Introduction

Top Abstract Introduction Results Discussion References

Cytochrome P450s (P450²) comprise a superfamily of enzymes that catalyze the oxidation of a wide variety of xenobiotic chemicals including drugs and carcinogens and steroids including sex hormones (Gonzalez, 1990; Guengerich, 1992; Niwa et al., 1998). CYP2C isoforms constitute about 20% of constitutively expressed hepatic P450s among humans (Shimada et al., 1994), and CYP2C9 and CYP2C19 plays important roles in the metabolism of a wide range of therapeutic agents (Goldstein and de Morais, 1994). Although only 43 of 490 amino acid residues differ between CYP2C9 and CYP2C19 (91% amino acid sequence identity), they have very distinctive substrate specificities (Romkes et al., 1991; Goldstein and de Morais, 1994; Miners et al., 2000; Goldstein, 2001; Tsao et al., 2001). For example, CYP2C19 is the principal enzyme responsible for the S-mephenytoin 4'-hydroxylation and omeprazole 5-hydroxylation (Goldstein et al., 1994; Lasker et al., 1998). On the other hand, CYP2C9 exhibits little activity toward these substrates but exhibits high activity toward the hydroxylation of phenytoin and diclofenac (Goldstein and de Morais, 1994; Lasker et al., 1998; Mancy et al., 1999).

Gotoh (1992) predicted six potential substrate recognition sites (SRS) in CYP2 family based on an alignment with bacterial CYP101 (P450cam), the substrate binding residues of which have been identified by X-ray crystallography and analysis of mutations that altered substrate specificity in experimental studies of the CYP2 subfamilies. Recently, Goldstein and colleagues reported that residues 286 and 289 of CYP2C9 are important in conferring specificity for diclofenac 4'-hydroxylation and ibuprofen hydroxylation (Klose et al., 1998), that amino acids 99, 220, and 221 are key residues to determine the specificity of CYP2C19 for omeprazole (Ibeanu et al., 1996), and that three residues in Helix I, Asp286, Ala292, and Leu295 of CYP2C19 are essential for S-mephenytoin 4'-hydroxylation (Tsao et al., 2001). However, the key residues of CYP2C9 and CYP2C19 for the metabolism of other substrates except for these chemicals are not clear.

Yamazaki and Shimada (1997) reported that the formation of androstenedione from testosterone by CYP2C19 was 5 times higher than that by CYP2C9, and we have reported that aminopyrine N-demethylation by CYP2C19 was 7 times higher than that by CYP2C9 (Niwa et al., 1999). On the other hand, it has been reported that the kinetic parameters of CYP2C19-mediated tolbutamide hydroxylation resembled those of the CYP2C9-catalyzed reaction (Lasker et al., 1998).

The present study was designed to elucidate the key regions or amino acid residues that determine the marked specificity of CYP2C9 and CYP2C19 for five kinds of substrates: diclofenac, tolbutamide, S-mephenytoin, aminopyrine, and testosterone. To investigate key amino acid residues responsible for the marked specificity of CYP2C9 and CYP2C19 for these substrates, we constructed 15 kinds of chimeras and mutant enzymes that replaced the residues of CYP2C9 or CYP2C19 with those of the other P450 (Kishimoto et al., 1995). The structures of the chimeras and mutants and their relationship to the SRS proposed by Gotoh (1992) are shown in Fig. 1.

View larger version (53K): [in this window] [in a new window]   Fig. 1.   Schematic representation of the chimeric constructs and mutants of CYP2C9 and CYP2C19. Putative SRS (numbered 1-6) are indicated at the appropriate sites of CYP2C9.

Experimental Procedures

Materials. Testosterone and aminopyrine were purchased from Wako Pure Chemical Industries (Osaka, Japan). Tolbutamide, estrone, and androstenedione were obtained from Sigma-Aldrich (St. Louis, MO). 4'-Hydroxydiclofenac, hydroxytolbutamide, S-mephenytoin, and 4-hydroxymephenytoin were purchased from Sumika Chemical Analysis Service (Osaka, Japan), and diclofenac was obtained from Ultrafine Chemicals (Manchester, UK). All other reagents were of the highest purity commercially available.

