- Email: [email protected]
Effects of Human Cytochrome b5on CYP3A4 Activity and Stabilityin Vivo
Archives of Biochemistry and Biophysics Vol. 366, No. 1, June 1, pp. 116 –124, 1999 Article ID abbi.1999.1192, available online at http://www.idealibrary.com on
Effects of Human Cytochrome b 5 on CYP3A4 Activity and Stability in Vivo 1 Mike W. Voice, Yan Zhang, C. Roland Wolf, Brian Burchell,* and Thomas Friedberg 2 Biomedical Research Centre, and *Department of Molecular and Cellular Pathology, Ninewells Hospital and Medical School, Dundee, DD1 9SY, United Kingdom
Received January 13, 1999, and in revised form March 1, 1999
Cytochrome P450s (P450) form a superfamily of membrane-bound proteins that play a key role in the primary metabolism of both xenobiotics and endogenous compounds such as drugs and hormones, respectively. To be enzymically active, they require the presence of a second membrane-bound protein, NADPH P450 reductase, which transfers electrons from NADPH to the P450. Because of the diversity of P450 enzymes, much of the work on individual forms has been carried out on purified proteins, in vitro, which requires the use of complex reconstitution mixtures to allow the P450 to associate correctly with the NADPH P450 reductase. There is strong evidence from such reconstitution experiments that, when cytochrome b 5 is included, the turnover of some substrates with certain P450s is increased. Here we demonstrate that allowing human P450 reductase, CYP3A4, and cytochrome b 5 to associate in an in vivolike system, by coexpressing all three proteins together in Escherichia coli for the first time, the turnover of both nifedipine and testosterone by CYP3A4 is increased in the presence of cytochrome b 5. The turnover of testosterone was increased by 166% in whole cells and by 167% in preparations of bacterial membranes. The coexpression of cytochrome b 5 also resulted in the stabilization of the P450 during substrate turnover in whole E. coli, with 109% of spectrally active CYP3A4 remaining in cells after 30 min in the presence of cytochrome b 5 compared with 43% of the original P450 remaining in cells in the absence of cytochrome b 5. © 1999 Academic Press
1 This work was supported financially by the Biotechnology and Biological Sciences Research Council, the Department of Trade and Industry, and the consortium of pharmaceutical companies within the LINK scheme: Astra, Glaxo–Wellcome, Janssen Pharmaceutica, Lilly, Novo Nordisk, Parke–Davis, Pfizer, Roche Products, Sanofi– Winthrop, Servier, SmithKline Beecham, Wyeth–Ayerst, and Zeneca. 2 To whom correspondence should be addressed. Fax: 44 1382 669993. E-mail: [email protected].
116
Key Words: cytochrome b 5; recombinant cytochrome P450; NADPH P450 reductase; heterologous expression.
The cytochrome P450s (P450) 3 constitute a superfamily of microsomal heme proteins that play a key role in the metabolism of a wide variety of both xenobiotic and endogenous compounds including drugs, hormones, and procarcinogens (1, 2). For monooxygenase activity, the P450s require the presence of NADPH P450 reductase, a microsomal flavoprotein, to supply electrons from NADPH to the heme moiety of the P450 (3). The large number of P450s and the similarity between family members make the investigation of the individual enzymes difficult. The characterization of individual P450s has, therefore, been confined largely to the use of purified proteins. The assay of purified P450s requires that they are reconstituted with NADPH P450 reductase in a complex mixture which includes detergent, phospholipids, and reduced glutathione (4). The conditions employed for reconstitution and the components of the reconstitution mixture have been shown to critically affect the activity of the resultant system (4 – 6) and can lead to significant variability in the reported activity of the enzymes [compare CYP3A4-dependent testosterone turnovers reported in (7, 8)]. Some in vitro reconstitution experiments have shown that, for a number of P450s, the inclusion of an additional heme protein, cytochrome b 5, can significantly increase substrate turnover by the monooxygenAbbreviations used: P450, cytochrome P450; IPTG, isopropyl-bPMSF, phenylmethylsulfonyl fluoride; DTT dithiothreitol; SDS, sodium dodecyl sulfate, Chaps, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; HRP, horseradish peroxidase; d-ALA, d-aminolevulinic acid; RT-PCR, reverse transcription polymerase chain reaction. 3
D-thiogalactopyranoside;
0003-9861/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
ase system (5, 7, 9 –12) by improving the coupling between the P450 and NADPH P450 reductase (10, 13, 14). This effect is both P450 enzyme and substrate specific; for example, with CYP3A4, one of the major forms found in human liver (15), the inclusion of cytochrome b 5 in the reconstitution mixture stimulates testosterone 6b-hydroxylation and nifedipine oxidation while not affecting the N-demethylation of erythromycin (16). The mechanism of action of cytochrome b 5 in these reconstituted systems is not clear (17). The participation of cytochrome b 5 in electron transport from NADPH P450 reductase has been suggested (7, 14, 17), but there is evidence that apocytochrome b 5 is as effective as the holoprotein in increasing P450 activity (8, 18). This suggests a structural role for cytochrome b 5, strengthening the interaction between the P450 and NADPH P450 reductase, improving the coupling in vitro. Evidence that the inclusion of certain other P450s in reconstitution reactions leads to increased CYP3A4-mediated testosterone turnover (11) supports the suggestion that the inclusion of specific types of membrane-binding peptides may merely be improving the conditions for reconstitution. Indirectly, however, evidence from experiments using specific antibodies raised against cytochrome b 5 (19, 20) in incubations with human liver microsomes show that cytochrome b 5 probably acts in vivo on P450 activity. In addition, in partial reconstitution systems, where microsomes prepared from either yeast (18) or insect (12) cells expressing a P450 together with either cytochrome b 5 or NADPH P450 reductase have been enriched with either NADPH P450 reductase or cytochrome b 5, respectively, the inclusion of cytochrome b 5 has been shown to enhance substrate turnover by certain P450s. There are two conflicting reports on the effects of human cytochrome b 5 when coexpressed together with a P450 and NADPH P450 reductase (21, 22). In one (21), cytochrome b 5 had no significant effect on the turnover of testosterone by CYP3A4 even though this P450 has been shown to be activated by cytochrome b 5 in in vitro reconstitutions (16), whereas, in the other (22), cytochrome b 5 significantly increased the turnover of p-nitrophenetole by CYP2B1. In the former study there was a significant level of endogenous cytochrome b 5, possibly saturating the system. In the latter study (22), however, the recombinant P450 was a rat enzyme while the NADPH P450 reductase and cytochrome b 5 were human forms. In a third report, (23) human cytochrome b 5 was coexpressed in yeast together with a number of human P450s; however, yeast NADPH P450 reductase was used to supply electrons. In this case, the expression of human cytochrome b 5 increased substrate turnover by CYP1A1 and CYP3A4, even in the presence of endogenous yeast cytochrome b 5.
117
Recently, CYP3A4 has been coexpressed in Escherichia coli together with human NADPH P450 reductase (24), allowing the proteins to associate in an in vivo-like system. Bacterial membranes prepared from the cells expressing the proteins exhibited high levels of activity associated with CYP3A4. This system provides the means to study the actions of individual P450s without introducing the variability associated with the reconstitution of purified proteins. In addition, in E. coli, the background of endogenous monooxygenases that may contribute to P450 activity is very low. Here we demonstrate that, using this system to express CYP3A4, human NADPH P450 reductase, and human cytochrome b 5 together for the first time in E. coli, the inclusion of cytochrome b 5 results in increased CYP3A4 activity in an in vivo-like system. In addition, in osmotically shocked cells, we show that the coexpression of cytochrome b 5 results in stabilization of CYP3A4 during substrate turnover. MATERIALS AND METHODS Unless otherwise stated, all chemicals were ANALAR grade and were obtained from Merck (Poole, Dorset, UK) and restriction enzymes and DNA modification enzymes were obtained from Promega Corporation (Southampton, U.K.) or GIBCO BRL (Paisley, U.K.). Isolation and manipulation of human cytochrome b 5 cDNA. The cDNA coding for human cytochrome b 5 was isolated from 5 mg total human liver RNA by reverse transcription polymerase chain reaction (RT-PCR) using the following oligonucleotides as primers: (forward) GGGAATTCCATATGGCTCACCATCATCACGCCATGGCAGAGCAGTCGGAC, (reverse) GCTCTAGATCAGTCCTCTGCCATGTATAGGC. The forward, 59 PCR primer used to amplify the cDNA contained extra 59 sequence to introduce four histidine residues at the N terminal of the recombinant protein to facilitate its purification and an NdeI restriction site at the ATG to ease subsequent cloning steps. The reverse primer introduced an XbaI site after the coding region to facilitate the cloning of the cDNA into pCW. The sequence in the primers specific for cytochrome b 5 is identical to the sequence in the EMBL database, Accession No. M22865. Reverse transcription was carried out using the random hexamer mix supplied with the first-strand synthesis kit (Pharmacia Biotech, Uppsala, Sweden) as primers. The entire first-strand synthesis reaction was used as template for the specific amplification of the cytochrome b 5 cDNA. The cDNA was cloned into the vector pT7 Blue (Novagen, Madison, WI) and sequenced using the Sequenase v2.0 system (Amersham International, Amersham, Bucks, UK) to check for PCRintroduced errors. The resultant coding sequence was excised from the PCR cloning plasmid and cloned into the expression plasmid pCW using the unique NdeI/XbaI sites to give the expression construct pHcb5 (Fig. 1). The expression cassette including the promoter and transcription terminator was then excised from pHcb5 using BclI/BglII and cloned into the plasmid pACYC184 (New England Biolabs, Beverly, MA) at the unique BamHI site to give pHcocb5 (Fig 1) to allow coexpression of the human cytochrome b 5 with the CYP3A4 and NADPH P450 reductase which were expressed from a pCW-based vector, pB216, as described in (24) (Fig. 1). Expression of recombinant human cytochrome b 5 alone and together with CYP3A4 and NADPH P450 reductase. Expression of the recombinant proteins was carried out as described previously (24). Briefly, 100 ml of Terrific broth was inoculated with 1 ml of an overnight culture of E. coli JM109 containing the relevant plasmids. Following growth at 30°C with shaking at 200 rpm to an OD 600 of
118
VOICE ET AL.
