Abstract
Variation in CYP2A6 levels and activity can be attributed to genetic polymorphism and, thus, functional characterization of allelic variants is necessary to define the importance of CYP2A6 polymorphism in humans. The aim of the present study was to investigate the reported alleles CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22, in terms of the functional consequences of their mutations on the enzyme catalytic activity. With use of the wild-type CYP2A6 cDNA as template, site-directed mutagenesis was performed to introduce nucleotide changes encoding K194E substitution in CYP2A6*15, R203S substitution in CYP2A6*16, K476R substitution in CYP2A6*21, and concurrent D158E and L160I substitutions in CYP2A6*22. Upon sequence verification, the CYP2A6 wild-type and mutant constructs were individually coexpressed with NADPH-cytochrome P450 reductase in Escherichia coli. A kinetic study using a coumarin 7-hydroxylase assay indicated that CYP2A6*15 exhibited higher Vmax than the wild type, whereas all mutant constructs, except for variant CYP2A6*16, exhibited higher Km values. Analysis of the Vmax/Km ratio revealed that all mutants demonstrated 0.85- to 1.05-fold differences from the wild type, with the exception of variant CYP2A6*22, which only portrayed 39% of the wild-type intrinsic clearance. These data suggested that individuals carrying the CYP2A6*22 allele are likely to have lower metabolism of CYP2A6 substrate than individuals expressing CYP2A6*15, CYP2A6*16, CYP2A6*21, and the wild type.
Human CYP2A6 remains one of the less well characterized members among the many known isoforms of human cytochromes P450 (Pelkonen et al., 2000). To date, it has been recognized as the major isoform involved specifically in the oxidative metabolism of nicotine, a major constituent in tobacco smoke (Rossini et al., 2008). CYP2A6 also significantly contributes to the catalytic metabolism of clinically used drugs such as methoxyflurane, halothane, losigamone, letrozole, valproic acid, disulfiram, and fadrozole and activates some procarcinogens such as 4-methylnitrosoamino-1-(3-pyridyl)-1-butanone and N-nitrosodiethylamine (Oscarson, 2001). Early studies of hepatic coumarin 7-hydroxylase activity have indicated pronounced interindividual differences in CYP2A6 expression levels and activity (Pelkonen et al., 1985; Camus et al., 1993; Shimada et al., 1996). High variations were detected with some livers found to be completely lacking the enzyme (Shimada et al., 1996). This variability in CYP2A6 shows important ethnic differences, with only 1% of whites being reported as poor metabolizers, but up to 20% of Asians (Rautio et al., 1992; Shimada et al., 1996; Ingelman-Sundberg et al., 2007). Thus, unequivocal elucidation of all functional alleles and global genotyping for CYP2A6 are vital because of the distinctive role played by the isoform in the metabolism of various substrates, especially pharmacologically and toxicologically relevant compounds. In tandem with nicotine and other tobacco-specific carcinogens being established as high-affinity substrates for CYP2A6, much attention has been focused on the toxicological and clinical significance of this highly polymorphic isoform in humans.
To date, 33 allelic variants, designated as CYP2A6*2 to CYP2A6*37, have been identified (the list of allelic variants is available at http://www.imm.ki.se/CYPalleles/). Among the first few isolated allele, CYP2A6*2 has been reported to encode a protein with an L160H substitution that renders the enzyme inactive (Hadidi et al., 1997). CYP2A6*3, which was initially reported as a hybrid allele of multiple gene conversions with a pseudogene, CYP2A7, is now believed to be an artifact because of the shortcoming of the previously used PCR-based genotyping assay for CYP2A6*1, CYP2A6*2, and CYP2A6*3 (Oscarson et al., 1999). Alleles CYP2A6*4, CYP2A6*5, and CYP2A6*20 were also found to display abolished functional activity due to gene deletion or point mutation in their primary sequence. Twelve CYP2A6 alleles (*6, *7, *10, *11, *12, *17, *18, *19, *23, *24, *26, and *27) have been ascertained to cause reduction in enzyme activity, whereas genetic polymorphisms in the promoter region of CYP2A6 have been implicated for the decreased transcriptional activity observed in alleles CYP2A6*1D, CYP2A6*1H, and CYP2A6*9 (Ariyoshi et al., 2001; Kitagawa et al., 2001; Daigo et al., 2002; Fukami et al., 2004, 2005; Ho et al., 2008).
