Search

Bacterial nitroreductase enzymes that can efficiently convert nitroaromatic prodrugs to a cytotoxic form have numerous applications in targeted cellular ablation. For example, the generation of cytotoxic metabolites that have low bystander potential (i.e., are largely confined to the activating cell) has been exploited for precise ablation of specific cell types in animal and cell-culture models; while enzyme-prodrug combinations that generate high levels of bystander cell killing are useful for anti-cancer strategies such as gene-directed enzyme-prodrug therapy (GDEPT). Despite receiving substantial attention for such applications, the canonical nitroreductase NfsB from Escherichia coli has flaws that limit its utility, in particular a low efficiency of conversion of most prodrugs. Here, we sought to engineer a superior broad-range nitroreductase, E. coli NfsA, for improved activity with three therapeutically-relevant prodrugs: the duocarmycin analogue nitro-CBI-DEI, the dinitrobenzamide aziridine CB1954 and the 5-nitroimidazole metronidazole. The former two prodrugs have applications in GDEPT, while the latter has been employed for targeted ablation experiments and as a precise ‘off-switch’ in GDEPT models to eliminate nitroreductase-expressing cells. Our lead engineered NfsA (variant 11_78, with the residue substitutions S41Y, L103M, K222E and R225A) generated reduced metabolites of CB1954 and nitro-CBI-DEI that exhibited high bystander efficiencies in both bacterial and 2D HEK-293 cell culture models, while no cell-to-cell transfer was evident for the reduced metronidazole metabolite. We showed that the high bystander efficiency for CB1954 could be attributed to near-exclusive generation of the 2-hydroxylamine reduction product, which has been shown in 3D cell culture to cause significantly greater bystander killing than the 4-hydroxylamine species that is also produced by NfsB. We similarly observed a high bystander effect for nitro-CBI-DEI in HCT-116 tumor spheroids in which only a small proportion of cells were expressing variant 11_78. Collectively, our data identify variant 11_78 as a broadly improved prodrug-activating nitroreductase that offers advantages for both targeted cellular ablation and suicide gene therapy applications.

Thousands of genes have been implicated in retinal regeneration, but only a few have been shown to impact the regenerative capacity of Müller glia—an adult retinal stem cell with untapped therapeutic potential. Similarly, among nearly 300 genetic loci associated with human retinal disease, the majority remain untested in animal models. To address the large-scale nature of these problems, we are applying CRISPR/Cas9-based genome modification strategies in zebrafish to target over 300 genes implicated in retinal regeneration or degeneration. Our intent is to enable large-scale reverse genetic screens by applying a multiplexed gene disruption strategy that markedly increases the efficiency of the screening process. To facilitate large-scale phenotyping, we incorporate an automated reporter quantification-based assay to identify cellular degeneration and regeneration-deficient phenotypes in transgenic fish. Multiplexed gene targeting strategies can address mismatches in scale between “big data” bioinformatics and wet lab experimental capacities, a critical shortfall limiting comprehensive functional analyses of factors implicated in ever-expanding multiomics datasets. This report details the progress we have made to date with a multiplexed CRISPR/Cas9-based gene targeting strategy and discusses how the methodologies applied can further our understanding of the genes that predispose to retinal degenerative disease and which control the regenerative capacity of retinal Müller glia cells.