FIG. 1. Plasmids used for the expression of recombinant CYP3A4, NADPH P450 reductase, and cytochrome b 5 in E. coli. The plasmids were constructed as described under Materials and Methods. pHcb5 was used for the expression of cytochrome b 5 alone and pHcob5 was used to coexpress cytochrome b 5 with CYP3A4 and NADPH P450 reductase.
between 0.6 and 0.9, the cultures were induced by the addition of IPTG to 1 mM and d-aminolevulinic acid to 0.5 mM. After 20 h the cells were harvested and resuspended in 1/20th culture volume of a solution comprising 100 mM Tris–acetate, 500 mM sucrose, 0.5 mM EDTA, pH 7.6 (2X TSE). The suspension was then diluted with an equal volume of distilled water. The P450 content of the cell suspension was determined using Fe 21–CO versus Fe 21 difference spectra as described in (24, 25) and the cytochrome b 5 content was determined using reduced versus oxidized difference spectra (10). Cell membranes were prepared from the bacteria by sonication followed by differential centrifugation as (26) and the CYP3A4 and cytochrome b 5 content was determined as described previously. The level of NADPH P450 reductase in the cell membranes was estimated using a spectrophotometric assay to determine the rate of NADPH-dependent reduction of cytochrome c (27). The cytochrome c reductase activity in whole cells was not determined due to the high background endogenous activity. The protein content of membrane preparations was determined by the method of (28) using reagent supplied by Bio-Rad (Hemel Hempsted, Herts). Purification of recombinant human cytochrome b 5. The recombinant protein was purified from bacterial membranes by solubilization in 1 mg Emulgen 911/ml for 1 h at 4°C followed by clarification by centrifugation at 100,000g. The solubilized cytochrome b 5 in the 100,000g supernatant was purified over a nickel Sepharose column (Pharmacia Biotech) in 20 mM Tris–Cl, pH 7.4, 500 mM KCl, 20% glycerol and eluted using 50 mM imidazole. Determination of nifedipine and testosterone turnover by CYP3A4 in osmotically shocked cells and cell membranes. For the determination of enzyme activity of whole cells, the cell suspension was diluted in 1X TSE to give a CYP3A4 level of 100 pmol/ml in a conical flask. The suspension was allowed to equilibrate to 37°C before either testosterone or nifedipine was added to a final concentration of 200 mM. The cells were incubated for 2 min at 37°C after which a 200-ml aliquot was withdrawn into 100 ml ice-cold methanol with 5 ml 60% perchloric acid in water. The precipitated protein was removed after a 10-min incubation on ice by centrifugation and the substrate and metabolite in the supernatant were separated by reverse-phase HPLC as described in (24). Substrate turnover by bacterial membranes was determined as in
(24). Membranes containing 10 pmol CYP3A4 were diluted to 160 ml with 50 mM Hepes, pH 7.4 (for testosterone metabolism), or 100 mM potassium phosphate, pH 7.9. Where appropriate, sufficient MgCl 2 was added to give a final concentration in the assay of 30 mM. Substrate was added from concentrated stock to give a final concentration of 200 mM for nifedipine or 100 mM for testosterone in the assay. The mixture was equilibrated at 37°C, after which the assay was started by the addition of 40 ml 5X NADPH-generating system (5 mM NADP, 25 mM glucose 6-phosphate, 1 unit glucose-6-phosphate dehydrogenase in assay buffer). After 10 min, the assay was stopped by the addition of 100 ml ice-cold methanol and 5 ml 60% perchloric acid in water. Following a 10-min incubation on ice, the precipitated proteins were removed by centrifugation. The substrate and metabolite in the supernatant were separated by reverse-phase HPLC (24). Immunoblotting. Immunoblotting was carried out as described in (29). Proteins were separated on 9% SDS–polyacrylamide gels when probing for CYP3A4 and P450 reductase and 18% gels when probing for cytochrome b 5. The blots were developed using polyclonal antibodies raised in rabbits against recombinant human CYP3A4 or purified human NADPH P450 reductase or cytochrome b 5 followed by horseradish-peroxidase linked donkey anti-rabbit immunoglobulin (SAPU). The bands were visualized using enhanced chemiluminescence (ECL, Amersham International, Amersham, Bucks).
RESULTS
Cloning and Expression of Human Cytochrome b 5 in E. coli The cDNA coding for cytochrome b 5 was isolated by RT-PCR and found to be 100% identical to the sequence in the EMBL database (Accession No. M22865) (30). The cDNA was then cloned into the vector pCW, giving the plasmid pHcb5 which was used for the expression of recombinant cytochrome b 5 (Fig. 1). The yield of recombinant cytochrome b 5 was 1250 nmol/liter culture (Fig. 2). When cells lacking the expression vec-
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
119
FIG. 2. Analysis of E. coli JM109 transformed with pHcb5. (A) The level of cytochrome b 5 was determined from reduced versus oxidized membrane spectra as described. Control cells were E. coli JM109 that had no pHcb5 (i.e., no expression plasmid). (B) Western blot analysis was carried out using an 18% polyacrylamide gel. Polyclonal antibodies raised in rabbits against human cytochrome b 5 were used as primary antibodies. Preparations were derived either from control cells (C) or from cells grown in the presence of IPTG (I).
tor were analyzed, they were found to have a background spectrum both in whole cells and in cell membranes, presumably due to endogenous bacterial heme proteins (Fig. 2). This value has been subtracted from all quoted recombinant cytochrome b 5 levels. Bacterial membranes isolated from E. coli expressing recombinant human cytochrome b 5 contained 3400 pmol cytochrome b 5/mg protein, representing an 87% recovery of the total expressed recombinant protein from whole cells. Western blotting confirmed the presence of an immunoreactive protein in induced cell membranes of the correct size (Fig. 2) (there is a nonspecific band seen in human liver at a much higher molecular weight). Following purification over nickel agarose, the ability of the recombinant human cytochrome b 5 to increase the turnover of CYP3A4 in a reconstituted system was confirmed. The reconstitution conditions were as described in (7). The inclusion of 20 pmol cytochrome b 5 in a reaction mixture containing 10 pmol CYP3A4 and 30 pmol NADPH P450 reductase resulted in the turnover of nifedipine being increased from 2.1 to 12.1
min 21, demonstrating that the recombinant human cytochrome b 5 was capable of interacting functionally with P450s. Coexpression of Human Cytochrome b 5 with CYP3A4 and NADPH P450 Reductase The cytochrome b 5 cDNA, promoter, and transcription terminator from pHcb5 were then cloned into the vector pACYC184, giving pHcob5 (Fig. 1). This plasmid has a different selection marker (chloramphenicol) and an alternative origin of replication to allow cotransformation of cells with expression plasmids encoding human CYP3A4, NADPH P450 reductase (on pCW), and cytochrome b 5. Interaction of cytochrome b 5 with CYP3A4 was studied using two strategies involving different N-terminal modifications of CYP3A4 for expression. In one strategy, CYP3A4 was modified as described in (24) by altering the 59 coding region yielding 17a3A4, the most commonly used form of recombinant CYP3A4. The second strategy used bacterial
120
VOICE ET AL. TABLE I
Yield of CYP3A4, P450 Reductase, and Cytochrome b 5 in Whole Cells and Bacterial Membranes Whole Cells Strain expressing 17aCYP3A4
Proteins expressed CYP3A4 1 P450 reductase CYP3A4, P450 reductase, b 5
Strain expressing ompA(12) CYP3A4
P450 (nmol/liter culture)
Reductase
Cytochrome b 5 (nmol/liter culture)
P450 (nmol/liter culture)
Reductase
Cytochrome b 5 (nmol/liter culture)
140 6 13
na a
na
162 6 17
na
na
na
405 6 54
na
807 6 163
82 6 14*
81 6 10*
Cell Membranes
CYP3A4 1 P450 reductase CYP3A4, P450 reductase, b 5
P450 (pmol/mg protein)
Reductase (pmol/mg protein)
b5 (pmol/mg protein)
P450 (pmol/mg protein)
Reductase (pmol/mg protein)
b5 (pmol/mg protein)
146 6 20
180 6 18
na
185 6 29
323 6 34
na
227 6 20
328 6 65
361 6 117
557 6 123
86 6 18*
86 6 13*
Note. P450 levels were determined spectrophotometrically from the Fe 21–CO versus Fe 21 difference spectra. Cytochrome b 5 levels were determined spectrophotometrically from reduced versus oxidized difference spectra. Values are the means of at least five independent expression experiments 6 SEM. a Not applicable. * P , 0.05 comparing values for 6cytochrome b 5.