Functional significance of some alleles such as CYP2A6*15 (K194E), CYP2A6*16 (R203S), CYP2A6*21 (K476R), and CYP2A6*22 (D158E and L160I) has yet to be investigated in detail. The occurrence of these four alleles in the human population was first revealed in 2002 and 2005 with variant CYP2A6*15 found in the Korean and Japanese populations at frequencies of approximately 1.2 and 1.5 to 2.2%, respectively (Kiyotani et al., 2002; Nakajima et al., 2006). Conversely, the frequency of CYP2A6*16 was prominent among whites (0.3–3.6%) and African Americans (1.7%), whereas this allele remained undetected in the Asian population (Kiyotani et al., 2002; Nakajima et al., 2006). Allelic variant CYP2A6*21 also has a higher occurrence in white subjects at 0.5 to 7.0% compared with Chinese (3.4%) and black subjects (0.6%), whereas Japanese subjects remained unaffected (Haberl et al., 2005; Al Koudsi et al., 2006; Nakajima et al., 2006). Comprehensive evaluation of CYP2A6 polymorphic alleles by Nakajima et al. (2006) in four ethnic populations did not detect any occurrence of CYP2A6*22 within the populations studied. So far, the variant was only reported at low frequency (0.3%) among whites (Haberl et al., 2005). Although no data are available for CYP2A6*15 and CYP2A6*16, haplotype analyses on CYP2A6*21 and CYP2A6*22 revealed that haplotypes carrying the two alleles occurred at low frequencies (at 0.6 and 0.3%, respectively) in whites (Haberl et al., 2005). Haplotype frequencies of these four alleles in other populations as well as their phenotypic association with protein level and activity, however, remain unknown at this stage.
Despite the existence of these four CYP2A6 variants in the human populations, functional characterization of the polymorphisms manifested has yet to be determined in detail. With one or more amino acid mutations in their primary sequences, it is likely that these mutations would have a certain degree of effects on their structural stability and catalytic activity. Structural and functional characterization of these polymorphic alleles is necessary because it contributes to our better understanding of the consequences of mutation and hence aids in defining the pharmacological and toxicological importance of CYP2A6 polymorphisms in humans.
Materials and Methods
Materials.
The QuikChange site-directed mutagenesis system was purchased from Stratagene (La Jolla, CA), and endonuclease restriction enzymes were obtained from New England Biolabs (Ipswich, MA). Mouse anti-human cytochrome P450 CYP2A6 monoclonal antibody and goat anti-mouse IgG-horseradish peroxidase were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). 5-Bromo-4-chloro-3-indolyl-β-d-galactopyranoside, isopropyl-β-d-thiogalactopyranoside, and Tris base were acquired from Promega (Madison, WI). Escherichia coli DH5α competent cells, oligonucleotide primers, and Luria-Bertani and Teriffic broth media were purchased from Invitrogen (Carlsbad, CA). All other chemicals and reagents were obtained from Sigma-Aldrich (St. Louis, MO).
In Vitro Site-Directed Mutagenesis of CYP2A6 cDNA.
Site-directed mutagenesis on CYP2A6 cDNA was performed using the QuikChange site-directed mutagenesis system according to the manufacturer's instructions. Basically, in this a supercoiled double-stranded DNA vector with an insert of interest and two synthetic oligonucleotide primers containing the desired mutation are used. The DNA template was a pCW-CYP2A6 vector, which was previously constructed in our laboratory (C. E. Ong, unpublished data). It contained N-terminal sequence modification, a P450 17α-derived MALLLAVF sequence as reported by Barnes et al. (1991) and the full coding sequence of human CYP2A6. The sequence has 100% identity to that of the reported wild-type CYP2A6 in GenBank (GenBank accession number NM_000762). Mutagenic primers that were exclusively designed for generating mutant alleles of CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22 are listed in Table 1. Single nucleotide substitution of AAG to GAG, CGC to AGC, and AAA to AGA (nucleotides that were changed to make the desired mutation are shown in bold) on the CYP2A6 primary sequence had each contributed to the amino acid substitution of K194E, R203S, and K476R, respectively, in the expressed bacterial membrane fragment of CYP2A6*15, CYP2A6*16, and CYP2A6*21 proteins. Likewise, primer possessing both GAC to GAG and CTC to ATC substitutions has permitted expression of mutant CYP2A6*22 with concurrent D158E and L160I substitutions on the primary sequence. Before full nucleotide sequencing of the entire cDNA coding frame of each clone, each mutant construct was subjected to restriction analyses by several endonucleases (EcoRI, FspI, PmlI, and Hpy99I). The preliminary analysis with restriction endonucleases was necessary as an indication that desired mutations had successfully taken place on the nucleotide strand of CYP2A6 cDNA. The full nucleotide sequencing of all four mutant cDNAs was further verified by a capillary-based sequencing method, which was outsourced to AITBIOTECH Pte Ltd. (Singapore).
