CYP2C9 constitutes approximately 20% of the cytochrome P450 protein content of human liver microsomes and is responsible for the metabolism of many clinically important drugs. These include drugs with a narrow therapeutic index such as warfarin and phenytoin and other routinely prescribed drugs such as acenocoumarol, tolbutamide, losartan, glipizide, and some nonsteroidal anti-inflammatory drugs (Lee et al., 2002).

The CYP2C9 gene is highly polymorphic. At least 13 CYP2C9 alleles have been identified to date and most of them are associated with reduced CYP2C9 activity. Among them, CYP2C9^*3, with an Ile359Leu mutation, has been most widely studied. In vitro studies show it has significantly impaired catalytic activity to various CYP2C9 substrates relative to the wild type (Takanashi et al., 2000). In vivo investigations show that individuals heterozygous and homozygous for CYP2C9^*3 have reduced intrinsic clearance of warfarin, phenytoin, and glipizide and are more at risk of clinical toxicity from these drugs (Aithal et al., 1999; Kidd et al., 1999; Ninomiya et al., 2000). Other CYP2C9 alleles such as CYP2C9^*2 (Arg144Cys), CYP2C9^*4 (Ile359Thr), CYP2C9^*5 (Asp360Glu), CYP2C9^*6 (null allele), CYP2C9^*11 (Arg335Trp), and CYP2C9^*12 (Pro489Ser) also show decreased enzymatic activity in vitro and in vivo (Crespi et al., 1997; Aithal et al., 1999; Imai et al., 2000; Dickmann et al., 2001; Kidd et al., 2001; Kirchheiner et al., 2002; Allabi et al., 2004; Blaisdell et al., 2004).

Recently, a new CYP2C9 allele designated CYP2C9^*13 has been identified in a Chinese poor metabolizer of lornoxicam. The allele possesses a T269C transversion in exon 2 of CYP2C9 that leads to a Leu90Pro substitution. Frequency analysis shows that approximately 2% of the Chinese populations carry the allele (Si et al., 2004). Genotyping of this poor lornoxicam metabolizer revealed a CYP2C9^*3/^*13 genotype with the two mutations located on separate alleles. His lornoxicam half-life of about 105 h was markedly longer than that of other CYP2C9^*1/^*3 and CYP2C9^*1/^*1 carriers (half-lives of 5.8–8.1 and 3.2–6.3 h, respectively; Zhang et al., 2005), suggesting that the CYP2C9^*13 allele has a larger effect on CYP2C9 metabolic capability than other alleles.

CYP2C9 has been shown to be the primary enzyme responsible for the biotransformation of the nonsteroidal anti-inflammatory drug lornoxicam to its major metabolite, 5′-hydroxylornoxicam, in human liver microsomes (Bonnabry et al., 1996; Kohl et al., 2000). Recently, it was reported that lornoxicam 5′-hydroxylation by the CYP2C9^*3 allele was markedly reduced compared with wild type, both in vitro and in vivo (Iida et al., 2004; Zhang et al., 2005). Thus, lornoxicam is an ideal substrate for the study of CYP2C9 enzyme activity. The purpose of this study was to compare the enzymatic activity of CYP2C9^*1, CYP2C9^*3, and CYP2C9^*13 toward lornoxicam both in vitro in appropriately transfected COS-7 cells and in vivo in subjects with CYP2C9^*1/^*3, CYP2C9^*1/^*13, and CYP2C9^*1/^*1 genotypes.

