Heterologous expression of human cytochrome P450 ...

Research article Received: 3 September 2012,

Revised: 19 December 2012,

Accepted: 24 December 2012

Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/bmc.2872

Heterologous expression of human cytochrome P450 (CYP) 2C19 in Escherichia coli and establishment of RP-HPLC method to serve as activity marker Yan Pana*, Joon Wah Maka and Chin Eng Ongb ABSTRACT: In this study, a simple and reliable reverse-phase high-performance liquid chromatography (RP-HPLC) method was established and validated to analyze S-mephenytoin 4-hydroxylase activity of a recombinant CYP2C19 system. This system was obtained by co-expressing CYP2C19 and NADPH-CYP oxidoreductase (OxR) proteins in Escherichia coli (E. coli) cells. In addition to RP-HPLC, the expressed proteins were evaluated by immunoblotting and reduced CO difference spectral scanning. The RP-HPLC assay showed good linearity (r2 = 1.00) with 4-hydroxymephenytoin concentration from 0.100 to 50.0 mM and the limit of detection was 5.00 102 mM. Intraday and interday precisions determined were from 1.90 to 8.19% and from 2.20 to 14.9%, respectively. Recovery and accuracy of the assay were from 83.5 to 85.8% and from 95.0 to 105%. Enzyme kinetic parameters (Km, Vmax and Ki) were comparable to reported values. The presence of CYP2C19 in bacterial membranes was confirmed by immunoblotting and the characteristic absorbance peak at 450 nm was determined in the reduced CO difference spectral assay. Moreover, the activity level of co-expressed OxR was found to be comparable to that of the literature. As a conclusion, the procedures described here have generated catalytically active CYP2C19 and the RP-HPLC assay developed is able to serve as CYP2C19 activity marker for pharmacokinetic drug interaction study in vitro. Copyright © 2013 John Wiley & Sons, Ltd. Keywords: HPLC; CYP2C19; in vitro; drug interaction

Introduction Cytochrome P450 (CYP) is a superfamily of membrane-bound heme proteins with characteristic absorption peak at 450 nm when reduced by CO. CYPs function as mono-oxygenases catalyzing numerous endogenous and exogenous substances. A significant number of clinically used drugs also undergo metabolism mediated by CYPs, and mainly CYP1, 2 and 3 families are involved. Concurrent intake of CYP substrates, inhibitors or inducers increases the risk of adverse drug reactions owing to drug–drug interactions. CYP2C19 belongs to the CYP2C subfamily and is responsible for metabolizing prescribed drugs, including omeprazole (Andersson et al., 1992), proguanil (Ward et al., 1991), certain barbiturates (Adedoyin et al., 1994), citalopram (Sindrup et al., 1993) and diazepam (Bertilsson et al., 1993). This isoform receives special attention with regard to its genetic polymorphism, with ethnic differences in poor metabolizer frequency, ranging from 2–5% in Caucasians to 18–23% in Asian population (Nakamura et al., 1985; Wilkinson et al., 1989). In vitro screening of drug–drug interaction potentials is always suggested before performing in vivo tests using human subjects. One important screening parameter is modulation of CYP catalytic activity. In order to determine the alteration of CYP activity, several types of enzyme sources are employed, including human liver microsomes, liver homogenates and hepatocytes. However, problems such as short supply of the tissue sources, ethical concerns in relation to patient consents and inconsistencies of population of CYP isoforms from different donors limit their use.

Biomed. Chromatogr. (2013)

Moreover, the inherent low catalytic activity of CYP2C19 is another potential problem in using human-derived enzymes. Heterologous expression of human CYPs has become an alternative enzyme source in in vitro predictions on drug–drug interactions. Human CYPs expressed from mammalian cells (Hyland et al., 2001), yeast (Eugster et al., 1990) and bacteria (Larson et al., 1991) have demonstrated sufficient activities in in vitro studies. Among these systems, the E. coli system has become one of the popular choices owing to its lower cost of maintenance, ease of use and high yield of protein within a relatively short period of culture time. Nevertheless, to achieve a high expression level in bacterial systems, native human CYP cDNA sequences need to be modified to overcome species differences of codon preferences between mammals and bacteria (Waterman, 1993). The first catalytically active CYP17a protein was expressed in E. coli cells after N-terminal modification (Barnes et al., 1991). Subsequently

* Correspondence to: Y. Pan, School of Medical Sciences, International Medical University, 126, Jalan 19/155B, Bukit Jalil, 57000 Kuala Lumpur, Malaysia. Email: [email protected] a

School of Medical Sciences, No. 126, Jalan Jalil Perkasa 19, Bukit Jalil, 57000 Kuala Lumpur, Malaysia

b

Jeffrey Cheah School of Medicine and Health Sciences, Monash University Sunway Campus, Jalan Lagoon Selatan, 46150 Bandar Sunway, Selangor Darul Ehsan, Malaysia Abbreviation used: CYP, cytochrome P450; OxR, oxidoreductase

Copyright © 2013 John Wiley & Sons, Ltd.

