Abstract
Histone 3 mutations at lysine 27 (H3K27M) are frequent drivers of midline gliomagenesis, occurring in ~80% of diffuse intrinsic pontine gliomas (DIPG) and leading to widespread H3K27me3 changes through PRC2 inhibition. Although H3K27M mutations appear to cooperate with additional mutations including TP53 and PDGFRA, the exact oncogenic function of H3K27M is unknown, and in vivo models have not yielded tumors driven by H3K27M alone. Here we created a genetic mouse model by microinjection, expressing H3.3K27M under control of the Fabp7 promoter, which is active in the brain from approximately E14.5, primarily in radial glia and astrocyte precursors, but also allows expression in non-brain developing and adult tissues. H3.3K27M expression in the developing brainstem led to SOX10 upregulation via loss of H3K27me3 at the Sox10 promoter, resulting in an expression profile consistent with K27M-mutant DIPG, including an oligodendrocyte precursor cell (OPC) signature along with RAS and EMT activation. H3.3K27M induced tumors in multiple organs that were driven by K27M alone and, when combined with Trp53 loss, led to primary high-grade gliomas. The tumors had a cell-type independent expression signature featuring RAS and MYC activation, which overlapped with human DIPG and pointed to a core K27M transcriptome. Furthermore, as in human DIPG, mouse tumors spontaneously mutated the RAS pathway and MYC to lock in pathway activation. Our data suggest that RAS and MYC are core pathways that will need to be targeted in order to effectively treat this devastating disease.