Materials and Methods

Materials. Lornoxicam was purchased from Shangdi Xinshiji Medical Academy (Beijing, China.). 5′-Hydroxylornoxicam was provided by the Laboratory of Microorganisms, Shenyang Pharmaceutical University (Shenyang, China). Dulbecco's modified Eagle's medium (DMEM), pcDNA3.1(+), and LipofectAMINE 2000 were purchased from Invitrogen (Carlsbad, CA). Fetal bovine serum was purchased from Tianjin, H&Y Bio Co. Ltd. (Tianjing, China). KpnI, XhoI, and DpnI enzymes were purchased from New England Biolabs (Beverly, MA). Pfu DNA polymerase was purchased from Bio Basic Inc. (Toronto, ON, Canada). A pREP9 plasmid containing human CYP2C9^*1 cDNA and E. coli Top 10 were provided by the Department of Pathophysiology and the Laboratory of Medical Molecular Biology, School of Medicine, Zhejiang University (Zhejiang, China). Rabbit anti-human cytochrome P450 2C9 antibody was purchased from Serotec (Oxford, UK). Alkaline phosphatase-labeled anti-rabbit IgG, 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) and bovine serum albumin were purchased from Beijing Dingguo Biotechnology Development Center (Beijing, China). NADPH was purchased from Roche Molecular Biochemicals (Basel, Switzerland). COS-7 cells were kindly donated by the Vaccination Center, Jilin University (Changchun, China). Human recombinant NADPH-P450 reductase (CPR) was purchased from MBL International Corporation (Woburn, MA). All other reagents were of analytical grade.

Construction of Expression Plasmids.CYP2C9^*1 cDNA in pREP9 plasmid was subcloned into pcDNA3.1(+) by digestion with KpnI and XhoI enzymes. Site-directed mutagenesis to introduce the A→C transition at position 1075 (CYP2C9^*3) and the T→C transition at position 269 (CYP2C9^*13) was performed using pcDNA3.1(+) plasmids carrying CYP2C9^*1 cDNA as the template for polymerase chain reaction amplification by Pfu DNA polymerase. The specific base transition was introduced into the amplification products by a pair of completely complementary primers containing substituted base. The mutagenic primers for CYP2C9^*3 and ^*13 were 5′-CGAGGTCCAGAGATACCTTGACCTTCTCCCCAC-3′ and 5′-GGAAGCCCTGATTGATCCTGGAGAGGAGTTTTCTG-3′, respectively (mutations underlined). After incubation with DpnI enzyme, the origin templates were digested, but the new amplified polymerase chain reaction products containing substituted base remained and were transformed to E. coli Top 10. Clones containing the desired nucleotide change were identified by sequencing carried out by Shanghai Sangon Biological Engineering Technology and Service Co. Ltd. (Shanghai, China).

Transfection of COS-7 Cells and Preparation of Postmitochondrial Supernatant (S9). COS-7 cells were seeded into 10-cm culture flasks in DMEM containing 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. When cells were 90 to 95% confluent, the culture medium was replaced with DMEM without penicillin and streptomycin, and the CYP2C9 expression plasmids (24 μg/flask), purified with a QIAGEN plasmid mini kit (QIAGEN, Valencia, CA), were transfected into COS-7 cells using LipofectAMINE 2000 at 60 μl/flask, as per the manufacturer's instructions. Forty-eight hours after transfection, cells were scraped from the culture flask and washed twice with Ca²⁺- and Mg²⁺-free Hanks' solution. The pellets were resuspended in 20 mM potassium phosphate buffer, pH 7.4, containing 0.2 mM EDTA, 1 mM dithiothreitol, and 20% glycerol, and sonicated with twelve 5-s pulses at 23% power of a Sonics Vibra-Cell sonicator (Sonics & Materials, Inc., Newtown, CT). The homogenate was centrifuged at 9000g, 4°C for 20 min and the postmitochondrial supernatant (S9 fraction) collected for assay or storage at -70°C. Protein concentrations in S9 were determined by the Bradford method (Bradford, 1976) using bovine serum albumin as standard.

Quantification of CYP2C9 Protein by Western Blotting. S9 fraction (50 μg) and human liver microsomes (10 μg) were separated on 10% sodium dodecyl sulfate-polyacrylamide gels and transferred to a polyvinylidene difluoride membrane (Millipore Corporation, Billerica, MA). The membrane was incubated with rabbit anti-human cytochrome P450 2C9 antibody as the primary antibody and then with alkaline phosphatase-labeled anti-rabbit IgG as the secondary antibody. Bands were visualized by incubation with BCIP/NBT and quantified by microsomes from insect cells expressing human CYP2C9 (Invitrogen) as a standard with ImageJ software (National Institutes of Health, Bethesda, MD).

