Cytochrome P450 2C9 polymorphism : interindividual differences in drug metabolism and phenotyping methodology

Abstract: Cytochrome P450 2C9 (CYP2C9) is one of the most important drug metabolizing enzymes in human liver. It catalyses the metabolism of a wide range of different drugs such as S-warfarin, acenocoumarol, losartan, irbesartan, phenytoin, sulphamethoxazole, tolbutamide, glipizide and numerous non-steroidal antiinflammatory drugs. Single nucleotide polymorphisms in the coding region of the CYP2C9 gene give rise to enzyme variants with different amino acid residues and enzymatic properties. The most common enzyme variant originates from the CYP2C9'1 allele, whereas CYP2C9'2 and CYP2C9'3 code for enzymes with amino acid substitutions R144C and 1359L, respectively. There is accumulating evidence that this genetic polymorphism give rise to clinically important interindividual differences in drug metabolism. The general aim of this project was to further characterize the relationship between genetic polymorphism of CYP2C9 and interindividual variation in enzyme activity, including the identification and validation of a sensitive and specific phenotyping probe for CYP2C9 activity in vivo and in vitro. Different PCR-based endonuclease protocols were initially validated regarding their specificity for analysis of the CYP2C9-gene. The frequency of the CYP2C9'2 and CYP2C9'3 alleles was analyzed in DNA-samples from 430 healthy Swedish volunteers, and found to be 0. 107 and 0.074, respectively. Enzyme kinetics was studied in vitro by analyzing the oxidation of losartan to E- 3174 and 4'-hydroxylation of diclofenac in genotyped human liver microsomes. Yeast microsomes with recombinant expression of the different CYP2C9 variants were also used. The oxidation of losartan to E-3174 was found to be completely dependent on CYP2C9 at low and therapeutically relevant concentrations (< 1 µM). The intrinsic clearance of losartan oxidation was significantly lower in human liver samples with the CYP2C9'1/'3 genotype compared to '1/'1, and almost 20-fold lower in a '3/'3 sample. In contrast, diclofenac 4'-hydroxylation did not dependent on genotype in human liver microsomes. In vivo-studies were performed where healthy volunteers with different CYP2C9 genotypes received a single oral dose of diclofenac and losartan, at separate occasions. Plasma kinetics and urinary excretion of parent compounds and metabolites were compared between the different CYP2C9 genotypes. For diclofenac, no significant differences in kinetic parameters were found between the genotypes. In contrast, the kinetics of losartan oxidation indicated important genotype differences. The plasma AUClosartan/AUCE-3174 ratio was higher not only in a subject genotyped as CYP2C9'/'3 (30-fold) but also in the CYP2C9'1/'3 and '2/'3 groups (approximately 2- and 3-fold respectively, as compared to '1/'1. The plasma ratios correlated significantly with the 0-8 hour urinary losartan/E-3174 ratios. The urinary ratio was significantly higher in CYP2C9'1/'3 (n=12) and '2/'3 (n=5) genotypes compared to '1/'1 (n=27) and approximately 50-fold higher in CYP2C9'3/'3 (n=3) subjects. There was a highly significant correlation between the losartan/E-3174 urinary ratios determined in the same subjects at two different occasions. In conclusion, the results strongly indicate that losartan may be used as a sensitive and specific phenotyping probe for CYP2C9 activity both in vivo and in vitro. This will be a useful tool in further characterization of interindividual differences in drug metabolism.