Do reactive oxygen species play a role in myeloid leukemias? by Paul Spencer Hole, Richard Lawrence Darley, and Alex Tonks Blood Volume 117(22):5816-5826 June 2, 2011 ©2011 by American Society of Hematology

Physiologic ROS homeostasis networks. Physiologic ROS homeostasis networks. Univalent reduction of oxygen results in the formation of superoxide (O2∙ −), which can occur as a result of NADPH oxidase (NOX) activity, and also as a by-product of oxidative phosphorylation, primarily at complex I in the mitochondrial electron transport chain (ETC). Superoxide may act as a reductant or an oxidant and is a key molecule in several subsequent physiologic reactions. Most of the superoxide generated in vivo is converted into hydrogen peroxide (H2O2) primarily by the actions of superoxide dismutases, which exist in cytosolic (SOD1), mitochondrial (SOD2), and extracellular (SOD3) isoforms. H2O2 levels are tightly regulated by several mechanisms, including the actions of catalase, the glutathione peroxidase (GPX) system, and peroxiredoxins (Prx). H2O2 may be further processed by the actions of myeloperoxidase (MPO) during an immune response to form hypochlorous acid (HOCl), which may in turn react with superoxide to form hydroxyl radicals. Hydroxyl radicals may also be formed from H2O2 by Fenton chemistry, which may occur in the presence of free metal cations such as Fe2+ or Cu+. Where superoxide production and production of nitric oxide (NO∙) are colocalized, reactive nitrogen species (RNS) may be formed, with the proximal species being peroxynitrite. Various RNS may then form via further chemical reactions with other ROS or RNS. This network of ROS and RNS production can be disrupted or biased in the presence of various compounds such as diphenyleneiodonium (DPI), which inhibits flavoproteins including the NOX oxidase family, ion chelators that can terminate Fenton chemistry cycles, and L-arginine analogs such as L-monomethyl arginine (L-NMMA), which inhibits nitric oxide synthase. Green represents molecular oxygen, blue are ROS derived from O2, and red represents nitric oxide and other RNS. Paul Spencer Hole et al. Blood 2011;117:5816-5826 ©2011 by American Society of Hematology

Antioxidant treatment versus pro-oxidant treatment as a therapy for hematologic malignancy. Antioxidant treatment versus pro-oxidant treatment as a therapy for hematologic malignancy. ROS production in myeloid leukemia cells may present an exploitable therapeutic target. Both antioxidant and pro-oxidant strategies may be effective, but both have potential advantages and disadvantages. The effects of antioxidant treatment on malignant cells (left side of figure) are likely to include reduced proliferative drive, which may reduce tumor burden and also protect nonmalignant cells from oxidative damage, particularly when administered in combination with chemotherapeutic (CT) agents. However, there are also concerns that suppression of cell cycle and antagonism of chemotherapy-induced ROS may adversely affect treatment efficacy. Pro-oxidant treatment (right side of figure) induces further ROS beyond that already produced by the malignant cell, either by depleting antioxidant defenses or augmenting ROS production. Treatment-induced oxidative stress combined with the intrinsic stress already present in the malignant cell leads to lipid peroxidation, oxidation of redox-sensitive residues within proteins, and DNA oxidation resulting in base-transversion and DSBs. Furthermore, damage to the electron transport chain and mutations in mitochondrial DNA can lead to a cycle of increased mitochondrial ROS. Elevated ROS has also been shown to contribute to cell-cycle progression in some contexts, which may increase tumor burden, but may simultaneously sensitize malignant cells to mainstay treatments. Although these factors may lead to induction of apoptosis in malignant cells, elevated ROS will increase the DNA mutation rate in any cells that fail to undergo apoptosis, possibly leading to selection of resistant clones. Furthermore, ROS generated by malignant cells may have paracrine effects on ROS signaling and oxidative damage in nonmalignant cells, and this effect would be augmented during treatment with pro-oxidants. Paul Spencer Hole et al. Blood 2011;117:5816-5826 ©2011 by American Society of Hematology