A growing number of cancers, especially in children and young adults, are due to epigenetic dysfunction, which promotes their genesis, progression and metastatic processes. Our group was one the first to identify a histone mutation in human disease, irrefutably linking epigenetic deregulation to cancer. High-frequency recurrent somatic
mutations at specific residues in histone H3 variants occur in a particularly lethal
form of brain tumor, high-grade gliomas (HGG) affecting children and young
adults. Since this initial discovery, these oncohistones as we label which occur on histone H3 variants, K-to M substitutions on residues 27 and 36 and H3.3 G34 R, V and W mutations have been shown to be at the core of tumorigenesis in several cancers including, bone tumours, leukemias, and head and neck cancers [1,2].
Our recent work shows these oncohistones act by reprogramming the epigenome leading to stalled development and impaired differentiation, the - “Peter Pan Syndrome” - as we name it. Indeed, in the context of K27M, building on seminal findings showing this substitution inhibits methyltransferase activity of the polycomb repressive complex 2 (PRC2), we showed that cells are stuck in a progenitor state where they indefinitely cycle as they are unable to spread the repressive H3K27me3 and me2 marks to achieve further differentiation. In H3K36M tumors, defective spread of the abundant H3K36me2 mark leads to pervasive invasion of the repressive H3K2me3 mark and dilution of PRC1 repression and prevents subsequent differentiation.
For both mutations, we show that redistribution of antagonistic marks from boundaries they are usually contained in, H3K36me2 and H3K27ac in K27M and H3K27me2/me3 in K36M, may participate in oncogenesis or generates exquisite vulnerabilities as we showed [3-5]. G34-mutants exquisitely target noncanonical H3.3, and specific mutations arise in specific cancers: G34R and rarely G34V mainly occur in HGG of the cerebral cortex in adolescent and young adults, while H3.3G34W arise in a specific bone cancer, giant cell tumors of the bone (GCT) and never occur in the brain. Our findings suggest different mechanisms of oncogenesis associated with specific G34 amimo acid substitutions. Last we helped uncover Enhancer of Zeste Inhibitor Protein (EZHIP), a K27M oncohistone -mimic, which may act to induce “PRC2 poising” during specific developmental windows and/or progenitor states to prevent untimely differentiation and repression of transcriptional developmental programs. We show this physiological stalling of PRC2 can possibly be co-opted by aberrant/persistent expression of this gene. Last, we identify a degree of viral mimicry in K-to-M mutagenesis that can lead to therapies by further engaging the immune system in these cancers. Collectively, oncohistones and oncohistone-mimics lead to the redistribution of key antagonistic chromatin marks stalling differentiation and normal development. We propose this new, previously unsuspected mechanism of oncogenesis leads to aberrant PRC1/PRC2 function and may provide exquisite therapeutic vulnerabilities.