Y. Pan et al. many human CYP isoforms have been expressed with the same modifications and used in in vitro assays. S-Mephenytoin is a hydantoin, used as an anticonvulsant. Its use as anticonvulsant, however, is obsolete as there are many other safer and more effective antiepileptic agents available. CYP2C19 was identified as the major CYP isoform that metabolizes S-mephenytoin 4-hydroxylation in humans; hence, S-mephenytoin has been commonly used as activity marker for CYP2C19 (Goldstein and Faletto, 1994). Here, we have described a procedure to co-express functional active human CYP2C19 and OxR from E. coli cells and the enzyme activity was characterized by a RP-HPLC assay, which was developed and validated in the present study as well.

Experimental Chemicals and materials Acetonitrile and dichloromethane (all HPLC-grade) were purchased from Fisher Scientific (Pittsburgh, PA, USA). The other chemicals [S-mephenytoin, 4-hydroxymephenytoin, phenytoin, omeprazole, b-nicotinamide adenine dinucleotide phosphate (NADP), D-glucose 6-phosphate (G-6-P), glucose6-phosphate dehydrogenase and magnesium chloride] were all purchased from Sigma (St Louis, MO, USA) with the highest grades available.

Table 1. Sequences of oligonucleotide primers used for PCR mutagenesis of the CYP2C19 N-terminus CYP form

Sequence of oligonucleotide primers

CYP2C19

Forward primer (NdeIa) M A L L 50 a gga att cat atg gct ctg tta L A V F L tta gca gtt ttt ctg tgt ctc tca tgt ttg ctt ctc 30 Reverse primer (HindIIIa) 50 agg gaa ttc aag ctt tca gac agg aat gaa gca cag ctg 30

Quantification of recombinant protein expression The CYP2C19 spectral content of the bacterial cell lysates was quantified using a dual-wavelength/double beam spectrophotometer (Shimadzu, Tokyo, Japan). This spectrophotometer also allowed us to estimate the OxR activity using cytochrome c as a substitute electron acceptor. Both experimental procedures were described in detail in our previously published paper (Pan et al., 2011).

The expression of CYP2C19 in the E. coli whole cells was demonstrated by immunoblotting analysis. CYP2C19 protein was analyzed using SDS– PAGE in 12.0% polyacrylamide gel based on Laemmli’s (1970) method. After running SDS-PAGE, the separated proteins were transferred to a nitrocellulose membrane. Primary antibody (rabbit antihuman CYP2C8/ 9/19 polyclonal antibody), which is able to bind CYP2C19 protein, was incubated with the nitrocellulose membrane followed by incubation with horseradish peroxidase-conjugated secondary antibody (polyclonal goat anti rabbit antibody). The presence of CYP protein was visualized using a chloronapthol developing solution (containing 15.0 mg of 4-cholro-1-napthol, 5.00 mL of ethanol, 50.0 mL of Tris-buffered saline and 50.0 mL of 30.0% hydrogen peroxide).

Restriction enzyme cutting sites are in bold type in the sequences.

+

Native CYP2C19 cDNA was generated by reverse-transcriptase polymeras chain reaction (RT-PCR) method and subcloned in the plasmid BluescriptW II SK + (pBS-2C19) by previous work in our laboratory (data unpublished; Pan et al., 2011). Bacterial expression vector pCWori+ and plasmid pACYC-ompA-OxR were kindly provided by Professors John Miners and Donald Birkett (Flinders University, Adelaide, Australia). 17a Sequence, MALLLAVFL, was introduced into the N-terminus of the CYP2C19 cDNA by PCR-based mutagenesis with the help of a mutagenic primer containing an NdeI site and a reverse primer completely complementary to a HindIII site located just outside CYP2C19 cDNA stop codon (Table 1). The PCR product and pCWori+ vector were digested with NdeI and HindIII and then ligated to form pCW-2C19. The general scheme is illustrated in Fig. 1. DNA sequencing of both strands at full length of the constructed pCW-2C19 was performed to ensure that no errors had been introduced during the amplification process. The verified plasmid was then co-transformed with pACYC-ompA-OxR into an E. coli DH5a bacterial cell. Successfully co-transformed cells were prepared and kept as 70.0% glycerol stocks in 80 C freezer. The cDNA for CYP2C19 were expressed in E. coli cells. Expression and purification of the proteins were carried out according to an established protocol in our laboratory (Singh et al., 2008).

