Mechanisms of targeting cellular self-renewal in glioblastoma

ABSTRACT RNAs are underexploited targets for small molecules or chemical probes of cellular functions. The RNA coding for overexpressed transcriptional repressors are promising yet invalidated targets for cancer therapy. Overexpression of B-cell-specific Moloney murine leukemia virus integration site-1 (BMI1), a master polycomb transcriptional repressor and driver of cellular self-renewal, is a common oncogenic event in many cancers, including glioblastoma (GBM). We and others have shown that targeting BMI1 could have widely effective anti- tumor responses. We used cellular and zebrafish reporter screens to identify several molecules that not only alter BMI1 post-transcriptional processes but also selectively reduce BMI1 protein levels, modulate BMI1 targets and functionally inhibit cellular self-renewal. Subsequent structure-activity relationship studies revealed the basic pharmacophore and potential to modulate RNA translation. Upon chemical optimization, we synthesized a set of novel small molecules, among them are RU-A15 and RU-A16 that target BMI1 and kill tumor stem-like cells from various cancers at low nM concentrations. We aim to i) establish the mechanism of action of RU-A15 and RU-A16; ii) study their in vivo effects; and iii) identify BMI1 targets to serve as markers for sensitivity or resistance to targeted therapy. To achieve this, we generated a set of unique models utilizing normal human astrocytes as controls, patient derived GBM spheres, organoids and orthotopic patient derived xenografts (PDXs) that more accurately represent GBM biology than traditional cell lines. We hypothesize that the BMI1 probes selectively interfere with the post-transcriptional regulation of BMI1 RNA processing, leading to depletion of BMI1 protein levels and potent anti-GBM activities, including abrogation of cellular self-renewal, inhibition of GBM initiation and/or growth and sensitization to standard therapy. We will utilize these unique models to reveal the effects of targeting BMI1 with RU-A15 and RU-A16 and validate these effects in vivo in bioluminescent GBM PDXs and established genetically engineered mouse model (GEMM) of glioma. In Aim 1, we will utilize physico-chemical assays to determine the effects of the BMI1 probes on wild-type or mutant RNA using mutate-and-map strategy and assess the effects of treatment on sphere cell cycle, chromatin binding and key BMI1 targets. In Aim 2, we will examine the pharmacokinetic, dynamic, potency and selectivity and assess the efficacy of the BMI1 probe in GBM PDXs vs temozolomide (TMZ) and irradiation (IR). In Aim 3, we will assess the effects of the combined BMI1 probe with TMZ/IR therapy in orthotopic PDXs and GEMM of glioma. From these advances, we will develop highly-validated BMI1 probes for RNA targeting and clarify the significance of BMI1 as a therapeutic target both in GBM and likely many cancers with aberrant BMI1 functions. Our strategy uses a collaborative translational approach to validate (or refute) future use of the BMI1 probe in clinical applications in large cohorts of patients.