The purpose of the present studies was to define the role of the I359L allelic variant of CYP2C9 in the metabolism of the low therapeutic index anticoagulant warfarin, by performing in vitro kinetic studies with the two enantiomers of the drug. To obtain sufficient quantities of these variants to perform kinetic studies at physiologically relevant substrate concentrations, methodology was established for the high-level expression, purification, and structural characterization of wild-type CYP2C9 and CYP2C9V1 using the baculovirus system. Both forms were expressed at levels up to 250 nmol/liter and purified in 50-55% yield to specific contents of 13-14 nmol holoenzyme/mg protein. The purified preparations were characterized by Edman degradation and electrospray-mass spectrometry. Both forms of the enzyme metabolized the pharmacologically more potent (S)-enantiomer of warfarin with the same regioselectivity; however, CYP2C9V1 exhibited a fivefold lower Vmax and a fivefold higher Km compared to the wild-type enzyme for this substrate. Neither form of the enzyme formed significant quantities of the (R)-warfarin phenols. Additional studies performed with prochiral arylalkyl sulfides provided confirmation of the low turnover rates catalyzed by CYP2C9V1 and demonstrated further that sulfoxide product stereochemistry did not differ significantly between the two variants. Therefore, decreased catalytic efficiency rather than a gross alteration in substrate orientation appears to be the consequence of this putative active-site mutation. The greatly decreased catalytic efficiency of the I359L variant suggests that leucine homozygotes would eliminate (S)-warfarin, and probably many other CYP2C9 substrates, at much slower rates in vivo than individuals expressing the wild-type enzyme.