The effect of functional groups on reduction and activation of quinone bioreductive agents by DT-diaphorase

Abstract.

Purpose: Bioreductive antitumor agents are an important class of anticancer drugs that include the clinically used drug, mitomycin C, and new agents such as EO9 and tirapazamine that have recently been tested in clinical trials. These agents require activation by reductive enzymes such as DT-diaphorase or NADPH:cytochrome P450 reductase. A major focus for improving cancer chemotherapy has been to increase the selectivity and targeting of antitumor drugs to tumor cells. Bioreductive antitumor agents are ideally suited to improving tumor selectivity by an enzyme-directed approach to tumor targeting. However, none of the bioreductive agents developed to date has been specific for activation by a single reductive enzyme. This is in part due to a lack of knowledge about structural factors that confer selectivity for activation by reductive enzymes. The purpose of this study was to investigate the ability of specific functional groups to modify reduction and activation of quinone bioreductive agents by DT-diaphorase. Methods: We used a series of model benzoquinone mustard (BM) bioreductive agents and compared the parent compound BM to MBM, which has a strong electron-donating methoxy group, MeBM, which has a weaker electron-donating methyl group, CBM, which has an electron-withdrawing chloro group, and PBM and its structural isomer, meta-PBM (m-PBM), which both have sterically bulky benzene rings attached to the quinone moiety. We determined the rate of reduction of these agents by purified human DT-diaphorase under hypoxic and aerobic conditions. We also measured the cytotoxic activity of these agents in human tumor cell lines with and without the DT-diaphorase inhibitor, dicoumarol. Results: Under hypoxic conditions in vitro, the t_1/2 values for reduction of the analogs by purified DT-diaphorase were 4, 6, 8, 9, 10 and 21 min for BM, MeBM, CBM, MBM, PBM and m-PBM, respectively. Under aerobic conditions the rank order of redox cycling after two-electron reduction by DT-diaphorase was MBM>MeBM>BM≈CBM≈PBM≈m-PBM. The rate of reduction by DT-diaphorase of HBM, a non-alkylating analog of BM, was similar to that of BM under hypoxic conditions, and the rate of redox cycling under aerobic conditions was comparable to that of BM, suggesting that structural changes to the cytotoxic group of these BMs do not affect DT-diaphorase-mediated reduction and redox cycling potential. MBM, MeBM and PBM were more toxic than BM in the NCI-H661 human non-small-cell lung cancer cells and SK-MEL-28 human melanoma cells, while CBM displayed significantly increased cytotoxic activity compared to BM only in the NCI H661 cells. m-PBM had similar cytotoxic activity compared with BM in both cell lines. These cell lines have moderate to high levels of DT-diaphorase activity. When cells were pretreated with the DT-diaphorase inhibitor, dicoumarol, the cytotoxic activity of BM increased while that of MBM decreased in both cell lines, suggesting that BM was inactivated by DT-diaphorase while MBM was activated by this enzyme. Pretreatment of the SK-MEL-28 melanoma cells with dicoumarol resulted in an increased cytotoxic activity of MeBM, but pretreatment of the NCI-H661 cells did not affect the cytotoxicity of MeBM. This suggests, that similar to the results with BM, DT-diaphorase is an inactivating enzyme for MeBM in the SK-MEL-28 cell line. Dicoumarol had no significant effect on the cytotoxicity of CBM, PBM or m-PBM in both cell lines. Conclusions: These studies demonstrated that functional groups can significantly affect the reduction and activation of bioreductive agents by DT-diaphorase. All the functional groups decreased the rate of reduction of the quinone group by DT-diaphorase. Since MeBM and MBM, with electron-donating functional groups, and CBM with an electron-withdrawing functional group had similar half-lives of reduction by DT-diaphorase, steric rather than electronic effects of the functional groups appear to be more important for modifying the rate of reduction by DT-diaphorase. Steric effects on reduction by DT-diaphorase were also influenced by the position of the functional group on the quinone ring moiety, as the reduction of m-PBM was much slower than the reduction of PBM. The electron-donating methoxy and methyl functional groups increased the ability of the reduced products of MBM and MeBM to undergo redox cycling. DT-diaphorase appeared to be an activating enzyme for MBM. This may have resulted in part from increased formation of reactive oxygen species resulting from the increased redox cycling by MBM. In contrast, DT-diaphorase was an inactivating enzyme for BM, and for MeBM in the SK-MEL-28 melanoma cells, possibly because the hydroquinone product of BM and MeBM may be less cytotoxic than the semiquinone produced by one-electron reduction by NADPH:cytochrome P450 reductase.