the generation of OATP1B1 specific antibodies. Importantly, the tag did not interfere with the functionality of the transporter. Compared to LLC-PK1 cells and cells which expressed only OATP1B1, the cell line that co-expressed MRP2 and OATP1B1 displayed high directional basolateral-to-apical transport of 17 beta-estradiol-17 beta-glucuronide and estrone-3-sulfate. Dehydroepiandrosterone sulfate already displayed a significant basolateral-to-apical transport in the parental cell line, which was further stimulated upon expression of both transporters. Transcellular flux of all steroid conjugates in the opposite direction (apical-to-basolateral) was much lower. By employing this cellular model we were able to demonstrate for the first time that OATP1B1 together with MRP2 mediates the trans-cellular transport of rifampicin. It is anticipated that the models established herein will greatly facilitate the identification of transporters involved in the disposition of novel drug candidates.

Proteasomal degradation of human CYP1B1: effect of the Asn453Ser polymorphism on the post-translational regulation of CYP1B1 expression.

Allelic variations in CYP1B1 are reported to modulate the incidence of several types of cancer. To provide a mechanistic basis for this association, we investigated the impact of nonsilent allelic changes on the intracellular levels and post-translational regulation of CYP1B1 protein. When transiently expressed in COS-1 cells, either in the presence or absence of recombinant cytochrome P450 reductase, the cellular level of the CYP1B1.4 allelic variant (containing a Ser at the amino acid position 453; Ser453) was 2-fold lower compared with the other four allelic CYP1B1 proteins (containing Asn453), as analyzed by both immunoblotting and ethoxyresorufin O-deethylase activity. This difference was caused by post-translational regulation; as in the presence of cycloheximide, the rate of degradation of immunodetectable and enzymatically active CYP1B1.4 was distinctly faster than that of CYP1B1.1. Pulse-chase analysis revealed that the half-life of CYP1B1.4 was a mere 1.6 h compared with 4.8 h for CYP1B1.1. The presence of the proteasome inhibitor MG132 [N-benzoyloxycarbonyl (Z)-Leu-Leuleucinal] increased the stability not only of immunodetectable CYP1B1, but also--unexpectedly given the size of the proteasome access channel--increased the stability of enzymatically active CYP1B1. The data presented herein also demonstrate that CYP1B1 is targeted for its polymorphism-dependent degradation by polyubiquitination but not phosphorylation. Our results importantly provide a mechanism to explain the recently reported lower incidence of endometrial cancer in individuals carrying the CYP1B1*4 compared with the CYP1B1*1 haplo-type. In addition, the mechanistic paradigms revealed herein may explain the strong overexpression of CYP1B1 in tumors compared with nondiseased tissues.

AMP-activated protein kinase mediates phenobarbital induction of CYP2B gene expression in hepatocytes and a newly derived human hepatoma cell line.

Phenobarbital (PB) administration is known to trigger pleiotropic responses, including liver hypertrophy, tumor promotion, and induction of genes encoding drug-metabolizing enzymes. The induction of human CYP2B6 and the rat (CYP2B1) and mouse (Cyp2b10) homologues by PB is mediated by the nuclear receptor constitutive androstane receptor (CAR). The study of CYP2B gene regulation and CAR activity by PB has been difficult due to the lack of a cellular model. In this study, we describe a novel differentiated human hepatoma cell line (WGA), derived from HepG2, which expresses CYP2B6 and CAR. WGA cells represent a powerful system to study the regulation of CYP2B6 gene expression by PB. There is evidence that CAR activity is regulated by phosphorylation and that regulation of some CYP genes depends on the nutritional status of cells. The AMP-activated protein kinase (AMPK) functions as an energy sensor and is activated when cells experience energy-depleting stresses. In this report, we show that addition of 5-amino-imidazole carboxamide riboside, an AMPK activator, to WGA and human hepatocytes induces CYP2B6 gene expression. Expression of a constitutively active form of AMPK mimics the PB induction of CYP2B6 and CYP2B1 gene expression. Conversely, the expression of a dominant negative form of AMPK inhibits the induction of these genes by PB. Finally, we demonstrate, for the first time, that AMPK activity increases in cells cultured with PB. Our data strongly support a role for AMPK in the PB induction of CYP2B gene expression and provide new insights into the regulation of gene expression by barbiturate drugs.

Glucuronidation as a mechanism of intrinsic drug resistance in colon cancer cells: contribution of drug transport proteins.

