Whole-genome sequencing of glioblastoma reveals enrichment of non-coding constraint mutations in known and novel genes

Abstract

Background: Glioblastoma (GBM) has one of the worst 5-year survival rates of all cancers. While genomic studies of the disease have been performed, alterations in the non-coding regulatory regions of GBM have largely remained unexplored. We apply whole-genome sequencing (WGS) to identify non-coding mutations, with regulatory potential in GBM, under the hypothesis that regions of evolutionary constraint are likely to be functional, and somatic mutations are likely more damaging than in unconstrained regions. Results: We validate our GBM cohort, finding similar copy number aberrations and mutated genes based on coding mutations as previous studies. Performing analysis on non-coding constraint mutations and their position relative to nearby genes, we find a significant enrichment of non-coding constraint mutations in the neighborhood of 78 genes that have previously been implicated in GBM. Among them, SEMA3C and DYNC1I1 show the highest frequencies of alterations, with multiple mutations overlapping transcription factor binding sites. We find that a non-coding constraint mutation in the SEMA3C promoter reduces the DNA binding capacity of the region. We also identify 1776 other genes enriched for non-coding constraint mutations with likely regulatory potential, providing additional candidate GBM genes. The mutations in the top four genes, DLX5, DLX6, FOXA1, and ISL1, are distributed over promoters, UTRs, and multiple transcription factor binding sites. Conclusions: These results suggest that non-coding constraint mutations could play an essential role in GBM, underscoring the need to connect non-coding genomic variation to biological function and disease pathology.

Characterization of the Genomic and Immunologic Diversity of Malignant Brain Tumors through Multisector Analysis

“…1d). This high frequency of TERT promoter mutations mirrors earlier work estimating frequencies upwards of 70-80% among GBM (32,33). Additionally, the frequency of TP53 alterations is consistent with prior studies observing mutations in 30-40% of GBM samples, while the frequency of EGFR mutations is slightly lower (21,33).…”

“…This high frequency of TERT promoter mutations mirrors earlier work estimating frequencies upwards of 70-80% among GBM (32,33). Additionally, the frequency of TP53 alterations is consistent with prior studies observing mutations in 30-40% of GBM samples, while the frequency of EGFR mutations is slightly lower (21,33). We next assessed the prevalence of copy number alterations (CNAs) within our cohort and identified classical GBM-associated changes such as frequent chromosome 7 amplification encompassing EGFR (17/18), chromosome 9 deletions of the CDKN2A locus (15/18), and chromosome 10 deletions of the PTEN locus (17/18) (Fig.…”

Human Endogenous Retroviruses in Glioblastoma Multiforme

“…For example, the MLT1M-int_LTR_ERVL-MaLR elements on chromosome 7 were downregulated in GBM, along with the downregulation of the adjacent gene, CNTNAP2 (contactin-associated protein 2). CNTNAP2 is an important neurogenesis gene that some studies suggest is a tumor suppressor gene in glioma [ 72 , 73 ]. This observation could provide some clues to the mechanisms and functional implication of differentially expressed elements.…”

Pan-genomic characterization of high-risk pediatric papillary thyroid carcinoma

“…For non-coding variants, a similar strategy to Sakthikumar et al was employed ( Sakthikumar et al 2020 ). First, false-positive somatic mutations were likely excluded by filtering out variants with allele frequencies of 0.1% or more in gnomAD ( Karczewski et al 2020 ).…”