ODZ1 allows glioblastoma to sustain invasiveness through a Myc-dependent transcriptional upregulation of RhoA.

Long-term survival remains low for most patients with glioblastoma (GBM), which reveals the need for markers of disease outcome and novel therapeutic targets. We describe that ODZ1 (also known as TENM1), a type II transmembrane protein involved in fetal brain development, plays a crucial role in the invasion of GBM cells. Differentiation of glioblastoma stem-like cells drives the nuclear translocation of an intracellular fragment of ODZ1 through proteolytic cleavage by signal peptide peptidase-like 2a. The intracellular fragment of ODZ1 promotes cytoskeletal remodelling of GBM cells and invasion of the surrounding environment both in vitro and in vivo. Absence of ODZ1 by gene deletion or downregulation of ODZ1 by small interfering RNAs drastically reduces the invasive capacity of GBM cells. This activity is mediated by an ODZ1-triggered transcriptional pathway, through the E-box binding Myc protein, that promotes the expression and activation of Ras homolog family member A (RhoA) and subsequent activation of Rho-associated, coiled-coil containing protein kinase (ROCK). Overexpression of ODZ1 in GBM cells reduced survival of xenografted mice. Consistently, analysis of 122 GBM tumour samples revealed that the number of ODZ1-positive cells inversely correlated with overall and progression-free survival. Our findings establish a novel marker of invading GBM cells and consequently a potential marker of disease progression and a therapeutic target in GBM.

Blockade of the NFκB pathway drives differentiating glioblastoma-initiating cells into senescence both in vitro and in vivo.

Glioblastoma multiforme is one of the most devastating cancers and presents unique challenges to therapy because of its aggressive behavior. Cancer-initiating or progenitor cells have been described to be the only cell population with tumorigenic capacity in glioblastoma. Therefore, effective therapeutic strategies targeting these cells or the early precursors may be beneficial. We have established different cultures of glioblastoma-initiating cells (GICs) derived from surgical specimens and found that, after induction of differentiation, the NFκB transcriptional pathway was activated, as determined by analyzing key proteins such as p65 and IκB and the upregulation of a number of target genes. We also showed that blockade of nuclear factor (NF)κB signaling in differentiating GICs by different genetic strategies or treatment with small-molecule inhibitors, promoted replication arrest and senescence. This effect was partly mediated by reduced levels of the NFκB target gene cyclin D1, because its downregulation by RNA interference reproduced a similar phenotype. Furthermore, these results were confirmed in a xenograft model. Intravenous treatment of immunodeficient mice bearing human GIC-derived tumors with a novel small-molecule inhibitor of the NFκB pathway induced senescence of tumor cells but no ultrastructural alterations of the brain parenchyma were detected. These findings reveal that activation of NFκB may keep differentiating GICs from acquiring a mature postmitotic phenotype, thus allowing cell proliferation, and support the rationale for therapeutic strategies aimed to promote premature senescence of differentiating GICs by blocking key factors within the NFκB pathway.