Cytomegalovirus infection induces a stem cell phenotype in human primary glioblastoma cells: prognostic significance and biological impact.

Glioblastoma (GBM) is associated with poor prognosis despite aggressive surgical resection, chemotherapy, and radiation therapy. Unfortunately, this standard therapy does not target glioma cancer stem cells (GCSCs), a subpopulation of GBM cells that can give rise to recurrent tumors. GBMs express human cytomegalovirus (HCMV) proteins, and previously we found that the level of expression of HCMV immediate-early (IE) protein in GBMs is a prognostic factor for poor patient survival. In this study, we investigated the relation between HCMV infection of GBM cells and the presence of GCSCs. Primary GBMs were characterized by their expression of HCMV-IE and GCSCs marker CD133 and by patient survival. The extent to which HCMV infection of primary GBM cells induced a GCSC phenotype was evaluated in vitro. In primary GBMs, a large fraction of CD133-positive cells expressed HCMV-IE, and higher co-expression of these two proteins predicted poor patient survival. Infection of GBM cells with HCMV led to upregulation of CD133 and other GSCS markers (Notch1, Sox2, Oct4, Nestin). HCMV infection also promoted the growth of GBM cells as neurospheres, a behavior typically displayed by GCSCs, and this phenotype was prevented by either chemical inhibition of the Notch1 pathway or by treatment with the anti-viral drug ganciclovir. GBM cells that maintained expression of HCMV-IE failed to differentiate into neuronal or astrocytic phenotypes. Our findings imply that HCMV infection induces phenotypic plasticity of GBM cells to promote GCSC features and may thereby increase the aggressiveness of this tumor.

Complexation of retrovirus with cationic and anionic polymers increases the efficiency of gene transfer.

Previously, we have demonstrated that chondroitin sulfate proteoglycans and glycosaminoglycans inhibit retrovirus transduction. While studying the mechanism of inhibition, we found that the combined addition of equal-weight concentrations (80 microg/ml) of Polybrene and chondroitin sulfate C to retrovirus stocks resulted in the formation of a high-molecular-weight retrovirus-polymer complex that could be pelleted by low-speed centrifugation. The pelleted complex contained more than 80% of the virus particles, but less than 0.3% of the proteins that were originally present in the virus stock. Surprisingly, the virus in the complex remained active and could be used to transduce cells. The titer of the pelleted virus, when resuspended in cell culture medium to the starting volume, was three-fold greater than the original virus stock. The selectivity (CFU/mg protein) of the process with respect to virus activity was more than 1000-fold. When the pelleted virus-polymer complex was resuspended in one-eighth of the original volume and used to transduce NIH 3T3 murine fibroblasts and primary human fibroblasts, gene transfer was increased 10- to 20-fold over the original unconcentrated retrovirus stock. The implications of our findings for the production, processing, and use of retrovirus stocks for human gene therapy protocols are discussed.