The G-113A polymorphism in CYP1A2 affects the caffeine metabolic ratio in a Chinese population

Objective: This study was designed to better understand genetic variation in the cytochrome P450 (CYP) gene CYP1A2 and its impact on CYP1A2 activity in Chinese subjects. Methods. CYP1A2 genetic polymorphisms were screened by direct sequencing in 27 selected Chinese subjects. Plasma 1,7-dimethylxanthine/caffeine ratios 5 hours after a 100-mg caffeine administration, used as an index of CYP1A2 in vivo activity, were determined in 422 healthy subjects. Five single-nucleotide polymorphism markers, including G-860A (CYP1A2*1C), T-3594G, G-3113A, A-163C (CYP1A2*1F), and C5347T (CYP1A2*1B), were selected and genotyped by either polymerase chain reaction-restriction fragment length polymorphism or direct sequencing. Results. Thirteen polymorphisms and 2 linkage disequilibrium blocks with a boundary around -2467 were identified at this locus. The allele frequency for -3860A, -3594G, -3113A, -163C, and 5347T was 0.21, 0.15, 0.10, 0.36, and 0.14, respectively, in the CYP1A2-phenotyped cohort. A significant difference in CYP1A2 activity was observed among genotypes of polymorphism G-3113A (P = .038), and CYP1A2 activity in subjects carrying the AA genotype was lower than that in those carrying the GA (P = .096) and GG genotypes (P = .036): -0.45 +/- 0.05 (mean +/- SD), -0.32 +/- 0.16, and -0.29 +/- 0.16, respectively. Further analysis based on haplotype pairs found a 1.92-fold variation (95% confidence interval, 1.13-2.71) in mean CYP1A2 activity between haplotype pairs 13 and 15, and the difference was significant (-0.19 +/- 0.15 versus -0.45 +/- 0.05, P = .016). As compared with haplotype pair 10, haplotype pairs 9 and 15 and most haplotype pairs heterozygous for the haplotype with an A allele at -3113, including pairs 5, 8, and 12, also showed significantly lower CYP1A2 activity (P = .015,.048,.008,.024, and .014 for pairs 5, 8, 9, 12, and 15, respectively). In addition, haplotype pairs 5, 9, and 12 also showed significantly lower CYP1A-2 activity than pair 13 (P = .034,.020, and .037 for pairs 5, 9, and 12, respectively). Conclusions: The G-3113A polymorphism is associated with decreased CYP1A2 activity, haplotype pairs 10 and 13 are responsible for high CYP1A2 activity, and haplotype pairs 5, 8, 9, 12, and 15 are responsible for low CYP1A2 activity in Chinese subjects.