Cancer dual-targeting of an infectivity-enchanced CRAd

DESCRIPTION (provided by applicant): A number of characteristics of the Adenovirus type 5 (Ad5) make it an optimal gene therapy/virotherapy vector suitable for a wide array of cancer therapeutic approaches. Despite these advantages, overall ifficacy remains limited by sub-optimal adenovirus delivery to cancer tissues. One conceptual approach to overcome this limitation is to achieve an amplification effect of adenovirus infectivity via selective replication of the delivered adenovirus post-infection, such that lateral spread of progeny viruses may occur. In this approach, a conditionally replication-competent virus replicates oncolytically in transduced tumor cells but not in normal tissue. Production of adenovirus progeny from transduced tumor cells would then allow infection of the neighboring tumor cells. Using conditionally replicative adenoviruses (CRAds) represent a method to achieve efficient tumor cell oncolysis and mitigate tumor cell infection limitations. However, given that Ad5-based CRAds have a natural liver tropism, any background replication activity in normal tissues (and subsequent liver toxicity) presents a particularly critical problem that warrants development of novel approaches to improve cancer specificity. In this proposal, we aim specifically to reduce background replication activity in non-cancer cells by introducing two different levels of cancer targeting for the Ad5 E1A gene expression. First, this can be achieved by placing E1A gene, which is essential for CRAd replication function, under under transcriptional control. To accomplish this, we will use the CXCR4 gene promoter, which has a distinct cancer-specific activation profile in squamous cell carcinomas of the head and neck (HNSCC). Second, we will introduce cancer-specific protein translational control by engineering a highly structured 5'-untranslated region (5'-UTR) sequence of the Fibroblast Growth Factor 2 (FGF2) upstream of the E1A mRNA coding sequence. This will render efficient translation of such a chimeric mRNA dependent upon the translation initiation factor elF-4E, which is expressed in limiting amounts in most normal tissues and is over expressed in most cancer cells. Thus, we hypothesize that the dual level of cancer-specific E1A mRNA transcription together with cancer-specific protein translation will significantly enhance target expression of E1A gene to cancer tissues beyond the use of either control element alone, and thereby dramatically improve cancer-specific replication of the CRAd. The successful completion of the proposed research to develop and validate the principle of dual-targeting of transgene expression in HNSCC cell lines, will result in a totally new class of CRAds for cancer virotherapy.