Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cytochrome P450 2D6 (CYP2D6) catalyzes the oxidative metabolism of various clinically important drugs (Rendic and Di Carlo, 1997). Interindividual and interethnic differences have been seen in the metabolic activity of CYP2D6, mainly due to the polymorphism of CYP2D6 gene (Marez et al., 1997; Sachse et al., 1997; Griese et al., 1998). The percentage of poor metabolizers that have a deficiency in CYP2D6 activity varies from 5 to 10% in Caucasian populations (Gonzalez et al., 1988; Sachse et al., 1997), whereas it is less than 1% in Asian populations (Yokoi et al., 1996). In Caucasian populations, CYP2D6*3, *4, *5, and *6 are mainly involved in the deficiency of CYP2D6 activity (Sachse et al., 1997), whereas CYP2D6*5 and *14 are mainly involved in the deficiency of CYP2D6 activity in Asian populations (Kubota et al., 2000; Nishida et al., 2000).

CYP2D6*14 is a mutant allele that was first found in a Chinese subject (Wang, 1992) and has since been found only in Asian populations (Wang et al., 1999; Kubota et al., 2000). The variant protein corresponding to CYP2D6*14 (CYP2D6 14) is thought to have little activity, because a Chinese subject carrying CYP2D6*5/*14 was shown to be the phenotype of a poor metabolizer (Wang et al., 1999). There are four amino acid substitutions in CYP2D6 14: P34S, G169R, R296C, and S486T (Daly et al., 1996; Wang et al., 1999). Among them, only G169R is unique to this variant protein (Human CYP Allele Nomenclature Committee, http://www.imm.ki.se/CYPalleles/cyp2d6.htm). Therefore, G169R substitution was considered to be mainly responsible for the deficient activity of CYP2D6 14. However, a previous study using a Rat-1 cell cDNA expression system showed that substitution of P34S caused a dramatic decrease in CYP2D6 activity, whereas the decrease induced by G169R substitution was not as dramatic as that induced by substitution of P34S. Based on these findings, Wang et al. (1999) speculated that G169R is not a critical substitution but that its combination with P34S, R296C, and/or S486T may abolish the activity of CYP2D6 14. However, this possibility has not been studied by direct construction of variant CYP2D6 protein having combined substitutions of G169R and others.

Therefore, we constructed mutated cDNAs that yield CYP2D6 variant proteins substituted with G169R and/or P34S, expressed in V79 cells, and compared their metabolic properties for the prototype substrates of CYP2D6. R296C and S486T substitutions were not examined in the present study, since they are present in CYP2D6 2, and subjects with CYP2D6*2/*2 show similar metabolic activities to those of wild type for various substrates of CYP2D6 (Johansson et al., 1993; Dahl et al., 1995; Tateishi et al., 1999). On the other hand, P34S substitution has been reported to cause a significant reduction in CYP2D6 activity in cDNA expression systems (Kagimoto et al., 1990; Johansson et al., 1994; Fukuda et al., 2000).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of Mammalian Cell Expression Vectors. A G1758A mutation for a Gly169-to-Arg169 change was introduced into wild-type CYP2D6 cDNA in a pUC19 plasmid vector using a transformer site-directed mutagenesis kit (BD Biosciences Clontech, Palo Alto, CA). The sequence of the oligonucleotide used for the mutation was 5'-CGCCAACCACTCTAGACGCCCC-3'. The mutation was confirmed by sequencing, and the mutated fragment was subcloned into a pBluescript II SK+ vector (Stratagene, La Jolla, CA). CYP2D6/pTARGET and CYP2D6 G169R/pTARGET expression vectors were obtained by ligating the wild-type CYP2D6 and mutated CYP2D6 G169R cDNA to a pTARGET Mammalian Cell Expression Vector (Promega, Madison, WI), respectively.

Mammalian Cell Culture and Expression of Recombinant Protein. Parental V79 cells (V79-4, CL93; American Type Cell Culture Collection, Manassas, VA) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin, and 100 µg/ml streptomycin in 5% CO₂ and 95% air at 37°C. The V79 cells were plated at 1 × 10⁶ cells/100-mm plate. On the following day, recombinant expression vectors were transfected into V79 cells using LIPOFECTAMINE Reagent and Plus Reagent (Invitrogen, Carlsbad, CA) according to manufacturer's recommendations. For negative control, V79 cells were transfected with an empty pTARGET vector. After two days of transfection, V79 cells were plated into a 100-mm plate with a medium containing the selective agent G418 (Invitrogen) at 800 µg/ml. Each medium was changed every 2 to 3 days, and cells were maintained for at least 3 weeks to obtain stable expression colonies. After selection with G418, V79 cells were harvested and cell homogenate was prepared.