ompA leader sequence to optimize expression (31) (M. Pritchard et al., manuscript in preparation). This strategy, where the DNA coding for the ompA leader is fused to the 59 end of the CYP3A4 cDNA and the protein is subsequently posttranslationally modified, leaves the N-terminal amino acid sequence of the protein unaltered except for the addition of two amino acids. On induction with IPTG for 20 h, all three proteins were expressed at high levels. The level of cytochrome b 5 was higher in cells coexpressing the ompA(12) CYP3A4 with a yield of 807 6 163 nmol/liter in whole cells compared with 405 6 54 nmol/liter when it was coexpressed with the 17aCYP3A4 construct. In membranes, the difference remained, with a yield of 557 6 123 pmol/mg cytochrome b 5 in membranes carrying the ompA(12) CYP3A4 protein compared with 328 6 65 pmol/mg when it was coexpressed with the 17aCYP3A4 construct. Although the yield of CYP3A4 was variable, in cells coexpressing cytochrome b 5, it was found, in general, to be 50% of that in cells expressing CYP3A4 and NADPH P450 reductase alone (Table I). This reduction in yield was found to carry through to membrane preparations from cells coexpressing cytochrome b 5 and was seen with both derivatives of CYP3A4 (Table I). The level of NADPH P450 reductase in bacterial membranes as measured by the rate of NADPH-dependent reduction of cytochrome c
was largely unaffected by the coexpression of cytochrome b 5. The 17aCYP3A4 construct exhibited a rate of cytochrome c reduction of 540 6 54 nmol/min/mg protein, and the ompA(12) construct, a rate of 970 6 101 nmol/min/mg protein in the absence of cytochrome b 5. These values rose very slightly to 681 6 60 and 1084 6 135 nmol/min/mg protein, respectively, on coexpression of cytochrome b 5. Visual inspection of a Western blot of bacterial membranes prepared from induced cells (Fig. 3) demonstrated that there was no obvious reduction in the level of CYP3A4 polypeptide on coexpression of cytochrome b 5, suggesting that the reduction in yield of spectrally active CYP3A4 may be due to heme depletion; however, increasing the concentration of d-aminolevulinic acid in the culture medium had no effect on the level of spectrally active CYP3A4 (data not shown). Effect of Cytochrome b 5 on the Turnover of Nifedipine and Testosterone by CYP3A4 In osmotically shocked bacteria, the turnover of both nifedipine and testosterone by ompA(12) CYP3A4 was higher than that seen with the 17aCYP3A4 (Table II) as previously demonstrated (M. Pritchard et al., manuscript in preparation). The coexpression of cytochrome b 5 resulted in significantly increased turnover of both
121
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
FIG. 3. Western blot analysis of E. coli expressing recombinant CYP3A4, NADPH P450 reductase, and cytochrome b 5. Ten micrograms of bacterial membrane proteins from E. coli JM109 expressing recombinant CYP3A4, NADPH P450 reductase 6 cytochrome b 5 or human liver microsomes was loaded on a 9% polyacrylamide gel (18% for analysis of cytochrome b 5). The proteins were detected using polyclonal antibodies raised in rabbits against the proteins indicated and horseradish peroxidase-linked donkey anti-rabbit IgG followed by enhanced chemiluminescence.
substrates with both forms of CYP3A4 (Table II), showing that it interacts equally well with the ompA(12) CYP3A4 and the most commonly used form(17aCYP3A4). In bacterial membranes, the presence of cytochrome b 5 had a similar effect, significantly increasing substrate turnover (Table III) by both forms of CYP3A4. It has been demonstrated previously that the inclusion of magnesium in the assay buffer can enhance the activity of CYP3A4 (16, 24). To determine whether magnesium and cytochrome b 5 act through the same mechanism, we assayed the activity of CYP3A4 6 cytochrome b 5 in the presence of 30 mM MgCl 2. MgCl 2 stimulated TABLE II
the activity of the system by a similar amount regardless of the presence of cytochrome b 5, suggesting that cytochrome b 5 and MgCl 2 act independently. Effects of Cytochrome b 5 on Stability of CYP3A4 Incubation of intact cells expressing CYP3A4 and NADPH P450 reductase in the presence of 200 mM testosterone at 37°C led to a significant reduction in the amount of spectrally active CYP3A4 (Fig. 4). How-
TABLE III
Turnover of Nifedipine and Testosterone by Bacterial Membranes Containing Recombinant CYP3A4 and NADPH P450 Reductase 6 Cytochrome b 5
Turnover of Testosterone and Nifedipine by Whole Cells Expressing CYP3A4 and NADPH P450 Reductase 6 Cytochrome b 5
Nifedipine oxidase (min 21) Control 1 Cytochrome b 5 Testosterone-6b-hydroxylase (min 21) Control 1 cytochrome b 5
17aCYP3A4
ompA(12) CYP3A4
9.4 1 0.7 17.4 6 2.5*
34.2 6 3.2 59.8 6 10.2*
17.8 6 1.7 25.3 6 2.5*
43.2 6 2.9 71.8 6 4.9*
Note. Turnover numbers are expressed as picomoles of product formed (nifedipine oxide and 6b-hyroxytestosterone, respectively) per picomole of CYP3A4 per minute at 37°C during a 2-min incubation. Values are the means from at least six independent expression experiments 6 SEM. * P , 0.05 comparing values for 6cytochrome b 5.
Nifedipine oxidase (min 21) Control 1 Cytochrome b 5 1 30 mM MgCl 2 1 Cytochrome b 5, 30 mM MgCl 2 Testosterone-6b-hydroxylase (min 21) Control 1 Cytochrome b 5 1 30 mM MgCl 2 1 Cytochrome b 5, 30 mM MgCl 2
17aCYP3A4
ompA(12) CYP3A4
25.7 6 1.6 46.3 6 5.2* 37.4 6 2.8 62.6 6 7.6*
43.4 6 4.3 71.5 6 2.7* 49.4 6 3.3 84.7 6 4.8*
2.1 6 0.2 4.9 6 0.7* 24.6 6 1.9 40.2 6 1.4*
2.1 6 0.4 3.5 6 0.3* 32.5 6 2.7 44.3 6 1.3*
Note. Turnover numbers are expressed as picomoles of product formed (oxidized nifedipine and 6b-hydroxytestosterone, respectively) per picomole of CYP3A4 per minute at 37°C during a 10-min assay. Values are the means from determinations of turnover in at least six independent membrane preparations each carried out in duplicate. * P , 0.02 comparing values for 6cytochrome b 5.
122
VOICE ET AL.
FIG. 4. Effect of cytochrome b 5 on the stability of spectrally active CYP3A4 in whole cells during substrate turnover in whole cells. Cells were diluted in 1X TSE to give 100 pmol/ml CYP3A4 in 10-ml final volume. The suspension was allowed to equilibrate to 37°C after which testosterone was added to 200 mM final concentration. Twomilliliter aliquots were withdrawn 0, 15, and 30 min after the addition of testosterone and the remaining spectrally active CYP3A4 was determined by Fe 21–CO versus Fe 21 difference spectra. h, Control, no substrate; ■, control, 200 mM testosterone; E, cytochrome b 5, no substrate; F, cytochrome b 5, 200 mM testosterone. Two-hundredmicroliter aliquots were withdrawn 2 min after the addition of testosterone to confirm substrate turnover by HPLC analysis. In addition, 5-ml aliquots were withdrawn at 30 min for analysis by SDS– polyacrylamide gel electrophoresis. The turnover of testosterone was 36 6 0.8 min 21 in control cells and 59 6 6 in cells containing cytochrome b 5. Values are expressed as the percentage CYP3A4 remaining and are the means of at least four incubations using independent expression cultures. *P , 0.05 comparing cytochrome b 5 coexpression 6 substrate, **P , 0.02 comparing values for 6cytochrome b 5.
ever, in cells expressing CYP3A4, NADPH P450 reductase, and cytochrome b 5, the level of spectrally active CYP3A4 remained unchanged, even after 30 min of incubation at 37°C in the presence of 200 mM testosterone (Fig. 4). Conversely, the presence of cytochrome b 5 caused a slight but statistically significant loss of spectrally active CYP3A4 when the cells were incubated in the absence of substrate (Fig. 4). Western blotting demonstrated that the loss of spectrally active CYP3A4 was not accompanied by a concomitant loss of immunologically reactive protein (Fig. 5A). In all incubations, the levels of spectrally active cytochrome b 5 were unchanged (data not shown). Incubation of bacterial membranes containing CYP3A4 and NADPH P450 reductase with and without cytochrome b 5 in the presence of substrate at 37°C showed that the protective effect of cytochrome b 5 was confined to whole-cell incubations. The loss of spectrally active CYP3A4 was similar in membranes with CYP3A4 and reductase and membranes with CYP3A4, P450 reductase, and cytochrome b 5, with around 20%
of the original level of spectrally active CYP3A4 remaining after 20 min (Table IV). Loss of spectrally active CYP3A4 from membranes in the presence of cytochrome b 5 and NADPH but in the absence of substrate was significantly increased compared with membranes with CYP3A4 and P450 reductase alone, with only 31% remaining after 20 min (Table IV). This was significantly less than in membranes lacking cytochrome b 5 which had 66% of the original spectrally active CYP3A4 remaining after 20 min. Western blotting demonstrated that the levels of immunoreactive CYP3A4 were not affected by the incubations (Fig. 5). As with the whole-cell experiments, the levels of spectrally active cytochrome b 5 were unchanged during the incubations (data not shown). As equal amounts of P450 were loaded on the gel, there is apparently more CYP3A4 in membranes from cells coexpressing cytochrome b 5 which have a higher level of apo-CYP3A4. DISCUSSION
Here, we have shown that when CYP3A4 and NADPH P450 reductase associate in an in vivo-like system the coexpression of cytochrome b 5 leads to increased turnover of both testosterone and nifedipine by CYP3A4 when the activity of the system is assayed in whole cells (Table I) or in vitro (Table II). This clearly demonstrates that cytochrome b 5 can modulate the activity of the P450s in an in vivo-like system where the possible confounding effects associated with reconstitution are absent. The action of cytochrome b 5 on the ompA(12) CYP3A4 demonstrates that it interacts on native P450s as well as the more commonly used 17aCYP3A4. Our data on the action of cytochrome b 5 contrast with a report (21) that CYP3A4, NADPH P450 reductase, and cytochrome b 5 were coexpressed in insect cells and the inclusion of cytochrome b 5 had no significant effect on the turnover of testosterone. There was, however, a low but significant amount of endogenous cytochrome b 5 in the cells used. The molar ratio of endogenous cytochrome b 5:CYP3A4 was around 0.3, a level that has since been shown in vitro (8) to fully activate testosterone turnover by CYP3A4. In our system, although there was a background heme spectrum in E. coli (Fig. 2), there was no cytochrome b 5-like protein detectable by Western blotting (Fig. 2), allowing us to determine the action of cytochrome b 5 in the absence of endogenous protein. It has been postulated that there may be a bacterial protein able to substitute for mammalian cytochrome b 5 (24). This was based on the high substrate turnover achieved when CYP3A4 and NADPH P450 reductase were coexpressed as compared with reconstituted systems in the absence of cytochrome b 5. Here we have shown that the coexpression of cytochrome b 5 with CYP3A4 and NADPH P450 reductase results in the
123
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
FIG. 5. Western blot analysis of CYP3A4 in whole cells and bacterial membranes following incubation with substrate and cofactor. (A) Aliquots of cell suspension 6 coexpressed cytochrome b 5 and 6 testosterone were taken at the times shown from the incubations described in Fig. 4. A volume containing the equivalent of 0.2 pmol CYP3A4 at the start of the incubation was loaded onto a 9% polyacrylamide gel which was blotted. (B) Aliquots of incubations described in Table IV were taken after 20 min incubation. A volume containing the equivalent of 1 pmol CYP3A4 at the start of the incubation was loaded onto a 9% polyacrylamide gel which was Western blotted. The blots were developed using polyclonal antibodies raised in rabbits against purified recombinant CYP3A4 and horseradish peroxidase-linked donkey anti-rabbit IgG.
