Cyanide-resistant alternative respiration is strictly correlated to intracellular peroxide levels in Acremonium chrysogenum

Long-wave ultraviolet radiation (UVA) may cause extensive DNA damage via reactive oxygen species (ROS). In this study we examined whether UVA- and H₂O₂-mediated DNA damage have equivalent effects on the induction of G2/M phase checkpoint and cell cycle progression in a transformed keratinocyte cell line HaCaT. By employing single cell gel electrophoresis (comet assay) we determined the equipotent doses of UVA and H₂O₂ with respect to the induction of alkali-labile sites (an indicator of oxidative DNA decay). However, in contrast to H₂O₂ which caused a pronounced G2/M cell cycle arrest 24h after treatment, UVA irradiation did not affect cell cycle progression. Increasing UVA doses up to 150 kJ/m² did not affect cell cycle and proliferation whereas increasing H₂O₂ concentrations caused a cell cycle block or cell death. Cytometric analysis revealed that G2/M cell cycle arrest took place beyond the cyclin B1 restriction point. We conclude that the DNA damage induced by UVA is easily repaired and does not perturb cell growth, whereas the H₂O₂-induced damage leads ultimately to cell cycle arrest or cell death.     

Light promotes the synthesis of lignin through the production of H2O2 mediated by diamine oxidases in soybean hypocotyls

In order to analyze the relationship between polyamine oxidative degradation induced by light and the lignin synthesis in cell walls, the activities of diamine oxidases and peroxidase, the contents of H₂O₂ and lignin, and the growth of hypocotyls in soybean [Glycine max (Linn.) Merr.] grown under light or in darkness were investigated. In comparison with the dark treatment, light irradiation significantly inhibited the growth of soybean hypocotyls and promoted the activities of diamine oxidases and peroxidase as well as the accumulation of H₂O₂ and lignin. Treatments with the different concentrations of diamine oxidase inhibitors (2-hydroxyethylhydrazine and aminoguanidine) under the light condition inhibited diamine oxidase activity, and decreased the contents of H₂O₂ and lignin. The results provide evidence for the hypothesis that light irradiation could promote the accumulation of H₂O₂ and lignin in cell walls by activating polyamine oxidative degradation mediated by diamine oxidases.     

Oxidant-mediated AA release from astrocytes involves cPLA(2) and iPLA(2)

Excessive generation of reactive oxygen species (ROS) in the central nervous system (CNS) is a leading cause of neuronal injury. Despite yet unknown mechanisms, oxidant compounds such as H₂O₂ have been shown to stimulate the release of arachidonic acid (AA) in a number of cell systems. In this study, H₂O₂ and menadione, a compound known to release H₂O₂ intracellularly, were used to examine the phospholipases A₂ (PLA₂) responsible for AA release from primary murine astrocytes. Both H₂O₂ and menadione dose-dependently stimulated AA release, and the release mediated by H₂O₂ was completely inhibited by catalase. H₂O₂ also stimulated phosphorylation of extracellular signal-regulated kinases (ERK1/2) and cytosolic phospholipase A₂ (cPLA₂). However, complete inhibition of cPLA₂ phosphorylation by U0126, an inhibitor for mitogen-activated protein kinase kinase (MEK) and GF109203x, a nonselective PKC inhibitor preferring the conventional and novel isoforms, only reduced H₂O₂-stimulated AA release by 50%. MAFP, a selective, active, site-directed, irreversible inhibitor of both cPLA₂ and the Ca²⁺-independent iPLA₂, nearly completely inhibited H₂O₂-mediated AA release; but, HELSS, a potent irreversible inhibitor of iPLA₂, only inhibited H₂O₂-mediated AA release by 40%. Along with the observation that H₂O₂-mediated AA release was only partially inhibited upon chelating intracellular Ca²⁺ by BAPTA, these results indicate the involvement of both cPLA₂ and iPLA₂ in H₂O₂-mediated AA release in murine astrocytes.     

Caspases are reversibly inactivated by hydrogen peroxide

Hydrogen peroxide (H₂O₂) is known to both induce and inhibit apoptosis, however the mechanisms are unclear. We found that H₂O₂ inhibited the activity of recombinant caspase-3 and caspase-8, half-inhibition occurring at about 17 μM H₂O₂. This inhibition was both prevented and reversed by dithiothreitol while glutathione had little protective effect. 100–200 μM H₂O₂ added to macrophages after induction of caspase activation by nitric oxide or serum withdrawal substantially inhibited caspase activity. Activation of H₂O₂-producing NADPH oxidase in macrophages also caused catalase-sensitive inactivation of cellular caspases. The data suggest that the activity of caspases in cells can be directly but reversibly inhibited by H₂O₂.     