Immunoblotting of expressed CYP2C19 protein

a

pBS-2C19

Construction of expression plasmids

Forward primer Reverse primer

Taq

NdeI

PCR

HindIII

PCR-2C19

PCR template

+ AmpR pCW-2C19

+

pCW ori+

AmpR

AmpR

T4 ligase

pCW ori+ HindIII

NdeI

Figure 1. General scheme illustrating the process of N-terminal modification of CYP2C19 cDNA to generate pCW-2C19 construct. The template used in the PCR was pBS-2C19 where CYP2C19 native cDNA was subcloned in the plasmid BluescriptW II SK+ by previous work (data unpublished).

wileyonlinelibrary.com/journal/bmc



Expression of CYP2C19 and establishment of RP-HPLC method RP-HPLC conditions The RP-HPLC assay for determining 4-hydroxymephenytoin was performed according to the published method with minor modifications (Ko et al., 2000) employing a Gilson HPLC system, which consisted of a Gilson 307 pump, a Gilson UV/vis-152 detector and an ODS HYPERSIL column (Thermo, 4.6 250 mm, 5 mm). The mobile phase consisted of 26.0% acetonitrile (in 0.0500 M phosphate buffer, pH 5.00) and was pumped at a flow rate of 1.50 mL/min at wavelength of 211 nm. The whole operation lasted 25 min. A sample volume of 50.0 mL of the solution was subjected to HPLC analysis.

expressed as the percentage of bias between the nominal and the calculated concentrations. Recovery. Recovery of 4-hydroxymephenytoin was determined by comparing peak area ratios of 4-hydroxymephenytoin to phenytoin from extracted incubation mixtures, with those from equivalent amounts of 4-hydroxymephenytoin spiked directly into extraction solvent (dichloromethane). The recovery was determined by comparing the calculated concentration with the known 4-hydroxymephenytoin concentration of spiked samples at concentrations of 0.500, 5.00 and 20.0 mM three times. The results were then expressed as the percentage of bias of determined concentrations between the incubation and the spiked mixtures.

Calibration curve The calibration curve was established by injecting different concentrations of 4-hydroxymephenytoin (0.100–50.0 mM) prepared from stock solution onto the HPLC system. The peak area ratios of 4-hydroxymephenytoin to phenytoin (internal standard) were calculated and used for constructing calibration curves.

Method validation The RP-HPLC assay mentioned in this study was validated for the following parameters: solution stability, specificity, linearity, detection limit, precision, accuracy and recovery. Stock and sample solution stability. The stock solutions of S-mephenytoin and 4-hydroxymephenytoin were prepared at 50.0 mM in acetonitrile. Phenytoin (internal standard) was prepared at 0.100 g/mL in methanol. All stock solutions were stored at 20 C, and were found to be stable for at least a month. Solution stability of all analytes (S-mephenytoin, 4-hydroxymephenytoin and phenytoin) after sample collection and preparation were found to be stable for up to 48 h. All our samples were analyzed by being injected onto HPLC system within 12 h of preparations on routine basis. Specificity of the assay condition. The specificity of the method developed was ascertained by comparing the separation of S-mephenytoin, 4-hydroxymephenytoin and phenytoin in sample incubation (containing CYP2C19 protein) with control incubation (containing control protein, bacterial membranes isolated from culture stock transformed with pCWori+ plasmid with no CYP2C19 cDNA). Triplicates of comparison were carried out to demonstrate the lack of endogenous interference and batch-to-batch variation. Linearity and limit of detection. Linear regression analysis was performed to determine the slope, intercept and r2 value of calibration curves obtained at seven 4-hydroxymephenytoin concentration levels (0.100–50.0 mM). The limit of detection (LOD) was estimated at a signalto-noise ratio of 3:1 by injecting a series of diluted solutions with known concentrations of 4-hydroxymephenytoin (2.00 103, 5.00 103, 1.00 102, 2.00 102 and 5.00 102 mM). Precision. Intra- and inter-day coefficients of variation (% CV) were determined by measuring the spiked 4-hydroxymephenytoin concentrations in incubation mixtures. This was performed by measuring the peak area ratios (4-hydroxymephenytoin over phenytoin), which was injected in three concentrations, namely 0.500, 5.00 and 20.0 mM. Each of these concentrations was injected in triplicates onto the system on the same day to obtain intraday precision. Interday precision determination involved injection of the same concentrations on three consecutive days. The data obtained were calculated as repeatability of recovered amounts, expressed by mean, standard deviation (SD), and coefficient of variation (% CV = SD/mean for each 4-hydroxymephenytoin concentration). Accuracy. Accuracy was determined by comparing the estimated amount with the known 4-hydroxymephenytoin concentration of spiked samples at 0.500, 5.00 and 20.0 mM in triplicate. The results were then