further activation of recombinant CYP3A4, indicating the absence of any bacterial protein fully able to substitute for cytochrome b 5. There is evidence that the P450:NADPH P450 reductase ratio can affect the rate of substrate turnover (17, 20). Although in our system the coexpression of cytochrome b 5 with CYP3A4 and NADPH P450 reductase resulted in an increased reductase:P450 ratio, the expression levels of CYP3A4 were sufficiently variable for the ratios to overlap. This meant that the P450:reductase ratio was similar in cultures containing low CYP3A4 levels in the absence of cytochrome b 5 compared with cultures containing high CYP3A4 levels in the presence of cytochrome b 5. In these cases, the inclusion of cytochrome b 5 still significantly stimulated testosterone and nifedipine turnover, demonstrating that the increased activity was not due to altered CYP3A4:NADPH P450 reductase ratios. We have also shown that, in addition to stimulating CYP3A4 activity, the presence of cytochrome b 5 prevents the loss of spectrally active CYP3A4 seen during the turnover of substrate in whole cells (Fig. 5). This could be a result of cytochrome b 5 improving the coupling between CYP3A4 and NADPH P450 reductase, reducing the levels of damaging reactive oxygen spe-
TABLE IV
Stability of CYP3A4 in Bacterial Membranes during Substrate Turnover % Spectrally active CYP3A4 remaining Incubation mixture
Control
1 cytochrome b 5
200 mM Testosterone NADPH-generating system 200 mM Testosterone 1 NADPH-generating system
105 6 6 66 6 6
101 6 1 31 6 1*
26 6 4
16 6 1
Note. Membranes were diluted in 50 mM Hepes, pH 7.4, 30 mM MgCl 2 6 250 mM testosterone to give 125 pmol/ml CYP3A4 in 1.76-ml final volume. The suspension was allowed to equilibrate to 37°C after which either 0.44 ml 50 mM Hepes, pH 7.4, 30 mM MgCl 2, or 0.44 ml 5X NADPH-generating system containing 30 mM MgCl 2 was added. After 20 min incubation at 37°C 200 ml was withdrawn to determine substrate turnover where appropriate. The turnover of testosterone was 23 6 0.5 min 21 in control membranes and 37 6 6 min 21 in membranes containing cytochrome b 5. The amount of spectrally active CYP3A4 in the remaining 2 ml was determined spectrophotometrically from the Fe 21–CO versus Fe 21 difference spectra. Values are expressed as the percentage of spectrally active CYP3A4 remaining after 20 min and are the means of incubations using at least three independent bacterial membrane preparations. * P , 0.02 comparing values for 6cytochrome b 5.
124
VOICE ET AL.
cies in the cells. Indeed, cytochrome b 5 has been shown in vitro to improve the coupling between a number of rat liver P450s and NADPH P450 reductase, with a concomitant reduction in H 2O 2 production (10, 13, 32). The protective effect was not observed with bacterial membranes (Table IV) which may be due to the loss of cytoplasmic components during the preparation of membranes. The presence of cytochrome b 5 led to an increased rate of loss of spectrally active CYP3A4 when either cells or membranes were incubated in the absence of substrate but in the presence of NADPH (endogenous in the case of cells). It is possible that, due to the increased coupling, in the absence of substrate, the CYP3A4 is involved in a futile cycle, generating reactive species that destroy the holoenzyme. In conclusion, we have shown that, using a system that allows CYP3A4, NADPH P450 reductase, and cytochrome b 5 to associate in an in vivo-like system, cytochrome b 5 increases the rate of turnover of testosterone and nifedipine by CYP3A4 and, in addition, protects CYP3A4 from damage during substrate turnover in whole cells. REFERENCES 1. Guengerich, F. P., and Parikh, A. (1997) Curr. Opin. Biotechnol. 8, 623– 628. 2. Friedberg, T., and Wolf, C. R. (1996) Adv. Drug Delivery Rev. 22, 187–213. 3. Smith, G. C. M., Tew, D. G., and Wolf, C. R. (1994) Proc. Natl. Acad. Sci. USA 91, 8710 – 8714. 4. Gillam, E. M. J., Guo, Z., Ueng, Y.-F., Yamazaki, H., Cock, I., Reilly, P. E. B., Hooper, W. D., and Guengerich, F. P. (1995) Arch. Biochem. Biophys. 317, 5. Shaw, P. M., Hosea, N. A., Thompson, D. V., Lenius, J. M., and Guengerich, F. P. (1997) Arch. Biochem. Biophys. 348, 107–115. 6. Imaoka, S., Imai, Y., Shimada, T., and Funae, Y. (1992) Biochemistry 31, 6063– 6069. 7. Yamazaki, H., Nakano, M., Imai, Y., Ueng, Y., Guengerich, F. P., and Shimada, T. (1996) Arch. Biochem. Biophys. 325, 174 –182. 8. Yamazaki, H., Johnson, W. W., Ueng, Y.-F., Shimada, T., and Guengerich, F. P. (1996) J. Biol. Chem. 271, 27438 –27444. 9. Jansson, I. J., Tamburini, P. P., Favreau, L. V., and Schenkman, J. B. (1985) Drug Metab. Dispos. 13, 453– 458. 10. Holmans, P. L., Shet, M. S., Martin-Wixtrom, C. A., Fisher, C. W., and Estabrook, R. W. (1994) Arch. Biochem. Biophys. 312, 554 –565.
11. Yamazaki, H., Gillam, E. M. J., Dong, M.-S., Johnson, W. W., Guengerich, F. P., and Shimada, T. (1997) Arch. Biochem. Biophys. 342, 329 –337. 12. Patten, C. J., and Koch, P. (1995) Arch. Biochem. Biophys. 317, 504 –513. 13. Gorsky, L. D., and Coon, M. J. (1986) Drug Metab. Dispos. 14, 89 –96. 14. Bell, L. C., and Guengerich, F. P. (1997) J. Biol. Chem. 272, 29643–29651. 15. Forrester, L. M., Henderson, C. J., Glancey, M. J., Back, D. J., Park, B. K., Ball, S. E., Kitteringham, N. R., McLaren, A. W., Miles, J. S., Skett, P., and Wolf, C. R. (1992) Biochem. J. 281, 359 –368. 16. Yamazaki, H., Ueng, Y.-F., Shimada, T., and Guengerich, F. P. (1995) Biochemistry 34, 8380 – 8389. 17. Morgan, E. T., and Coon, M. J. (1984) Drug Metab. Dispos. 12, 358 –364. 18. Auchus, R. J., Lee, T. C., and Miller, W. L. (1998) J. Biol. Chem. 273, 3158 –3165. 19. Sakai, Y., Yanase, T., Hara, T., Takayanagi, R., Haji, M., and Nawata, H. (1994) Clin. Endocrinol. 40, 205–209. 20. Yamazaki, H., Nakano, M., Gillam, E. M. J., Bell, L. C., Guengerich, F. P., and Shimada, T. (1996) Biochem. Pharmacol. 52, 301–309. 21. Lee, C. A., Kadwell, S. H., Kost, T. H., and Serabjit-Singh, C. J. (1995) Arch. Biochem. Biophys. 319, 157–167. 22. Aoyama, T., Nagata, K., Yamazoe, Y., Kato, R., Matsunaga, E., Gelboin, H. V., and Gonzalez, F. J. (1990) Proc. Natl. Acad. Sci. USA 87, 5425–5429. 23. Truan, G., Cullin, C., Reisdorf, P., Urban, P., and Pompon, D. (1993) Gene 125, 49 –55. 24. Blake, J. A. R., Pritchard, M., Ding, S., Smith, G. C. M., Burchell, B., Wolf, C. R., and Friedberg, T. (1996) FEBS Lett. 397, 210 –214. 25. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2379 –2385. 26. Renaud, J. P., Peyronneau, M. A., Urban, P., Truan, G., Cullin, C., Pompon, D., Beaune, P., and Mansuy, D. (1993) Toxicology 82, 39 –52. 27. Strobel, H. W., and Dignam, D. J. (1978) Methods Enzymol. 52, 89 –96. 28. Bradford, M. M. (1976) Anal. Biochem. 72, 248 –254. 29. Steinberg, P., Lafranconi, M., Wolf, C. R., Waxman, D. J., Oesch, F., and Friedberg, T. (1987) Mol. Pharmacol. 32, 463– 470. 30. Yoo, M., and Steggles, A. W. (1988) Biochem. Biophys. Res. Commun. 156, 576 –580. 31. Pritchard, M. P., Ossetian, R., Li, D. N., Henderson, C. J., Burchell, B., Wolf, C. R., and Friedberg, T. (1997) Arch. Biochem. Biophys. 345, 342–354. 32. Jansson, I., and Schenkman, J. B. (1987) Drug Metab. Dispos. 15, 344 –348.
