Adenovirus is an important model system for understanding basic aspects of cell function. It is also being used as a vector for gene therapy, and adenovirus mutants and recombinants are being developed for use as therapeutic agents to replicate in and kill neoplastic cells, causing a spreading infection in tumor, without spreading through normal tissue. Our goal is to understand the function of the adenovirus E1B-55K protein during a productive viral infection and how this knowledge can be applied to make a more effective therapeutic agent. An E1B-55K null mutant developed in our lab, dl1520 (aka ONYX-015), has been used in clinical trials for the treatment of human tumors, with some efficacy in some patients and not others. The original rationale for this approach was that since E1B-55K functions to inactivate p53, and most tumors are defective in p53 or in the p53-pathway, a mutant virus that cannot inactivate p53 would be expected to be restricted for replication in normal p53+ cells, but able to replicate in p53-minus cells. However, further study revealed that the ability of the mutant to replicate to varying extents in different tumor cell lines was related to the expression of viral late proteins that comprise the virus particle, and not the p53-status of the cells. We found that E1B-55K assembles a ubiquitin-protein ligase with another viral protein, E4orf6, and several cellular proteins that polyubiquitinates p53, marking it for degradation by proteosomes. Others found that this adenovirus ubiquitin-protein ligase probably also causes the degradation of the cellular MRN complex involved in DNA double-strand break repair, and that if the MRN complex is not inactivated, viral DNA replication is inhibited and viral DNA is concatenated into long linear DNA molecules too long to package in the virion. Recently, we found that the activity of this ubiquitin-ligase complex is also required to stimulate viral late gene expression. Studies with cell mutants suggest that it is the MRN complex that must be inactivated to allow the normal level of viral late mRNA export from the nucleus to the cytoplasm and translation in the cytoplasm. The results imply that the MRN ? ATM/ATR ? downstream targets pathway induces an anti-viral response that inhibits viral mRNA nuclear export and translation of late viral mRNAs. We will study how ATM activation inhibits viral mRNA nuclear export and translation. Our findings may allow us to genetically characterize tumors to determine which ones might be effectively treated with dl1520.