Quantification of mRNA Content in V79 Cells. Total RNA was extracted from V79 cells using an RNeasy mini kit and QIA shredder column (QIAGEN GmbH, Hilden, Germany), and then first-strand cDNA was prepared. CYP2D6 mRNA contents in V79 cells were analyzed in a GeneAmp 5700 sequence detection system (Applied Biosystems, Foster City, CA) using fluorescence detection for SYBR-Green II. The nucleotide sequences of the primers used for the amplification were 5'-GCAGCACTTCAGCTTCTCGG-3' and 5'-CTCACCAGGAAAGCAAAGACACCAT-3'. Each PCR product was confirmed by the dissociation curve and agarose gel electrophoresis. The GAPDH mRNA contents in each cell were also measured, and CYP2D6 mRNA content was normalized to GAPDH mRNA content. The nucleotide sequences of reverse transcription-PCR products were confirmed.

Enzyme Assay and Kinetic Analysis of CYP2D6 Activity in V79 Cells Bufuralol 1'-hydroxylation. The activity of bufuralol 1'-hydroxylation was measured by the method of Kronbach et al. (1987) with slight modifications. Reaction mixtures containing the cell homogenates (0.22-0.29 mg/ml) and 0.5 to 80 µM bufuralol were incubated for 60 min at 37°C. The reaction was performed in a linear range with respect to protein concentration and incubation time and was stopped by the addition of cold acetonitrile. The determination of 1'-hydroxybufuralol was carried out using the following HPLC methods. The HPLC system consisted of a model L-7100 pump, a model L-7485 fluorescence detector, a model L-7200 autosampler, a model D-7500 integrator (Hitachi, Tokyo, Japan), and a 4.6 × 150-mm CAPCELL PAK C₁₈ UG120 column (Shiseido, Tokyo, Japan). The mobile phase consisted of acetonitrile and water at the ratio of 20:80 (v/v) in 0.01 M citric acid buffer (pH 3.4), and it was delivered at a flow rate of 1.0 ml/min. The elute was monitored at an excitation wavelength of 252 nm and emission wavelength of 302 nm. A calibration curve was generated from 0.002 to 0.2 µM by processing the authentic standard substance through the entire procedure. The detection limit of this analytical method was 0.04 pmol/ml.

Dextromethorphan O-Demethylation. The activity of dextromethorphan O-demethylation was measured by the method of von Moltke et al. (1998) with slight modifications. The reaction mixture containing the cell homogenates (0.22 to 0.29 mg/ml) and 10 µM dextromethorphan was incubated for 60 min at 37°C. The reaction was performed in a linear range, with respect to protein concentration and incubation time, and stopped by the addition of cold acetonitrile. The determination of dextrorphan was carried out by the same HPLC system and mobile phase used for the determination of 1'-hydroxybufuralol. The eluate was monitored at excitation wavelength of 270 nm and emission wavelength of 312 nm. A calibration curve was generated from 0.01 to 1.0 µM by processing the authentic standard substance through the entire procedure. The detection limit of this analytical method was 0.37 pmol/ml.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Expression of Wild-Type and Mutated CYP2D6 in V79 Cells. The expression of CYP2D6 protein in V79 cells transfected with recombinant CYP2D6 cDNA was confirmed by Western blot analysis (Fig. 1A). There were no significant differences between CYP2D6 protein levels in V79 cells expressing wild-type (2D6WT) and G169R-substituted CYP2D6 (2D6G169R). On the other hand, CYP2D6 protein levels in V79 cells expressing P34S-substituted CYP2D6 (2D6P34S) and those expressing both G169R- and P34S-substituted CYP2D6 (2D6G169R/P34S) decreased to approximately 26% that of 2D6WT (Fig. 1B). No significant differences were found between mRNA levels of CYP2D6 in V79 cells transfected with wild-type and mutated CYP2D6 cDNAs (Fig. 2).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   CYP2D6 apoprotein contents in homogenate of V79 cells expressing wild-type CYP2D6 or CYP2D6 substituted with G169R and/or P34S. V79 cell homogenates (1.6 µg/well) were separated by SDS-PAGE and transferred onto a nitrocellulose sheet and then stained immunochemically as described under Materials and Methods. A, Western blot analysis; B, densitometric quantification of CYP2D6 apoprotein levels. Values in columns are means ± S.D. calculated from six independent analyses. Marked columns are significantly different from CYP2D6 wild type (*, P < 0.005) or G169R-substituted CYP2D6 (, P < 0.005) determined by Student's t test. PAGE, polyacrylamide gel electrophoresis; HLM, human liver microsome; WT, wild type; N.D., not detected.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   CYP2D6 mRNA contents in V79 cells expressing wild-type CYP2D6 or CYP2D6 substituted with G169R and/or P34S. CYP2D6 mRNA contents in V79 cells were quantified as described under Materials and Methods. CYP2D6 mRNA content was normalized to GAPDH mRNA content. WT, wild type.