We have recently shown that drug conjugation catalysed by UDP-glucuronosyltransferases (UGTs) functions as an intrinsic mechanism of resistance to the topoisomerase I inhibitors 7-ethyl-10-hydroxycamptothecin and NU/ICRF 505 in human colon cancer cells and now report on the role of drug transport in this mechanism. The ability of transport proteins to recognise NU/ICRF 505 as a substrate was evaluated in model systems either transfected with breast cancer-resistance protein 1 (Bcrp1), multidrug-resistance protein 2 (Mrp2) or Mrp3, or overexpressing MRP1 or P-170 glycoprotein. Results from chemosensitivity assays suggested that NU/ICRF 505 was not a substrate for any of the above proteins. In drug accumulation studies in human colon cancer cell lines NU/ICRF 505 was taken up avidly and retained in cells lacking UGTs (HCT116), whereas, following equally rapid uptake, it was cleared rapidly from cells displaying UGT activity (HT29) as glucuronide metabolites. HT29 cells were shown to express MRP1 and 3, but not P-170 glycoprotein, MRP2 or breast cancer-resistance protein. The major glucuronide of NU/ICRF 505 inhibited ATP-dependent transport of estradiol 17-beta-glucuronide in Sf9 insect cell membrane vesicles containing MRP1 or MRP3, while co-incubation of HT29 cells with the MRP antagonist, MK571, significantly restored intracellular concentrations of NU/ICRF 505. These data lead us to conclude that the presence of a glucuronide transporter is essential for glucuronidation to represent a major de novo resistance mechanism and that UGTs will contribute more as a primary resistance mechanism when the parent drug (e.g. NU/ICRF 505) is not itself recognised by transport proteins.

An amperometric biosensor with human CYP3A4 as a novel drug screening tool.

We developed a biosensor based on the redox properties of human CYP3A4 to directly monitor electron transfer to the heme protein. Enzyme films were assembled on gold electrodes by alternate adsorption of a CYP3A4 layer on top of a polycation layer. Direct, reversible electron transfer between the electrode and CYP3A4 was observed with voltammetry under anaerobic conditions. In the presence of oxygen, the oxidation peak of the hemoprotein disappeared, and the reduction peak increased 2- to 3-fold. Addition of CYP3A4 substrates (verapamil, midazolam, quinidine, and progesterone) to the oxygenated solution caused a concentration-dependent increase in the reduction current in cyclic voltammetric and amperometric experiments. Product analyses after electrolysis with the enzyme film showed catalytic activity of the biosensor depending on substrate concentration, its inhibition by ketoconazole, and a minor contribution of H(2)O(2) to the catalytic cycle. These results suggest that electron exchange between the electrode and the immobilized CYP3A4 occurred, and that metabolic activity of the enzyme was maintained. Thus, important requirements for the application of human CYP biosensors in order to identify drugs or drug candidates as substrates or inhibitors to the attached enzyme are fulfilled.

In vitro metabolism of genistein and tangeretin by human and murine cytochrome P450s.

Recombinant cytochrome P450 (CYP) 1A2, 3A4, 2C9 or 2D6 enzymes obtained from Escherichia coli and human liver microsomes samples were used to investigate the ability of human CYP enzymes to metabolize the two dietary flavonoids, genistein and tangeretin. Analysis of the metabolic profile from incubations with genistein and human liver microsomes revealed the production of five different metabolites, of which three were obtained in sufficient amounts to allow a more detailed elucidation of the structure. One of these metabolites was identified as orobol, the 3'-hydroxylated metabolite of genistein. The remaining two metabolites were also hydroxylated metabolites as evidenced by LC/MS. Orobol was the only metabolite formed after incubation with CYP1A2. The two major product peaks after incubation of tangeretin with human microsomes were identical with 4'-hydroxy-5,6,7,8-tetramethoxyflavone and 5,6-dihydroxy-4',7,8-trimethoxyflavone, previously identified in rat urine in our laboratory. By comparison with UV spectra and LC/MS fragmentation patterns of previously obtained standards, the remaining metabolites eluting after 14, 17 and 20 min. were found to be demethylated at the 4',7-, 4',6-positions or hydroxylated at the 3'- and demethylated at the 4'-positions, respectively. Metabolism of tangeretin by recombinant CYP1A2, 3A4, 2D6 and 2C9 resulted in metabolic profiles that qualitatively were identical to those observed in the human microsomes. Inclusion of the CYP1A2 inhibitor fluvoxamine in the incubation mixture with human liver microsomes resulted in potent inhibition of tangeretin and genistein metabolism. Other isozymes-selective CYP inhibitors had only minor effects on tangeretin or genistein metabolism. Overall the presented observations suggest major involvement of CYP1A2 in the hepatic metabolism of these two flavonoids.

Endogenous drug transporters in in vitro and in vivo models for the prediction of drug disposition in man.