In Vitro Lornoxicam 5′-Hydroxylation of Recombinant CYP2C9 Protein. Lornoxicam 5′-hydroxylation activity of the recombinant CYP2C9 protein was determined as described previously (Kohl et al., 2000) with minor modifications. S9 fraction containing CYP2C9s was incubated with 100 mM Tris buffer (pH 7.5), 200 μM NADPH, and lornoxicam at 37°C for 1 h in the presence or absence of CPR. The reaction was stopped by addition of 500 μl of methanol and stored overnight at -20°C to allow complete protein precipitation. After centrifugation for 30 min at 12,000 rpm, the supernatants were reduced to 100 μl by warming at 65°C and subjected to high performance liquid chromatography assay. High performance liquid chromatography was carried out on a SB-300A C₁₈ 10-μm column (4.6 × 200 mm, Agilent Technologies, Palo Alto, CA) using a mobile phase of 0.1 M NaH₂PO₄, pH 6.0/acetonitrile (7:3), at a flow rate of 1 ml/min. Detection was by UV absorption at 371 nm. Under these conditions, retention times of 5′-hydroxylornoxicam and lornoxicam were 6.9 and 11.8 min, respectively. A six-point standard curve was used to quantify 5′-hydroxylornoxicam.

In Vivo Lornoxicam Metabolism. The study was approved by the Independent Ethics Committee of the People's Hospital of Liaoning Province (Shenyang, China). Genotyping of CYP2C9^*3 and CYP2C9^*13 was carried out as described previously (Si et al., 2004; Zhang et al., 2005). Thirteen CYP2C9^*1 homozygotes, 7 CYP2C9^*3 heterozygotes, and 3 CYP2C9^*13 heterozygotes participated in the phenotyping study. All subjects were in good health and were required to refrain from all medication and alcohol prior to the pharmacokinetic study. In vivo lornoxicam metabolism was performed according to the method described previously (Zhang et al., 2005). In brief, after a single oral dose of 8 mg of lornoxicam, blood samples were collected before dosing, and at 0.5, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 10, 13, and 24 h postdose. Plasma concentrations of lornoxicam were determined using a validated liquid chromatography-tandem mass spectrometry method reported elsewhere (Zeng et al., 2004).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

CYP2C9 protein levels in postmitochondrial supernatant (S9) from COS-7 cells expressing wild-type and variant CYP2C9s. A, immunoblot analysis of recombinant human CYP2C9 protein. S9 fraction (50 μg) isolated from COS-7 cells transfected with the wild-type and variant CYP2C9 cDNA was utilized for immunoblotting using anti-human CYP2C9 antibody. Human liver microsome (10 μg) was used as positive control. B, protein levels of CYP2C9 were quantified by densitometric analysis. The results are expressed as a percentage of the level of CYP2C9^*1. Each bar represents the mean ± S.E.M. of three independent experiments.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Michaelis-Menten kinetics of lornoxicam by postmitochondrial supernatant (S9) from COS-7 cells expressing wild-type and variant CYP2C9s. S9 fraction corresponding to 500 μg of protein was incubated with different concentrations of lornoxicam in the absence of CPR. Experimental conditions are described under Materials and Methods. Each point represents the mean of three independent experiments. ▪, CYP2C9^*1; ▴, CYP2C9^*3; ▾, CYP2C9^*13.

Data Analysis. Michaelis-Menten analysis was performed by nonlinear regression curve fitting using the computer program Prism v4.0 (GraphPad Software Inc., San Diego, CA). Pharmacokinetic parameters were calculated using standard noncompartmental methods. Student's t test was used for intergroup comparison. A value of P < 0.05 was considered to be statistically significant.

