Epigenetic landscape of gliomas and their implications for tumor biology and treatment of cancer
Description: Glioblastoma is one of the most difficult tumors to treat, notorious for its resistance to treatment. Here we propose to target the tumor epigenome in glioblastoma taking two different approaches: The failures of single agents tested in clinical trials have dramatically shown that combination therapies are required for improvement of outcome of glioblastoma patients. However, the rational choice for successful combination therapy remains a challenge. For our first approach, we propose an innovative strategy to identify potentially cooperating drugs. We aim at identifying druggable vulnerabilities of cancer relevant pathways revealed upon disturbing the tumor cells by Bromodomain inhibitors (BETi). BET inhibitors are a new class of drugs that act on Bromodomain and extra-terminal (BET) family proteins, such as BRD4, which are important regulators of enhancers of cancer relevant genes. We have identified several gene signatures of interest that were disturbed upon treatment with BETi and require mechanistic investigation and evaluation of efficacy in in vivo glioblastoma models. A signature suggesting histone deacetylase (HDAC) inhibitors as second drug has proven promising, yielding synergistic treatment effects in vitro. The identification of BETi-mediated suppression of a glioblastoma related interferon signature may be if interest in the context of immune check-point inhibition. Mechanistic follow-up will comprise chromatin immunoprecipitation in order to determine the association of H3K27 acetylation, BRD4 occupancy and gene expression (Pol2) under relevant conditions of the distinct signatures. These experiments will also shed light on other pathways and genes, e.g. the potential druggability of DNA damage response in combination with BETi. Most interestingly, MGMT was downregulated by BETi, in contrast to many other DDR genes. Thus, BETi may sensitize MGMT unmethylated patients to TMZ treatment, which would be of great clinical use. In vivo experiments are planned, to confirm the relevance of in vitro findings. Furthermore, RNA-seq of short-term treated orthotopic Xenografts will inform simultaneously on changes of the transcriptome of tumor cells (human) and the invaded brain (mouse, including micro-glia and macrophages). Respective signatures will be of interest as biomarkers for future phase 0 studies. In a second approach, we aim at investigating the potential tumor suppressor function of HIV-1 Tat Interactive Protein 2 (HTATIP2) in modulating nuclear import of GBM-relevant proteins. In a screen of the glioblastoma methylome, we have identified functional inactivation of HTATIP2 by promoter methylation in a subset of GBM. HTATIP2 is involved in the regulation of nuclear transport of proteins by binding to importin beta. Interestingly, we found that the BER enzyme N-Methylpurine-DNA Glycosylase (MPG) that repairs DNA damage induced by alkylating agents, is retained in the cytoplasm when HTATIP2 is expressed. Loss of HTATIP2 expression leads to nuclear localization of MPG, which has been associated with resistance to treatment with TMZ as we have reported previously. Hence, loss of HTATIP2 expression is involved in MPG mediated drug resistance. This interesting new mechanism cannot be directly assessed by analyzing OMICs data. It is expected that other proteins are also affected by loss of HTATIP2 expression and more readily enter the nucleus, thereby modulating the biology of the cells, potentially contributing to tumor malignancy. Here we propose to first quantify the effect HTATIP2 mediated (Tet-On) localization of MPG by Image Stream and confocal microscopy, and its effect on DNA repair as a proof of principle. Then we aim at identifying new candidates using the HTATIP2 inducible system and RNA-seq. This should allow identification of changes in gene expression, to uncover new mechanisms and pathways involved in tumor malignancy that may inform us on potential “Achilles heels” of the tumor that potentially can be targeted. Validation tools comprise tissue micro arrays of glioblastoma. The development of bioinformatics tools are planned to extract respective information from OMICs data.
Amount: CHF 786,200
Funding type: research
Funder scheme: Project funding
Start date: 2019-02-01
End date: 2023-04-30
Investigator: Hegi, Monika
Institution: University of Lausanne – LA
Funded outputs count: 8
Funder award ID: 182821
OpenAlex ID: G3507804253