Effects of Human Cytochrome b 5 on CYP3A4 Activity and Stability in Vivo 1 Mike W. Voice, Yan Zhang, C. Roland Wolf, Brian Burchell,* and Thomas Friedberg 2 Biomedical Research Centre, and *Department of Molecular and Cellular Pathology, Ninewells Hospital and Medical School, Dundee, DD1 9SY, United Kingdom
Received January 13, 1999, and in revised form March 1, 1999
Cytochrome P450s (P450) form a superfamily of membrane-bound proteins that play a key role in the primary metabolism of both xenobiotics and endogenous compounds such as drugs and hormones, respectively. To be enzymically active, they require the presence of a second membrane-bound protein, NADPH P450 reductase, which transfers electrons from NADPH to the P450. Because of the diversity of P450 enzymes, much of the work on individual forms has been carried out on purified proteins, in vitro, which requires the use of complex reconstitution mixtures to allow the P450 to associate correctly with the NADPH P450 reductase. There is strong evidence from such reconstitution experiments that, when cytochrome b 5 is included, the turnover of some substrates with certain P450s is increased. Here we demonstrate that allowing human P450 reductase, CYP3A4, and cytochrome b 5 to associate in an in vivolike system, by coexpressing all three proteins together in Escherichia coli for the first time, the turnover of both nifedipine and testosterone by CYP3A4 is increased in the presence of cytochrome b 5. The turnover of testosterone was increased by 166% in whole cells and by 167% in preparations of bacterial membranes. The coexpression of cytochrome b 5 also resulted in the stabilization of the P450 during substrate turnover in whole E. coli, with 109% of spectrally active CYP3A4 remaining in cells after 30 min in the presence of cytochrome b 5 compared with 43% of the original P450 remaining in cells in the absence of cytochrome b 5. © 1999 Academic Press
1 This work was supported financially by the Biotechnology and Biological Sciences Research Council, the Department of Trade and Industry, and the consortium of pharmaceutical companies within the LINK scheme: Astra, Glaxo–Wellcome, Janssen Pharmaceutica, Lilly, Novo Nordisk, Parke–Davis, Pfizer, Roche Products, Sanofi– Winthrop, Servier, SmithKline Beecham, Wyeth–Ayerst, and Zeneca. 2 To whom correspondence should be addressed. Fax: 44 1382 669993. E-mail: [email protected].
116
Key Words: cytochrome b 5; recombinant cytochrome P450; NADPH P450 reductase; heterologous expression.
The cytochrome P450s (P450) 3 constitute a superfamily of microsomal heme proteins that play a key role in the metabolism of a wide variety of both xenobiotic and endogenous compounds including drugs, hormones, and procarcinogens (1, 2). For monooxygenase activity, the P450s require the presence of NADPH P450 reductase, a microsomal flavoprotein, to supply electrons from NADPH to the heme moiety of the P450 (3). The large number of P450s and the similarity between family members make the investigation of the individual enzymes difficult. The characterization of individual P450s has, therefore, been confined largely to the use of purified proteins. The assay of purified P450s requires that they are reconstituted with NADPH P450 reductase in a complex mixture which includes detergent, phospholipids, and reduced glutathione (4). The conditions employed for reconstitution and the components of the reconstitution mixture have been shown to critically affect the activity of the resultant system (4 – 6) and can lead to significant variability in the reported activity of the enzymes [compare CYP3A4-dependent testosterone turnovers reported in (7, 8)]. Some in vitro reconstitution experiments have shown that, for a number of P450s, the inclusion of an additional heme protein, cytochrome b 5, can significantly increase substrate turnover by the monooxygenAbbreviations used: P450, cytochrome P450; IPTG, isopropyl-bPMSF, phenylmethylsulfonyl fluoride; DTT dithiothreitol; SDS, sodium dodecyl sulfate, Chaps, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; HRP, horseradish peroxidase; d-ALA, d-aminolevulinic acid; RT-PCR, reverse transcription polymerase chain reaction. 3
D-thiogalactopyranoside;
0003-9861/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
ase system (5, 7, 9 –12) by improving the coupling between the P450 and NADPH P450 reductase (10, 13, 14). This effect is both P450 enzyme and substrate specific; for example, with CYP3A4, one of the major forms found in human liver (15), the inclusion of cytochrome b 5 in the reconstitution mixture stimulates testosterone 6b-hydroxylation and nifedipine oxidation while not affecting the N-demethylation of erythromycin (16). The mechanism of action of cytochrome b 5 in these reconstituted systems is not clear (17). The participation of cytochrome b 5 in electron transport from NADPH P450 reductase has been suggested (7, 14, 17), but there is evidence that apocytochrome b 5 is as effective as the holoprotein in increasing P450 activity (8, 18). This suggests a structural role for cytochrome b 5, strengthening the interaction between the P450 and NADPH P450 reductase, improving the coupling in vitro. Evidence that the inclusion of certain other P450s in reconstitution reactions leads to increased CYP3A4-mediated testosterone turnover (11) supports the suggestion that the inclusion of specific types of membrane-binding peptides may merely be improving the conditions for reconstitution. Indirectly, however, evidence from experiments using specific antibodies raised against cytochrome b 5 (19, 20) in incubations with human liver microsomes show that cytochrome b 5 probably acts in vivo on P450 activity. In addition, in partial reconstitution systems, where microsomes prepared from either yeast (18) or insect (12) cells expressing a P450 together with either cytochrome b 5 or NADPH P450 reductase have been enriched with either NADPH P450 reductase or cytochrome b 5, respectively, the inclusion of cytochrome b 5 has been shown to enhance substrate turnover by certain P450s. There are two conflicting reports on the effects of human cytochrome b 5 when coexpressed together with a P450 and NADPH P450 reductase (21, 22). In one (21), cytochrome b 5 had no significant effect on the turnover of testosterone by CYP3A4 even though this P450 has been shown to be activated by cytochrome b 5 in in vitro reconstitutions (16), whereas, in the other (22), cytochrome b 5 significantly increased the turnover of p-nitrophenetole by CYP2B1. In the former study there was a significant level of endogenous cytochrome b 5, possibly saturating the system. In the latter study (22), however, the recombinant P450 was a rat enzyme while the NADPH P450 reductase and cytochrome b 5 were human forms. In a third report, (23) human cytochrome b 5 was coexpressed in yeast together with a number of human P450s; however, yeast NADPH P450 reductase was used to supply electrons. In this case, the expression of human cytochrome b 5 increased substrate turnover by CYP1A1 and CYP3A4, even in the presence of endogenous yeast cytochrome b 5.
117
Recently, CYP3A4 has been coexpressed in Escherichia coli together with human NADPH P450 reductase (24), allowing the proteins to associate in an in vivo-like system. Bacterial membranes prepared from the cells expressing the proteins exhibited high levels of activity associated with CYP3A4. This system provides the means to study the actions of individual P450s without introducing the variability associated with the reconstitution of purified proteins. In addition, in E. coli, the background of endogenous monooxygenases that may contribute to P450 activity is very low. Here we demonstrate that, using this system to express CYP3A4, human NADPH P450 reductase, and human cytochrome b 5 together for the first time in E. coli, the inclusion of cytochrome b 5 results in increased CYP3A4 activity in an in vivo-like system. In addition, in osmotically shocked cells, we show that the coexpression of cytochrome b 5 results in stabilization of CYP3A4 during substrate turnover. MATERIALS AND METHODS Unless otherwise stated, all chemicals were ANALAR grade and were obtained from Merck (Poole, Dorset, UK) and restriction enzymes and DNA modification enzymes were obtained from Promega Corporation (Southampton, U.K.) or GIBCO BRL (Paisley, U.K.). Isolation and manipulation of human cytochrome b 5 cDNA. The cDNA coding for human cytochrome b 5 was isolated from 5 mg total human liver RNA by reverse transcription polymerase chain reaction (RT-PCR) using the following oligonucleotides as primers: (forward) GGGAATTCCATATGGCTCACCATCATCACGCCATGGCAGAGCAGTCGGAC, (reverse) GCTCTAGATCAGTCCTCTGCCATGTATAGGC. The forward, 59 PCR primer used to amplify the cDNA contained extra 59 sequence to introduce four histidine residues at the N terminal of the recombinant protein to facilitate its purification and an NdeI restriction site at the ATG to ease subsequent cloning steps. The reverse primer introduced an XbaI site after the coding region to facilitate the cloning of the cDNA into pCW. The sequence in the primers specific for cytochrome b 5 is identical to the sequence in the EMBL database, Accession No. M22865. Reverse transcription was carried out using the random hexamer mix supplied with the first-strand synthesis kit (Pharmacia Biotech, Uppsala, Sweden) as primers. The entire first-strand synthesis reaction was used as template for the specific amplification of the cytochrome b 5 cDNA. The cDNA was cloned into the vector pT7 Blue (Novagen, Madison, WI) and sequenced using the Sequenase v2.0 system (Amersham International, Amersham, Bucks, UK) to check for PCRintroduced errors. The resultant coding sequence was excised from the PCR cloning plasmid and cloned into the expression plasmid pCW using the unique NdeI/XbaI sites to give the expression construct pHcb5 (Fig. 1). The expression cassette including the promoter and transcription terminator was then excised from pHcb5 using BclI/BglII and cloned into the plasmid pACYC184 (New England Biolabs, Beverly, MA) at the unique BamHI site to give pHcocb5 (Fig 1) to allow coexpression of the human cytochrome b 5 with the CYP3A4 and NADPH P450 reductase which were expressed from a pCW-based vector, pB216, as described in (24) (Fig. 1). Expression of recombinant human cytochrome b 5 alone and together with CYP3A4 and NADPH P450 reductase. Expression of the recombinant proteins was carried out as described previously (24). Briefly, 100 ml of Terrific broth was inoculated with 1 ml of an overnight culture of E. coli JM109 containing the relevant plasmids. Following growth at 30°C with shaking at 200 rpm to an OD 600 of
118
VOICE ET AL.
