Integrative analysis of DNA methylation suggests down-regulation of oncogenic pathways and reduced somatic mutation rates in survival outliers of glioblastoma

Abstract

The study of survival outliers of glioblastoma can provide important clues on gliomagenesis as well as on the ways to alter clinical course of this almost uniformly lethal cancer type. However, there has been little consensus on genetic and epigenetic signatures of the long-term survival outliers of glioblastoma. Although the two classical molecular markers of glioblastoma including isocitrate dehydrogenase 1 or 2 (IDH1/2) mutation and O6-methylguanine DNA methyltransferase (MGMT) promoter methylation are associated with overall survival rate of glioblastoma patients, they are not specific to the survival outliers. In this study, we compared the two groups of survival outliers of glioblastoma with IDH wild-type, consisting of the glioblastoma patients who lived longer than 3 years (n = 17) and the patients who lived less than 1 year (n = 12) in terms of genome-wide DNA methylation profile. Statistical analyses were performed to identify differentially methylated sites between the two groups. Functional implication of DNA methylation patterns specific to long-term survivors of glioblastoma were investigated by comprehensive enrichment analyses with genomic and epigenomic features. We found that the genome of long-term survivors of glioblastoma is differentially methylated relative to short-term survivor patients depending on CpG density: hypermethylation near CpG islands (CGIs) and hypomethylation far from CGIs. Interestingly, these two patterns are associated with distinct oncogenic aspects in gliomagenesis. In the long-term survival glioblastoma-specific sites distant from CGI, somatic mutations of glioblastoma are enriched with higher DNA methylation, suggesting that the hypomethylation in long-term survival glioblastoma can contribute to reduce the rate of somatic mutation. On the other hand, the hypermethylation near CGIs associates with transcriptional downregulation of genes involved in cancer progression pathways. Using independent cohorts of IDH1/2- wild type glioblastoma, we also showed that these two patterns of DNA methylation can be used as molecular markers of long-term survival glioblastoma. Our results provide extended understanding of DNA methylation, especially of DNA hypomethylation, in cancer genome and reveal clinical importance of DNA methylation pattern as prognostic markers of glioblastoma.

Keywords: DNA methylation; Genome-wide analyses; Glioblastoma; Long-term survivor.

Fig. 1

Differentially methylated sites between LTS-GBM and STS-GBM. a Heatmap of DNA methylation levels measured as beta values: color gradients from blue to red correspond to beta values from 0 to 1. Hierarchical clustering was performed for both of glioblastoma patients (columns) and the differentially-methylated sites between LTS-GBM and STS-GBM (rows). Here, hierarchical clustering of the most significant sites (−log₁₀(FDR) < 5: 13,049 sites) with Euclidean distance and complete linkage was performed for convenient visualization. b Distribution of the mean increased level of DNA methylation in LTS relative to STS (labelled as “LTS-STS” in y-axis) across the sites categorized by their locations from the closest CpG Island (CGI) (see materials and methods). The group of selected significant sites are denoted by pink while the other sites are described by grey. If the distribution is shifted to the positive in y-axis (for example, at “Island”), it means that LTS has in general higher DNA methylation compared to STS while the shift to the negative in y-axis indicates the opposite case (for example, in “OpenSea”). c Same as (b) except that categorization of sites were done by distances from the nearest transcription start site (TSS) (see materials and methods)

Fig. 2

Enrichment of differentially-methylated sites in regulatory histone marks. a Enrichment patterns of differentially-methylated sites with histone marks. For a given histone mark (x axis), Y axis value describes the differences of enrichment proportion between the significantly- differential methylation sites and the insignificant sites. The red denotes hypermethylated sites in island while the blue describes hypomethylated sites in open sea. The size of dot indicates the range of enrichment proportion of the significantly- differential sites. We grouped the proportion to 4 regions for clear visualization (4 dot sizes): 0~0.25, 0.25~0.5, 0.5~0.75, 0.75~1. b The relation between the ChIP-seq signal of H3K27ac and the DNA methylation level (beta) for the hyper-methylated sites in U87MG cell line. Each dot denotes the site. c The same plot as (b) for the hyper-methylated sites in H1 cell line

Fig. 3

Functional implication of hyper- and hypo-methylated sites in LTS-GBM. a Distribution of Pearson correlation coefficients between DNA methylation level (beta) and gene expression measured by RNA-seq (FPKM) across 32 GBM samples in TCGA. The red describes the genes (number of genes: 366) corresponding to the selected hyper-methylated sites in this study while the gray shows the genes (number of genes: 4413) matched to insignificant sites associated with the island. b DNA methylation level around somatic mutations found in 136 TCGA GBM samples by whole exome sequencing. The hypo-methylated sites in LTS-GBM (N = 1802) were compared with the insignificant sites in open sea (N = 5032)

Fig. 4

Hyper- and hypo-methylation in two independent cohorts (TCGA and ‘Australian’). a Heatmap of DNA methylation levels measured as beta values: hierarchical clustering was performed for both of samples (columns) and the sites (rows) either hyper- or hypo-methylation identified in the discovery cohort (“SNU”). The color gradients from blue to red correspond to beta values from 0 to 1. b Summary scores in terms of hyper- and hypo-methylation for each sample in two test cohorts: Each sample, denoted by a dot is assigned to two simple arithmetic averages (x and y axes values) of beta values in hyper- and hypo- methylation sites. The dashed line indicates 0.2 as a decision threshold for LTS-GBM

Fig. 5

Genome-wide DNA methylation pattern of glioblastoma. The genomes of long-term survivors in glioblastoma are differentially methylated relative to short-term survival patients depending on CpG density: hypermethylation near CpG islands (CGIs) and hypomethylation far from CGIs (open sea). The hypermethylation at CGIs frequently occurs around regions with histone marks of active transcription such as H3K27ac, correlating with downregulation of gene expression in cancer progression pathways. The hypomethylated region at open sea are enriched with a histone mark of heterochromatin, H3K9me3. The rate of de novo mutation is high in this region when it is methylated, implying survival advantage of hypomethylation of the region in glioblastoma. In the figure, we highlighted genic regions such as first exon and gene body to emphasize potential effect of perturbed DNA methylation in glioblastoma