Whole-epigenome analysis in multiple myeloma reveals DNA hypermethylation of B cell-specific enhancers

Abstract

While analyzing the DNA methylome of multiple myeloma (MM), a plasma cell neoplasm, by whole-genome bisulfite sequencing and high-density arrays, we observed a highly heterogeneous pattern globally characterized by regional DNA hypermethylation embedded in extensive hypomethylation. In contrast to the widely reported DNA hypermethylation of promoter-associated CpG islands (CGIs) in cancer, hypermethylated sites in MM, as opposed to normal plasma cells, were located outside CpG islands and were unexpectedly associated with intronic enhancer regions defined in normal B cells and plasma cells. Both RNA-seq and in vitro reporter assays indicated that enhancer hypermethylation is globally associated with down-regulation of its host genes. ChIP-seq and DNase-seq further revealed that DNA hypermethylation in these regions is related to enhancer decommissioning. Hypermethylated enhancer regions overlapped with binding sites of B cell-specific transcription factors (TFs) and the degree of enhancer methylation inversely correlated with expression levels of these TFs in MM. Furthermore, hypermethylated regions in MM were methylated in stem cells and gradually became demethylated during normal B-cell differentiation, suggesting that MM cells either reacquire epigenetic features of undifferentiated cells or maintain an epigenetic signature of a putative myeloma stem cell progenitor. Overall, we have identified DNA hypermethylation of developmentally regulated enhancers as a new type of epigenetic modification associated with the pathogenesis of MM.

Figure 1.

Whole-genome DNA methylation data of neoplastic and normal plasma cells. (A) Principal component analysis of HumanMethylation450 BeadChip data in MGUS, MM, and NPC from tonsils or bone marrow. (B) Unsupervised hierarchical clustering analysis of MM and MGUS patient samples as well as NPC samples from tonsil or bone marrow from healthy donors (the 20,000 CpG sites with most variable methylation values were used for this analysis). (HyperD) Hyperdiploid MM sample; [t(IGH)] MM sample with IGH translocation; [del(TP53)] MM sample with deletion of TP53. (C) Bar plot showing the coefficient of variation of median methylation values per case in ALL, CLL, DLBCL, MGUS, and MM. (D) Circular representation of DNA methylation levels of purified plasma cells of two MM patient samples (MM1 and MM2) compared to purified NPC obtained from bone marrow from healthy donors. Histograms represent CpG methylation levels averaged in 10-Mbp genomic windows. This representation with low resolution indicates that the shift of methylation levels in MM1 and MM2, as compared to NPCs, takes place throughout the genome and not only in particular chromosomal regions. (E,F) Scatter plots and density color codes for DNA methylation data of all autosomes. Pairwise comparisons of MM1 sample to NPC (E) and MM2 sample to NPC (F) are shown.

Figure 2.

Differential DNA methylation in MM and MGUS as compared to NPCs. (A,B) Differentially methylated CpGs using WGBS data in (A) MM1 versus NPC and (B) MM2 versus NPC. (C,D) Differentially methylated CpGs identified using the HumanMethylation450 BeadChip data in (C) MGUS versus NPC samples and (D) MM versus NPC samples. (E) Heatmap of significantly hypomethylated (left) or hypermethylated CpGs (right) in MM samples versus NPC samples. (F) Relative distribution of differentially hypo- or hypermethylated CpGs across different functional chromatin states of the genome using WGBS data (left) or HumanMethylation450 BeadChip data (right) as compared to their respective backgrounds (first column). The numbers inside each cell point to the percentage of CpGs belonging to a particular chromatin state. (WG) Whole-genome; (HyperM) hypermethylated CpGs; (HypoM) hypomethylated CpGs.

Figure 3.

Functional and transcriptional analysis of hypermethylated enhancer regions in MM. (A) Percentage of hypermethylated CpGs associated with enhancer regions in MGUS and MM patient samples using HumanMethylation450 BeadChip data. (B) Density plot of correlation coefficients between methylation levels of hypermethylated enhancers and the expression of their associated genes (cyan) using RNA-seq data. As control, intronic CpGs of the same genes were studied excluding the enhancer associated ones (black dotted line). For this analysis, we used 663 CpGs (out of 794) annotated to 574 genes with available gene expression, and 8956 CpGs in nonenhancer intronic regions of the same genes. (C) A snapshot of the UCSC Genome Browser showing the promoter (left) and the 3′ intronic enhancer region (right) of the SLC15A4 gene. Displayed tracks include the chromatin state characterization in IMBCs and ChIP-seq data for H3K27ac, H3K4me1, and H3K4me3. DNA methylation levels of NPCs and MM patient samples measured by WGBS and HumanMethylation450 BeadChip are also shown. (D–F) Correlation analysis between DNA methylation levels of the hypermethylated enhancers and expression of the associated gene (left). In addition, we display the luciferase reporter activity data of the analyzed enhancer region (right) located in the intron of (D) SLC15A4, (E) PVT1, and (F) NCOR2. (MP) Minimal promoter; (MET) methylated; (UNMET) unmethylated.

Figure 4.

DNA methylation and chromatin features of hypermethylated enhancers in MM in the context of normal B-cell differentiation. (A) DNA methylation levels of 794 enhancer-associated CpGs in seven B-cell differentiation stages, MGUS and MM patient samples, as well as ESCs (H1), IMBCs (GM12878), and U266-MM cell lines. (B) ChIP-seq levels of H3K4me1 and H3K27ac, and DNase-seq data of 794 enhancer-associated CpGs in ESCs, IMBCs, NPCs, and the U266 MM cell line. (C) Density plot of H3K4me1, H3K27ac, and DNase levels in ESCs, IMBCs, NPCs, and the U266 MM cell line. Among the 794 enhancer-associated CpGs in MM, those hypermethylated in U266 are shown in the top panel, whereas those unmethylated in this cell line appear at the bottom. (MPP-CLP) Hematopoietic multipotent progenitors–common lymphoid progenitors; (ESCs) embryonic stem cells; (IMBCs) immortalized mature B cells; (NPCs) normal plasma cells; (U266-MM) multiple myeloma derived cell line U266.

Figure 5.

B cell-specific TFs' expression correlates with DNA methylation levels of their binding sites in MM. (A) The upper part shows that hypermethylated enhancers are enriched for binding sites of B cell-specific TFs. The lower part displays the correlation coefficients between TF expression and mean DNA methylation level of their respective binding sites with hypermethylation in MM. (*) P < 0.05. (B–D) Scatter plots showing the association between TF expression and mean methylation level of their respective binding sites. Normal plasma cells are shown in blue and multiple myeloma samples are depicted in red. (E) Correlation matrix of expression levels of TFs among the 11 MM cases with available RNA-seq data. Only those TFs from A with at least 1.5-fold enrichment (in log₂) were used for this analysis.