Expression of Human P450s, Construction of Chimeras and Mutants, and Preparation of Microsomes. Expression of human hepatic CYP2C9 and CYP2C19 in recombinant Saccharomyces cerevisiae and preparation of microsomal fractions from the cells were carried out according to the methods described previously (Imaoka et al., 1996; Niwa et al., 1998). The chimeras were constructed using BamHI, SmaI, PstI, and SphI sites at residues 212, 228, 292, and 339, that are shared by both enzymes' cDNA (Kishimoto et al., 1995; Ibeanu et al., 1996; Klose et al., 1998; Tsao et al., 2001). The mutagenesis procedure was essentially as described by Kishimoto et al. (1995) and Ibeanu et al. (1996). Amino acid changes were introduced in CYP2C9 and CYP2C19 using synthetic oligonucleotides (Table 1) containing the desired point mutations and a second primer. The pAAH5N vectors containing CYP2C9/19-related cDNA were used for transformation of the Saccharomyces cerevisiae AH22 (a, leu2-3, leu2-112, his4-519, can1)(cir+) strain. The P450 contents in yeast microsomes for CYP2C9 and CYP2C19 were 127 and 49 pmol/mg of protein, respectively. The contents (114-270 pmol/mg of protein) for the chimeras and mutants that have residues 1-98 of CYP2C9 were comparable to that of CYP2C9, and the contents (13-61 pmol/mg of protein) for other chimeras and mutants were similar to that for CYP2C19. The contents of P450s were determined as described by Omura and Sato (1964).

Assay of Diclofenac 4'-Hydroxylase Activity. Diclofenac 4-hydroxylase activity was measured by the method described previously (Niwa et al., 2002). Briefly, the incubation mixture consisted of yeast microsomes (8-40 pmol P450), 2 to 500 µM diclofenac, 1 mM NADPH, and 100 mM Tris-HCl buffer (pH 7.5) in a final volume of 0.5 ml. After a 2-min preincubation at 37^oC, the reaction was started by adding NADPH. Incubation was carried out at 37^oC for 3 to 5 min, and the reaction was terminated by the addition of 3 ml of ethyl acetate, and 50 µl of 100 µM estrone was added as an internal standard. After ethyl acetate extraction, the organic phase (2.5 ml) was evaporated under nitrogen, and the residue was dissolved in 200 µl of the HPLC mobile phase for HPLC analysis. The HPLC system consisted of an L-7100 pump, an L-7400 UV-detector set at 282 nm, and a D-7500 integrator (Hitachi, High-Technologies Corp., Tokyo, Japan). A column (4.6 × 250 mm) packed with Cosmosil 5C18-AR-II (Nacalai Tesque Inc., Kyoto, Japan) was used. The mobile phase consisted of 0.01% triethylamine in 0.75 µM phosphoric acid/acetonitrile (55:45), and the flow rate was 1 ml/min.

Assay of S-Mephenytoin 4'-Hydroxylase Activity. The incubation mixture consisted of yeast microsomes (30 pmol P450), 233 µM S-mephenytoin, 3 mM NADPH, and 100 mM potassium phosphate buffer (pH 7.4) in a final volume of 0.19 ml. After a 5-min preincubation at 37^oC, the reaction was started by adding yeast microsomes. Incubation was carried out at 37^oC for 5 min, and the reaction was terminated by the addition of 50 µl of ice-cold 2N HCl. After the mixtures were shaken and centrifuged at 10,000g for 5 min, the resulting supernatant (5 µl) was injected onto an HPLC apparatus with a Cosmosil 5C18 column (4.6 × 150 mm, Waters Corp., Milford, MA). The mobile phase was 10 mM potassium phosphate buffer (pH 7.4) as eluent A and methanol as eluent B; the flow rate was 1 ml/min. Gradient conditions were 0 to 5 min, 20% B; 5 to 10 min, 20 to 50% B; 10 to 15 min, 50% B; 15 to 20 min 50 to 100% B; and 10 to 20 min, 100% B (linear gradient). The UV-detector was set at 230 nm.