CYP2C19-mediated S-mephenytoin 4-hydroxylase assay Basically, reactions were initiated by the addition of membrane protein and were carried out in air at 37 C in a metabolic shaker for 60 min. Each reaction mixture (250 mL) contained bacterial membrane protein, an NADPH generating system (1.00 mM NADP, 10.0 mM G-6-P, 2.00 IU G-6-P dehydrogenase and 5.00 mM MgCl2) in 0.100 M phosphate buffer, pH 7.400 and S-mephenytoin. The reaction was terminated by addition of 130 mL of acetonitrile (containing 30.0 mL of 20.0 mg/mL phenytoin as internal standard) and cooling on ice followed by extraction with 1.00 mL of dichloromethane via votexing. The organic layer was then collected and dried. The residue was dissolved in 120 mL of mobile phase, and 50.0 mL of the solution was subjected to HPLC analysis.

Protein and time curves The effects of protein concentration and incubation time were examined by constructing the protein and time curves. Protein curve was generated by incubating a series of concentrations (0.200–4.00 mg/mL) of expressed CYP2C19 with S-mephenytoin at the final concentration of 100 mM at 37 C for 60 min. The time curve was constructed by incubating 0.200 mg/mL CYP2C19 with S-mephenytoin (final concentration 100 mM) at 37 C for various periods (10–180 min).

Kinetic characterization of CYP2C19 In order to characterize the catalytic activity of the CYP2C19, S-mephenytoin was incubated in various concentrations (ranging from 1.00 to 1.00 103 mM) with 5.50 102 mg CYP2C19 protein in volumes of 250 mL for 60 min. Using these data, saturation and Eadie–Hofstee plots were created and Km and Vmax values were determined with the EZ-fitW kinetic software (Perrella Scientific Inc., Amherst, NH, USA).

Results and discussion Heterologous expression of CYP2C19 in E. coli In this study, sequence (MALLLAVFL) was added to the N-terminus of CYP2C19 cDNA and ligated to pCWori+ plasmid, which was subsequently transformed into E. coli DH5a cells for expression. This modification strategy was adopted to overcome low expression level of human CYP protein achieved by bacteria owing to species difference in the preferred codon. Moreover, the E. coli cells do not possess an endogenous electron transport system to support the full catalytic activity of CYP enzymes; co-expression of CYP and OxR in E. coli cells has been found to be necessary to achieve efficient catalytic ability. We have reported co-expressed pCW-2D6 or pCW-3A4 with pACYC-OxR as separate proteins, demonstrating satisfactory expression levels and catalytic activities (Pan et al., 2011). Here we employed the same strategy to co-express CYP2C19 and OxR for use in the subsequent study.



