The role of PTEN in epigenetic and metabolic regulation of IDH-mutant gliomas

ABSTRACT Astrocytic tumors of the central nervous system (CNS) are intractable tumors and harbor dismal outcomes due to a lack of targeted treatment options. More than 70% of astrocytic tumors bear mutations in the TCA cycle- related gene, isocitrate dehydrogenase 1 (IDH1). Wildtype IDH1 catalyzes the oxidative decarboxylation of

isocitrate to generate α-Ketoglutarate (αKG), while mutated IDH1 (IDH1m) metabolizes αKG into the oncometabolite D-2-hydroxyglutarate (D2HG). Several epigenetic dioxygenases including histone and DNA demethylases require αKG to demethylate histone residues and DNA CpG islands. D2HG competitively

inhibits these enzymes due to its structural similarity to αKG. This results in an increase in histone and DNA methylation referred to as the glioma-CpG island methylation phenotype (G-CIMP). However, DNA methylation G-CIMP levels are not uniform in IDH1m astrocytomas. IDH1m astrocytomas with low versus high levels of G-

CIMP bear a particularly grim prognosis. My preliminary data identifies PTEN loss of heterozygosity (LOH) as the top genetic alteration in G-CIMP low versus G-CIMP high tumors. Moreover, PTEN LOH independently related with a poor prognosis. However, it remains unknown how PTEN LOH drives tumorigenicity in IDH1m

astrocytomas. Our premise is based on the observation that PTEN is a critical tumor suppressor. In other cancers, PTEN LOH activates PI3-kinase (PI3K)/AKT-mTOR signaling to drive tumor growth through various mechanisms including metabolic reprogramming. My preliminary data demonstrates increased cell proliferation

in IDH1m astrocytic cell lines with partial PTEN knockdown (KD) accompanied by a reduction in D2HG levels. Based on my premise and preliminary data, I hypothesize that PTEN LOH activates PI3K/AKT-mTOR signaling to metabolically reprogram IDH1m astrocytomas to reduce D2HG levels and thereby mediate lowered DNA

methylation and G-CIMP levels. Subsequently, targeting PI3K/AKT-mTOR signaling represents a promising candidate for therapeutic development. To test this hypothesis, I propose two specific aims. Aim 1 will map alterations in metabolic pathways that generate D2HG in human and murine-derived IDH1m astrocytomas with

or without PTEN reduction using a novel in-utero electroporation (IUE) Pten +/- IDH1m model. In parallel, I will assess corresponding genome wide epigenetic alterations including DNA methylation using next generation sequencing-based epigenetic assays in relation to changes in gene expression. Aim 2 will determine if

targeting PTEN LOH-driven PI3K/AKT-mTOR activation is therapeutic in vitro and proof-of-principle in animal models. Together, my work will address a critical gap in our knowledge by defining how PTEN LOH drives a subset of aggressive IDH1m astrocytomas via metabolic reprograming and lay the groundwork for developing

targeted and effective therapies for malignant IDH1m astrocytomas.