NEW YORK – New research has uncovered non-coding mutations that can help drive the formation of glioblastoma (GBM) brain tumors through regulatory interactions that alter expression of new and known GBM-related genes, including non-coding variants that affect the promoter of the semaphorin-coding gene SEMA3C.
"Our results show that a specific, evolutionarily conserved, mutation in the vicinity of SEMA3C disrupts the binding of certain proteins whose task is to bind genes and regulate their activity," senior author Karin Forsberg-Nilsson, an immunology, genetics, and pathology researcher at Uppsala University, said in a statement, noting that these and other findings from the study "confirm the importance of the association between genetic alterations in non-coding regions, their biological function, and disease pathology."
As she and her colleagues reported in Genome Biology on Tuesday, they did whole-genome sequencing on tumor and matched normal samples from dozens of treatment-naïve or relapsed wild-type IDH1 GBM cases, including 35 yet-to-be-treated primary GBM cases and four relapsed cases. Within that collection, which included tumors from classical, mesenchymal, and proneural subtypes, they considered both the coding and the non-coding mutations, somatic copy number changes, and other alterations that may contribute to tumor formation, particularly at evolutionarily constrained sites in the genome.
"We apply whole-genome sequencing 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," Forsberg-Nilsson and her co-authors wrote.
Along with alterations affecting coding genes linked to GBM previously, the team's analyses pointed to an over-representation of non-coding mutations falling near 78 GBM-related genes. While non-coding variants appeared to be less common near protein-coding genes not implicated in GBM in past studies, the analysis highlighted suspicious changes in non-coding sequences falling near almost 1,800 genes.
"The catalog of exome mutations across our samples largely mirrors what has previously been described, thus serving to validate the cohort," the authors wrote. "In addition, we report here a range of non-coding constraint mutations that may have functional impact on the disease."
The researchers focused much of their follow-up work on SEMA3C and the dynein cytoplasmic 1 intermediate chain 1 gene DYNC1I1 — genes that are reportedly expressed at higher- and lower-than-usual levels, respectively, in prior studies of glioma — along with more than 40 of the genes that were not implicated in GBM in prior studies.
The authors noted that SEMA3C and DYNC1I1 "show the highest frequencies of alterations, with multiple mutations overlapping transcription factor binding sites."
Consistent with transcription factor binding predictions for the non-coding constraint mutation sites considered, the team noted that available chromatin immunoprecipitation sequence data pointed to a dip in DNA binding in the SEMA3C promoter region in the presence of GBM-related non-coding mutations.
From these and other findings, they suggested that "a better understanding of the non-coding genome of GBM tumors will help in elucidation of functional genetic and epigenetic alterations and consequently may unlock therapeutic opportunities for personalized treatment strategies."