Abstract B202: Identification of new molecular targets in glioblastoma.

Abstract Glioblastoma multiformae (GBM) is the most common primary brain tumor and one of the most malignant human cancers (median survival 14 months from diagnosis despite combined surgery/irradiation/chemotherapy). Evidence from our and other groups indicates that glioblastoma stem-like cells (GSCs) contained within these tumors are critically important in GBM initiation, progression, and resistance to therapy. To identify new targets for glioblastoma treatment, we have systematically compared gene expression in nine primary GBM cell cultures to that in five primary neural stem cell cultures from the adult human brain (ahNSCs), using microarrays. Rigorous bioinformatic filtering identified 20 genes whose RNA expression levels were very high in primary GSC cultures but were undetectable in ahNSCs. The identified genes are involved in cell-cycle/division, epigenetic regulation, signaling and down-regulation of tumor-suppressors. Several of the candidate genes are implicated in cancer (breast, ovary and colon), while others have no known associations to cancer or have unknown functions. A total of nine new GBM primary cultures were further analyzed by real-time polymerase chain reaction (qPCR), Western blot and Immunohistochemistry on tissue sections and cell cultures. qPCR confirmed the increased expression of 17 of the candidate genes. These results were in good accordance with the information provided by public data bases (Rembrant, TCGA). Western blot performed on 17 candidates showed increased protein levels in 9 cases thus demarcating these as best candidates for molecular targeting. The rest of the proteins featured aberrant isoforms, probably due to alternate splicing, modifications, cleavage or degradation. Further bioinformatics analysis identified a subset of 5 candidates that are particularly highly up-regulated in GBM when compared to low-grade gliomas. To explore the functional importance of the potential target genes we established lentiviral-based shRNA delivery and started to test the potential of gene knock-downs (KDs) to inhibit growth of tumor cells. Our preliminary results look very promising. Three different lentiviral vectors, designed to silence a candidate gene encoding a tumor-suppressor binding protein kinase, were tested in three different primary GBM cultures and achieved a KD efficiency of 60–70%. We used proliferation essays to describe the growth speed of GBMs. To estimate relative numbers of sphere-forming cells (widely referred to as cancer stem cells) we performed sphere-forming assays (SFAs). The KD of the protein kinase gene resulted both in dramatic reduction of growth speed (50–70%) and in the number of sphere-forming cells (50–70%). In addition, the volume of the spheres was reduced to 50%. Cell cycle analysis revealed that there was doubling of the number of KD cells arrested in G2/M phase compared to non-silencing controls. There was also an increase (20–65%) in the proportion of apoptotic cells in 2 out of 3 GBM KD cultures. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B202.