Introduction

The cytochrome P450 enzymes (CYP) metabolize many therapeutic drugs, endogenous hormones and xenobiotic toxins/carcinogens. Genetic polymorphisms in the CYP genes may cause differences in enzyme activities or gene expression, which may be responsible for interindividual and interethnic variabilities in drug response and carcinogenetic susceptibility.1, 2, 3

CYP2C19 is one of the most polymorphic CYP genes in diverse racial groups.4, 5 Two variant alleles account for the majority of the defective genotypes, namely CYP2C19*2 (rs4244285), which carries a 681G>A change in exon 5 causing an aberrant splice site, and CYP2C19*3 (rs4986893), which has a 636G>A change in exon 4 producing a premature stop codon. It has been recognized that the prediction of CYP2C19 phenotypes can be achieved by genotyping CYP2C19*2 and CYP2C19*3 in a population.6 On the basis of CYP2C19 genotypes, individuals can be grouped into poor metabolizer genotypes (PMs, the sum of homozygous and heterozygous genotypes of CYP2C19*2/*2, *3/*3 and *2/*3) and extensive metabolizer genotypes (EMs, CYP2C19*1/*1, *1/*2 and *1/*3).7 CYP2C19 PMs may suffer undesirable adverse effects with a normal dose of a drug inactivated by CYP2C19, and may also show decreased responses to drugs that need to be activated by CYP2C19.

CYP2E1 is known to metabolize and activate many low molecular weight compounds, drugs and procarcinogens. The two single-nucleotide polymorphisms (SNPs) (−1053C>T, rs3813867 and −1293G>C, rs2031920) in the 5′ flanking transcription regulatory region of this gene are in linkage disequilibrium, namely as CYP2E1*5B. This allele is associated with a higher transcriptional level of CYP2E1 expression and an increased risk of hepatocarcinogenetics.8 Moreover, another SNP of CYP2E1*6 (7632T>A, rs6413432) was found in intron 6 of the CYP2E1 gene, and the higher frequency of CYP2E1*6 was associated with the alcohol-related hepatic disease.9 Recently, it has been shown that the frequencies of the CYP2E1*5B and *6 alleles in Orientals were significantly higher than those in Caucasians.10

In our earlier studies, the frequencies of genetic polymorphisms in the CYP3A5 and MDR1 (multidrug resistance 1) genes in the Chinese Uygur and Kazakh populations, who reside in the Xinjiang Autonomous Region of northwestern China with genetic admixture between Orientals and Caucasians, were observed and compared with that of the Chinese Han population.11 The purpose of this study was to further investigate the frequencies of the functional SNPs in the CYP2C19 and CYP2E1 genes in the three mainland Chinese Kazakh, Uygur and Han populations, and to compare them with those previously reported in the Caucasian population.

Materials and methods

A total of 359 unrelated mainland Chinese healthy participants, including 103 Chinese Han, 107 Chinese Kazakh and 149 Chinese Uygur, were recruited for the genotyping study. Each participant had the same ethnic origin for at least three generations. All the Chinese Uygur and Kazakh participants resided in the Xinjiang Autonomous Region of China. The study protocol was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Beijing University. Written informed consent was obtained from all participants.

CYP2C19 and CYP2E1 genotyping were performed by PCR followed by restriction fragment length polymorphism similar to previously described methods with little modifications.5, 10 Fragments of the genomic regions of the selected SNPs were amplified by a PTC-100 Peltier Thermal Cycler (MG Research Inc., Waltham, MA, USA). The detailed information regarding primers, conditions and products of PCR amplification, and restriction endonuclease digestion for genotyped SNPs is listed in Table 1.

Table 1 Selected primers, conditions and products of PCR amplification and restriction endonuclease digestion for the CYP2C19 and CYP2E1 genotyping
Full size table

Distributions of genotypes and alleles were compared by χ2 and Fisher's exact tests between different ethnic groups. P<0.05 was considered to be significant in the comparison.

Results and discussion

Both frequencies of the CYP2C19 and CYP2E1 genotypes met the Hardy–Weinberg equilibrium in three ethnic groups. Similarly, frequencies of the CYP2C19 and CYP2E1 alleles were within the 95% confidence interval.