DNA Single Strand Breakage by H2O2 and Ferric or Cupric Ions: Its Modulation by Histidine

The role of histidine on DNA breakage induced by hydrogen peroxide (H₂O₂) and ferric ions or by H₂O₂ and cupric ions was studied on purified DNA. L-histidine slightly reduced DNA breakage by H₂O₂ and Fe³⁺ but greatly inhibited DNA breakage by H₂O₂ and Cu²⁺. However, only when histidine was present, the addition of EDTA to H₂O₂ and Fe³⁺ exhibited a bimodal dose response curve depending on the chelator metal ratio. The enhancing effect of histidine on the rate of DNA degradation by H₂O₂ was maximal at a chelator metal ratio between 0.2 and 0.5, and was specific for iron. When D-histidine replaced L-histidine, the same pattern of EDTA dose response curve was observed. Superoxide dismutase greatly inhibited the rate of DNA degradation induced by H₂O₂, Fe³⁺, EDTA and L-histidine involving the superoxide radical.

These studies suggest that the enhancing effect of histidine on the rate of DNA degradation by H₂O₂ and Fe³⁺ is mediated by an oxidant which could be a ferrous-dioxygen-ferric chelate complex or a chelate-ferryl ion.     

The role of H2O2 as a mediator of UVB-induced apoptosis in keratinocytes

Apoptosis is an active form of cell death that is initiated by a variety of stimuli, including reactive oxygen species (ROS) and ultraviolet (UV) radiation. Previously, it has been reported that UVB-irradiation of keratinocytes leads to intracellular generation of hydrogen peroxide (H₂O₂) and that antioxidants can inhibit ROS-induced apoptosis. Although both UVB-irradiation and H₂O₂-incubation led to increased intracellular H₂O₂ levels, the antioxidants catalase and glutathione monoester (GME), inhibited apoptosis only when induced by H₂O₂, not by UVB. Furthermore, extracellular signal-regulated kinase (ERK), a prominent member of the mitogen-activated protein kinase (MAPK) family, was found to be activated by treatment with both UVB and H₂O₂. Inhibition of ERK phosphorylation by pre-treatment with PD98059 resulted in enhanced apoptosis after H₂O₂-exposure. However, no significant difference of apoptosis was observed between cells with and without inhibitor pre-treatment upon UVB-irradiation. DNA damage in the form of cyclobutane pyrimidine dimers was observed after exposure to UVB, but no photoproducts were found in H₂O₂-treated cells. These results suggest a ROS-independent pathway of UVB-induced apoptosis. Although UVB-irradiation causes moderate increase in H₂O₂, the generation of H₂O₂ does not contribute to the induction of apoptosis. Moreover, activation of ERK only blocks H₂O₂-dependent apoptosis but has no impact on UVB-induced apoptosis.     

Does hydrogen peroxide exist “free” in biological systems?

Hydrogen peroxide (H₂O₂) can diffuse far from the site of production to intracellular locations where biological effects may be greater. The diffusion range is extended by H₂O₂ carriers formed spontaneously by hydrogen bonding with monomeric and polymeric compounds, including amino and dicarboxylic acids, peptides, proteins, nucleic acid bases, and nucleosides. Hydrogen peroxide adducts (HPAs) are readily synthesized, e.g., crystalline histidine (His)-H₂O₂ adducts. An equilibrium exists between an adduct-forming compound and H₂O₂. The detection and relative stabilities of HPAs are measured by the degree of decomposition of H₂O₂ as influenced by test compounds in buffered solution competing with glucose or fructose for H₂O₂. The HPAs delay decomposition of H₂O₂ up to several hundredfold. The overall charge on an HPA, i.e., its ability to penetrate cell membranes, influences the cytotoxic and clastogenic effects of H₂O₂. Growth inhibition of Salmonella typhimurium LT₂ by H₂O₂ is enhanced by neutral HPAs but decreased by anionic HPAs. Addition of catalase 1, 10, or 30 min after inoculation of S. typhimurium LT₂ reduces or nearly eliminates partial growth inhibition by H₂O₂, but a neutral HPA, expecially his-H₂O₂, transported H₂O₂ into the cells within 1 min, and in about 10 min completely inhibited growth. The stability of HPAs decreases with increasing pH or increasing temperature, while added Fe(II) in the presence and absence of EDTA accelerates H₂O₂ and HPA decomposition. Calculations indicate H₂O₂ hydrogen bonds with nucleic acid-base pairs with no apparent bond strain and energy stabilization comparable to normal hydrogen bonding.     

