Chemicals. All chemicals and reagents used were of the highest available commercial grade. Diltiazem, testosterone, dextromethorphan, (±)-propranolol, (±)-metoprolol, diazepam, tolbutamide, ibuprofen, and -nicotinamide adenine dinucleotide phosphate, reduced form (-NADPH) were purchased from Sigma Chemical Co. (Poole, UK). (±)-Verapamil was purchased from Aldrich Chemical Co. Ltd. (Gillingham, UK). Omeprazole was synthesized at AstraZeneca R&D Charnwood (Loughborough, UK).
Source of Cytochrome P450. The LINK consortium, a collaboration between UK-based academia and industry, provided stocks of transformed cells with human CYP1A2, CYP2C9, CYP2D6, and CYP3A4 individually coexpressed with human NADPH-P450 reductase in E. coli as described previously (McGinnity et al., 1999). All experiments with CYP1A2, CYP2C9, CYP2D6, and CYP3A4 utilized the E. coli membrane source. All transformed cells were stored as glycerol stocks at 80°C. Expression of the recombinant proteins and preparation of the respective E. coli membranes were carried out as described previously (McGinnity et al., 1999).
Microsomes prepared from insect cells infected with a baculovirus containing the cDNA for human CYP2C19 and rabbit NADPH-P450 reductase were purchased from PanVera Corp. (Madison, WI). All experiments with CYP2C19 utilized this enzyme source. Pooled HLM were purchased from IIAM (Leicester, UK) and In Vitro Technologies (Baltimore, MD). Table 1 displays the CYP isoform characterization of the individual HLM pools as determined by the commercial supplier.
Cytochrome P450 contents were estimated spectrally by the method of Omura and Sato (1964). Protein concentrations were measured using the Randox Laboratories Ltd. (Crumlin, UK) protein kit based on pyrogallol red complexing with protein in an acid environment containing molybdate ions (Watanabe et al., 1986), using bovine serum albumin as a standard.
Probe Substrates. Ten commercially available drugs were selected as probe substrates to establish the suitability of this approach. The compounds were selected from the literature as marketed drugs for which the relative CYP-dependent metabolic formation was known and their metabolism by the five CYPs was adequately represented: tolbutamide (Back et al., 1988; Bourrie et al., 1996; Jung et al., 1997; Wester et al., 2000); diazepam (Ono et al., 1996); metoprolol (Otton et al., 1988; Mautz et al., 1995); ibuprofen (Hamman et al., 1997); propranolol (Otton et al., 1990; Yoshimoto et al., 1995); dextromethorphan (Dayer et al., 1989; Broly et al., 1990; Jacqz-Aigrain et al., 1993; Kerry et al., 1994; Von Moltke et al., 1998); omeprazole (Andersson et al., 1993; Kobayashi et al., 1994; Yamazaki et al., 1997); diltiazem (Pichard et al., 1990; Sutton et al., 1997); testosterone (Waxman et al., 1988; Wang et al., 1997); and verapamil (Kroemer et al., 1993; Tracy et al., 1999).
Automated CYP CLint Determination. CYP CLint determination assays were fully automated and performed by a robotic sample processor (RSP) (Genesis RSP 150; Tecan, Reading, UK). Assays performed by the RSP were done using a program created by the user and not by a default program supplied with the hardware.3 The primary stock of all probe substrates was prepared manually in dimethyl sulfoxide or acetonitrile at 100-fold final incubation concentration. The final concentration of organic solvent in the incubation was 1% v/v. At this concentration dimethyl sulfoxide has been shown to reduce the activities of CYP2C9/19 (Chauret et al., 1998; Hickman et al., 1998), although this effect appears to be substrate-dependent. All substrates were incubated at 3 µM except tolbutamide (CLint calculated by determining Vmax and Km), ibuprofen (10 µM), and testosterone (10 µM). The RSP was programmed to add chilled quenching solvent (100 µl of acetonitrile) to 96-well refrigerated blocks, and compound stocks were then prediluted in 100 mM potassium phosphate buffer, pH 7.4. E. coli membranes and microsomes prepared from baculovirus coexpressing individual CYPs and NADPH-reductase were added to incubation tubes (100 pmol of CYP · ml1 final concentration) located in a 96-well heated block (37°C). A subaliquot was removed to produce a 0-min time point, and the assay was initiated via addition of NADPH (1 mM final concentration). Aliquots (50 µl) were removed at 5, 10, 15, and 20 min and quenched in acetonitrile. Samples were subsequently removed from the RSP, frozen for 1 h at 20°C, and then centrifuged at 3500 rpm for 20 min. The supernatants were removed and transferred into HPLC vials using the RSP.
Automated Human Liver Microsome CLint Determination. HLM were diluted in 100 mM potassium phosphate buffer, pH 7.4 (1 mg · ml1 final). Probe substrates were incubated at identical concentrations as the CYP CLint assay, and incubations were carried out on the RSP as described above. Reactions were again initiated by addition of NADPH (1 mM final concentration), and several aliquots were taken over 45 min.
HPLC Methods. Aliquots (20 µl) were analyzed by HPLC-UV or HPLC-fluorescence for either parent loss or metabolite appearance using a model 1100 Chemstation (Hewlett-Packard, Palo Alto, CA) and a Hewlett-Packard 1046A fluorescence detector. A symmetry shield RP8 3.9- × 50-mm cartridge (Waters, Watford, UK) and a mobile phase of 0.025% (w/v) ammonium acetate (solvent 1A) and acetonitrile (solvent 1B) was used for the chromatography of most analytes. Testosterone required a mobile phase of 0.025% ammonium acetate:methanol (95:5, v/v) (solvent 2A) and acetonitrile:methanol (95:5, v/v) (solvent 2B). The flow rate for all methods was 1.5 ml · min1. Diazepam, metoprolol, propranolol, omeprazole, diltiazem, and verapamil eluted using a 5-min linear gradient from 80% solvent 1A to 20% solvent 1A, tolbutamide 99% to 65% over 5 min, dextromethorphan 80% to 20% over 3.5 min, and ibuprofen 85% to 20% over 5 min. Testosterone was eluted using a linear gradient from 85% solvent 2A to 75% solvent 2A over 12 min, 75% to 20% over 3 min followed by isocratic conditions (20:80) for 2 min. UV detection was performed for omeprazole (302 nm), diltiazem (237 nm), ibuprofen (222 nm), and testosterone (254 nm) and for metabolites of diazepam (229 nm) and tolbutamide (230 nm). Fluorometric detection was performed for metoprolol (Excitation 222 nm and Emission 320 nm), dextromethorphan (270, 312 nm), propranolol (205, 340 nm) and verapamil (280 nm, 310 nm).
Metabolite Identification. HLM or recombinant CYPs were diluted in 100 mM potassium phosphate buffer, pH 7.4 (1 mg · ml1 or 100 pmol · ml1, respectively). Probe substrates were incubated at 30 µM, reactions were initiated by addition of NADPH (1 mM), and aliquots were quenched in 1:1 (v/v) methanol at 0 and 45 min. Aliquots (20 µl) were analyzed by liquid chromatography-mass spectrometry using the Hewlett-Packard 1100 Chemstation with a symmetry shield RP8 3.9- × 50-mm cartridge and a mobile phase of 0.025% (w/v) ammonium acetate (solvent 3A) and methanol (solvent 3B). Analytes were eluted using a gradient of 95% solvent 3A to 10% solvent 3A over 7 min. Metabolites were detected using a TSQ 7000 mass spectrophotometer (Finnigan MAT, San Diego, CA) with an atmospheric pressure chemical ionization ion source and a triple quadrupole mass analyzer in full scan mode. The molecular ion (either M + H+ or M H+ depending on the orifice polarity) was detected for each metabolite.
Data Analysis. Throughout this study, several approaches were adopted for quantifying intrinsic clearance:
Metabolite appearancelow turnover compounds.
if S Km
Parent loss. Because dose/C0 gives a term for the volume of the incubation (expressed in ml · pmol of CYP1) and the elimination rate constant k = 0.693/T1/2, an equation expressing CLint in terms of T1/2 of parent loss can be derived:
The contribution of individual CYP to HLM CLint
was estimated as follows:
shows the average percentage content of the five major isoforms in human hepatic microsomes.
All individual data represent means from at least duplicate determinations.