Glioblastoma epigenome profiling identifies SOX10 as a master regulator of molecular tumour subtype - cell line gene expression microarray data

Glioblastomas in adult patients are classified into four subtypes, IDH, MES, RTK I, and RTK II, based on DNA-methylation and RNA-expression data. Tumour subtype transitions are common during treatment, and transitions to the mesenchymal (MES) subtype are associated with therapy resistance and adverse prognosis. Here, we present DNA methylome and histone modification data of glioblastoma primary tumours and find that glioblastoma subtypes differ in their enhancer landscapes. Using Core Regulatory Circuitry analysis of chromatin and orthogonal analysis of RNA-derived gene regulatory networks, we identified 38 subtype Master Regulators whose cell population-specific activities we further mapped in single-cell RNA sequencing data. These analyses identified the oligodendrocyte precursor marker and chromatin modifier SRY-Box 10 (SOX10) as a master regulator in RTK I tumours. In vitro functional studies demonstrated that SOX10 loss causes a subtype switch analogous to the proneural-mesenchymal Transition observed in patients at the transcriptomic, epigenetic and phenotypic levels. This subtype transition is dependent on the activity of the SOX10 and enhancer co-factor Bromodomain Containing 4 (BRD4). Sox10 repression in an in vivo syngeneic graft glioblastoma mouse model results in increased tumour invasion, immune cell infiltration and significantly reduced survival, reminiscent of progressive human glioblastoma. These results identify SOX10 as a bona fide master regulator of the RTK I subtype, with both tumour cell-intrinsic and microenvironmental effects, highlighting that both glioblastoma cell plasticity and their tumour-microenvironment interactions are important contributors to tumour phenotypes. Two RTK I-like cell lines were transduced with experimental systems that allow repression of the SOX10 gene: LN229 with CRISPRi, and ZH487 with shRNA. In each case, 3 independent SOX10-targeting reagents were used. In this series, expression microarray profiling was used to compare the transcriptomes of the two models with and without repression of SOX10 expression. To identify SOX10-regulated genes, limma differential expression analysis was performed, comparing the expression of control (mock and NT) to the SOX10-repressed samples.