Results

A representative immunoblot of S9 fraction prepared from COS-7 cells expressing CYP2C9^*1, CYP2C9^*3, and CYP2C9^*13 proteins is presented in Fig. 1. All constructs yielded immunodetectable CYP2C9 protein, as did human liver microsomes. The expressed protein level of CYP2C9^*1 was 7.51 pmol/mg S9 protein, and the expression levels of variant CYP2C9^*3 and ^*13 were 69.9% and 35.5%, respectively, of that of CYP2C9^*1.

The effect of exogenous CPR on lornoxicam 5′-hydroxylation was studied by incubation with recombinant CYP2C9^*1. S9 fraction containing 3.75 pmol of CYP2C9^*1 was mixed with varying amounts of CPR and 10 μM lornoxicam. When CPR was 0, 5, 12.5, or 25 μM, the mean value of V_max for CYP2C9^*1 was 0.66, 0.74, 0.68, or 0.68 pmol/min/pmol, respectively. There was no significant alteration in catalytic efficiency of CYP2C9^*1 with the increasing CPR concentration, indicating that endogenous reductase is enough for lornoxicam metabolism in the COS-7 expression system. Thus, exogenous CPR was not utilized in the following kinetic study.

Michaelis-Menten kinetics of lornoxicam for wild-type and mutant CYP2C9 is shown in Fig. 2. Corresponding kinetic parameters are summarized in Table 1. On the basis of protein levels of S9 fraction, both CYP2C9^*3 and CYP2C9^*13 exhibited lower intrinsic clearance of lornoxicam (P < 0.01) than did wild-type CYP2C9^*1 resulting from higher values of K_m (1.3-fold and 2.3-fold, respectively) and lower values of V_max (76% and 90%, respectively). When the enzymatic activities were normalized to CYP2C9 protein levels, intrinsic clearance values (V_max/K_m) for variant CYP2C9^*3 and CYP2C9^*13 were also separately decreased to 28% and 12% compared with wild type. Clearly, the activity of CYP2C9^*13 is lower than that of CYP2C9^*3.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Plasma concentration-time curves of lornoxicam in healthy subjects with genotype CYP2C9^*1/^*1 (♦) (n = 13), CYP2C9^*1/^*3 (□) (n = 7), and CYP2C9^*1/^*13 (▵) (n = 3) after a single oral dose of 8 mg of lornoxicam.

In the in vivo study, the CYP2C9 genotype significantly affected the pharmacokinetics of lornoxicam (Fig. 3; Table 2). The AUC of lornoxicam increased 1.9-fold and CL/F was decreased by 44% in CYP2C9^*1/^*13 individuals compared with CYP2C9^*1/^*1 individuals. Similar changes were found in CYP2C9^*1/^*3 individuals.

Discussion

To investigate the catalytic activity of the CYP2C9^*13 allele in vitro, we established a COS cell expression system. This has been widely applied for functional characterization of P450 alleles containing CYP2C9, CYP2D6, CYP2B6, and CYP2E1 (Veronese et al., 1993; Hu et al., 1997; Marcucci et al., 2002; Jinno et al., 2003). However, due to the low levels of expression in COS-7 cells, we failed to quantify CYP2C9 holoenzyme contents by CO-difference spectroscopy. Thus, CYP2C9 proteins were quantified by immunoblotting with microsomes from insect cells expressing human CYP2C9 as standard. This kind of method quantifying P450 was also used for the functional characterization of CYP2D6, CYP2B6, and CYP2E1 allelic variants in the COS expression system (Marcucci et al., 2002; Hanioka et al., 2003; Jinno et al., 2003). Our results show that COS-7 cells can efficiently express active CYP2C9 protein. The protein levels of cDNA-expressed CYP2C9^*3 and ^*13 in S9 fraction from COS-7 cells were lower than those in wild-type CYP2C9^*1. The reduced protein levels in the CYP2C9 variants may contribute to lower transcription, translation efficiency, and protein stability. Otherwise, although polyclonal antibody was used, the immunoreactivity of mutant CYP2C9s may be altered by mutagenization. Therefore, the enzymatic activities were assayed in two ways, on the basis of S9 protein level and CYP2C9 protein level. These data obtained using the COS expression system need to be confirmed by a baculovirus system, which is better suited to obtain the quantity of P450 by spectral analysis. These studies are currently under investigation in our laboratory.

