AUTHOR=Zhang Qiuyi , Gao Ying , Lin Shuchun , Goldin Lynn R. , Wang Yonghong , Stevenson Holly , Edelman Daniel C. , Killian Keith , Marti Gerald , Meltzer Paul S. , Xiang Song , Caporaso Neil E. 

TITLE=Genome-wide DNA methylation profiling in chronic lymphocytic leukaemia

JOURNAL=Frontiers in Genetics

VOLUME=Volume 13 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2022.1056043

DOI=10.3389/fgene.2022.1056043

ISSN=1664-8021

ABSTRACT=Background: DNA methylation aberrations are widespread among the malignant B lymphocytes of patients with chronic lymphocytic leukaemia (CLL), suggesting that DNA methylation might contribute to the pathogenesis of CLL. 
Aim: We aimed to explore the different methylation positions (DMPs) associated with CLL and screen the differentially methylated and expressed genes (DMEGs) by combining public databases. We aimed to observe the direction of each DMEG in CLL based on the DMPs in promoter and the body respectively to narrow down DMEGs. We also aimed to explore the methylation heterogeneity of CLL subgroups and the effect of B cell maturation on CLL.
Methods: We reported a genome-wide DNA methylation association study using the Infinium HumanMethylation450 BeadChip, profiling the DNA methylation of CD19+ B cells from 48 CLL cases and 28 healthy controls. By integrating methylation and expression data, gene sets were jointly screened, and then the relationship between methylation sites in promoter and body region and expression was explored. Support vector machine (SVM) classification algorithm was used to identify subgroups of CLL cases based on methylation pattern, and the effect of B-cell differentiation related CpGs on CLL-related sites was observed.
Results: We identified 34 797 DMPs related to CLL across the genome, most of which were hypomethylated; the majority were located in gene body regions. By combining these DMPs with published DNA methylation and RNA sequencing data, we detected 26 244 replicated DMPs associated with 1 130 genes whose expression were significantly different in CLL cases. Among these DMEGs, 9 low expressed DMEGs were selected with hypermethylated in promoter and hypomethylated in body region, and 83 high expressed DMEGs were selected with both hypomethylated in promoter and body region. The 48 CLL cases were divided into 3 subgroups based on methylation site by SVM algorithm. Nearly 90% of the sites associated with B cell subtypes were found in CLL-related DMPs.
Conclusions: The DNA methylation pattern was altered across the genome in CLL patients. Further studies are warranted to confirm our findings and identify the underlying mechanisms through which these methylation markers are associated with CLL.