Nitroreductase-CB1954 [5-(aziridin-1-yl)-2,4-dinitrobenzamide] is a promising gene- directed enzyme-prodrug therapy for the treatment of a variety of cancers. This treatment involves the use of exogenous nitroreductase expressed in cancer cells to convert a non-toxic prodrug into an activated cytotoxic agent, which then targets tumor cells. The major advantage of this therapy includes its ability to target both dividing and nondividing cancer cells, independent of the cell cycle. Both the 4-nitro and 2-nitro of CB1954 can be activated by the nitroreductase. When the 4-nitro is reduced, a DNA interstrand crosslink agent 5-(aziridin-1-yl)-4-hydroxyamino-2-nitrobenzamide is produced and triggers the apoptosis of cancer cells. On the other hand, the reduction of the CB1954’s 2-nitro yields a metabolite that exhibits a substantially higher bystander effect to spread to other cancerous cells. In the human body, DT-diaphorase (NAD(P)H dehydrogenase) is able to activate CB1954, but it is significantly less efficient in reducing CB1954 than Escherichia coli nitroreductase (NTR). Despite it, low enzyme efficiency in reducing CB1954 is always a problem for cancer treatment. Unfortunately, since 2008, there has been no further update about this gene-directed enzyme-prodrug therapy or any advances in increasing its efficiency. To address this critical issue in gene-directed enzyme-prodrug therapy, the objective of this project is to optimize the catalytic efficiency of nitroreductase toward CB1954. Instead of engineering the extensively studied E. coli NTR, I chose a NTR from a human commensal bacterium, Haemophilic influenzae (HiNfsB). The Ding lab recently revealed that HiNfsB effectively metabolizes almost all clinically used nitroimidazole antibiotics. In addition, his lab also reported the X-ray structure of HiNfsB and showed that HiNfsB metabolized CB1954 more effectively than E. coli NTR. In this project, we will further engineer HiNfsB to improve its efficiency in metabolizing CB1954 to produce anticancer drug molecules. To achieve the objective, we will first dock CB1954 into the active site of HiNfsB to identify key residues for the binding of CB1954. Next, we will perform site-directed mutagenesis on selected residues to screen for mutants that show improved nitroreduction activities toward CB1954. The products will be identified in LC-MS analysis. Finally, the improved HiNfsB mutants will be tested with tumor cells to evaluate their performance in gene-directed enzyme-prodrug therapy. The results will lay a solid basis toward the development of improved directed enzyme-prodrug therapy for cancer treatment.