FIG. 1. Plasmids used for the expression of recombinant CYP3A4, NADPH P450 reductase, and cytochrome b 5 in E. coli. The plasmids were constructed as described under Materials and Methods. pHcb5 was used for the expression of cytochrome b 5 alone and pHcob5 was used to coexpress cytochrome b 5 with CYP3A4 and NADPH P450 reductase.
between 0.6 and 0.9, the cultures were induced by the addition of IPTG to 1 mM and d-aminolevulinic acid to 0.5 mM. After 20 h the cells were harvested and resuspended in 1/20th culture volume of a solution comprising 100 mM Tris–acetate, 500 mM sucrose, 0.5 mM EDTA, pH 7.6 (2X TSE). The suspension was then diluted with an equal volume of distilled water. The P450 content of the cell suspension was determined using Fe 21–CO versus Fe 21 difference spectra as described in (24, 25) and the cytochrome b 5 content was determined using reduced versus oxidized difference spectra (10). Cell membranes were prepared from the bacteria by sonication followed by differential centrifugation as (26) and the CYP3A4 and cytochrome b 5 content was determined as described previously. The level of NADPH P450 reductase in the cell membranes was estimated using a spectrophotometric assay to determine the rate of NADPH-dependent reduction of cytochrome c (27). The cytochrome c reductase activity in whole cells was not determined due to the high background endogenous activity. The protein content of membrane preparations was determined by the method of (28) using reagent supplied by Bio-Rad (Hemel Hempsted, Herts). Purification of recombinant human cytochrome b 5. The recombinant protein was purified from bacterial membranes by solubilization in 1 mg Emulgen 911/ml for 1 h at 4°C followed by clarification by centrifugation at 100,000g. The solubilized cytochrome b 5 in the 100,000g supernatant was purified over a nickel Sepharose column (Pharmacia Biotech) in 20 mM Tris–Cl, pH 7.4, 500 mM KCl, 20% glycerol and eluted using 50 mM imidazole. Determination of nifedipine and testosterone turnover by CYP3A4 in osmotically shocked cells and cell membranes. For the determination of enzyme activity of whole cells, the cell suspension was diluted in 1X TSE to give a CYP3A4 level of 100 pmol/ml in a conical flask. The suspension was allowed to equilibrate to 37°C before either testosterone or nifedipine was added to a final concentration of 200 mM. The cells were incubated for 2 min at 37°C after which a 200-ml aliquot was withdrawn into 100 ml ice-cold methanol with 5 ml 60% perchloric acid in water. The precipitated protein was removed after a 10-min incubation on ice by centrifugation and the substrate and metabolite in the supernatant were separated by reverse-phase HPLC as described in (24). Substrate turnover by bacterial membranes was determined as in
(24). Membranes containing 10 pmol CYP3A4 were diluted to 160 ml with 50 mM Hepes, pH 7.4 (for testosterone metabolism), or 100 mM potassium phosphate, pH 7.9. Where appropriate, sufficient MgCl 2 was added to give a final concentration in the assay of 30 mM. Substrate was added from concentrated stock to give a final concentration of 200 mM for nifedipine or 100 mM for testosterone in the assay. The mixture was equilibrated at 37°C, after which the assay was started by the addition of 40 ml 5X NADPH-generating system (5 mM NADP, 25 mM glucose 6-phosphate, 1 unit glucose-6-phosphate dehydrogenase in assay buffer). After 10 min, the assay was stopped by the addition of 100 ml ice-cold methanol and 5 ml 60% perchloric acid in water. Following a 10-min incubation on ice, the precipitated proteins were removed by centrifugation. The substrate and metabolite in the supernatant were separated by reverse-phase HPLC (24). Immunoblotting. Immunoblotting was carried out as described in (29). Proteins were separated on 9% SDS–polyacrylamide gels when probing for CYP3A4 and P450 reductase and 18% gels when probing for cytochrome b 5. The blots were developed using polyclonal antibodies raised in rabbits against recombinant human CYP3A4 or purified human NADPH P450 reductase or cytochrome b 5 followed by horseradish-peroxidase linked donkey anti-rabbit immunoglobulin (SAPU). The bands were visualized using enhanced chemiluminescence (ECL, Amersham International, Amersham, Bucks).
RESULTS
Cloning and Expression of Human Cytochrome b 5 in E. coli The cDNA coding for cytochrome b 5 was isolated by RT-PCR and found to be 100% identical to the sequence in the EMBL database (Accession No. M22865) (30). The cDNA was then cloned into the vector pCW, giving the plasmid pHcb5 which was used for the expression of recombinant cytochrome b 5 (Fig. 1). The yield of recombinant cytochrome b 5 was 1250 nmol/liter culture (Fig. 2). When cells lacking the expression vec-
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
119
FIG. 2. Analysis of E. coli JM109 transformed with pHcb5. (A) The level of cytochrome b 5 was determined from reduced versus oxidized membrane spectra as described. Control cells were E. coli JM109 that had no pHcb5 (i.e., no expression plasmid). (B) Western blot analysis was carried out using an 18% polyacrylamide gel. Polyclonal antibodies raised in rabbits against human cytochrome b 5 were used as primary antibodies. Preparations were derived either from control cells (C) or from cells grown in the presence of IPTG (I).
tor were analyzed, they were found to have a background spectrum both in whole cells and in cell membranes, presumably due to endogenous bacterial heme proteins (Fig. 2). This value has been subtracted from all quoted recombinant cytochrome b 5 levels. Bacterial membranes isolated from E. coli expressing recombinant human cytochrome b 5 contained 3400 pmol cytochrome b 5/mg protein, representing an 87% recovery of the total expressed recombinant protein from whole cells. Western blotting confirmed the presence of an immunoreactive protein in induced cell membranes of the correct size (Fig. 2) (there is a nonspecific band seen in human liver at a much higher molecular weight). Following purification over nickel agarose, the ability of the recombinant human cytochrome b 5 to increase the turnover of CYP3A4 in a reconstituted system was confirmed. The reconstitution conditions were as described in (7). The inclusion of 20 pmol cytochrome b 5 in a reaction mixture containing 10 pmol CYP3A4 and 30 pmol NADPH P450 reductase resulted in the turnover of nifedipine being increased from 2.1 to 12.1
min 21, demonstrating that the recombinant human cytochrome b 5 was capable of interacting functionally with P450s. Coexpression of Human Cytochrome b 5 with CYP3A4 and NADPH P450 Reductase The cytochrome b 5 cDNA, promoter, and transcription terminator from pHcb5 were then cloned into the vector pACYC184, giving pHcob5 (Fig. 1). This plasmid has a different selection marker (chloramphenicol) and an alternative origin of replication to allow cotransformation of cells with expression plasmids encoding human CYP3A4, NADPH P450 reductase (on pCW), and cytochrome b 5. Interaction of cytochrome b 5 with CYP3A4 was studied using two strategies involving different N-terminal modifications of CYP3A4 for expression. In one strategy, CYP3A4 was modified as described in (24) by altering the 59 coding region yielding 17a3A4, the most commonly used form of recombinant CYP3A4. The second strategy used bacterial
120
VOICE ET AL. TABLE I
Yield of CYP3A4, P450 Reductase, and Cytochrome b 5 in Whole Cells and Bacterial Membranes Whole Cells Strain expressing 17aCYP3A4
Proteins expressed CYP3A4 1 P450 reductase CYP3A4, P450 reductase, b 5
Strain expressing ompA(12) CYP3A4
P450 (nmol/liter culture)
Reductase
Cytochrome b 5 (nmol/liter culture)
P450 (nmol/liter culture)
Reductase
Cytochrome b 5 (nmol/liter culture)
140 6 13
na a
na
162 6 17
na
na
na
405 6 54
na
807 6 163
82 6 14*
81 6 10*
Cell Membranes
CYP3A4 1 P450 reductase CYP3A4, P450 reductase, b 5
P450 (pmol/mg protein)
Reductase (pmol/mg protein)
b5 (pmol/mg protein)
P450 (pmol/mg protein)
Reductase (pmol/mg protein)
b5 (pmol/mg protein)
146 6 20
180 6 18
na
185 6 29
323 6 34
na
227 6 20
328 6 65
361 6 117
557 6 123
86 6 18*
86 6 13*
Note. P450 levels were determined spectrophotometrically from the Fe 21–CO versus Fe 21 difference spectra. Cytochrome b 5 levels were determined spectrophotometrically from reduced versus oxidized difference spectra. Values are the means of at least five independent expression experiments 6 SEM. a Not applicable. * P , 0.05 comparing values for 6cytochrome b 5.
