Prodrug gene therapy (PGT) is a therapeutic strategy in which tumor cells are transfected with a 'suicide' gene that encodes a metabolic enzyme capable of converting a nontoxic prodrug into a potent cytotoxin. Several enzyme/prodrug combinations are under active investigation. This strategy is inherently limited by inefficient delivery of the gene to cancer cells (in effect, replacing the problem of drug delivery with the problem of gene delivery). To offset this significant issue, the pharmacokinetic properties of the enzyme (its stability, half-life and kinetic activity), the prodrug (its toxicity and metabolism) and combination of the two (their uniqueness to transfected cells) must be optimized for maximum therapeutic efficacy. In this project, three collaborating laboratories are engineering and optimizing two nucleoside salvage/synthesis enzymes for PGT: cytosine deaminase (CD) and deoxycytidine kinase (dCK). CD (a microbial enzyme) is being engineered to efficiently convert 5-fluorocytosine (5-FC) to 5-fluorouracil (5-FU), which is a metabolic inhibitor of DNA synthesis and RNA function. In contrast, dCK (a human enzyme) catalyzes the ?-phosphorylation of pyrimidine nucleosides, and is being engineered to efficiently activate pyrimidine analogues such as gemcitabine and decitabine. In both cases, the project follows a 'design cycle' of crystallographic structure determination, computational protein engineering, directed evolution and subsequent kinetic and structural analyses. The ability of the best enzyme variants to induce sensitivity to the prodrug is assayed in tumor cell lines, animal models and ongoing clinical trials. Our data from the previous funding cycle demonstrate that either the stability or the substrate-specific activity and specificity of a given enzyme/prodrug combination can be limiting for performance in prodrug therapy. Furthermore, either limitation can be overcome by design and selection of improved enzyme constructs. Based on suggestions from previous review of this renewal application, we now describe a set of revised specific aims for this project as follows: (1) We will determine whether optimization of yCD or bCD leads to significant therapeutic efficacy gains via recognizable mechanisms of increased enzyme expression and/or drug production in tumor cells. (2) We will create a new enzyme/prodrug combination (dCK and decitabine, which is a potent cytoxin but is both unstable and inefficiently phosphorylated by dCK). We will compare the results of enzyme redesign for enhanced activity against decitabine to parallel experiments with gemcitabine (which, in contrast, is an efficient substrate for dCK). In addition to adding a new enzyme/prodrug combination to the PGT arsenal, these experiments will examine limitations on an enzyme's performance in PGT that are substrate-dependent. Our hypothesis is that decitabine should ultimately couple with engineered enzyme variants to yield improvements in the performance of dCK, due to the lack of this activity in non-cancerous tissues. ? ? ?