DIPG-14. INTEGRATED DIFFUSE INTRINSIC PONTINE GLIOMA (DIPG) NEXT-GENERATION SEQUENCING REVEALS EPIGENETIC DYSREGULATION OF GENE EXPRESSION AND BROMODOMAIN INHIBITION AS A NOVEL THERAPEUTIC TARGET

Amanda Saratsis,Andrea Piunti,Tina Huang,Elizaeth Bartom,Patrick A. Ozark,Stacy A. Marshall,Jin Qi,Rintaro Hashizume,Rishi Lulla,C. David James,Ali Shilatifard
DOI: https://doi.org/10.1093/neuonc/noy059.107
2018-01-01
Neuro-Oncology
Abstract:Diffuse intrinsic pontine glioma (DIPG) has the highest mortality of all pediatric solid tumors. Histone H3 mutation (H3K27M) occurs in 80%, altering chromatin structure and function. To characterize the effects of H3K27M as an oncogenic driver, we analyzed the transcriptome and epigenome in DIPG cell lines. Transcriptomes (RNA-Seq) and genomic enrichment of Histone proteins (ChIP-Seq) were characterized in H3K27M DIPG (n=7), wild-type pediatric high-grade glioma (n=3), neural stem cells (n=1) and astrocyte cell lines (n=1). Reads were aligned (UCSC hg19, Tophat), normalized and quantified (HTSeq), and formatted \ for peak calling (SICER) and genomic enrichment mapping (Ensembl). Differential expression and enrichment patterns were identified (edgeR) then analyzed for biological relevance (TopGO, IPA). Epigenetic therapeutic targets were studied in vitro and in vivo. Unsupervised clustering revealed two DIPG subgroups, distinct from control and H3 wild-type lines. Increased EZH2 (p<0.05), BRD4 (p<0.05) and pol II activity at proximal promoter and enhancer regions (p<0.01) were identified in H3K27M lines relative to wild-type. H3K27M co-localized with H3K27ac, pol II, BRD2 and BRD4 at actively transcribed genes, with PRC2 excluded from H3K27M occupied regions. Neuronal differentiation was repressed in H3K27M lines (p<0.01), consistent with morphologic changes observed after EZH2 inhibition in vitro. BRD inhibition suppressed tumor growth in vitro and in vivo. We present the largest epigenetic analysis of rare DIPG cell lines, implicating tumorgenetic pathways in H3K27M mutants. Our results indicate that H3K27M alters the balance between H3K27me3 and H3K27ac, creating a pro-oncogenic chromatin state that promotes DIPG, and highlighting novel epigenetic therapeutic targets.
What problem does this paper attempt to address?