The use of genome-wide DNA methylation microarray to study both the common and rare diseases

Abstract

DNA methylation plays many important roles in human physiology such as imprinting and X chromosome inactivation (XCI), and therefore disruption in DNA methylation can lead to disease development. The objective of this study is to study the role of DNA methylation in both the common and rare disease, using Systemic Lupus Erythematosus (SLE) and chromosome X translocation as the example respectively. The genome-wide DNA methylation analysis was studied by Illumina HumanMethylation450 BeadChip, which is the microarray that allows the detection of more than 480,000 CpG sites across 99% of reference sequence genes. Numerous studies have shown that hypomethylation occurred in specific genes of SLE patients, and genome-wide DNA methylation analysis has never been performed in Chinese before. Since DNA methylation can be ethnicity-specific, we aim to identify the Chinese-specific DNA methylation pattern in SLE patients. Microarray results showed that differential DNA methylation changes were loci-specific, where 36 CpG sites showed loss of DNA methylation while 8 of them showed gain of it, representing 24 genes and 7 genes respectively. In order to replicate the microarray findings, bisulfite pyrosequencing was performed in an additional cohort of 100 patients and 100 controls on four hypomethylated genes, which were selected based on their relevance with immunity. Bisulfite pyrosequencing confirmed the microarray result that hypomethylation occurred in SLE, and were associated with increased in the corresponding gene’s mRNA expressions. Gene ontology analysis revealed that hypomethylated genes identified in the microarray study were overrepresented in type I interferon pathway, where type I interferon has long been implicated in SLE pathogenesis. Therefore this study also support the importance of type I interferon in SLE from the epigenetic point of view. X;autosome translocation is a rare condition and the autosome with chromosome X translocated on can be inactivated by XCI, but DNA methylation change is rarely used to investigate the spread of XCI. In this study, we aim to identify genes subjected to XCI in X;15 translocation using the DNA methylation microarray. Results of microarray showed that 586 CpG sites spanning the long arm of chromosome 15 had DNA methylation gain of more than 20%. Since genes subjected to XCI are known to have gain of DNA methylation in their corresponding CpG-island promoters, the analysis was then focused on CpG sites in these regions, and a total of 75 sites representing 24 genes were hypermethylated. Nearly all of these CpG sites are located in region proximal to the breakpoint, from 15q11.2 to 15q21.3 accounting for 35Mb, suggesting that XCI was spread to the proximal region of 15q. Gain of DNA methylation, especially in the CpG-island promoter, can result in functional inactivation of genes, and therefore could explain the worsen phenotype of the patient. In conclusion, we successfully demonstrated the use of genome-wide DNA methylation microarray in different diseases, allowing the identification of genes or pathways important in diseases and opened the door for further investigation of the effect of these differentially methylated genes on disease.