ompA leader sequence to optimize expression (31) (M. Pritchard et al., manuscript in preparation). This strategy, where the DNA coding for the ompA leader is fused to the 59 end of the CYP3A4 cDNA and the protein is subsequently posttranslationally modified, leaves the N-terminal amino acid sequence of the protein unaltered except for the addition of two amino acids. On induction with IPTG for 20 h, all three proteins were expressed at high levels. The level of cytochrome b 5 was higher in cells coexpressing the ompA(12) CYP3A4 with a yield of 807 6 163 nmol/liter in whole cells compared with 405 6 54 nmol/liter when it was coexpressed with the 17aCYP3A4 construct. In membranes, the difference remained, with a yield of 557 6 123 pmol/mg cytochrome b 5 in membranes carrying the ompA(12) CYP3A4 protein compared with 328 6 65 pmol/mg when it was coexpressed with the 17aCYP3A4 construct. Although the yield of CYP3A4 was variable, in cells coexpressing cytochrome b 5, it was found, in general, to be 50% of that in cells expressing CYP3A4 and NADPH P450 reductase alone (Table I). This reduction in yield was found to carry through to membrane preparations from cells coexpressing cytochrome b 5 and was seen with both derivatives of CYP3A4 (Table I). The level of NADPH P450 reductase in bacterial membranes as measured by the rate of NADPH-dependent reduction of cytochrome c
was largely unaffected by the coexpression of cytochrome b 5. The 17aCYP3A4 construct exhibited a rate of cytochrome c reduction of 540 6 54 nmol/min/mg protein, and the ompA(12) construct, a rate of 970 6 101 nmol/min/mg protein in the absence of cytochrome b 5. These values rose very slightly to 681 6 60 and 1084 6 135 nmol/min/mg protein, respectively, on coexpression of cytochrome b 5. Visual inspection of a Western blot of bacterial membranes prepared from induced cells (Fig. 3) demonstrated that there was no obvious reduction in the level of CYP3A4 polypeptide on coexpression of cytochrome b 5, suggesting that the reduction in yield of spectrally active CYP3A4 may be due to heme depletion; however, increasing the concentration of d-aminolevulinic acid in the culture medium had no effect on the level of spectrally active CYP3A4 (data not shown). Effect of Cytochrome b 5 on the Turnover of Nifedipine and Testosterone by CYP3A4 In osmotically shocked bacteria, the turnover of both nifedipine and testosterone by ompA(12) CYP3A4 was higher than that seen with the 17aCYP3A4 (Table II) as previously demonstrated (M. Pritchard et al., manuscript in preparation). The coexpression of cytochrome b 5 resulted in significantly increased turnover of both
121
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
FIG. 3. Western blot analysis of E. coli expressing recombinant CYP3A4, NADPH P450 reductase, and cytochrome b 5. Ten micrograms of bacterial membrane proteins from E. coli JM109 expressing recombinant CYP3A4, NADPH P450 reductase 6 cytochrome b 5 or human liver microsomes was loaded on a 9% polyacrylamide gel (18% for analysis of cytochrome b 5). The proteins were detected using polyclonal antibodies raised in rabbits against the proteins indicated and horseradish peroxidase-linked donkey anti-rabbit IgG followed by enhanced chemiluminescence.
substrates with both forms of CYP3A4 (Table II), showing that it interacts equally well with the ompA(12) CYP3A4 and the most commonly used form(17aCYP3A4). In bacterial membranes, the presence of cytochrome b 5 had a similar effect, significantly increasing substrate turnover (Table III) by both forms of CYP3A4. It has been demonstrated previously that the inclusion of magnesium in the assay buffer can enhance the activity of CYP3A4 (16, 24). To determine whether magnesium and cytochrome b 5 act through the same mechanism, we assayed the activity of CYP3A4 6 cytochrome b 5 in the presence of 30 mM MgCl 2. MgCl 2 stimulated TABLE II
the activity of the system by a similar amount regardless of the presence of cytochrome b 5, suggesting that cytochrome b 5 and MgCl 2 act independently. Effects of Cytochrome b 5 on Stability of CYP3A4 Incubation of intact cells expressing CYP3A4 and NADPH P450 reductase in the presence of 200 mM testosterone at 37°C led to a significant reduction in the amount of spectrally active CYP3A4 (Fig. 4). How-
TABLE III
Turnover of Nifedipine and Testosterone by Bacterial Membranes Containing Recombinant CYP3A4 and NADPH P450 Reductase 6 Cytochrome b 5
Turnover of Testosterone and Nifedipine by Whole Cells Expressing CYP3A4 and NADPH P450 Reductase 6 Cytochrome b 5
Nifedipine oxidase (min 21) Control 1 Cytochrome b 5 Testosterone-6b-hydroxylase (min 21) Control 1 cytochrome b 5
17aCYP3A4
ompA(12) CYP3A4
9.4 1 0.7 17.4 6 2.5*
34.2 6 3.2 59.8 6 10.2*
17.8 6 1.7 25.3 6 2.5*
43.2 6 2.9 71.8 6 4.9*
Note. Turnover numbers are expressed as picomoles of product formed (nifedipine oxide and 6b-hyroxytestosterone, respectively) per picomole of CYP3A4 per minute at 37°C during a 2-min incubation. Values are the means from at least six independent expression experiments 6 SEM. * P , 0.05 comparing values for 6cytochrome b 5.
Nifedipine oxidase (min 21) Control 1 Cytochrome b 5 1 30 mM MgCl 2 1 Cytochrome b 5, 30 mM MgCl 2 Testosterone-6b-hydroxylase (min 21) Control 1 Cytochrome b 5 1 30 mM MgCl 2 1 Cytochrome b 5, 30 mM MgCl 2
17aCYP3A4
ompA(12) CYP3A4
25.7 6 1.6 46.3 6 5.2* 37.4 6 2.8 62.6 6 7.6*
43.4 6 4.3 71.5 6 2.7* 49.4 6 3.3 84.7 6 4.8*
2.1 6 0.2 4.9 6 0.7* 24.6 6 1.9 40.2 6 1.4*
2.1 6 0.4 3.5 6 0.3* 32.5 6 2.7 44.3 6 1.3*
Note. Turnover numbers are expressed as picomoles of product formed (oxidized nifedipine and 6b-hydroxytestosterone, respectively) per picomole of CYP3A4 per minute at 37°C during a 10-min assay. Values are the means from determinations of turnover in at least six independent membrane preparations each carried out in duplicate. * P , 0.02 comparing values for 6cytochrome b 5.
122
VOICE ET AL.
FIG. 4. Effect of cytochrome b 5 on the stability of spectrally active CYP3A4 in whole cells during substrate turnover in whole cells. Cells were diluted in 1X TSE to give 100 pmol/ml CYP3A4 in 10-ml final volume. The suspension was allowed to equilibrate to 37°C after which testosterone was added to 200 mM final concentration. Twomilliliter aliquots were withdrawn 0, 15, and 30 min after the addition of testosterone and the remaining spectrally active CYP3A4 was determined by Fe 21–CO versus Fe 21 difference spectra. h, Control, no substrate; ■, control, 200 mM testosterone; E, cytochrome b 5, no substrate; F, cytochrome b 5, 200 mM testosterone. Two-hundredmicroliter aliquots were withdrawn 2 min after the addition of testosterone to confirm substrate turnover by HPLC analysis. In addition, 5-ml aliquots were withdrawn at 30 min for analysis by SDS– polyacrylamide gel electrophoresis. The turnover of testosterone was 36 6 0.8 min 21 in control cells and 59 6 6 in cells containing cytochrome b 5. Values are expressed as the percentage CYP3A4 remaining and are the means of at least four incubations using independent expression cultures. *P , 0.05 comparing cytochrome b 5 coexpression 6 substrate, **P , 0.02 comparing values for 6cytochrome b 5.
ever, in cells expressing CYP3A4, NADPH P450 reductase, and cytochrome b 5, the level of spectrally active CYP3A4 remained unchanged, even after 30 min of incubation at 37°C in the presence of 200 mM testosterone (Fig. 4). Conversely, the presence of cytochrome b 5 caused a slight but statistically significant loss of spectrally active CYP3A4 when the cells were incubated in the absence of substrate (Fig. 4). Western blotting demonstrated that the loss of spectrally active CYP3A4 was not accompanied by a concomitant loss of immunologically reactive protein (Fig. 5A). In all incubations, the levels of spectrally active cytochrome b 5 were unchanged (data not shown). Incubation of bacterial membranes containing CYP3A4 and NADPH P450 reductase with and without cytochrome b 5 in the presence of substrate at 37°C showed that the protective effect of cytochrome b 5 was confined to whole-cell incubations. The loss of spectrally active CYP3A4 was similar in membranes with CYP3A4 and reductase and membranes with CYP3A4, P450 reductase, and cytochrome b 5, with around 20%
of the original level of spectrally active CYP3A4 remaining after 20 min (Table IV). Loss of spectrally active CYP3A4 from membranes in the presence of cytochrome b 5 and NADPH but in the absence of substrate was significantly increased compared with membranes with CYP3A4 and P450 reductase alone, with only 31% remaining after 20 min (Table IV). This was significantly less than in membranes lacking cytochrome b 5 which had 66% of the original spectrally active CYP3A4 remaining after 20 min. Western blotting demonstrated that the levels of immunoreactive CYP3A4 were not affected by the incubations (Fig. 5). As with the whole-cell experiments, the levels of spectrally active cytochrome b 5 were unchanged during the incubations (data not shown). As equal amounts of P450 were loaded on the gel, there is apparently more CYP3A4 in membranes from cells coexpressing cytochrome b 5 which have a higher level of apo-CYP3A4. DISCUSSION
Here, we have shown that when CYP3A4 and NADPH P450 reductase associate in an in vivo-like system the coexpression of cytochrome b 5 leads to increased turnover of both testosterone and nifedipine by CYP3A4 when the activity of the system is assayed in whole cells (Table I) or in vitro (Table II). This clearly demonstrates that cytochrome b 5 can modulate the activity of the P450s in an in vivo-like system where the possible confounding effects associated with reconstitution are absent. The action of cytochrome b 5 on the ompA(12) CYP3A4 demonstrates that it interacts on native P450s as well as the more commonly used 17aCYP3A4. Our data on the action of cytochrome b 5 contrast with a report (21) that CYP3A4, NADPH P450 reductase, and cytochrome b 5 were coexpressed in insect cells and the inclusion of cytochrome b 5 had no significant effect on the turnover of testosterone. There was, however, a low but significant amount of endogenous cytochrome b 5 in the cells used. The molar ratio of endogenous cytochrome b 5:CYP3A4 was around 0.3, a level that has since been shown in vitro (8) to fully activate testosterone turnover by CYP3A4. In our system, although there was a background heme spectrum in E. coli (Fig. 2), there was no cytochrome b 5-like protein detectable by Western blotting (Fig. 2), allowing us to determine the action of cytochrome b 5 in the absence of endogenous protein. It has been postulated that there may be a bacterial protein able to substitute for mammalian cytochrome b 5 (24). This was based on the high substrate turnover achieved when CYP3A4 and NADPH P450 reductase were coexpressed as compared with reconstituted systems in the absence of cytochrome b 5. Here we have shown that the coexpression of cytochrome b 5 with CYP3A4 and NADPH P450 reductase results in the
123
COEXPRESSION OF CYP3A4, NADPH P450 REDUCTASE, AND CYTOCHROME b 5 IN E. coli
FIG. 5. Western blot analysis of CYP3A4 in whole cells and bacterial membranes following incubation with substrate and cofactor. (A) Aliquots of cell suspension 6 coexpressed cytochrome b 5 and 6 testosterone were taken at the times shown from the incubations described in Fig. 4. A volume containing the equivalent of 0.2 pmol CYP3A4 at the start of the incubation was loaded onto a 9% polyacrylamide gel which was blotted. (B) Aliquots of incubations described in Table IV were taken after 20 min incubation. A volume containing the equivalent of 1 pmol CYP3A4 at the start of the incubation was loaded onto a 9% polyacrylamide gel which was Western blotted. The blots were developed using polyclonal antibodies raised in rabbits against purified recombinant CYP3A4 and horseradish peroxidase-linked donkey anti-rabbit IgG.
further activation of recombinant CYP3A4, indicating the absence of any bacterial protein fully able to substitute for cytochrome b 5. There is evidence that the P450:NADPH P450 reductase ratio can affect the rate of substrate turnover (17, 20). Although in our system the coexpression of cytochrome b 5 with CYP3A4 and NADPH P450 reductase resulted in an increased reductase:P450 ratio, the expression levels of CYP3A4 were sufficiently variable for the ratios to overlap. This meant that the P450:reductase ratio was similar in cultures containing low CYP3A4 levels in the absence of cytochrome b 5 compared with cultures containing high CYP3A4 levels in the presence of cytochrome b 5. In these cases, the inclusion of cytochrome b 5 still significantly stimulated testosterone and nifedipine turnover, demonstrating that the increased activity was not due to altered CYP3A4:NADPH P450 reductase ratios. We have also shown that, in addition to stimulating CYP3A4 activity, the presence of cytochrome b 5 prevents the loss of spectrally active CYP3A4 seen during the turnover of substrate in whole cells (Fig. 5). This could be a result of cytochrome b 5 improving the coupling between CYP3A4 and NADPH P450 reductase, reducing the levels of damaging reactive oxygen spe-
TABLE IV
Stability of CYP3A4 in Bacterial Membranes during Substrate Turnover % Spectrally active CYP3A4 remaining Incubation mixture
Control
1 cytochrome b 5
200 mM Testosterone NADPH-generating system 200 mM Testosterone 1 NADPH-generating system
105 6 6 66 6 6
101 6 1 31 6 1*
26 6 4
16 6 1
Note. Membranes were diluted in 50 mM Hepes, pH 7.4, 30 mM MgCl 2 6 250 mM testosterone to give 125 pmol/ml CYP3A4 in 1.76-ml final volume. The suspension was allowed to equilibrate to 37°C after which either 0.44 ml 50 mM Hepes, pH 7.4, 30 mM MgCl 2, or 0.44 ml 5X NADPH-generating system containing 30 mM MgCl 2 was added. After 20 min incubation at 37°C 200 ml was withdrawn to determine substrate turnover where appropriate. The turnover of testosterone was 23 6 0.5 min 21 in control membranes and 37 6 6 min 21 in membranes containing cytochrome b 5. The amount of spectrally active CYP3A4 in the remaining 2 ml was determined spectrophotometrically from the Fe 21–CO versus Fe 21 difference spectra. Values are expressed as the percentage of spectrally active CYP3A4 remaining after 20 min and are the means of incubations using at least three independent bacterial membrane preparations. * P , 0.02 comparing values for 6cytochrome b 5.
124
VOICE ET AL.
cies in the cells. Indeed, cytochrome b 5 has been shown in vitro to improve the coupling between a number of rat liver P450s and NADPH P450 reductase, with a concomitant reduction in H 2O 2 production (10, 13, 32). The protective effect was not observed with bacterial membranes (Table IV) which may be due to the loss of cytoplasmic components during the preparation of membranes. The presence of cytochrome b 5 led to an increased rate of loss of spectrally active CYP3A4 when either cells or membranes were incubated in the absence of substrate but in the presence of NADPH (endogenous in the case of cells). It is possible that, due to the increased coupling, in the absence of substrate, the CYP3A4 is involved in a futile cycle, generating reactive species that destroy the holoenzyme. In conclusion, we have shown that, using a system that allows CYP3A4, NADPH P450 reductase, and cytochrome b 5 to associate in an in vivo-like system, cytochrome b 5 increases the rate of turnover of testosterone and nifedipine by CYP3A4 and, in addition, protects CYP3A4 from damage during substrate turnover in whole cells. REFERENCES 1. Guengerich, F. P., and Parikh, A. (1997) Curr. Opin. Biotechnol. 8, 623– 628. 2. Friedberg, T., and Wolf, C. R. (1996) Adv. Drug Delivery Rev. 22, 187–213. 3. Smith, G. C. M., Tew, D. G., and Wolf, C. R. (1994) Proc. Natl. Acad. Sci. USA 91, 8710 – 8714. 4. Gillam, E. M. J., Guo, Z., Ueng, Y.-F., Yamazaki, H., Cock, I., Reilly, P. E. B., Hooper, W. D., and Guengerich, F. P. (1995) Arch. Biochem. Biophys. 317, 5. Shaw, P. M., Hosea, N. A., Thompson, D. V., Lenius, J. M., and Guengerich, F. P. (1997) Arch. Biochem. Biophys. 348, 107–115. 6. Imaoka, S., Imai, Y., Shimada, T., and Funae, Y. (1992) Biochemistry 31, 6063– 6069. 7. Yamazaki, H., Nakano, M., Imai, Y., Ueng, Y., Guengerich, F. P., and Shimada, T. (1996) Arch. Biochem. Biophys. 325, 174 –182. 8. Yamazaki, H., Johnson, W. W., Ueng, Y.-F., Shimada, T., and Guengerich, F. P. (1996) J. Biol. Chem. 271, 27438 –27444. 9. Jansson, I. J., Tamburini, P. P., Favreau, L. V., and Schenkman, J. B. (1985) Drug Metab. Dispos. 13, 453– 458. 10. Holmans, P. L., Shet, M. S., Martin-Wixtrom, C. A., Fisher, C. W., and Estabrook, R. W. (1994) Arch. Biochem. Biophys. 312, 554 –565.
11. Yamazaki, H., Gillam, E. M. J., Dong, M.-S., Johnson, W. W., Guengerich, F. P., and Shimada, T. (1997) Arch. Biochem. Biophys. 342, 329 –337. 12. Patten, C. J., and Koch, P. (1995) Arch. Biochem. Biophys. 317, 504 –513. 13. Gorsky, L. D., and Coon, M. J. (1986) Drug Metab. Dispos. 14, 89 –96. 14. Bell, L. C., and Guengerich, F. P. (1997) J. Biol. Chem. 272, 29643–29651. 15. Forrester, L. M., Henderson, C. J., Glancey, M. J., Back, D. J., Park, B. K., Ball, S. E., Kitteringham, N. R., McLaren, A. W., Miles, J. S., Skett, P., and Wolf, C. R. (1992) Biochem. J. 281, 359 –368. 16. Yamazaki, H., Ueng, Y.-F., Shimada, T., and Guengerich, F. P. (1995) Biochemistry 34, 8380 – 8389. 17. Morgan, E. T., and Coon, M. J. (1984) Drug Metab. Dispos. 12, 358 –364. 18. Auchus, R. J., Lee, T. C., and Miller, W. L. (1998) J. Biol. Chem. 273, 3158 –3165. 19. Sakai, Y., Yanase, T., Hara, T., Takayanagi, R., Haji, M., and Nawata, H. (1994) Clin. Endocrinol. 40, 205–209. 20. Yamazaki, H., Nakano, M., Gillam, E. M. J., Bell, L. C., Guengerich, F. P., and Shimada, T. (1996) Biochem. Pharmacol. 52, 301–309. 21. Lee, C. A., Kadwell, S. H., Kost, T. H., and Serabjit-Singh, C. J. (1995) Arch. Biochem. Biophys. 319, 157–167. 22. Aoyama, T., Nagata, K., Yamazoe, Y., Kato, R., Matsunaga, E., Gelboin, H. V., and Gonzalez, F. J. (1990) Proc. Natl. Acad. Sci. USA 87, 5425–5429. 23. Truan, G., Cullin, C., Reisdorf, P., Urban, P., and Pompon, D. (1993) Gene 125, 49 –55. 24. Blake, J. A. R., Pritchard, M., Ding, S., Smith, G. C. M., Burchell, B., Wolf, C. R., and Friedberg, T. (1996) FEBS Lett. 397, 210 –214. 25. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2379 –2385. 26. Renaud, J. P., Peyronneau, M. A., Urban, P., Truan, G., Cullin, C., Pompon, D., Beaune, P., and Mansuy, D. (1993) Toxicology 82, 39 –52. 27. Strobel, H. W., and Dignam, D. J. (1978) Methods Enzymol. 52, 89 –96. 28. Bradford, M. M. (1976) Anal. Biochem. 72, 248 –254. 29. Steinberg, P., Lafranconi, M., Wolf, C. R., Waxman, D. J., Oesch, F., and Friedberg, T. (1987) Mol. Pharmacol. 32, 463– 470. 30. Yoo, M., and Steggles, A. W. (1988) Biochem. Biophys. Res. Commun. 156, 576 –580. 31. Pritchard, M. P., Ossetian, R., Li, D. N., Henderson, C. J., Burchell, B., Wolf, C. R., and Friedberg, T. (1997) Arch. Biochem. Biophys. 345, 342–354. 32. Jansson, I., and Schenkman, J. B. (1987) Drug Metab. Dispos. 15, 344 –348.