PM-14 * MODELING GBM IN MICE USING THE CRISPR NUCLEASE SYSTEM TO FUNCTIONALLY VALIDATE NOVEL CANDIDATE DRIVER GENES AND TEST NEW THERAPIES

Abstract

Glioblastoma multiforme (GBM) is the most common malignant primary brain tumor and has a dismal survival rate of less than 20% in pediatric patients and less than 5% in adults. Identifying and functionally validating driver genes is needed to develop novel therapies to better treat GBM. To this end, we developed a novel method to induce de novo GBM in mice via in vivo gene transfer of the CRISPR nuclease system to functionally validate and/or screen candidate drivers. The CRISPR system is composed of the Cas9 nuclease that uses a RNA guide molecule (gRNA) to identify specific DNA sequences and induce double strand breaks (DSBs). These DSB are repaired by either the error prone process of non-homologous end joining (NHEJ) or perfectly repaired by homologous recombination (HR), allowing the opportunity for gene knockout (KO) or introduction of new sequences into the genome, respectively. To test our system, we utilized information from a pre-validated set of genes shown to induce GBM via in vivo gene transfer of Sleeping Beauty transposons harboring NRASG12V and shTrp53. Thus, Polyethylenimine (PEI) complexed Cas9, Trp53-gRNA, SB100X transposase, and NRASG12V transposon plasmids were injected into the lateral ventricles of neonatal mice and aged for tumor development. Animals injected with Cas9 and Trp53-gRNA or SB100X and NRASG12V alone did not develop tumors whereas all animals receiving both developed rapid onset GBM in 60 ± 16 days. Analysis of the Trp53-gRNA target site from tumors demonstrated clonal frame-shift mutations. These results demonstrate that the CRISPR is highly efficient in vivo to KO gene function. Currently, we are replacing the NRASG12V transposon with the use of HR to incorporate the G12V mutation at the endogenous Nras locus. Future studies will focus on validating candidate GBM driver genes identified from human genomics data and forward genetic screens and attempts to model GBM subgroups.