DESCRIPTION (provided by applicant): Glioblastomas are aggressive and invasive brain tumors that generally lead to death within a year of diagnosis. No cure is available. Current treatments prolong life by only a few months, often at the expense of quality of life. Here we test the general hypothesis that a novel recombinant chimeric virus will target and kill gliomas with no detectable adverse effect to the brain. Of the large number of viruses we have tested, vesicular stomatitis virus (VSV) appears to be one of the most effective for targeting and destroying brain tumors. However, VSV has the potential unwanted side effect of infecting neurons, and half of our efforts in the last few years have focused on reducing or controlling potential neurotoxicity of VSV. To avoid complications and toxicity of the VSV G-protein, particularly its binding to neurons, we have compared a number of recombinant chimeric viruses in which the VSV G-protein gene was replaced by genes coding for binding glycoproteins of non-related viruses including rabies, lymphocytic choriomeningitis, Marburg, Ebola, and Lassa viruses. Of these chimeric viruses tested, one stood out as a clear superlative safe oncolytic candidate: a chimeric virus consisting of a gene coding for the Lassa glycoprotein together with genes coding for the VSV N,P,M, and L proteins, and a GFP reporter gene which further attenuates the virus. Lassa-VSV is safe, both in rodents and primates. Of considerable importance, our direct injection of Lassa-VSV into the brains of normal mice or rats, or even into the brains of immunodeficient mice has not resulted in any detectable adverse effects, whereas injections of native or other attenuated VSVs generated neurological complications sometimes resulting in death. Equally important, in our preliminary experiments in vitro and in vivo, Lassa-VSV targets and destroys gliomas completely without damage to the host brain, and extends survival of tumor-bearing mice indefinitely. In the first set of experiments, we test the hypothess that Lassa-VSV successfully targets and kills glioblastoma cells that are transplanted into the brain, both after an intratumoral virus injection, and after intravenous inoculation. We use both glioma cell lines and primary human gliomas transplanted into immunodeficient mice. We also test syngeneic mouse glioma implanted into immunocompetent mouse host brains. Tumors are detected by expression of a red fluorescent reporter and virally infected cells are detected by expression of a reporter gene coding for green fluorescent protein. In the second Aim, we address the hypothesis that the mechanism behind the safety and selectivity of Lassa-VSV in the brain is that the virus either does not bind to receptors on neurons or normal glia, or is not internalized, whereas binding and internalization in glioma is robust. The lack of virus infection of neurons is studied by blocking or enhancing various steps in the infectious pathwa...