The epithelial canine and porcine kidney cell lines MDCK, MDCKII and LLC-PK1, respectively are employed to establish recombinant models of drug transport. Endogenous drug carriers in these cells may contribute to the activities of recombinant drug transporters, thus making it difficult to assess their properties. We analysed the expression of endogenous transporters in these cell lines by RT-PCR and by determining drug transporter activities. Concerning drug efflux, multidrug resistance protein 1 (MDR1) and MRP1 mRNAs were found in all lines. MRP2 mRNA was expressed in all cell lines except MDCK. Transepithelial transport of vinblastine and its modulation by a MDR1-specific inhibitor or by the MDR1- and MRP-inhibitor verapamil, indicated that MDCKII cells have, in comparisons to the other cell lines, relatively high levels of functional MDR1 while vinblastine transport in MDCK cells is likely to be mediated more by MRP1. Notably, LLC-PK1 cells displayed little activity attributable to either MDR1 and MRP1, thus making them suitable for the expression of these efflux pumps. Of the drug uptake carriers, OATP-A mRNA was only expressed in MDCK cells. OATP-C mRNA was barely detectable in MDCK cells and absent in MDCKII and LLC-PK1 cells. In agreement with transcriptional profiling, the OATP-mediated uptake of either estradiol-glucuronide or estrone-sulfate was either absent or barely detectable in all cell lines thus implying that they are suitable to establish recombinant models for human OATP's. Transcriptional profiling was also performed on porcine and canine tissues and revealed that MRP1 was expressed in canine but not in human or porcine liver, whereas surprisingly OATP-C was expressed in canine kidney but only in human and porcine liver. The findings presented are relevant to the use of porcine and canine models for drug disposition.

The N-demethylation of the doxepin isomers is mainly catalyzed by the polymorphic CYP2C19.

PURPOSE: This study was conducted to identify the cytochrome P450s (CYPs) responsible for the metabolism of the cis- and trans-isomers of the tricyclic antidepressant doxepin to its pharmacologically active N-desmethylmetabolite by in vitro techniques. METHODS: The doxepin N-demethylation was studied by means of pooled human liver microsomes and chemical inhibitors, recombinant human (rh)-CYPs, and geno- and phenotyped human liver microsomes. RESULTS: The N-demethylation of both isomers was inhibited most prominently by tranylcypromine (CYP2C19) to more than 50%. Furafylline (CYP1A2) and sulfaphenazole (CYP2C9) inhibited the N-demethylation to a lesser extent while quinidine (CYP2D6) or troleandomycine (CYP3A4) had no effect. Rh-CYP2C19, -CYP1A2, and -CYP2C9 were able to N-demethylate cis- and trans-doxepin. Only traces of trans-desmethyldoxepin were detectable when CYP3A4 was used. The maximum velocity in the cis- and transdoxepin N-demethylation was significantly (P < 0.05) lower in microsomes with low CYP2C19 activity (345 +/- 44 and 508 +/- 75 pmol/min/ mg protein, respectively) compared to those with high CYP2C19 activity (779 +/- 132 and 1,189 +/- 134 pmollmin/mg). CONCLUSION: The present study demonstrates a significant contribution of the polymorphic CYP2C19 to the N-demethylation of doxepin. CYP2C9 and CYP1A2 play a minor role and CYP3A4 does not contribute substantially.

CYP3A4 active site volume modification by mutagenesis of leucine 211.

The leucine 211 --> phenylalanine (L211F) and leucine 211 --> tyrosine (L211Y) mutant forms of cytochrome P450 3A4 have been generated by site-directed mutagenesis and expressed functionally in Escherichia coli. Substrate binding affinities (S50 values) for testosterone and 7-benzyloxy-4-trifluoromethylcoumarin (BFC) were similar for the mutants and wild-type CYP3A4 (49 and 21 microM for L211F, 35 and 20 microM for L211Y, and 33 and 20 microM for the wild type, respectively). For erythromycin, however, the K(m) values determined for the L211F and L211Y mutants were 2.4- and 10.5-fold higher than for the wild type. Furthermore, IC50 values for the inhibition of testosterone 6 beta-hydroxylation by erythromycin and troleandomycin for L211F were 2.4- and 3.7-fold higher, and those for L211Y were 3.4- and 9.2-fold higher than those measured for the wild type. Conversely, small inhibitors, such as diazepam, exhibited no significant difference in IC50 values between the wild type and the L211F and L211Y mutants. It is proposed that large substrates bound in the catalytic center of CYP3A4 with molecular volumes greater than approximately 600 A(3) were less well accommodated in the altered active sites, resulting in lower association energies and increased IC50 values.