Y. Pan et al. 0.09 0.08 0.07

Absorbance

pCW-2C19 and pACYC-ompA-OxR were successfully co-expressed from E. coli DH5a cells. The yield of membrane protein following routine culture protocol was high with values of 21.0 3.62mg membrane protein per litre culture. OxR expression level was 980 143 nmol/min/mg membrane determined from cytochrome c reductase assay, which fell within the reported range in the literatures (Blake et al., 1996; Pritchard et al., 1998) and these showed sufficient activities, allowing optimal activity for CYPs in oxidizing various substrates. The harvested and fractionated membrane proteins were analyzed using western immunoblotting (Fig. 2), showing a band with a molecular mass of approximately 50 kDa, which was in accordance with the size reported in the literature (Schulz-Utermoehl et al., 2000a, 2000b). In the meantime, control protein expressed from DH5a cells transformed with pCWori+ plasmid without any foreign gene did not demonstrate detectable band at this site. Nevertheless, the western blotting result only revealed the level of full-length total protein but not the type of the protein (whether they were apoprotein or holoprotein). Hence, reduced CO difference spectral scanning was carried out, which showed an absorbance maximum at 450 nm while the control cells that were transformed with the blank pCWori+ vector did not show any absorbance peak (Fig. 3). This indicated the successful expression of functional protein (i.e. in the form of holoenzyme). Our result once again demonstrated that E. coli expression system constructed with co-transforming OxR and CYP was able to obtain high output capacity. More than one foreign protein was expressed from DH5a cells with negligible effects with regard to competition between the genes of interest for the transcriptional and translational machinery of the cells. The result obtained is in line with findings from other researchers as well as our previous study, where high levels of OxR and CYP proteins were simultaneously achieved (Gillam 1998; Pan et al., 2011).

0.06 0.05 0.04 0.03 0.02 400

420

440

460

480

500

Wavelength (nm) Figure 3. Reduced CO difference spectra showing expression of CYP2C19 (solid line) and the control (dotted line). The arrow indicates the absorbance peak at 450 nm wavelength.

mean linear regression (n = 3) was y = 5.82 102x + 1.23 102 with an r2 value of 1.00. Sequentially diluted solutions of 4-hydroxymephenytoin injected directly onto the HPLC system revealed that 5.00 102 mM was the LOD for this assay. Specificity. Figure 4 shows representative chromatograms obtained from incubation mixtures containing the expressed CYP2C19 and OxR in comparison to that of the control cell.

(A) 1

HPLC method development Calibration curve, linearity and limit of detection. The linearity determined by linear regression analysis was found to be quite satisfactory and reproducible with time. The equation of 1

2

2

3

(B)

103KD 77KD

1

50KD 34.3KD 28.8KD 3

2

20.7KD Figure 2. Immunoblotting analyses of CYP2C19 expression in DH5a cells. The E. coli membrane proteins (150 mg) were electrophoresed on a 12.0% SDS–polyacrylamide gel, transferred to a nitrocellulose filter and reacted with appropriateantibody to CYP2C19. The labeled proteins were developed using peroxidase-conjugated goat anti-rabbit IgG. Lane 1, molecular weight markers (prestained protein standards, low range, BioRad, USA); lane 2, membrane fractions of the control cell carrying blank pCWori+ plasmid; lane 3, membrane fractions isolated from DH5a culture stocks carrying the co-transformed OxR and CYP2C19 plasmids.


Figure 4. Representative HPLC chromatograms of incubations of S-mephenytoin (100 mM) with (A) 5.50 102 mg/250 mL control cell membranes at 37 C for 60 min. Peak 1, S-mephenytoin (elution time = 12.5 min); peak 2, internal standard (elution time = 22.5 min); (B) 5.50 102 mg/250 mL bacterial membranes (expressing CYP2C19 and OxR) at 37 C for 60 min. Peak 3, Hydroxymephenytoin (elution time = 4.8 min); peak 1, S-mephenytoin (elution time = 12.5 min); peak 2, internal standard (elution time = 22.5 min).



Expression of CYP2C19 and establishment of RP-HPLC method As shown in the figure, the peaks for 4-hydroxymephenytoin (the metabolite), S-mephenytoin and phenytoin (internal standard), eluted at retention times of 4.8, 12.5 and 22.5 min, respectively, exhibited adequate symmetry and were well separated. No detectable 4-hydroxymephenytoin peak was observed in the control incubations. The metabolite peak was proven to be that of 4-hydroxymephenytoin as injection of pure metabolite onto the system yielded a peak with the same retention time.

Protein and time curves. The linear range for expressed CYP2C19 protein was from 0.200 to 0.800 mg/mL. The reaction rate stabilized thereafter at higher protein concentrations. The curve also shows that the linear range for incubation time was from 10 to approximately 120 min, after which the velocity curve started to level and stabilize. Hence 0.220 mg/mL bacterial membrane protein was incubated for 60min for subsequent routine reactions.

