Dissecting the role of NADPH:cytochrome P450 oxidoreductase in the activation of hypoxia-activated prodrugs using zinc finger nucleases

Abstract

Tumour hypoxia results in aggressive tumours with increased metastatic potential and resistance to therapy. Exploitation of tumour hypoxia can be achieved using hypoxiaactivated prodrugs (HAP), which are selectively activated by reductases in a process inhibited by oxygen. NADPH:cytochrome P450 oxidoreductase (POR) is widely considered to be the major enzyme responsible, but this conclusion is based on studies using purified POR or forced over-expression in cell lines. Thus the enzymology of HAP activation is still not fully understood. To examine the role of endogenous POR in HAP activation, POR knockout clones were generated by targeted mutation in exon 8 of POR using custom-designed zinc finger nucleases (ZFN) in HCT116 and SiHa cell lines. The POR knockout clones were validated for loss of POR expression by Western blotting, mutations were confirmed at the ZFN target site and a significant loss of POR enzyme activity was measured by cytochrome c assay and activation of a novel fluorogenic probe; FSL-61. The knockout models were compared to isogenic wild-type and POR over-expressing cells using metabolism and cytotoxicity assays across a panel of 12 HAP. POR over-expression significantly increased the sensitivity of HCT116 and SiHa cells to aromatic N-oxides (tirapazamine, SN29751, SN30000) and nitro compounds SN24349 and nitracrine under aerobic and anoxic conditions (p<0.05). Surprisingly, the effect of POR knockout on anoxic resistance to HAP was modest, with a statistically significant resistance to tirapazamine, SN30000, CB1954 and SN24349, in only one of the two HCT116 and SiHa POR knockout clones. Further interrogation with PR- 104A and SN30000 showed trends to lower formation of anoxic metabolites in the POR knockouts compared to wild-type cells, but the decreases were not statistically significant. In addition, the POR knockout clones did not demonstrate increased resistance to these HAP by clonogenic assay. These observations confirm that most HAP are substrates for POR, but indicate that POR at endogenous levels of expression plays only a limited role in the activation of these prodrugs, with clearer signals of its contribution in SiHa than HCT116. This suggests the existence of redundant one-electron reductases that are able to affect prodrug activation in the absence of POR.