Assay of Tolbutamide Hydroxylase Activity. The incubation mixture consisted of yeast microsomes (100 pmol P450), 532 µM tolbutamide, 3 mM NADPH, and 100 mM potassium phosphate buffer (pH 7.4) in a final volume of 0.19 ml. After a 5-min preincubation at 37^oC, the reaction was started by adding yeast microsomes. Incubation was carried out at 37^oC for 5 min, and the reaction was terminated by the addition of 50 µl of ice-cold 2N HCl. After the mixtures were shaken and centrifuged at 10,000g for 5 min, the resulting supernatant (5 µl) was injected onto an HPLC apparatus with a Cosmosil 5C18 column (4.6 × 150 mm, Waters). The mobile phase was 0.01% trifluoroacetic acid as eluent A and 0.01% trifluoroacetic acid in methanol as eluent B; the flow rate was 1 ml/min. Gradient conditions were 0 to 10 min, 40 to 50% B; and 10 to 20 min, 50 to 80% B (linear gradient). The UV-detector was set at 230 nm.

Assay of Androstenedione Formation from Testosterone. The incubation mixture consisted of yeast microsomes (50 pmol P450), 20 to 250 µM testosterone, 1 mM NADPH, and 100 mM Tris-HCl buffer (pH 7.5) in a final volume of 0.5 ml. After a 1-min preincubation at 37^oC, the reaction was started by adding NADPH. Incubation was carried out at 37^oC for 10 min, and the reaction was terminated by the addition of 4 ml of ethyl acetate. After ethyl acetate extraction, the organic phase (3 ml) was evaporated under nitrogen, and the residue was dissolved in 200 µl of the HPLC mobile phase. Thereafter, 50 µl of the solution was injected onto an HPLC apparatus with a Cosmosil 5C18-AR-II column (4.6 × 250 mm; Nacalai Tesque Inc.). The mobile phase consisted of methanol/acetonitrile/water (57:5:38), and the flow rate was 0.75 ml/min. The metabolite was monitored at 254 nm.

Assay of Aminopyrine N-Demethylase Activity. Aminoipyrine N-demethylase activity was measured by the method described previously (Niwa et al., 1999). For standard incubation, the final reaction mixture (0.5 ml) contained yeast microsomes (50 pmol P450), 0.1 to 5 mM aminopyrine, 1 mM NADPH, and 100 mM Tris-HCl buffer (pH 7.5). After a 2-min preincubation at 37^oC, the reaction was started by adding NADPH. Incubation was carried out at 37^oC for 5 min, and the reaction was stopped by the addition of 0.25 ml of 20% trichloroacetic acid. Aminopyrine N-demethylase activity was determined by the estimation of formaldehyde production using the Nash reagent (Nash, 1953).

Data Analysis. In preliminary experiments, the linearity of reaction with P450 concentration and incubation time was confirmed for each assay conditions. All data were analyzed using the mean of duplicate determinations; the deviation in each value was less than 15% of the mean. V_max and K_m values for diclofenac 4'-hydroxylation, aminopyrine N-demethylation, and androstenedione formation from testosterone were determined by fitting to Michaelis-Menten kinetics by nonlinear regression analysis (MULTI; Yamaoka et al., 1981).

	Results

Top Abstract Introduction Results Discussion References

Diclofenac 4'-Hydroxylase Activity. Kinetic parameters for diclofenac 4'-hydroxylation were assessed by CYP2C9, CYP2C19, and their chimeras and mutants (Table 2). CYP2C9 had an 18 times lower K_m and a 9 times higher V_max than CYP2C19. Therefore, the intrinsic clearance (V_max/K_m) for CYP2C9 was 161 times higher than that of CYP2C19. The V_max and K_m values for the CYP2C9(1-339)/CYP2C19(340-490) chimera were comparable to those for CYP2C9. The K_m values for the CYP2C9(1-288)/CYP2C19(289-490) chimera and the CYP2C9(I99H,V292A, F295L, I331V) mutant were comparable to those for CYP2C19. The K_m value for the CYP2C19(1-288)/CYP2C9(289-490) chimera was much higher than that for CYP2C9, and the V_max value for the chimera was 3 times higher than that for CYP2C19 and 36% of that for the CYP2C9. On the other hand, the K_m and V_max values for the CYP2C19(H99I) mutant were 3.5 times higher and 3.7 times lower than those for CYP2C19, respectively; V_max/K_m value for the mutant was 12.5 times lower than that for CYP2C19. The 4'-hydroxylated metabolite by CYP2C9(1-230)/CYP2C19(231-490) chimera was not detected (less than 0.5 nmol/min/nmol P450) even when using 500 µM diclofenac.