Heterologous Expression of CYP2A6 Wild-Type and Mutant Constructs in Bacterial Expression System.
CYP2A6 plasmid constructs harboring the desired mutations, pCW-CYP2A6*15, pCW-CYP2A6*16, pCW-CYP2A6*21, and pCW-CYP2A6*22 were individually cotransformed into E. coli DH5α cells together with the pACYC-OxR plasmid, the essential NADPH-cytochrome P450 oxidoreductase (OxR) coenzyme. CYP2A6 and OxR protein expression in bacterial cells and the subsequent membrane preparation was determine as described previously (Singh et al., 2008). The membrane fragments of E. coli were stored at −80°C in a 1:1 mixture of pH 7.6 TES buffer (100 mM Tris, 0.5 mM EDTA, and 500 mM sucrose) and ice-cold distilled water before analysis in the enzyme assay reaction.
Coumarin 7-Hydroxylase Assay.
Enzyme kinetic activities of wild-type CYP2A6 and mutants were assessed by a fluorescence-based coumarin 7-hydroxylase assay with slight modifications based a published protocol (Ghosal et al., 2003; Donato et al., 2004). Coumarin 7-hydroxylation was measured in a reaction mixture consisting of 50 μg of expressed CYP2A6, an NADPH-generating system (1.3 mM NADP, 3.5 mM glucose 6-phosphate, 2 IU of glucose-6-phosphate dehydrogenase, and 5 mM MgCl2) and 0.313 to 40 μM coumarin in 100 mM Tris-HCl buffer (pH 7.5). Coumarin was dissolved in acetonitrile with the final concentration of the organic solvent in each incubation mixture at 1% (v/v) or less. Reactions were initiated by addition of 50 μg of CYP2A6 protein after prewarming at 37°C in a metabolic shaker for 10 min and later were terminated by 50 μl of 500 mM Tris base after 25 min of incubation at 37°C. Quantification of 7-hydroxycoumarin was performed immediately by using an Infinite 200 series microplate reader (Tecan, Männerdorf, Switzerland) at the excitation wavelength of 365 nm and emission wavelength of 450 nm. Standard curves of 7-hydroxycoumarin were constructed in the range of 15.63 to 2000 pmol/well, and the metabolite formation rate was calculated based on the curves. All samples and standards were incubated in duplicate.
Kinetic Analysis.
Enzyme kinetic data were analyzed by the nonlinear least-squares regression analysis software EZ-Fit (Perrella Scientific, Anherst, NH), and the kinetic parameters, Michaelis-Menten constant (Km), and maximum velocity (Vmax) were determined over the substrate range studied. Statistical analyses were performed by using the SPSS statistical program (SPSS Inc., Chicago, IL).
SDS-Polyacrylamide Gel Electrophoresis and Immunoblotting of CYP2A6 Proteins.
In brief, bacterial membrane fractions (50 μg) expressing wild-type or mutant CYP2A6 proteins were heat-inactivated and separated on a 10% polyacrylamide gel before being transferred electrophoretically to a nitrocellulose membrane. The membrane was then treated with 5% nonfat skimmed milk in Tris-buffered saline at room temperature for 1 h. Immunoblotting was later performed using a 1:200 dilution of the mouse anti-human CYP2A6 monoclonal antibody at 37°C for 1 h, followed by incubation with a 1:3000 dilution of peroxidase-labeled goat anti-mouse IgG as the secondary antibody. Binding of the antibody to the expressed protein was detected using the 4-chloro-1-naphthol developing solution.
Other Methods.