The Action of Hydrogen Peroxide on the Formation of Thiobarbituric Acid-Reactive Material From Microsomes, Liposomes Or From Dna Damaged By Bleomycin Or Phenanthroline. Artefacts in the Thiobarbituric Acid Test

Incubation of rat-liver microsomes, previously azide-treated to inhibit catalase, with H₂O₂ caused a loss of cytochrome P-450 but not of cytochrome b₅. This loss of P-450 was not prevented by scavengers of hydroxyl radical, chain-breaking antioxidants or metal ion-chelating agents. Application of the thiobarbituric acid (TBA) assay to the reaction mixture suggested that H₂O₂ induces lipid peroxidation, but this was found to be due largely or completely to an effect of H²O₂ on the TBA assay. By contrast, addition of ascorbic acid and Fe(III) to the microsomes led to lipid peroxidation and P-450 degradation: both processes were inhibited by chelating agents and chain-breaking antioxidants, but not by hydroxyl radical scavengers. H₂O₂ inhibited ascorbate/Fe (III)-induced microsomal lipid peroxidation, but part of this effect was due to an action of H₂O₂ in the TBA test itself. H₂O₂ also decreased the colour measured after carrying out the TBA test upon authentic malondialdehyde, tetraethoxypropane, a DNA-Cu²⁺/o-phenanthroline system in the presence of a reducing agent, ox-brain phospholipid liposomes in the presence of Fe(III) and ascorbate, or a bleomycin-iron ion/DNA/ascorbate system. Caution must be used in interpreting the results of TBA tests upon systems containing H₂O₂.     

Inactivation of Yeast Glutathione Reductase by Fenton Systems: Effect of Metal Chelators, Catecholamines and Thiol Compounds

Oxygen radical generating systems, namely, Cu(II)/ H₂O₂, Cu(II)/ascorbate, Cu(II)/NAD(P)H, Cu(II)/ H₂O₂/catecholamine and Cu(II)/H₂O₂/SH-compounds irreversibly inhibited yeast glutathione reductase (GR) but Cu(II)/H₂O₂ enhanced the enzyme diaphorase activity. The time course of GR inactivation by Cu(II)/H₂O₂ depended on Cu(II) and H₂O₂ concentrations and was relatively slow, as compared with the effect of Cu(II)/ascorbate. The fluorescence of the enzyme Tyr and Trp residues was modified as a result of oxidative damage. Copper chelators, catalase, bovine serum albumin and HO^˙ scavengers prevented GR inactivation by Cu(II)/H₂O₂ and related systems. Cysteine, N-acetylcysteine, N-(2-dimercaptopropi-onylglycine and penicillamine enhanced the effect of Cu(II)/H₂O₂ in a concentration- and time-dependent manner. GSH, Captopril, dihydrolipoic acid and dithiotreitol also enhanced the Cu(II)/H₂O₂ effect, their actions involving the simultaneous operation of pro-oxidant and antioxidant reactions. GSSG and try-panothione disulfide effectively protected GR against Cu(II)/H₂O₂ inactivation. Thiol compounds prevented GR inactivation by the radical cation ABTS*⁺. GR inactivation by the systems assayed correlated with their capability for HO* radical generation. The role of amino acid residues at GR active site as targets for oxygen radicals is discussed.     

Degradation of Hyaluronic Acid by Photosensitized Riboflavin In Vitro. Modulation of the Effect by Transition Metals, Radical Quenchers, and Metal Chelators