Tolbutamide Hydroxylase Activity. Lasker et al. (1998) reported that the kinetic parameters of CYP2C19-mediated tolbutamide hydroxylation (K_m, 650 µM; V_max, 3.71 nmol/min/nmol P450) resembled those of the CYP2C9-catalyzed reaction (K_m, 178-407 µM; V_max, 2.95-7.08 nmol/min/nmol P450). Therefore, we investigated the activities by CYP2C9, CYP2C19, and their chimeras and mutants at 532 µM around the expected K_m as a substrate concentration (Fig. 2). Both CYP2C9 and CYP2C19 showed comparable activity. The activities for the CYP2C9(1-98)/CYP2C19(99-490) and CYP2C19(1-339)/CYP2C9(340-490) chimeras and the CYP2C19(H99I) mutant were 36 to 40% of those for CYP2C9, and the activities for the CYP2C9(1-212)/CYP2C19(213-490) and CYP2C9(1-230)/CYP2C19(231-490) chimeras were much lower than those for the CYP2C9 or CYP2C19. The activities by other chimeras and mutants were comparable to those of CYP2C9 and CYP2C19.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Tolbutamide hydroxylase activity of CYP2C9, CYP2C19, and their chimeras and mutants expressed in yeast. Tolbutamide at 532 µM substrate concentration was incubated with CYP2C9, CYP2C19, and their chimeras and mutants (50 pmol) at 37°C for 5 min. Results represent mean of duplicate determinations. CYP, cytochrome P450.

S-Mephenytoin 4'-Hydroxylase Activity. Tsao et al. (2001) reported that K_m values for S-mephenytoin hydroxylation by CYP2C19, and the 9/Sma-Sph (228-339), I99H, S220P, P221T and 9/Eco-Sph (283-339), I99H, S220P, P221T chimeras were 49.77, 81.70, and 46.20 µM, respectively, indicating that the mutants had K_m values similar to CYP2C19. Therefore, we investigated the activities by CYP2C9, CYP2C19, and their chimeras and mutants at 233 µM as a substrate concentration; the activities at this concentration are comparable to V_max. Although the activity by CYP2C19 was 14.9 nmol/min/nmol P450, the hydroxylation by CYP2C9 was not detected (less than 2 nmol/min/nmol P450) (Fig. 3). The activities for the CYP2C19(1-288, 340-490)/CYP2C9(289-339) and CYP2C19(1-339)/CYP2C9(340-490) chimeras and the CYP2C19(H99I) mutant were 17-31% of those for CYP2C19. The CYP2C9(1-339)/CYP2C19(340-490), CYP2C9(1-288)/CYP2C19(289-490), CYP2C9(1-212)/CYP2C19(213-490), and CYP2C19(1-212, 289-490)/CYP2C9(213-288) chimeras showed no detectable activity, whereas other chimeras and mutants had comparable activities with CYP2C19.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   S-Mephenytoin 4'-hydroxylase activity of CYP2C9, CYP2C19, and their chimeras and mutants expressed in yeast. S-Mephenytoin at 233 µM substrate concentration was incubated with CYP2C9, CYP2C19, and their chimeras and mutants (30 pmol) at 37°C for 5 min. Results represent mean of duplicate determinations. CYP, cytochrome P450.

Androstenedione Formation from Testosterone. The kinetic parameters for androstenedione formation from testosterone by CYP2C9, CYP2C19, and their chimeras and mutants are summarized in Table 3. Although the K_m for CYP2C19 was comparable to that for CYP2C9, CYP2C19 showed 41 times higher V_max value than CYP2C9. The K_m values in CYP2C9, CYP2C19, and all the chimeras and mutants constructed were in the same order (34.0-72.6 µM). A replacement of residues 1-98 of CYP2C19 with those of CYP2C9 (CYP2C9(1-98)/CYP2C19(99-490)) and a mutation at 343 (CYP2C19(G343S)) had no marked effect on the K_m and V_max values. On the other hand, the V_max values for the CYP2C19(1-339)/CYP2C9(340-490) and CYP2C19(1-288)/CYP2C9(289-490) chimeras and the CYP2C19(H99I) mutant were 21 to 35% of those for CYP2C19, and the V_max values for the CYP2C9(1-339)/CYP2C19(340-490) and CYP2C9(1-230)/CYP2C19(231-490) chimeras were comparable to those of CYP2C9.