Protein concentrations were measured according to the method of Bradford (1976). The P450 content of the membranes was determined by carbon monoxide difference spectra (Omura and Sato, 1964). The level of P450 reductase was estimated in membranes using a spectrophotometric assay to measure cytochrome c reduction (Strobel and Dignam, 1978).
Results
Site-Directed Mutagenesis of CYP2A6.
Four CYP2A6 mutant alleles had been successfully generated in this study using the QuikChange site-directed mutagenesis system. Four mutants (CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22) were constructed to harbor one or more nucleotide changes at positions as reported in the literature (Kiyotani et al., 2002; Haberl et al., 2005; Mwenifumbo et al., 2008). Isolated digestions were performed with the designated restriction endonucleases, namely EcoRI for CYP2A6*15 (Fig. 1A), FspI for CYP2A6*16 (Fig. 1B), PmlI for CYP2A6*21 (Fig. 1C), and Hpy99I for CYP2A6*22 (Fig. 1D). All restriction reactions gave the expected band patterns, indicating that the desired mutations have successfully taken place on the nucleotide strand of CYP2A6 cDNAs. Moreover, nucleotide sequencing evidently showed that all mutant constructs possessed the desired nucleotide changes, precisely AAG to GAG substitution in CYP2A6*15 (Fig. 2A), CGC to AGC substitution in CYP2A6*16 (Fig. 2B), AAA to AGA substitution in CYP2A6*21 (Fig. 2C), and both GAC to GAG and CTC to ATC substitutions in CYP2A6*22 (Fig. 2D). Observations from the electropherograms have detected no other nucleotide changes in the whole stretch of these mutant cDNAs (data not shown). Both restriction and sequencing analyses have evidently verified that the desired mutations have been successfully incorporated on the constructs at the desired positions and were ready for protein expression of the mutants and subsequent functional analyses.
Expression of CYP2A6 Proteins in E. coli.
Bacterial expressions of CYP2A6 mutant proteins were determined accordingly via SDS-polyacrylamide gel electrophoresis and immunoblotting. Immunodetection with monoclonal mouse anti-human CYP2A6 revealed the presence of a molecular mass with approximately 49 kDa in all the recombinant mutant proteins (Fig. 3, lanes *15, *16, *21, and *22), which was in accordance with the reported molecular weight of CYP2A6 (Maurice et al., 1991). The levels of CYP2A6 mutant protein expressions appeared to be comparable to that of the wild-type CYP2A6, indicating consistency in the expression of these proteins in the bacterial membrane fragments. As predicted, absence of bands was noted in the control protein expressing the original pCWori+ plasmid (Fig. 3, lane C). Cytochrome c reductase activity was determined in all mutant clones and also in the wild-type CYP2A6. The OxR protein expression level in all mutants was revealed to be insignificantly different (data not shown). Reduced CO difference spectroscopy was also performed for the four mutant CYP2A6 proteins. By performing three independent spectral determinations, we found no substantial deviations in expression levels between the wild-type and mutant proteins. Figure 4 shows the typical spectra obtained for the wild type as well as for CYP2A6*15. The other three mutants (CYP2A6*16, CYP2A6*21, and CYP2A6*22) also showed spectra of similar pattern (data not shown). These results have indicated that the amino acid exchanges had no influence on protein stability or expression level.
Kinetic Study of the CYP2A6 Wild-Type and Mutant Proteins.
Enzyme catalytic activity was determined for CYP2A6 variant proteins by a fluorescent metabolite detection assay as described previously. Each variant protein was measured at substrate concentration of 0.63 to 80.0 μM, of which kinetics for coumarin 7-hydroxylation was fitted to the Michaelis-Menten equation. Reproducibility of the data was confirmed with three independent determinations. The apparent Km values of variant CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22 are summarized in Table 2. Multiple comparisons of Km values among the wild-type CYP2A6 and variants revealed the differences to be highly significant for some mutants. Whereas no difference was observed between CYP2A6*16 and the wild type, the rest of the mutants, CYP2A6*15, CYP2A6*21, and CYP2A6*22, showed 1.5- to 2.9-fold higher values than the wild type. As for the Vmax values, all the variants appeared to show higher values than the wild type, with fold differences of 1.8, 1.05, 1.4, and 1.15 in CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22, respectively. However, except for CYP2A6*15, the differences for the mutants were not statistically different from the wild type. Analysis of Vmax/Km ratios, which represent the intrinsic clearance of the enzymes, indicated that variants CYP2A6*15, CYP2A6*16, and CYP2A6*21 exhibited Vmax/Km similar to that of the wild type (0.85- to 1.05-fold difference). CYP2A6*22 showed a significantly lower Vmax/Km (39% of the wild type), which indicates compromised catalytic activity. The lower value was mainly due to a significantly higher apparent Km value of this mutant and a much lesser degree of increase seen in the Vmax.