The effect of photoexcited riboflavin (RF) on the viscosity of hyaluronic acid (HA) solutions has been investigated. UV irradiation of RF causes under aerobic conditions fragmentation of HA and a decrease in the viscosity of its solutions. A decrease of HA viscosity occurs in PO₄-buffered solutions and is accelerated by high pH, Fe²⁺ (but much less so by Fe³⁺), certain metal chelators, and horseradish peroxidase (HRP); it is partially inhibited by catalase and less so by superoxide dismutase (SOD). The reactivity of the system was completely blocked by Tris, ethanol, aspirin, d-manitol, dimethylthiourea (DMTU), dimethylsulfoxide (DMSO), and sodium azide. These results indicate that the most likely chemical species involved in the reaction is the hydroxyl radical. Singlet oxygen (¹0₂) generation is suggested by the ability of NaN₃ and DMSO to completely inhibit the reactivity of the system. These two agents, however, may also interact with OH^√ radical, as well and suppress the reactivity of the system. H₂O₂ and seem also to be produced in significant amounts, because catalase and SOD partially block the reactivity of the system. The effect of HRP may be due to hydrogen subtraction from HA and H₂O₂ reduction to water. Photoexcitation of RF may potentially occur in vitro and in vivo in the organs and tissues that are permeable to light, such as the eye or skin, and damage HA and other cell-matrix components causing inflammation and accelerating aging.  © 1997 Elsevier Science Inc.     

Catalase catalyzes nitrotyrosine formation from sodium azide and hydrogen peroxide

Sodium azide (NaN₃) is known as an inhibitor of catalase, and a nitric oxide (NO) donor in the presence of catalase and H₂O₂. We showed here that catalase-catalyzed oxidation of NaN₃ can generate reactive nitrogen species which contribute to tyrosine nitration in the presence of H₂O₂. The formation of free-tyrosine nitration and protein-bound tyrosine nitration by the NaN₃/catalase/H₂O₂ system showed a maximum level at pH 6.0. Free-tyrosine nitration induced by peroxynitrite was inhibited by ethanol and dimethyl-sulfoxide (DMSO), and augmented by superoxide dismutase (SOD). However, free-tyrosine nitration induced by the NaN₃/catalase/H₂O₂ system was not affected by ethanol, DMSO and SOD. NO^-₂ and NO donating agents did not affect free-tyrosine nitration by the NaN₃/catalase/H₂O₂ system. The reaction of NaN₃ with hydroxyl radical generating system showed free-tyrosine nitration, but no formation of nitrite and nitrate. The generation of nitrite (NO^-₂) and nitrate (NO^-₃) by the NaN₃/catalase/H₂O₂ system was maximal at pH 5.0. These results suggested that the oxidation of NaN₃ by the catalase/H₂O₂ system generates unknown peroxynitrite-like reactive nitrogen intermediates, which contribute to tyrosine nitration.     

The Generation of Ferryl or Hydroxyl Radicals During Interaction of Haemproteins with Hydrogen Peroxide

The oxidation of 2-keto-4-thiomethyl butyric acid (KTBA) and methionine to ethylene has been used to evaluate generation of ferryl species or hydroxyl radicals by H₂O₂-_-activated haemproteins or free ferric ions. Hydrogen peroxide was generated by a glucose oxidase-glucose system at a rate of 1 μM/min. Free ferric in the presence of H₂O₂ oxidizes KTBA, and this was highly inhibited by hydroxyl radical scavengers, caeruloplasmin, superoxide dismutase (SOD) and EDTA. However, when metmyoglobin, methaemoglobin (MtHb) or horseradish peroxidase (HRP) were tested in the same model system, hydroxyl radical scavengers suppressed partially KTBA oxidation and caeruloplasmin, SOD and EDTA failed to inhibit the reaction. Cytochrome-c was found to be a weak promoter of KTBA oxidation in the presence of H₂O₂. Methionine was oxidized to ethylene by an active system which generates hydroxyl radicals, but not by H₂O₂-_-activated metmyoglobin. Ferric ions chelated to membranes or ADP in the presence of H₂O₂ generated enzymatically, initiated membranal lipid peroxidation only in the presence of ascorbic acid, and this was inhibited by EDTA. In contrast, metmyoglobin and methaemoglobin activated by H₂O₂ generated by the same system, initiated membranal lipid peroxidation and this was not inhibited by EDTA. It is concluded that ferryl and not HO. is the main oxidant in systems containing myoglobin and haemoglobin activated by low concentrations of H₂O₂.     

