CYP2C9 amino acid residues influencing phenytoin turnover and metabolite regio- and stereochemistry

Abstract

Phenytoin has been an effective anticonvulsant agent for over 60 years, although its clinical use is complicated by nonlinear pharmacokinetics, a narrow therapeutic index, and metabolically based drug-drug interactions. Although it is well established that CYP2C9 is the major cytochrome P450 enzyme controlling metabolic elimination of phenytoin through its oxidative conversion to (S)-5-(4-hydroxyphenyl)-5-phenylhydantoin (p-HPPH), nothing is known about the amino acid binding determinants within the CYP2C9 active site that promote metabolism and maintain the tight stereocontrol of hydroxy metabolite formation. This knowledge gap was addressed here through the construction of nine active site mutants at amino acid positions Phe100, Arg108, Phe114, Leu208, and Phe476 and in vitro analysis of the steady-state kinetics and stereochemistry of p-HPPH formation. The F100L and F114W mutants exhibited 4- to 5-fold increases in catalytic efficiency, whereas the F100W, F114L, F476L, and F476W mutants lost >90% of their phenytoin hydroxylation capacity. This pattern of effects differs substantially from that found previously for (S)-warfarin and (S)-flurbiprofen metabolism, suggesting that these three ligands bind within discrete locations in the CYP2C9 active site. Only the F114L, F476L, and L208V mutants altered phenytoin's orientation during catalytic turnover. The L208V mutant also uniquely demonstrated enhanced 6-hydroxylation of (S)-warfarin. These latter data provide the first experimental evidence for a role of the F-G loop region in dictating the catalytic orientation of substrates within the CYP2C9 active site.

Scheme 1.

In vitro metabolic pathways of phenytoin metabolism by CYP2C9. CYP2C19 is also known to catalyze formation of both (S)-p-HPPH and (R)-p-HPPH.

Fig. 1.

Metabolic turnover of wild-type CYP2C9 and Phe mutant enzymes with 0.1 mM phenytoin (A) and 0.5 mM (S)-warfarin (B) (Mosher et al., 2008). Bars equal the sum of major and minor metabolites. Values for R108L and R108F are omitted because of low turnover. Error bars, S.D. of triplicate incubations.

Fig. 2.

Fold change from wild-type CYP2C9 formation of percentage m-HPPH of total HPPH (wild-type percentage was 4 ± 0.02) (A), percentage (R)-p-HPPH of total p-HPPH (wild-type percentage was 3 ± 0.1) (B), 6-/7-hydroxywarfarin (C) (wild-type ratio was 0.3:1), or percentage 4′-hydroxywarfarin of total (S)-warfarin (D) (wild-type percentage was 3 ± 0.2). (Mosher et al., 2008). F476W values in A and B and R108L and R108F values in A to D are not reported because of low turnover. Error bars, propagated error from triplicate incubations.

Fig. 3.

LC-MS/MS chromatograms showing separation of (R)- and (S)-p-HPPH from turnover of phenytoin by wild-type CYP2C9 (A) or L208V (B). Only one channel of the MRM analysis is shown (m/z 269 > 198). The peaks are as follows: 8.3 min, m-HPPH; 9.0 min, (R)-p-HPPH; and 9.7 min, (S)-p-HPPH.

Fig. 4.

UCSF Chimera rendering of phenytoin in the CYP2C9 active site. Phenytoin is pictured in green surrounded by residues that appear to influence its orientation in the active site, including Phe114 (blue), Phe476 (purple), Arg108 (yellow), and Leu208 (orange). The heme is red.