The genotype distributions of CYP2C19 in the three mainland Chinese ethnic groups are shown in Table 2. The frequencies of the CYP2C19 PM genotypes in the Chinese Kazakh (7.5%, P<0.05) and Uygur (8.1%, P<0.05) populations were significantly lower than that in the Chinese Han population (16.5%), but higher than that reported in the Caucasian population (1.7%, P<0.01). The present data support earlier results7, 12, 13, 14 and suggest that the frequencies of the CYP2C19 PM genotypes in the mainland Chinese Kazakh and Uygur populations exist between those in the Chinese Han and Caucasian populations.15

Table 2 Frequencies of the CYP2C19 genotypes and alleles in mainland Chinese Han (n=103), Kazakh (n=107) and Uygur (n=149) ethnic populations, compared with those reported earlier in Caucasians (n=360)5
Full size table

As shown in Table 2, the frequencies of the prevalent defective allele of CYP2C19*2 in both the Chinese Kazakh (15.4%, P<0.01) and Uygur (16.1%, P<0.01) populations were significantly lower than that in the Chinese Han population (28.8%). It was reported that the allele frequency of CYP2C19*3 had been regarded as a unique Asian variant allele that accounted for the CYP2C19 PM phenotype in Oriental populations, which are a rare occurrence in the Caucasian populations.5 In the present results, the CYP2C19*3 allele was observed in the Chinese Kazakh (8.0%) and Uygur (9.4%) populations, and both frequencies were similar to that of the Chinese Han population (7.2%), but significantly higher than that in the Caucasian population (0%, P<0.01). We speculated that the possibly higher frequencies of CYP2C19*3 in the Chinese Uygur and Kazakh participants might be due to the genetic Eurasian admixture between the Caucasoid and Mongoloid populations, namely genetic excursion induced by emigration along the ancient Silk Road.16

As shown in Table 3, at the CYP2E1 polymorphic sites digested by Rsa I or Pst I endonucleases, frequencies of the CYP2E1 genotypes associated with higher gene transcription (the sum of genotypes of heterozygous CYP2E1*1A/*5B and homozygous CYP2E1*5B/*5B) in the Chinese Han population (35.9%) were significantly higher than that reported earlier in the Caucasian population (6.8%, P<0.01).10 Intermediated frequencies were observed in the Chinese Kazakh (19.6%, P<0.01) and Uygur (22.8%, P<0.05) populations. Similarly, for CYP2E1*6 digested by the Dra I enzyme, compared with the frequency in the Chinese Han population (44.7%), lower frequencies of CYP2E1 variants (the sum of CYP2E1*1A/*6 and CYP2E1*6/*6 genotypes) were observed in the Chinese Kazakh (28.0%, P<0.05) and Chinese Uygur populations (33.6%).

Table 3 Frequencies of the CYP2E1 genotypes and alleles in mainland Chinese Han (n=103), Kazakh (n=107) and Uygur (n=149) ethnic populations, compared with those reported earlier in Caucasians (n=264)10
Full size table

Furthermore, the allele frequency of CYP2E1*5B in the Chinese Han (19.4%, P<0.01), Kazakh (11.2%, P<0.01) and Uygur (12.1%, P<0.01) populations was significantly higher than that reported in the Caucasian population (3.8%). However, compared with the CYP2E1*6 allele frequency in the Caucasian population (11.4%), significant differences were observed in the mainland Chinese Han (26.2%, P<0.01) and Chinese Uygur (18.8%, P<0.05) populations, but no significant difference was observed in the Chinese Kazakh population (14.5%), leading to speculation that this discrepancy might be due to the ethnic complexity of this Eurasian admixed population. The present results support the proposition that the P450 genetic distances were well correlated with the geographic distances across Eurasian populations,17 and indicate potential relevance to clinical decision making.

In summary, the present research shows that frequencies of the functional SNPs in the CYP2E1 and CYP2C19 genes vary among the mainland Chinese Han, Kazakh and Uygur populations. Both of the latter were origins of Eurasian admixture between the Oriental and Caucasian populations. These results suggest that disease susceptibilities and drug responses associated with enzyme activities of CYP2C19 and CYP2E1 may vary in the diverse ethnic populations in mainland China.