Abstract 2197: A robotic genome-scale shRNA screen identifies metabolic drug targets and a glioblastoma gene signature

Abstract

Abstract Proteins encoded by oncogenes represent suitable drug targets and several have been validated clinically. However, certain non-oncogenes, which are downstream effectors of oncogenes, may be similarly useful as potential targets of therapy. We have developed a high-throughput strategy using a shRNA genomic library and tumor cells addicted to the mTOR pathway to identify genes (oncogenes or non-oncogenes) which control the oncogenic phenotype. mTOR addicted cells were generated from IL-3 dependent cells following frame-shift mutagenesis and selection for IL-3 independence. mTOR addicted cells obtained from this selection carried frame-shift mutations in the PTEN gene. mTOR addiction was revealed by sensitivity to rapamycin and confirmed biochemically by examining phosphorylation of the mTOR targets S6K and PKB. Interestingly (and of importance to our shRNA-based screening strategy), addition of IL-3 rescued cells from mTOR addiction, as this growth factor antagonized the apoptotic effect of rapamycin. Using robotics (BioMek-NX) and a genomic retroviral shRNA library we evaluated 14’000 genes for a potential role in maintaining mTOR addicted growth. Our strategy was to screen cells in the presence and absence of IL-3 to identify shRNAs that antagonize growth of cells in the absence of IL-3. We identified around 300 genes required for mTOR addicted growth. This set of genes was enriched in highly significant manner for genes known to be involved in cancer, providing proof of concept for the strategy. Examples of identified genes include ras, raf, PKB and mTOR. Also, genes with no cancer link and genes with unknown function were identified. Use of Ingenuity software allowed to place many of the identified genes into functionally distinct groups or pathways. In addition numerous hits affecting metabolic functions, redox functions, and ROS generating systems were identified including the cytoplasmic NADPH complex and complex I-IV of the respiratory chain. Combining the genes et with an Affymetrix gene overexpression analysis in the same cells and analysing by Oncomine software, we identified a glioblastoma gene signature consisting of 13 genes (p-value 2.4E-5) containing 3 oncogenes and several signaling elements. Downregulation of several of these genes in human cancer lines induced apoptosis. Some of he genes identified represent novel potential drug targets. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2197.