Aminopyrine N-Demethylase Activity. The kinetic parameters of aminopyrine N-demethylation by CYP2C9 and CYP2C19 were compared (data not shown). CYP2C19 (97 nmol/min/nmol P450) had 3 times higher V_max value than did CYP2C9 (31 nmol/min/nmol P450), whereas the K_m values in CYP2C9 (0.39 mM) and CYP2C19 (0.44 mM) were closely similar. Therefore, the intrinsic clearance (V_max/K_m, µl/min/nmol P450) was calculated to be 79 and 222 for CYP2C9 and CYP2C19, respectively.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Aminopyrine N-demethylase activity of CYP2C9, CYP2C19, and their chimeras and mutants expressed in yeast. Aminopyrine at 2 mM substrate concentration was incubated with CYP2C9, CYP2C19, and their chimeras and mutants (50 pmol) at 37°C for 5 min. Results represent mean of duplicate determinations. CYP, cytochrome P450.

	Discussion

Top Abstract Introduction Results Discussion References

The CYP2C subfamily in human liver comprises four members: CYP2C8, CYP2C9, CYP2C18, and CYP2C19. Among those, CYP2C9 and CYP2C19 are 91% identical in their amino acid sequences (Romkes et al., 1991), although they have very distinctive substrate specificities (Goldstein and de Morais, 1994; Miners et al., 2000). Six SRSs for CYP2C9 and other CYP2C subfamily proteins have been identified (Gotoh, 1992), and these SRSs span residues 96-117, 198-205, 233-240, 286-304, 359-369, and 470-477 (Tang et al., 2000). Ibeanu et al. (1996) reported that the replacement of Ile99 of CYP2C9 with His99, which is the only difference between CYP2C9 and CYP2C19 in SRS-1, increased omeprazole 5'-hydroxylase activity markedly. Additionally, Tsao et al. (2001) reported that three residues in Helix I, Asp286, Ala292, and Leu295 of CYP2C19, which are included in SRS 4 are essential for S-mephenytoin 4'-hydroxylation. On the other hand, the deletion of the 283-340 region, containing SRS 4, resulted in a drastic reduction of CYP2C9-dependent diclofenac and tolbutamide hydroxylase activities and a complete loss of ibuprofen hydroxylase activity. In contrast, diclofenac hydroxylase activity was affected only minimally, and some degree of hydroxylation of ibuprofen was maintained in a chimera lacking residues 228-282 of CYP2C9 containing SRS 3, suggesting that this region was less important than the SRS 4 region (Klose et al., 1998). Additionally, Ridderstrom et al. (2000) reported that arginines 97 and 108 in CYP2C9 are important for diclofenac 4'-hydroxylation. They observed that the R97D mutant had a 13 times higher K_m value whereas the V_max was in the same order as the wild type and that the Arg108 mutant had a 100 times lower activity toward diclofenac than the wild-type enzyme.