Precision. The intra- and interday precisions of 4-hydroxymephenytoin were obtained at three different concentrations of 0.500, 5.00 and 20.0 mM. The intraday precision ranged from 1.90 to 8.19% while interday precision ranged from 2.20 to 14.9%, which fell within the limit (i.e. below the upper limit of 15.0%) usually accepted for HPLC assay validation, indicating that the method developed has allowed precise and reliable measurement of 4-hydroxymephenytoin within the concentration range investigated.

Kinetics of CYP2C19-mediated S-mephenytoin 4-hydroxylation Figure 5(A) shows that 4-hydroxymephenytoin formation followed Michaelis–Menten kinetics {v = Vmax[S]/(Km+[S])}, where the velocity (v) of the reaction follows a hyperbolic pattern and approaches the maximum velocity (Vmax) with increasing substrate concentrations ([S]) and Km is the substrate concentration at which the reaction rate is half of its maximum. Similarly the Eadie–Hofstee plot (Fig. 5B) exhibited a straight line, indicating involvement of a single enzyme in the incubation reactions, which is expected as CYP2C19 was the only enzyme expressed in the bacterial membranes. The Km value determined from the plot, 108 9.20 mM, resides close to and within the variability of reported values for CYP2C19 in the literature (Table 2), indicating that the expressed CYP2C19 from this study was as active as those reported by others. However, there was a big difference between the Vmax (1.20 103 18.4 pmol/min/mg protein) determined from this study and reported values (Table 2). There will undoubtedly be differences between reports of Vmax values in human liver microsomes and various cDNA-expressed proteins

Accuracy and recovery. The accuracies determined using standard 4-hydroxymephenytoin at concentrations of 0.50, 5.00 and 20.0 mM were 101 4.30%, 95.0 6.20% and 105 15.9%, respectively. The calculated recoveries were 83.5 6.30% (0.500 mM), 85.8 14.8% (5.00 mM) and 85.2 2.50% (20.0 mM). Both results fell within acceptable range (80.0–120%). Thus, no significant difference was found between calculated concentrations and known concentrations while no significant differences were also noted between extracted samples with directly spiked samples, indicating that the method developed was accurate enough and the extraction procedure did not affect the assay.

(A)

(B)

Saturation Plot

Eadie-Hofstee Plot

2000 1000 1500

600

1000

V

v

V

800

400 500 200 0

0 0

500

1000

1500

2000

2500

3000

3500

4000

4500

0

5

S

10

15

V/S

Figure 5. Representative velocities vs substrate concentrations plot (A) and Eadie–Hofstee plot (B) for hydroxymephenytoin formation in DH5a membranes expressing the CYP2C19. Points are experimentally determined values while solid lines are the computer-generated curves of the best fit. V, Velocity in pmol/min/mg; S, S-mephenytoin concentration in (mM).

Table 2. Km and Vmax values collected from literatures for CYP2C19 mediated S-mephenytoin 4-hydroxylation Report 1 2 3 4 5 6

Km (mM) 57.2 2.20 35.0 7.00a 23.1 72.4 40.4a 52.7 59.0–143 a

Vmax (pmol/min/mg protein)

Reference

58.3 0.800 20.0 1.00a 69.4 70.5 48.0a 55.0 2.00a 12.7-80.8

Walsky and Obach (2004) Di Marco et al. (2007) Li et al. (2003) Wrighton et al. (1993) Hickman et al. (1998) Meier et al. (1985)

a

Data are represented as means SD.

a




Y. Pan et al. in heterologous expression system because these values are a function of the level of expression of enzyme in different system and the incubation conditions adopted in different laboratories. On the other hand, Km values should not differ greatly from preparation to preparation and should be comparable from one report to the next, since the parameter is an inherent property of the enzyme and should only potentially differ when the amino acid sequences of the enzymes show genetic variation, or if the activity being measured is not highly selective for one enzyme and has varying contributions from other enzymes. Nevertheless, it should be noted that Km values measured for the same reaction using different CYP2C19 sources still demonstrated some differences especially between pooled human liver microsomes and recombinant enzymes. Such differences are not uncommon and a number of factors could contribute to this variability. These include differences in protein concentrations used in vitro, differences in ratios of OxR and/or cytochrome b5 vs CYP in protein preparations, and differences in phospholipid composition of microsomes from expression systems vs that of liver tissue.