Cocoa procyanidins protect PC12 cells from hydrogen-peroxide-induced apoptosis by inhibiting activation of p38 MAPK and JNK

Oxidative stress induced by reactive oxygen species has been strongly associated with the pathogenesis of neurodegenerative disorders, including Alzheimer's disease. In this study, we investigated the possible protective effects of a cocoa procyanidin fraction (CPF) and procyanidin B2 (epicatechin-(4β-8)-epicatechin) – a major polyphenol in cocoa – against apoptosis of PC12 rat pheochromocytoma (PC12) cells induced by hydrogen peroxide (H₂O₂). CPF (1 and 5 μg/ml) and procyanidin B2 (1 and 5 μM) reduced PC12 cell death caused by H₂O₂, as determined by MTT and trypan blue exclusion assays. CPF and procyanidin B2 attenuated the H₂O₂-induced fragmentation of nucleus and DNA in PC12 cells. Western blot data demonstrated that H₂O₂ induced cleavage of poly(ADP-ribose)polymerase (PARP), downregulated Bcl-X_L and Bcl-2 in PC12 cells. Pretreatment with CPF or procyanidin B2 before H₂O₂ treatment diminished PARP cleavage and increased Bcl-X_L and Bcl-2 expression compared with those only treated with H₂O₂. Activation of caspase-3 by H₂O₂ was inhibited by pretreatment with CPF or procyanidin B2. Furthermore, H₂O₂-induced rapid and significant phosphorylation of c-Jun N-terminal protein kinase (JNK) and p38 mitogen-activated protein kinase (MAPK), and both of these effects were attenuated by CPF or procyanidin B2 treatment. These results suggest that the protective effects of CPF and procyanidin B2 against H₂O₂-induced apoptosis involve inhibiting the downregulation of Bcl-X_L and Bcl-2 expression through blocking the activation of JNK and p38 MAPK.     

p38 MAPK and Ca2+ contribute to hydrogen peroxide-induced increase of permeability in vascular endothelial cells but ERK does not

To examine the involvement of p38 mitogen-activated protein kinase (p38 MAPK) and extra-cellular signal-regulated kinase (ERK) in the oxidative stress-induced increase of permeability in endothelial cells, the effects of a p38 MAPK inhibitor (SB203580) and ERK inhibitor (PD90859) on the H₂O₂-induced increase of permeability in bovine pulmonary artery endothelial cells (BPAEC) were investigated using a two-compartment system partitioned by a semi-permeable filter. H₂O₂ at 1 mM caused an increase of the permeation rate of fluorescein isothiocyanate (FITC)-labeled dextran 40 through BPAEC monolayers. SB203580 inhibited the H₂O₂-induced increase of permeability but PD98059 did not, though activation (phosphorylation) of both p38 MAPK and ERK was observed in H₂O₂-treated cells in Western blot analysis. An H₂O₂-induced increase of the intracellular Ca²⁺ concentration ([Ca²⁺]_i) was also observed and an intracellular Ca²⁺ chelator (BAPTA-AM) significantly inhibited the H₂O₂-induced increase of permeability. However, it showed no inhibitory effects on the H₂O₂-induced phosphorylation of p38 MAPK and ERK. The H₂O₂-induced increase of [Ca²⁺]_i was not influenced by SB203580 and PD98059. These results indicate that the activation of p38 MAPK and the increase of [Ca²⁺]_i are essential for the H₂O₂-induced increase of endothelial permeability and that ERK is not.     

Inactivation of alcohol dehydrogenase by piroxicam-derived radicals

Alcohol dehydrogenase (ADH) was used as a marker molecule to clarify the mechanism of gastric mucosal damage as a side effect of using piroxicam. Piroxicam inactivated ADH during interaction of ADH with horseradish peroxidase and H₂O₂ (HRP-H₂O₂). The ADH was more easily inactivated under aerobic than anaerobic conditions, indicating participation by oxygen. Superoxide dismutase, but not hydroxyl radical scavengers, inhibited inactivation of ADH, indicating participation by superoxide. Sulfhydryl (SH) groups in ADH were lost during incubation of piroxicam with HRP-H₂O₂. Adding reduced glutathione (GSH) efficiently blocked ADH inactivation. Other SH enzymes, including creatine kinase and glyceraldehyde-3-phosphate dehydrogenase, were also inactivated by piroxicam with HRP-H₂O₂. Thus SH groups in the enzymes seem vulnerable to piroxicam activated by HRP-H₂O₂. Spectral change in piroxicam was caused by HRP-H₂O₂. ESR signals of glutathionyl radicals occurred during incubation of piroxicam with HRP-H₂O₂ in the presence of GSH. Under anaerobic conditions, glutathionyl radical formation increased. Thus piroxicam free radicals interact with GSH to produce glutathionyl radicals. Piroxicam peroxyl radicals or superoxide, or both, seem to inactivate ADH. Superoxide may be produced through interaction of peroxyl radicals with H₂O₂. Thus superoxide dismutase may inhibit inactivation of ADH through reducing piroxicam peroxyl radicals or blocking interaction of SH groups with O₂^-, or both. Other oxicam derivatives, including isoxicam, tenoxicam and meloxicam, induced ADH inactivation in the presence of HRP-H₂O₂.     