Discussion
Genetic polymorphism of CYP2A6 is believed to be the major cause of interindividual variation in enzymatic activity for various CYP2A6 substrates and it is thereby critical to characterize the enzymatic properties caused by the allelic polymorphism (Kitagawa et al., 2001). Therefore, construction of mutant alleles, CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22 was accomplished in the present study with the purpose of exploring the functional consequences of polymorphism in these alleles. Currently, allelic frequencies of the four alleles have been characterized in different populations, but the functional consequences of point mutations in these variants, particularly in relation to catalytic activity, have not been investigated thoroughly. A recent study on CYP2A6*21 demonstrated very little or negligible impact of the mutation of the allele on in vivo nicotine metabolism in white subjects (Al Koudsi et al., 2006). Studies on the other three alleles, either in vitro or in vivo, have generally been lacking, and further exploration is warranted.
E. coli was chosen as the expression system in the present study because it is one of the most commonly used hosts for performing structural, biophysical, and kinetic studies for P450s. A host of P450 isoforms have been heterologously expressed in E. coli for the study of structure-function relationships by site-directed mutagenesis, including those of CYP2A6 (Kim et al., 2005). It is conceivable that the bacterial system is not amenable for studying the post-translational event of the expressed protein, and this can only be investigated using other host cells such as yeast and mammalian cells. Experiments using these cell lines are currently in progress in our laboratory.
Of the four CYP2A6 mutants investigated in this study, the amino acid mutations were found not only in substrate recognition sites (SRSs) but also in other regions of the open reading frame. A single point mutation that causes the K194E substitution in variant CYP2A6*15 is located not within any particular SRS site but adjacent to helix F, which partially embraces SRS-2 (Fig. 5). A kinetics study of the K194E mutation revealed that the single nucleotide substitution affected the Km of coumarin 7-hydroxylation more than the Vmax (i.e., 2.1-fold increase in Km but only 1.8-fold increase in Vmax). However, catalytic efficiency of variant CYP2A6*15 was only slightly lower compared with that of the wild type (0.85-fold of the wild-type value). This result indicates that carriers of the CYP2A6*15 allele would not be expected to exhibit differences in their drug clearance for CYP2A6 substrates.
Variant CYP2A6*16, on the other hand, exhibits a single amino acid mutation, R203S, which is located within helix F and is directly positioned within the highly conserved SRS-2 of the CYP2A family (Fig. 5). Thus, it would be possible that mutation in helix F or specifically in SRS-2 may affect the folding of neighboring secondary structures of the active site, hence affecting the protein stability as well as the local heme and substrate binding. Moreover, replacement of basic Arg with a small and neutral residue Ser, a nonconservative substitution, would be expected to alter morphology of the active site. However, our results indicated that R203S did not alter substrate binding (i.e., no change in the Km) or catalytic capacity (i.e., no change in Vmax), resulting in similar catalytic efficiency for CP2A6. From a molecular modeling point of view, it has been reported that coumarin orientates for 7-hydroxylation via hydrogen-bonded contacts with Gln104 and His477 and is π stacking with Phe209 (Lewis et al., 1999). In addition, Thr212 could be involved in directing the access of coumarin to the binding site (Fukami et al., 2004). From such analyses, it is likely that the Arg203 residue does not have direct bonding contact with the coumarin molecule for 7-hydroxylation; hence, no change was observed in the catalytic behavior of this mutant.
Substitution of Lys476 with another strongly basic residue, Arg, in allele CYP2A6*21 did not affect catalytic efficiency compared with that of the wild-type enzyme (i.e., 0.91-fold change in Vmax/Km), even though the mutation is located in SRS-6 of the β-sheet 4 (Fig. 5). In other words, the subtle difference in catalytic efficiency between variant CYP2A6*21 and the wild type is not appreciable despite alteration in the highly conserved region of SRS-6. This finding seems to suggest that residue changes at the 476 site do not disrupt the conformation structure of CYP2A6*21 protein as the local residue polarity remained unchanged. Our finding seems to be consistent with data obtained by Al Koudsi et al. (2006), who concluded that CYP2A6*21 did not have a detectable impact on in vivo metabolism of nicotine, another CYP2A6 substrate.