Enzyme Activities. Bufuralol 1'-hydroxylation activity of 2D6G169R (1.00 pmol/min/pmol CYP2D6) was slightly lower than that of 2D6WT, whereas 2D6P34S (0.05 pmol/min/pmol CYP2D6) was much lower than that of 2D6WT (1.23 pmol/min/pmol CYP2D6), which corresponds to 4% of 2D6WT (Fig. 3A). However, the activity of 2D6G169R/P34S was under the detection limit of our analytical method (Fig. 3A).

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Enzyme activities of V79 cells expressing wild-type CYP2D6 or CYP2D6 substituted with G169R and/or P34S. A, bufuralol 1'-hydroxylation; B, dextromethorphan O-demethylation. The protein levels of wild-type and substituted CYP2D6 determined here were those of the apoprotein and not of the native protein. The data are means of two or three independent experiments. WT, wild type; N.D., not detected.

Kinetic Analysis of Bufuralol 1'-Hydroxylation. Kinetic parameters for bufuralol 1'-hydroxylation in V79 cells expressing wild-type and mutated CYP2D6 are shown in Table 1. While the K_m value was not affected, the V_max value slightly decreased in 2D6G169R. On the other hand, the K_m value and V_max value of 2D6P34S were 5.4-times higher and 7.2-times lower than those of 2D6WT, respectively, thus the V_max/K_m value of 2D6P34S was 40-times lower than that of 2D6WT. The kinetic parameters could not be determined for 2D6G169R/P34S because of the extremely low activity of bufuralol 1'-hydroxylation.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study clearly showed that a combination of G169R and P34S substitutions diminishes the activities of CYP2D6 to less than the detection limit of our analytical method for bufuralol 1'-hydroxylation and dextromethorphan O-demethylation. Considering that the activities of CYP2D6 are not completely abolished by the substitution of P34S (4 and 32% of 2D6WT for bufuralol 1'-hydroxylation and dextromethorphan O-demethylation, respectively), the findings suggest that simultaneous substitution of G169R and P34S is crucial for almost completely abolishing the activity of CYP2D6 at least in a cDNA expression system using V79 cells.

P34S is a key substitution of CYP2D6 10 encoded by CYP2D6*10, which has been found to reduce but not abolish the activity of CYP2D6 by in vitro studies (Ramamoorthy et al., 2001; Senda et al., 2001; Shimada et al., 2001). In vivo studies have also shown that the metabolic capacity of homozygous CYP2D6*10 is between those of extensive and poor metabolizers for various substrates of CYP2D6, and homozygous CYP2D6*10 has therefore been categorized as an intermediate metabolizer (Wang et al., 1993; Dahl et al., 1995; Lai et al., 1995; Tseng et al., 1996). In addition, heterozygous subject with CYP2D6*10 and a defective allele (e.g., CYP2D6*5/*10) has also been categorized as intermediate metabolizer (Griese et al., 1998). However, a heterozygous subject with CYP2D6*5 and *14 showed a metabolic ratio of debrisoquine of more than 12.6, indicating that the subject can be categorized as a poor metabolizer (Wang et al., 1999). Based on that finding, Wang et al. (1999) concluded that the suppressive effect of CYP2D6*14 on CYP2D6 activity is more pronounced than that of CYP2D6*10 and that CYP2D6*14 is a mutant allele causing defective activity of CYP2D6.