Hydrogen peroxide is necessary for abscisic acid-induced senescence of rice leaves

The role of H₂O₂ in abscisic acid (ABA)-induced rice leaf senescence is investigated. ABA treatment resulted in H₂O₂ production in rice leaves, which preceded the occurrence of leaf senescence. Dimethylthiourea, a chemical trap for H₂O₂, was observed to be effective in inhibiting ABA-induced senescence, ABA-increased malondialdehyde (MDA) content, ABA-increased antioxidative enzyme activities (superoxide dismutase, ascorbate peroxidase, glutathione reductase and catalase), and ABA-decreased antioxidant contents (ascorbic acid and reduced glutathione) in rice leaves. Diphenyleneiodonium chloride (DPI) and imidazole (IMD), inhibitors of NADPH oxidase, and KCN and NaN₃, inhibitors of peroxidase, prevented ABA-induced H₂O₂ production, suggesting NADPH oxidase and peroxidase are H₂O₂-generating enzymes in ABA-treated rice leaves. DPI, IMD, KCN, and NaN₃ also inhibited ABA-promoted senescence, ABA-increased MDA contents, ABA-increased antioxidative enzyme activities, and ABA-decreased antioxidants in rice leaves. These results suggest that H₂O₂ is involved in ABA-induced senescence of rice leaves.     

Inactivation of Heart Dihydrolipoamide Dehydrogenase by Copper Fenton Systems. Effect of Thiol Compounds and Metal Chelators

Copper Fenton systems (Cu(II)/H₂O₂ and Cu(II)/Asc) inactivated the lipoamide reductase and enhanced the diaphorase activity of pig-heart lipoamide dehydrogenase (LADH). Cupric ions alone were less effective. As a result of Cu(II)/H₂O₂ treatment, the number of titrated thiols in LADH decreased from 6 to 1 per subunit. NADH and ADP (not NAD⁺ or ATP) enhanced LADH inactivation by Cu(II). NADH also enhanced the effect of Cu(II)/H₂O₂. Dihydrolipoamide, dihydrolipoic acid, Captopril, acetylcysteine, EDTA, DETAPAC, histidine, bathocuproine, GSSG and trypanothione prevented LADH inactivation. 100 μM GSH, DL-dithiothreitol, N-(2-mercaptopropionylglicine) and penicillamine protected LADH against Cu(II)/Asc and Cu(II), whereas 1.0 mm GSH and DL-dithiothreitol also protected LADH against Cu(II)/H₂O₂. Allopurinol provided partial protection against Cu(II)/H₂O₂. EthanoI, mannitol, Na benzoate and superoxide dismutase failed to prevent LADH inactivation by Cu(II)/H₂O₂ or Cu(II). Catalase (native or denaturated) and bovine serum albumin protected LADH but that protection should be due to Cu binding. LADH inhibited deoxyribose oxidation and benzoate hydroxylation by Cu(II)/H₂O₂. It is concluded that site-specifically generated HO, radicals were responsible for LADH inactivation by Cu(II) Fenton systems. The latter effect is discussed in the context of ischemia-reoxygenation myocardial injury.     

Modification of the in vitro Cytotoxicity of Hydrogen Peroxide by Iron Complexes

The effect of a range of iron chelates on the cytotoxicity of H₂O₂ was studied on a mammalian epithelial cell line. Iron complexes which were internalised enhanced the cytotoxicity of H₂O₂ measured by delayed thymidine incorporation. Iron complexed to 8-hydroxyquinoline (Fe/8-HQ) potentiated the cytotoxicity of 50 µM by 38% and Fe/dextran by 23%. Pre-exposure of cells to Fe/dextran at 4°C did not result in any potentiation of H₂O₂-induced cytotoxicity which we ascribe to failure of the Fe/dextran to be endocytosed at low temperature. Iron complexes which are slowly taken up or remain extracellular protected the cells from H₂O₂-induced cytotoxicity. Thus, Fe/EDTA inhibited the cytotoxicity of 50 µM H₂O₂ by 33%; Fe/ADP by 80% and Fe/ATP by 88%, suggesting mutual extracellular detoxification.