In this study, we showed that the S-mephenytoin 4'-hydroxylase and aminopyrine N-demethylase activities for the CYP2C19(H99I) mutant, CYP2C9(1-230)/CYP2C19(231-490), and the chimeras that replaced residues 231-340, which includes SRS 3 and 4 [i.e., CYP2C19(1-288)/CYP2C9(289-490), CYP2C19(1-288, 340-490)/CYP2C9(289-339), CYP2C19(1-212, 289-490)/CYP2C9(213-288), CYP2C9(1-212)/CYP2C19(213-490), CYP2C9(1-288)/CYP2C19(289-490), and CYP2C9(1-339)/CYP2C19(340-490) chimeras], were much lower than those for CYP2C19 or comparable to those for CYP2C9 except that the aminopyrine N-demethylase activities for the CYP2C19(1-288)/CYP2C9(289-490) and CYP2C19(1-288, 340-490)/CYP2C9(289-339) chimeras remained 56 to 68% of those for CYP2C19, whereas the activities for other chimeras and mutants were comparable to those for CYP2C19 (Figs. 3 and 4). On the other hand, the K_m values of androstenedione formation from testosterone by CYP2C19 and their chimeras and mutants were comparable to those for CYP2C9, although the V_max of CYP2C19 was 30 times higher than CYP2C9. The V_max values for the CYP2C19(H99I) mutant and the CYP2C19(1-339)/CYP2C9(340-490) and CYP2C19(1-288)/CYP2C9(289-490) chimeras were 21 to 35% of those for CYP2C19, and the V_max values for the CYP2C9(1-339)/CYP2C19(340-490) and CYP2C9(1-230)/CYP2C19(231-490) chimeras were comparable to those of CYP2C9. These results in S-mephenytoin 4'-hydroxylation, aminopyrine N-demethylation, and androstenedione formation from testosterone support the previous finding for omeprazole 5'-hydroxylation and S-mephenytoin 4'-hydroxylation (Ibeanu et al., 1996; Tsao et al., 2001). Thus, it is suggested that the residues 99 and 231-339, the latter of which contains SRS 3 and 4, are important for the substrate specificity or the substrate binding of CYP2C19. Furthermore, although the tolbutamide hydroxylase activity by CYP2C19 was comparable to that by CYP2C9, the activities for the CYP2C19(H99I) mutant and the CYP2C9(1-212)/CYP2C19(213-490) and CYP2C9(1-230)/CYP2C19(231-490) chimeras were much lower than those for CYP2C19 (Fig. 2). On the other hand, the activities for the CYP2C19(1-288, 340-490)/CYP2C9(289-339) and CYP2C19(1-212, 289-490)/CYP2C9(213-288) chimeras, which includes SRS 3 or 4 of CYP2C19, were comparable to those for CYP2C19. Therefore, it can be speculated that His99 is important in the substrate binding of CYP2C19, whereas the residues 231-339 are essential for the substrate specificity. Replacement of residues 340-490 of CYP2C19, which includes SRS 5 and 6, with those of CYP2C9 also decreased the activities or the V_max value toward S-mephenytoin, aminopyrine, and testosterone by 31, 78, and 35%, respectively (Figs. 3 and 4, Table 3). Therefore, these results suggest that the region containing SRS 5 and/or 6 of CYP2C19 is also important in the substrate specificity.

For the 4'-hydroxylation of diclofenac, CYP2C9 had an 18 times lower K_m and a 9 times higher V_max than CYP2C19; the intrinsic clearance (V_max/K_m) for CYP2C9 was 161 times higher than that of CYP2C19 (Table 2). The V_max and K_m values for the CYP2C9(1-339)/CYP2C19(340-490) chimera were comparable to those for CYP2C9, suggesting that the carboxyl terminal region of CYP2C9 has minor influence of capacity to metabolize diclofenac. On the other hand, although the V_max values for the CYP2C9(1-288)/CYP2C19(289-490) chimera and CYP2C9(I99H, V292A, F295L, I331V) mutant were comparable to those for CYP2C9, the K_m values for the mutant were comparable to those for CYP2C19. These results suggest that the residues 292, 295, and/or 331 of CYP2C9, which are partially included in SRS 4, are essential for the recognition of substrate in CYP2C9. Additionally, the V_max value for the CYP2C19(1-288)/CYP2C9(289-490) chimera was 26% of that for CYP2C9 with the K_m values higher than those for CYP2C19, and the activity by CYP2C9(1-230)/CYP2C19(231-490) chimera was negligible. Therefore, these results suggest that residues 231-288 of CYP2C9, which includes SRS 3, are important for the metabolic capacity of CYP2C9. On the other hand, although the tolbutamide hydroxylase activities by CYP2C19 and the CYP2C9(1-339)/CYP2C19(340-490) and CYP2C9(1-288)/CYP2C19(289-490) chimeras were comparable to those by CYP2C9, and the activities for the CYP2C9(1-230)/CYP2C19(231-490) and CYP2C9(1-212)/CYP2C19(213-490) chimeras were much lower than those for CYP2C9 (Fig. 2), the decrease of the activities by these chimeras is speculated to be due to the replacement of residue 231-288 of CYP2C9, which includes SRS 3, with those of CYP2C19. Therefore, these results suggest that SRS 3 is important in the substrate binding of CYP2C9.

In summary, the present study suggests that SRS 3 and 4 of CYP2C19 are essential to determine the specificity of CYP2C19 toward S-mephenytoin 4'-hydroxylation, aminopyrine N-demethylation, and androstenedione formation from testosterone, whereas His99 is important for the substrate binding of CYP2C19. In addition, our results suggest that the residues 292, 295, and/or 331 of CYP2C9, which are included in SRS 4, are essential for the recognition of its substrate, diclofenac, and that the residues 231-288, including SRS 3, are important for the metabolic capacity of CYP2C9.