Conclusion As a conclusion, N-terminal modification introduced into CYP2C19 cDNA has allowed successful co-expression of CYP2C19 protein together with OxR protein in E. coli DH5a cells. This was judged by immunoblotting and reduced CO difference spectral assay, as well as cytochrome c reductase assay. In this work, we also developed and validated an RP-HPLC assay to determine the amount of 4-hydroxymephenytoin, which met the criteria for various validation characteristics for analytical procedures, including specificity, linearity, lower limit of detection, precision, accuracy and recovery. This validated assay was applied to characterize kinetics for CYP2C19 by incubating expressed protein with S-mephenytoin and co-factors. Enzyme kinetic parameters obtained demonstrated values that were either close to or within the range of values previously reported in literatures. In a nutshell, the protein expression strategy and the validated HPLC method described in this study were able to serve as activity markers for investigating drug metabolism and interactions involving CYP2C19.

Acknowledgments The authors are thankful to the International Medical University, Malaysia (grant IMU 091-05), as well as the Malaysian Ministry of Science, Technology and Innovation (grant eScienceFund 02-0209-SF0005) for financial support.

References Adedoyin A, Prakash C, O’Shea D, Blair IA and Wilkinson GR. Stereoselective disposition of hexobarbital and its metabolites: relationship to the S-mephenytoin polymorphism in Caucasian and Chinese subjects. Pharmacogenetics 1994; 4: 27–38. Andersson T, Regardh CG, Lou YC, Zhang Y, Dahl ML and Bertilsson L. Polymorphic hydroxylation of S-mephenytoin and omeprazole metabolism in Caucasian and Chinese subjects. Pharmacogenetics 1992; 2: 25–31. Barnes HJ, Arlotto MP and Waterman MR. Expression and enzymatic activity of recombinant cytochrome P450 17 alpha-hydroxylase in Escherichia coli. Proceedings of the National Academy of Sciences 1991; 88: 5597–5601. Bertilsson L, Dahl ML, Sjoqvist F, Abergwistedt A, Humble M, Johansson I, Lundqvist E and Ingelman-Sundberg M. Molecular basis for rational