The fourth allelic construct, CYP2A6*22, has been generated to carry two point mutations in accordance with the reported sequence in the literature. Both D158E and L160I substitutions are located in the D-helix, which appears to be exterior to the putative active site of CYP2A6 (Fig. 5). However, a mutational effect was observed in coumarin 7-hydroxylation with catalytic efficiency of CYP2A6*22 declining to 39% of that of the wild type, mainly by affecting the coumarin Km value (2.9-fold increase in Km versus 1.15-fold increase in Vmax). The effects of these substitutions suggest that structural elements outside the active site may play roles in changing the catalytic activity through a variety of changes proposed in the previous studies, such as blocking of substrate or product access channels, reductase binding, and the motion of protein during conformational changes (Kim and Guengerich, 2004a,b). The elucidation of the crystal structure of CYP2A6 (Yano et al., 2005) has given us better understanding that Asp158 and Leu160 reside on the surface of the CYP2A6 enzyme and do not make direct contact with coumarin. It is possible that mutations in helix D at positions 158 and 160 may have affected the folding of the neighboring secondary structures, hence the effect on substrate binding and catalytic efficiency. It is likely that Asp158 and Leu160 together with other residues in helix D in their native configuration are able to provide a folding motif that locks other neighboring secondary structures for proper heme and substrate binding. The three-dimensional structure of CYP2A6 as revealed by the X-ray crystallography (Yano et al., 2005) illustrates that helix D is located adjacent to many other structures that form part of the substrate access and binding sites, including the helices B, C, F, and F′ as well as the B-C loop. Consequently, alteration at the 158 and 160 positions may have disrupted this part of the protein, causing altered association and spatial repositioning of these associated structures. In other words, Asp158 and Leu160 may be involved in “long-range” interactions that are transmitted to residues in the surrounding structures that contact the heme and substrates. This type of interaction is not unique and has been reported for other P450 isoforms. For example, Glu351, a residue located in helix K of CYP21A2, has been found to be important for heme binding, even though it is located far from the heme moiety in the active site. E351K mutation introduced to the isoform was shown to result in the loss of enzyme activity toward the substrate progesterone (Krone et al., 2005). Results from this study are consistent with our findings and indicate that residues not involved in substrate and heme binding may play important functional or structural role in P450 catalysis.
In summary, kinetic data obtained from this project have indeed given us insights, particularly on the functional consequences of genetic polymorphisms in the four selected alleles (CYP2A6*15, CYP2A6*16, CYP2A6*21, and CYP2A6*22). Kinetic analyses of these polymorphic alleles of CYP2A6 indicated that point mutations harbored in these variants have encoded enzymes that were metabolically active toward coumarin oxidation, with the exception of CYP2A6*22, which has reduced but not inactive metabolic activity. Individual carriers of the homologous CYP2A6*22 allele would be expected to have decreased clearance of coumarin. However, it is not clear whether other substrates of CYP2A6 such as nicotine would also be affected by this polymorphism because it is known that catalytic activity of alleles may differ according to the substrates being investigated. Further study is currently being undertaken in our laboratory to elucidate enzyme kinetics of this variant with other substrates.
Acknowledgments.
We express our thanks and gratitude to Professor John Miners (Flinders University, Adelaide, Australia) and Professor Donald Birkett (Johnson and Johnson Research Pty. Ltd., Sydney, Australia) for their kind gifts of vectors pCWori+ and pACYC-OxR.
Footnotes
This work was supported by the Malaysia Toray Science Foundation [Science and Technology Research Grant 269817-K]; and by the International Medical University, Kuala Lumpur, Malaysia [International Medical University Research Fund IMU 112/2006].
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.109.031054.
-
ABBREVIATIONS:
- P450
- cytochrome P450
- OxR
- NADPH-cytochrome P450 oxidoreductase
- SRS
- substrate recognition site.
- Received November 13, 2009.
- Accepted February 5, 2010.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics