A pontine-specific axonal niche supports de novo gliomagenesis

Abstract

Diffuse intrinsic pontine gliomas (DIPGs), a major type of pediatric high-grade gliomas located in the pons, are the leading cause of death in children with brain cancer. A subset (20-25%) of DIPGs harbor a lysine 27-to-methionine (K27M) mutation in HIST1H3B, which encodes histone H3.1, and an activating ACVR1 mutation. The occurrence of this pair of mutations in DIPGs, but not in pediatric gliomas in other anatomical locations, suggests the existence of a pontine-specific niche that favors DIPG gliomagenesis. Unfortunately, the identity of the underlying pontine niche remains elusive as available mouse models fail to recapitulate the anatomic specificity that characterizes DIPGs. Herein we show that the trigeminal root entry zone (TREZ), a pontine structure where several major axon tracts intersect, is enriched with proliferating oligodendrocyte-lineage cells during brainstem development. Introducing both H3.1K27M and activating Acvr1 and Pik3ca mutations (which co-occur frequently with H3.1K27M in human DIPGs) into the mouse brain leads to rapid gliomagenesis. This pathology recapitulates the pons specificity of DIPGs as glioma cells proliferate on axon tracts at the TREZ. We further show that a hyaluronan receptor important for cell stemness (HMMR) plays a key role in glioma cell proliferation at the TREZ. We propose that H3.1K27M and its co-occurring mutations drive pontine specific gliomagenesis by inducing a proliferative response of oligodendrocyte-lineage cells with enhanced stemness on large TREZ axon tracts.