Abstract
Thiosemicarbazones are a group of compounds that have received comprehensive investigation as anticancer agents. The antitumor activity of the thiosemicarbazone, 3-amino-2-pyridinecarboxaldehyde thiosemicarbazone (3-AP; triapine), has been extensively assessed in more than 20 phase I and II clinical trials. These studies have demonstrated that 3-AP induces methemoglobin (metHb) formation and hypoxia in patients, limiting its usefulness. Considering this problem, we assessed the mechanism of metHb formation by 3-AP compared with that of more recently developed thiosemicarbazones, including di-2-pyridylketone-4,4-dimethyl-3-thiosemicarbazone (Dp44mT). This was investigated using intact red blood cells (RBCs), RBC lysates, purified oxyhemoglobin, and a mouse model. The chelation of cellular labile iron with the formation of a redox-active thiosemicarbazone-iron complex was found to be crucial for oxyhemoglobin oxidation. This observation was substantiated using a thiosemicarbazone that cannot ligate iron and also by using the chelator, desferrioxamine, that forms a redox-inactive iron complex. Of significance, cellular copper chelation was not important for metHb generation in contrast to its role in preventing tumor cell proliferation. Administration of Dp44mT to mice catalyzed metHb and cardiac metmyoglobin formation. However, ascorbic acid administered together with the drug in vivo significantly decreased metHb levels, providing a potential therapeutic intervention. Moreover, we demonstrated that the structure of the thiosemicarbazone is of importance in terms of metHb generation, because the DpT analog, di-2-pyridylketone-4-cyclohexyl-4-methyl-3-thiosemicarbazone (DpC), does not induce metHb generation in vivo. Hence, DpC represents a next-generation thiosemicarbazone that possesses markedly superior properties. This investigation is important for developing more effective thiosemicarbazone treatment regimens.