Research on genetic variant characteristics in ADME genes based on whole-exome sequencing in the Han Chinese population

Background

Genetic variants in absorption, distribution, metabolism, and excretion (ADME) genes affect drug efficacy and side effects. Whole-exome sequencing (WES) effectively identifies these variants. Findings from Western populations may not apply to the Han Chinese population due to ethnic differences.

Objective

This study aimed to investigate genetic variants, metabolic phenotypes, and drug impacts related to ADME genes in the Han Chinese population using WES data.

Methods

ADME genes and drug profiles were sourced from multiple databases. WES data were collected from 357 samples at Third Xiangya Hospital and 5,000 samples from the “HuaBiao Project.” After WES, fastp was used for quality control; BWA, samtools, and the Picard software were used for comparison, sequencing, and de-duplication; GATK was used for base quality score recalibration; and variant quality score recalibration was carried out after detecting the variants, and the variants were annotated using ANNOVAR. ADME genes and drug profiles were obtained through PharmGKB and other databases, and then all variants in the exonic regions of ADME genes were extracted. The distributional characteristics of these variants, ethnic differences, and metabolic phenotype distribution of important ADME genes, such as CYP2B6, were analyzed. Prediction of deleterious variants of ADME gene employed the ADME gene variant prediction framework, and western blotting and enzyme assays were used to validate the impact of harmful variants.

Results

We integrated 604 ADME genes and 972 drugs. There were 33,925 single nucleotide polymorphisms (SNPs) and 484 insertion-deletions (InDels) in 5,357 Han Chinese WES samples; 88.9 % were rare. Of the SNPs, 60.9 % are new functional mutations in enzyme and transporter genes; InDels also affected these genes. We discovered Chinese-specific variants in key ADME genes and ethnic-specific metabolic phenotypes that could affect drug use. Finally, we screened 3,657 potentially harmful mutations; 45.6 % are novel. Western blotting and enzyme activity assays confirmed several harmful mutations significantly reduced CYP2C19 gene expression and function (P < 0.01).