Comparison of linkage disequilibrium patterns and haplotype structure of eight single nucleotide polymorphisms across the CYP1A2 gene between the Korean, and other populations registered in the International HapMap database

The aim of this study was to determine the frequencies of CYP1A2 gene polymorphisms, analyze Linkage disequilibrium (LD) blocks and haplotypes in a Korean population, and compare them with those in African, European, Japanese and Chinese populations. We searched across diverse studies conducted in Korea and the Knowledge Base for Korean Pharmacogenomics Research Network operated by Seoul National University to determine the frequency of single nucleotide polymorphisms (SNPs) of the CYP1A2 gene and to choose frequently occurring SNPs in a Korean population. We analyzed and confirmed the frequencies of CYP1A2 SNPs that are inferred as MAF >0.05 in 400 healthy Korean subjects, using direct sequencing and a TaqMan assay. The LD block and haplotypes were constructed from the SNP databases in the other races registered in the International HapMap (Europeans, Chinese, Japanese and Africans) and the haplotype frequencies in each race were compared with those in the Korean population. We found 12 SNPs with minor allele frequency (MAF) values above 5% in the 5' regulatory regions, the exon, and surrounding introns of CYP1A2 gene based on previous reports in Koreans. In this study, two of twelve SNPs were lower than 5% in frequency (n = 400). The CYP1A2 haplotypes were analyzed based on 10 SNPs, confirmed to have MAF >0.05 in this study. Four haplotypes (H1, H2, H3 and H4) represented most of the Korean population (>94%). The haplotype frequencies among the five ethnic groups revealed that haplotype distributions in Koreans were similar to those of the Japanese and Chinese, but were quite different to those of the Africans and Europeans. These LD and haplotype data should be useful in drug development and in understanding genetic associations of CYP1A2 with adverse drug effects. These inter-ethnic differences in frequencies of SNPs and haplotypes may help to explain inconsistencies that have been reported in association studies and could contribute to predict the pharmacokinetics and pharmacodynamics of drugs that are metabolized by CYP1A2.