megaprescribing in ultrarapid hydroxylators of debrisoquine. Lancet 1993; 341: 63; doi: 10.1016/0140-6736(93)92546-6. Blake JAR, Pritchard M, Ding S, Smith GCM, Burchell B, Wolf CR and Friedberg T. Coexpression of a human P450 (CYP3A4) and P450 reductase generates a highly functional monooxygenase system in Escherichia coli. FEBS Letters 1996; 397: 210–214; doi: 10.1016/ S0014-5793(96)01196-9. Eugster HP, Sengstag C, Meyer UA, Hinnen A and Würgler FE. Constitutive and inducible expression of human cytochrome P450IA1 in yeast Saccharomyces cerevisiae: an alternative enzyme source for in vitro studies. Biochemical and Biophysical Research Communications 1990; 172: 737–744; doi: 10.1016/ 0006-291X(90)90736-7. Gillam EM. Human cytochrome P450 enzymes expressed in bacteria: reagents to probe molecular interactions in toxicology. Clinical and Experimental Pharmacology and Physiology 1998; 25: 877–886; doi: 10.1111/j.1440-1681.1998.tb02338.x. Goldstein JA, Faletto MB, Romkes -Sparks M, Sullivan T, Kitaree wan S, Raucy JL, Lasker JM and Ghanayem BI. Evidence that CYP2C19 is the major (S)-mephenytoin 40 -hydroxylase in humans. Biochemistry 1994; 33: 1743–1752; doi: 10.1021/bi00173a017. Hickman D, Wang JP, Wang Y and Unadkat JD. Evaluation of the selectivity of in vitro probes and suitability of organic solvents for the measurement of human cytochrome P450 monooxygenase activities. Drug Metabolism and Dispositions 1998; 26: 207–215. Hyland R, Roe EG, Jones BC and Smith DA. Identification of the cytochrome P450 enzymes involved in the N-demethylation of sildenafil. British Journal of Clinical Pharmacology 2001; 51: 239–248; doi: 10.1046/j.1365-2125.2001.00318.x. Ko JW, Desta Z, Soukhova NV, Tracy T and Flockhart DA. In vitro inhibition of the cytochrome P450 (CYP450) system by the antiplatelet drug ticlopidine: potent effect on CYP2C19 and CYP2D6. British Journal of Clinical Pharmacology 2000; 49: 343–351; doi: 10.1046/j.13652125.2000.00175.x. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680–685. Larson JR, Coon MJ and Porter TD. Alcohol-inducible cytochrome P-450IIE1 lacking the hydrophobic NH2-terminal segment retains catalytic activity and is membrane-bound when expressed in Escherichia coli. The Journal of Biological Chemistry 1991; 266: 7321–7324. Li XQ, Bjorkman A, Andersson TB, Gustafsson LL and Masimirembwa CM. Identification of human cytochrome P(450)s that metabolise antiparasitic drugs and predictions of in vivo drug hepatic clearance from in vitro data. European Journal of Clinical Pharmacology 2003; 59: 429–442; doi: 10.1007/s00228-003-0636-9. Meier UT, Dayer P, Male PJ, Kronbach T and Meyer UA. Mephenytoin hydroxylation polymorphism: characterization of the enzymatic deficiency in liver microsomes of poor metabolisers phenotyped in vivo. Clinical Pharmacology and Therapeutics 1985; 38: 488–494; doi:10.1038/clpt.1985.213. Nakamura K, Goto F, Ray WA, McAllister CB, Jacqz E, Wilkinson GR and Branch RA. Interethnic differences in genetic polymorphism of debrisoquin and mephenytoin hydroxylation between Japanese and Caucasian populations. Clinical Pharmacology and Therapeutics 1985; 38: 402–408; doi: 10.1038/clpt.1985.194. Pan Y, Abd-Rashid BA, Ismail Z, Ismail R, Mak JW and Ong CE. Heterologous expression of human cytochromes P450 2D6 and CYP3A4 in Escherichia coli and their functional characterization. The Protein Journal 2011; 30: 581–591; doi: 10.1007/s10930-0119365-6. Pritchard MP, Glancey MJ, Blake JA, Gilham DE, Burchell B, Wolf CR and Friedberg T. Functional co-expression of CYP2D6 and human NADPH-cytochrome P450 reductase in Escherichia coli. Pharmacogenetics 1998; 8: 33–42. Schulz-Utermoehl T, Edwards RJ and Boobis AR. Affinity and potency of proinhibitory antipeptide antibodies against CYP2D6 is enhanced using cyclic peptides as immunogens. Drug Metabolism and Dispositions 2000a; 28: 544–551. Schulz-Utermoehl T, Mountfield RJ, Madsen K, Jørgensen PN and Hansen KT. Selective and potent inhibition of human CYP2C19 activity by a conformationally targeted antipeptide antibody. Drug Metabolism and Dispositions 2000b; 28: 715–717. Sindrup SH, Brosen K, Hansen MG, Aaes-Jorgensen T, Overo KF and Gram LF. Pharmacokinetics of citalopram in relation to the sparteine and



Expression of CYP2C19 and establishment of RP-HPLC method the mephenytoin oxidation polymorphisms. Therapeutic Drug Monitoring 1993; 15: 11–17. Singh R, Ting JG, Pan Y, Teh LK, Ismail R and Ong CE. Functional role of Ile264 in CYP2C8: mutations affect haem incorporation and catalytic activity. Drug Metabolism and Pharmacokinetics 2008; 23: 165–174; doi: 10.2133/dmpk.23.165. Walsky RL and Obach RS. Validated assays for human cytochrome P450 activities. Drug Metabolism and Dispositions 2004; 32: 647–660; doi: 10.1124/dmd.32.6.647. Ward SA, Helsby NA, Skjelbo E, Brosen K, Gram LF and Breckenridge AM. The activation of the biguanide antimalarial proguanil cosegregates with the mephenytoin oxidation polymorphism – a


panel study. British Journal of Clinical Pharmacology 1991; 31: 689–692. Waterman MR. Heterologous expression of cytochrome P-450 in Escherichia coli. Biochemical Society Transactions 1993; 21: 1081–1085; doi: 10.1042/bst0211081. Wilkinson GR, Guengerich FP and Branch RA. Genetic polymorphism of S-mephenytoin hydroxylation. Pharmacology and Therapeutics 1989; 43: 53–76; doi: 10.1016/0163-7258(89)90047-8. Wrington SA, Stevens JC, Becker GW and VandenBranden M. Isolation and characterization of human liver cytochrome P450 2C19: correlation between 2C19 and S-mephenytoin 4'-hydroxylation. Archives of Biochemistry and Biophysics 1993; 306: 240–245.

