Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Multiple forms of cytochrome P-450 (P-450)¹ exist in mammals, and these P-450 enzymes play important roles in the oxidation of structurally diverse xenobiotic chemicals and endobiotics (Guengerich and Shimada, 1991; Guengerich, 1995; Nelson et al., 1996). P-450s are not self-sufficient enzymes, and the microsomal enzymes require a NADPH-P-450 reductase (NPR) as an electron carrier to function as monooxygenases. In human livers, levels of each of the P-450 forms are different and roles in various substrate oxidations vary. CYP3A4 is the major P-450 enzyme involved in the oxidation of a large number of compounds (Wrighton and Stevens, 1992; Gonzalez and Gelboin, 1994; Shimada et al., 1994).

Recently, recombinant P-450 enzymes from different sourcese.g., microsomes of human lymphoblastoid cells (Gonzalez et al., 1991; Crespi, 1995), yeast (Renaud et al., 1993; Imaoka et al., 1996), and insect cells infected with baculovirus systems (Buters et al., 1994; Lee et al., 1995) and reconstitution systems containing purified P-450 enzymes from Escherichia coli membranes (Gillam et al., 1993; Shaw et al., 1997) have been used widely for drug metabolism research. We have shown that a 1:2:1 molar ratio of CYP3A4/NPR/cytochrome b₅ (b₅) is optimal in the reconstitution of drug oxidation (Yamazaki et al., 1996b; Shimada and Yamazaki, 1998) using purified CYP3A4. However, the marker activities or kinetic parameters of CYP3A4 reported from different research laboratories and commercial enzymes manufacturers are not always similar. In this study, we have determined optimal catalytic activities for testosterone 6-hydroxylation, nifedipine oxidation, and midazolam 4- and 1'-hydroxylation by CYP3A4/NPR membranes (obtained using a bacterial bicistronic CYP3A4 expression system) fortified with b₅ or NPR and compared the activities of several types of recombinant CYP3A4 preparations before and after addition of b₅ and/or NPR. Enhancement of CYP3A4-dependent activities in membranes by addition of b₅ from different sources, apolipoprotein b₅ (apo b₅), and kinetic parameters for testosterone 6-hydroxylation are also reported.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. Midazolam and its metabolites were kindly donated by Yamanouchi Pharmaceuticals Co., Ltd. (Tokyo, Japan). Testosterone, nifedipine, and their metabolites and reagents used in this study were obtained from sources described previously or were of the highest qualities commercially available (Yamazaki et al., 1996b; Shimada and Yamazaki, 1998).

Enzyme Preparations. Membranes were prepared from E. coli into which CYP3A4 and NPR cDNAs had been introduced as described previously (Parikh et al., 1997). Briefly, single transformed colonies of E. coli strains DH5 and JM109 were used to inoculate starter Luria-Bertani medium (LB)/ampicillin (25 µg/ml) cultures. The starter cultures were incubated for 7 to 15 h at 37°C, with shaking at 170 rpm, and then diluted 1:100 into Terrific Broth (TB)/ampicillin (100 µg/ml) medium containing additives (0.5 mM -aminolevulinic acid, 1.0 mM isopropyl -D-thiogalactoside, trace salts, and 1.0 mM thiamine; Guengerich et al., 1996). The expression cultures (100 ml) were grown at 30°C with shaking at 120 to 180 rpm for 24 to 32 h in 500-ml triple-baffled flasks. Membrane fractions were prepared from the bacterial pellets by a series of fractionation and high-speed centrifugation steps (Guengerich et al., 1996) and suspended in one volume of 10 mM Tris-HCl buffer (pH 7.4) containing 0.10 mM EDTA and 20% glycerol (v/v). Recombinant monocistronic CYP3A4, CYP1A2, and NPR were purified from membranes as described elsewhere (Gillam et al., 1993; Dong et al., 1996; Parikh et al., 1997). Rabbit NPR (Guengerich et al., 1981) and rabbit (Strittmatter et al., 1978), rat, and human b₅ (Shimada et al., 1986) were purified from liver microsomes by the methods described. Recombinant human NPR and b₅ were purified using similar methods. Apo b₅ was prepared from rabbit b₅ as described previously (Yamazaki et al., 1996a). Horse heart cytochrome c was purchased from Sigma. Human liver microsomes were prepared in 10 mM Tris-HCl buffer (pH 7.4) containing 0.10 mM EDTA and 20% glycerol (v/v) as described previously (Guengerich, 1994; Inoue et al., 1997).

Enzyme Assays. Testosterone hydroxylation, nifedipine oxidation, and midazolam hydroxylation activities were determined as described (Kronbach et al., 1989; Brian et al., 1990) with slight modifications (Shimada and Yamazaki, 1998). The standard incubation mixture (final volume of 0.25 ml) contained 0.010 µM recombinant CYP3A4, 0.020 µM NPR, 0.010 µM b₅, and 100 mM potassium phosphate buffer (pH 7.4), an NADPH-generating system consisting of 0.5 mM NADP⁺, 5 mM glucose 6-phosphate, and 0.5 U of glucose 6-phosphate dehydrogenase/ml, and 200 µM testosterone (or nifedipine) or 100 µM midazolam. In some cases, human liver microsomes were used at the same CYP3A4 concentrations as the recombinant system. Incubations were carried out at 37°C for 10 min and terminated by adding 1.5 ml of CH₂Cl₂ and 0.3 M NaCl. Reactions with nifedipine were incubated at 37°C for 5 min and terminated by adding 1.5 ml of CH₂Cl₂, 0.2 M NaCl, and 0.1 M Na₂CO₃. Organic phases were evaporated under a nitrogen stream, and product formation was determined by HPLC with a C₁₈ (5-µm) analytical column (4.6 × 150 mm). Reactions with midazolam were incubated at 37°C for 5 min and terminated by adding 0.25 ml of CH₃OH. The elution was conducted with a mixture of 64% CH₃OH/36% H₂O (v/v) at a flow rate of 1.2 ml/min, and the detection was by UV absorbance at 240 nm (testosterone) and 254 nm (nifedipine). The elution of midazolam metabolites was conducted with a mixture of 27% CH₃OH/18% CH₃CN/55% 10 mM potassium phosphate buffer (pH 7.4) (v/v) at a flow rate of 1.5 ml/min, and detection was by UV absorbance at 220 nm.

Other Assays. Concentrations of P-450 and b₅ and protein were estimated spectrally by the described methods (Lowry et al., 1951; Omura and Sato, 1964). NADPH-cytochrome c reduction activities were determined as described (Williams and Kamin, 1962; Yasukochi and Masters, 1976) using ₅₅₀ = 21.1 mM¹ cm¹ and an assumed specific activity of 3.0 µmol reduced cytochrome c/min/nmol NPR based on purified human and rabbit NPR preparations (Parikh et al., 1997). The contents of CYP3A4 in human liver microsomes were estimated by coupled SDS-polyacrylamide gel electrophoresis/immunochemical development (Western blotting) (Guengerich et al., 1982).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Recovery of CYP3A4 and NPR in Membranes of E. coli. CYP3A4 and NPR were coexpressed in E. coli from a bicistronic vector using six different culture conditions (Table 1). With regard to CYP3A4 expression levels, long incubation for both LB and TB cultures with vigorous shaking resulted in good yields (Table 1, lot F). On the other hand, the final yield of NPR was highest when the culture time in LB was 7 h, followed by 32 h with mild shaking, in TB medium (Table 1, lot B). Catalytic activities of CYP3A4/NPR membranes for testosterone 6-hydroxylation were higher in samples B and A. 

Effects of Exogenous b₅ and NPR on Catalytic Rates of CYP3A4/NPR Membranes. Because a 1.1:1 molar ratio of NPR/CYP3A4 in membrane preparations appeared to give highest catalytic activities for CYP3A4-dependent testosterone 6-hydroxylation among the conditions tested (Table 1), the effects of b₅ on activity were investigated mainly using this preparation. The effect of b₅ was shown to be concentration-dependent (Fig. 1A). Both human b₅ in E. coli membranes and purified recombinant human b₅ enhanced the testosterone hydroxylation activities about 2-fold; a 1 to 2:1 molar ratio of b₅/CYP3A4 produced the highest activities.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Concentration dependence of the effects of human b5 and NPR in membranes on testosterone 6-hydroxylation by CYP3A4/NPR membranes. A, testosterone (200 µM) was incubated with CYP3A4/NPR membranes (Table 1, preparation B, 0.010 µM CYP3A4) fortified with membranes expressing human NPR (final, 0.020 µM) in the presence of recombinant human b5 added in E. coli membranes () or as purified b5 (). B, testosterone (200 µM) was incubated with CYP3A4/NPR membranes containing CYP3A4 (0.010 µM) and recombinant human NPR (0.011 µM) that were fortified further with different amounts of NPR, added in bacterial membranes to the final ratios shown, in the absence () or presence of added, purified human b5 () or b5 in membranes () (0.010 µM). At the "zero" molar ratio of NPR to CYP3A4, E. coli membranes expressing only CYP3A4 were used.

1

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Concentration dependence of the effects of cholate and purified NPR on testosterone 6-hydroxylation by CYP3A4/NPR membranes. A, cholate (0-1.0 mM) was added to CYP3A4/NPR membranes containing CYP3A4 (0.010 µM) and human NPR (0.011 µM), which were fortified further with exogenous rabbit NPR (0.010 µM) with or without purified human b5 (0.010 µM). These components were mixed and allowed to stand 10 min at room temperature, followed by addition of other reaction components. Testosterone 6-hydroxylation activities (200 µM testosterone) were determined. B, exogenous rabbit NPR (0-0.080 µM) was added to CYP3A4/NPR membranes containing CYP3A4 (0.010 µM) and human NPR (0.011 µM), which were fortified with purified human b5 (0.010 µM) and/or cholate (0.25 mM). Testosterone 6-hydroxylation activities were determined as described in A.

Comparison of Bacterial CYP3A4/NPR Membranes with Other Recombinant Proteins and Human Liver Microsomes. To compare testosterone 6-hydroxylation activities among recombinant CYP3A4 systems, we determined the rates of other recombinant CYP3A4 systems and human liver microsomes after addition of purified b₅ and/or NPR with cholate (Table 4). Catalytic activities of microsomes of lymphoblastoid cells were increased 3-fold by the addition of a 2-fold excess of b₅ over CYP3A4. However, the rates of this system fortified with NPR were lower than those obtained with CYP3A4 expressed using the bacterial CYP3A4/NPR membranes. The activities of one of the microsomal systems from insect cells with baculovirus systems (Baculosomes; PanVera) containing a 1:4.6 molar ratio of CYP3A4 to NPR were improved by the addition of a 2-fold excess of b₅ to CYP3A4. The activities of another baculovirus system containing CYP3A4/NPR/b₅ (1:12:16) (Supersomes; Gentest) were not affected by further addition of b₅ and/or NPR. Human liver microsomes containing 71% CYP3A4 (of total P-450) also were used. Based on CYP3A4 contents, the testosterone 6-hydroxylation activity of this sample was 85 nmol/min/nmol CYP3A4. The molar ratio of CYP3A4/NPR/b₅ in this human liver microsomal preparation was 1:0.06:0.52, and testosterone 6-hydroxylation was only minimally affected by the addition of recombinant b₅ or NPR.

1

Kinetic Analysis of Activity Catalyzed by CYP3A4/NPR Membranes Plus b₅ and Human Liver Microsomes. Because recombinant CYP3A4 (expressed in bacterial membranes using the bacterial bicistronic system) premixed at a molar ratio of 1:2:1 of CYP3A4 to human NPR to human b₅ appeared to be a suitable model for human liver microsomal CYP3A4 with regard to catalytic activities, kinetic parameters were compared with those of human liver microsomes. Testosterone 6-hydroxylation was dependent on CYP3A4 concentration (Fig. 3); linearity of product formation was obtained in a narrow range (0-0.010 µM). Kinetic analysis for testosterone 6-hydroxylation over a substrate concentration range of 10 to 500 µM yielded K_m and V_max values of 80 ± 23 µM and 105 ± 12 nmol/min/nmol CYP3A4 for the recombinant bacterial CYP3A4 system plus b₅ (1:2:1 molar ratio of CYP3A4/NPR/b₅) and 49 ± 14 µM and 105 ± 10 nmol/min/nmol P-450 for human liver microsomes, respectively.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Effects of concentration of CYP3A4/NPR membranes premixed with b5 on testosterone 6-hydroxylation. Testosterone (200 µM) was incubated with different concentrations of CYP3A4/NPR membranes (0.005-0.160 µM CYP) premixed with human NPR and b5 (1:2:1 molar ratio of CYP3A4/NPR/b5).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

A number of studies have shown that CYP3A4 is a major P-450 enzyme involved in the oxidation of many clinically used drugs in human liver microsomes (Guengerich, 1995; Wilkinson, 1996; Li et al., 1997). Prediction of microsomal oxidation of drugs in human livers has been studied recently using activities or kinetic parameters obtained from recombinant systems (Iwatsubo et al., 1997; Ito et al., 1998). A "relative activity factor" was proposed by Crespi (1995), and calculations have been reported with other factors (Kobayashi et al., 1997; Nakajima et al., 1998; Venkatakrishnan et al., 1998). Omeprazole oxidation by human liver microsomes was predicted by using a combination of liver microsomal contents of CYP2C19 and CYP3A4 (immunochemically determined) and kinetic parameters obtained from experiments using recombinant CYP2C19 and CYP3A4 (Yamazaki et al., 1997b), using the methods outlined by Iwatsubo et al. (1997). In these studies, catalytic activities and kinetic parameters of marker drug oxidations catalyzed by recombinant P-450 enzymes are very important for prediction; however, different activities have been reported using several expression systems for CYP3A4 (Shaw et al., 1997; Yamazaki et al., 1997b). Drug oxidation activities of purified CYP3A4 have been studied extensively, and it has been shown that some CYP3A4 activities are dependent on b₅, specific lipid mixtures, cholate, and buffer and salt compositions (Yamazaki et al., 1996a; Shimada and Yamazaki, 1998). We have shown (using the purified recombinant CYP3A4 obtained from a monocistronic bacterial expression system) that a 1:2:1 molar ratio of CYP3A4/NPR/b₅ was suitable for reconstitution of drug oxidation (Yamazaki et al., 1996b).

The present results indicate that a 1:2:1 molar ratio of CYP3A4/NPR/b₅ gives apparently optimal activities for testosterone 6-hydroxylation, nifedipine oxidation, and midazolam 4- and 1'-hydroxylations by recombinant CYP3A4 expressed in bacterial membranes using a bicistronic system. The molar ratio of the three proteins in membranes was the same as in a reconstituted system containing purified CYP3A4. Enhancing effects of exogenous b₅ were observed with either b₅ purified from human, rabbit, or rat liver microsomes, with recombinant human b₅ added in bacterial membranes or with apo b₅, devoid of heme. Effects of b₅ on CYP3A4 activities also were observed in microsomes from lymphoblastoid cells and insect cells (lacking endogenous b₅). Cholate (0.25 mM) was needed to enhance rates when purified NPR was added to CYP3A4 in bacterial membranes; however, cholate appeared not to be necessary when NPR was supplemented from bacterial membranes. Testosterone 6-hydroxylation (~80 min¹), nifedipine oxidation (~30 min¹), and midazolam 4- and 1'-hydroxylation (~10 and 20 min¹) activities were similar among human liver microsomes and/or baculovirus systems based on CYP3A4 contents. These results indicate that the bacterial CYP3A4/NPR membranes plus b₅ should be a simple and suitable model of drug oxidation study for microsomal CYP3A4-dependent reactions in human livers.

NPR is essential to P-450-dependent drug oxidation, and many catalytic activities of CYP3A4 require b₅ for optimal rates. We have shown that b₅ can stimulate some CYP3A4-catalyzed oxidations by complexing with CYP3A4 in a synthetic phospholipid mixture, cholate, and MgCl₂ and enhancing its reduction by NPR without directly transferring electrons to P-450 (Yamazaki et al., 1996a). In the present study, purified b₅, membrane b₅ as well as apo b₅ enhanced the catalytic activities of CYP3A4 in bacterial membranes. Exogenous b₅ also was effective when added to microsomal CYP3A4 systems from lymphoblastoid and insect cells (lacking coexpressed b₅). This suggested that insertion of rabbit, rat, or human b₅ into the phospholipid bilayers is facile, and b₅ can make a suitable complex with CYP3A4 and NPR for efficient electron transfer. Similarly, the membrane-associated NPR was able to mix with membranes containing CYP3A4 and enhance catalytic activities; however, catalytic function of exogenous-purified NPR was dependent on the presence of added b₅ and cholate. Both cholate and b₅ may be necessary for insertion of purified NPR into phospholipid membranes or for complexation of purified NPR with CYP3A4. Cholate may not be an essential component for all of the activities catalyzed by CYP3A4 in reconstituted system; however, it sometimes has been helpful for reconstitution (Yamazaki et al., 1997a). Similar enhancing effects of cholate were observed at the same concentrations (0.01% w/v, or 0.25 mM) on catalytic activities of CYP1B1 in a reconstituted system (Shimada et al., 1998).

In vitro assays are used to predict in vivo drug metabolism in humans. It has been reported that assay conditions, such as different buffers, salts, ionic strengths, detergent level, and components of NADPH-generating systems may affect the CYP3A4-mediated reactions (Yamazaki et al., 1997a; Mäenpää et al., 1998; Shimada and Yamazaki, 1998). Because effects of buffer and salt concentrations are likely to be more significant in reconstituted systems containing CYP3A4 than in human liver microsomes (Yamazaki et al., 1996b), we used 100 mM potassium phosphate buffer (pH 7.4) for drug metabolism reactions with CYP3A4/NPR membranes in this study.

In conclusion, the present study suggested that CYP3A4 coexpressed with NPR in bacterial membranes supplemented with additional NPR plus b₅ is a very useful model for prediction of the rates for microsomal CYP3A4-dependent drug oxidations in human livers. The ratio of NPR to CYP3A4 in human liver microsomes is low and very different from that found to give apparent maximal activities using the bacterial CYP3A4/NPR membranes. We reported previously that CYP3A4-dependent testosterone 6-hydroxylation is stimulated by CYP1A2 in reconstitution systems (Yamazaki et al., 1997a). In this study, CYP1A2 also enhanced catalytic activities of CYP3A4/NPR membranes. These findings indicate the potential for one P-450 enzyme to influence the catalytic characteristics of another P-450 enzyme when a low amount of NPR is present in human liver microsomes. Further studies are necessary for optimization of recombinant P-450 enzyme systems for use as human liver models.