RSS-Feed abonnieren
DOI: 10.1055/s-0038-1641819
Critical evaluation of DNA methylation markers for type-2-diabetes risk prediction
Publikationsverlauf
Publikationsdatum:
26. April 2018 (online)
Recent epigenome wide association studies (EWAS) indicate a benefit for type-2-diabetes (T2D) risk prediction when leucocyte DNA methylation information is known. Especially methylation at two CpG loci, cg06500161 (ABCG1) and cg11024682 (SREBF1), are promising EWAS-markers to improve risk scoring. Therefore, we evaluated these beyond population-based approaches in the context of personalized medicine.
By bisulfite pyrosequencing, methylation of both CpG sites was measured in blood of obese and lean subjects without T2D (n = 177), and in liver and visceral adipose tissue (VAT) of obese subjects with or without T2D (n = 99). Linear regression analysis was included to access a relationship between DNA methylation and other metabolic parameters. A methylation risk score (MRS) was calculated as model of additive effects based on both positions. For optimal risk modelling, we included analysis of single nucleotide polymorphisms (SNPs) in both genes.
Our results show a tissue specific correlation between methylation at cg11024682 and body mass index (BMI) in blood and liver. Leucocyte methylation of both loci is significantly increased in subjects with high HOMA-IR (> 2.5) and correlates with fasting glucose. The neighboring SNP rs9982016 influences methylation at cg06500161 without being associated with T2D itself. Calculation of an MRS using DNA methylation information from blood or VAT enabled a stratification of insulin resistant and sensitive subjects.
Our results highlight the tissue specificity of DNA methylation. Nevertheless, methylation at both loci in blood proved to be a promising approach for risk assessment and could be valuable for personalized T2D risk prediction in the future.
#