There are four amino acid substitutions (P34S, G169R, R296C, and S486T) in CYP2D6 14. Two of them (P34S and S486T) are overlapped with substitution in CYP2D6 10. Since S486T does not affect the suppressive effect of P34S on the activity of CYP2D6 (Fukuda et al., 2000; Tsuzuki et al., 2001), the pronounced effect of CYP2D6*14 appears to be derived from the combination of either G169R and P34S or R296C and P34S. The results of the present study support the former possibility. However, the possibility that interaction between R296C and P34S or among R296, S486T, and P34S plays a role in the pronounced effect of CYP2D6*14 cannot be ruled out, since CYP2D6*41 allele contains the substitutions R296C and S486T and has a significantly lower in vivo activity when compared with both *1 and *2 (Raimundo et al., 2000).

The precise mechanism underlying the combined effects of G169R and P34S could not be determined in the present study. However, based on three-dimensional models of bacterial P450 CYP102, G169R substitution of CYP2D6 corresponds to Ala135 of CYP102, which is located in the turn region between the D and E helices (Lewis et al., 1997; Lewis, 1998). Since Gly in a turn region is important for maintenance of a tertiary structure of a protein, it is assumed that substitution of Gly169 to Arg alters the conformation of the CYP2D6 protein. On the other hand, Pro34 of CYP2D6 is the first residue in the proline-rich region, and this region is followed by the N-terminal signal anchor region (Yamazaki et al., 1993), which is highly conserved in the mammalian CYP2 family (Yamazaki et al., 1993). Although the proline-rich region is far from the catalytic domain of CYP2D6 (Lewis, 1998) and not involved in the substrate-binding sites of CYP2D6 (Modi et al., 1996), P34S substitution has been reported to alter the catalytic properties of CYP2D6 for bufuralol 1'-hydroxylation, venlafaxine O-demethylation, and bunitrolol 4-hydroxylation (Fukuda et al., 2000; Tsuzuki et al., 2001). The present study also showed that the K_m value and V_max value of 2D6P34S for bufuralol 1'-hydroxylation were 5.4-times higher and 7.2-times lower than those of 2D6WT, respectively, thus the V_max/K_m value of 2D6P34S was 40-times lower than that of 2D6WT (Table 1). These findings suggest that the substitution of P34S, although it is not in a substrate-recognition site, affects the tertiary structure of CYP2D6 and reduces the activities of CYP2D6. Therefore, we are tempted to speculate that combined substitutions of G169R and P34S cause a more drastic change in the structure of CYP2D6 than that caused by a single substitution of P34S, which results in a more pronounced reduction in CYP2D6 activity. Similar combined effects of substitutions have also been reported for CYP2D6*17 (Oscarson et al., 1997). Alternatively, combined substitutions of G169R and P34S may deteriorate the incorporation of heme into the CYP2D6 apoprotein, although this possibility could not be assessed in the present study, since the levels of CYP2D6 protein expressed in V79 cells were too low to be determined by difference CO-reduced spectra.

Regarding the level of CYP2D6 protein, a combined substitution of G169R and P34S decreased it substantially in V79 cells (Fig. 1). However, the extent of decrease was not different from that caused by a single substitution of P34S (Fig. 1). This finding suggests that G169R substitution does not accelerate the decreasing effect of P34S substitution on the protein level of CYP2D6 reported previously (Johansson et al., 1994; Fukuda et al., 2000; Tsuzuki et al., 2001).

The present study also showed that there were no major differences between the levels of CYP2D6 mRNA in V79 cells transfected with wild-type and mutated CYP2D6 cDNA (Fig. 1). The finding suggests that substitutions of P34S and G169R do not appear to affect the transcriptional efficiency of CYP2D6 cDNA introduced to V79 cells.

In conclusion, the present study showed that a combination of G169R and P34S substitutions reduced the activities of CYP2D6 to undetectable levels despite the fact that the activities of CYP2D6 were not completely abolished by the substitution of P34S. The findings suggest that simultaneous substitution of G169R and P34S is required for the activity of CYP2D6 to be almost abolished, at least in V79 cells. This combined effect may be a factor responsible for the poor metabolism of debrisoquine found in a subject carrying CYP2D6*14 (Wang, 1992; Wang et al., 1999).