Research Project
10269904
PROJECT SUMMARY We were one of two groups to first identify amino acid substitutions on histone H3 variants (oncohistones) in human disease. In the past funding cycle, we showed that, beyond their initial identification in specific subtypes of high-grade gliomas and bone tumors, these oncohistones also occur in subgroups of sarcomas, head and neck squamous cell carcinomas, and acute myeloid leukemias. Working with members of this Program Project, we showed how H3K27M and H3K36M (K-to-M substitutions) lead to stalled development and blocked differentiation through epigenome rewiring as they promote aberrant spreading/redistribution of key chromatin marks from their initial boundaries, which in turn facilitate oncogenesis. Furthermore, our collaboration with Lewis/Garcia labs uncovered that, EZHIP, a testis-specific protein aberrantly expressed in posterior fossa group A ependymomas, biochemically mirrors H3K27M oncohistone. Our preliminary data indicate that EZHIP is also aberrantly activated in subgroups of osteosarcomas. In contrast to the growing information on K-to-M mutations, however, there is limited knowledge on the oncogenic function, molecular mechanisms and tumor-specific vulnerabilities of the oncohistone-mimic EZHIP and mutations affect H3.3G34. In large part, this is due to a paucity of resources and models that faithfully recapitulate the cellular effects mediated by these alterations. Therefore, a major goal of this project is to develop and employ novel patient sample-, cell culture- and animal model-based systems to model oncohistone-associated cancers and investigate the underlying pathogenic mechanisms. Nearly all G34 mutations affect histone variant H3.3, and while H3.3G34R/V are frequently found in pediatric gliomas, giant cell tumors are predominantly defined by H3.3G34W mutations. This dichotomy is reflected by the distinct phenotypes of the H3f3a G34W, R or V knock-in mouse models we have developed. We aim to delineate the mechanisms behind the codon-specific phenotypes resulting from H3.3G34 mutations, the role of specific mutational partnership for G34R/V-driven tumorigenesis, and the necessity of H3.3 variant in mediating G34-dependent phenotypes (Aim 1). We will investigate the role of EZHIP in development, how its aberrant expression is achieved and promotes oncogenesis in osteosarcomas and ependymomas - comparing and contrasting to H3K27M mutation (Aim 2). Lastly, we will build on our findings of the viral mimicry induced by K-to-M mutations to assess the degree and nature of immune infiltration in these tumors at baseline and upon treatment of epigenetic drugs (Aim 3). Notably, these studies are enabled by the unique syngeneic mouse models and CRISPR/Cas9-edited isogenic patient-derived tumor cell lines we have generated for various oncohistones and EZHIP. This project will synergize with other efforts of this Program Project to provide mechanistic insights and relevant and reliable pre-clinical models to the community, building a knowledge and resource base upon which targets amenable to therapy that can be conceived and validated.