In this study, the presence of the CYP2C9^*3 allele impairs both intrinsic clearance and systemic clearance of lornoxicam. A recent report shows that CYP2C9^*1/^*3 individuals have a 55% decrease in CL/F and a 1.9-fold increase in AUC of lornoxicam compared with CYP2C9^*1/^*1 individuals (Zhang et al., 2005). The magnitude of these changes in pharmacokinetic parameters is consistent with our in vivo results. Iida et al. (2004) reported that CYP2C9^*3 expressed in baculovirus-infected insect cells significantly decreased lornoxicam 5′-hydroxylation relative to wild type with a 2.3-fold increase in K_m (P < 0.05) and 76% decrease in V_max. In our COS-7 cells, the K_m showed a 1.3-fold increase (P > 0.05) and the V_max a 66% decrease on the basis of CYP2C9 protein level relative to wild type. The reason for the discrepancy is the difference in the heterologous cell expression system. The addition of exogenous reductase did not significantly affect the lornoxicam 5′-hydroxylation by CYP2C9^*1, indicating that the endogenous reductase in COS-7 cells is not limiting for lornoxicam metabolism.

In addition to the CYP2C9^*3 variant, the catalytic activity of a recently identified CYP2C9^*13 variant that contains a Leu90Pro substitution was investigated in this study. Compared with wild-type CYP2C9^*1, the CYP2C9^*13 variant also has lower intrinsic clearance for lornoxicam 5′-hydroxylation due to a 2.3-fold increase in K_m and 73% decrease in V_max on the basis of CYP2C9 protein level. The results are consistent with our in vivo observation that individuals with CYP2C9^*1/^*13 genotype have an impaired clearance of lornoxicam compared with individuals with CYP2C9^*1/^*1 genotype. Interestingly, in our study, individuals with CYP2C9^*1/^*3 and CYP2C9^*1/^*13 genotypes reveal the same extent of reduction in oral clearance of lornoxicam despite the fact that, in vitro, CYP2C9^*13 is associated with a lower intrinsic clearance of lornoxicam than is CYP2C9^*3.

Given the small number of CYP2C9^*1/^*13 subjects studied (n = 3), and in the absence of any individuals homozygous for the CYP2C9^*13 allele, we recognize that further in vivo studies are required to draw firm conclusions about the role of the CYP2C9^*13 allele. According to a crystal structure of CYP2C9 published by Williams et al. (2003) and Wester et al., (2004), Leu90 is located in the B-B′ loop, which is not the heme-binding region and far from the binding pocket of substrate. Thus, the reason for the increase in K_m for lornoxicam 5′-hydroxylation is not clear. Homology modeling based on the crystal structure of human CYP2C9 is ongoing in our laboratory (Wester et al., 2004).

Our results show that the activities of both CYP2C9^*3 and CYP2C9^*13 toward lornoxicam in vitro are compatible with their activities in vivo, and there is a reasonable correlation between in vitro activity and in vivo metabolic clearance of lornoxicam. Recently, it was reported that individuals carrying the CYP2C9^*3 allele are at risk of experiencing drug toxicity, especially of drugs with a narrow therapeutic index such as warfarin and phenytoin (Aithal et al., 1999; Kidd et al., 1999; Ninomiya et al., 2000). By extrapolation, one may speculate that carriers of the CYP2C9^*13 allele would experience greater risk from these drugs. Therefore, genotyping for CYP2C9^*13 may be important to allow individualization of dosing for CYP2C9 substrate drugs.

In conclusion, the Leu90Pro substitution of CYP2C9^*13 markedly decreases the intrinsic clearance of lornoxicam in vitro and in vivo. The reduction in activity due to CYP2C9^*13 is greater than that due to CYP2C9^*3 in vitro. Whether carriers of the CYP2C9^*13 allele may be at greater risk of toxicity from CYP2C9 substrate drugs with a narrow therapeutic index remains to be confirmed by further in vivo studies.